INTRODUCTION
I am proud and pleased to present what is probably the world's first serious effort to rate and rank all of the current NBA head coaches. The first edition of these annual ratings was published in October 2008. The second edition was published (slightly late) in early December, 2009.
Why should the coaches hide behind a black curtain? Concerning coaches, there is virtually a total lack of the kind of statistical comparing and contrasting that goes on with players 24/7. I for one think it is way overdue that coaches be fairly and systematically compared and contrasted.
I can pretty much guarantee you that no one has ever, even with the capabilities created by the Internet age, put in as much effort and thought as I have into fairly comparing NBA coaches with widely different lengths of time spent in professional head coaching. And this system CAN be used in other Leagues, other countries, and on other planets. If there are any other basketball planets, that is!
For convenience, this Guide is developed into main sections and subsections. The main sections are:
--Mechanics of Real Coach Ratings
--Usage of Real Coach Ratings
--Cautions Regarding Real Coach Ratings
Subsections are in caps as shown.
======== MECHANICS OF REAL COACH RATINGS =========
POSITIVE FACTORS THAT AFFECT REAL COACH RATINGS
1. Number of Regular Season Games Coached: The Experience Factor:
One Point is given for each regular season game coached up to 600 games, which is almost 7 1/2 seasons worth of games. If a Coach has not learned just about everything he needs to by this point, he most likely never will, so the award for experience is sharply reduced for all games coached beyond 600. 0.2 points is given for games 601 through 1,000. Nothing at all is given for any games coached beyond 1,000 games. If a coach has not learned everything he can learn after 1,000 games, he is never going to learn it.
What about rookie and near rookie coaches? Just because they have never coached in the NBA, should their experience rating be zero? No, I don't believe so. They either have substantial coaching experience in other Leagues, or they were extremely talented and/or intelligent players, or both, or else they would not have been hired to be a head Coach in the NBA. So any coach who has coached for fewer than 200 games is given exactly 200 points for experience. So rookie coaches start out with Real Coach Ratings of 200.
2. Number of Playoff Season Games Coached: the Playoff Experience Factor:
Three points are awarded for every playoff game coached regardless of result. The limit is going to be 300 such games. Probably no one will ever reach the limit except for Phil Jackson. He exactly reached 300 playoff games coached after he won his 10th ring in June 2009. So Jackson will fail to get any more playoff experience points when he coaches more playoff games in 2010.
3. Number of Games Coached With Current Team:
This is a supplementary experience score which credits coaches who have gone the longest without being fired by their current teams. The points given are 0.25 (1/4 of a point) for all games coached with the team the Coach is currently working for.
The one side of the coin regarding this is that the coach must be doing what the organization wants to avoid being fired, and he can't be a total failure basketball wise, so he deserves credit in proportion to how long he has kept his post. The other side of the coin is that the more experience a Coach has with a particular team, the more valuable he is to that franchise, because he knows everybody and everything concerned with the franchise better and better with each passing year. Generally speaking, the more successive games a Coach has coached with the same team, the more effectively and efficiently he can help the team squeeze out wins that would otherwise be losses.
Jerry Sloan, who coming in to 2009-10 had coached a mind boggling 1,668 games for the Jazz, is the ultimate example of a Coach who due to his many years with the same team is going to be more effective and efficient than he would be if he had just switched to a different team. Due partly to this factor, do not be surprised if the Jazz become a losing team shortly after Sloan finally retires.
Another name for this factor might be "franchise specific experience." This year the Washington Wizards hired a new head Coach, Flip Saunders, who has a lot of prior experience with other teams. But he is brand new to the Wizards, so be careful not to expect miracles or even to assume that his coaching is going to be as good as it has been in the past from the get go. Look instead for the Wizards to get a little better as the season goes along. Because Saunders needs time to merge his skills and abilities with the specific factors involved with making the Wizards a winning team.
4. Regular Season Wins
5 points per regular season win.
5. Playoff Wins:
25 points per playoff win. Very slightly more than half the teams make the playoffs in the current NBA. Theoretically, unless he is stuck with a truly lousy roster, any truly good coach can get his team into the playoffs. For a good coach, it really is not much of an accomplishment at all. But only the really good coaches can win in the playoffs. In the NBA, the regular season is quite honestly nothing more than the preseason for the "playoff season," which is the season which really matters when all is said and done.
Also, obviously, playoff games are generally more intense in all respects: individual players, team play as a whole, and coaching efforts made.
For all of these reasons, it is necessary to factor playoff games as being worth five times as much as regular season games.
6. Championships
12 Points is added for each Championship appearance. Since Championships average about 6 games, this is roughly equivalent to adding 2 experience points for coaching in Championship series games. Therefore, the total experience points for each Championship game is approximately five.
NEGATIVE FACTORS THAT AFFECT REAL COACH RATINGS
1. Regular Season Losses:
6.5 points is charged for each regular season loss.
2. Playoff Losses:
32.5 points is charged for each playoff loss.
Now there will be some who leap out of their seats and say "this guy is off his rocker" when they see that the penalty for losing a playoff game is 32.5 points. I can assure you, ye of little faith, that I know exactly what I am doing and that this is precisely correct. I have already explained why playoff games must be valued at least five times the valuation put on regular season games. A regular season loss is 6.5 points, and 5 times 6.5 is 32.5.
Now consider the true underlying net positive and negative scores for the four types of games and results, which you get by combining the experience award and the winning or losing number:
TRUE NET SCORES COMBINING EXPERIENCE AND EVENT SCORES TOGETHER
Regular Season Win: 6 Points: 5 points for the win and 1 point for the experience. But it is 5.2 points for coaches with between 600 and 1,000 games coached and it is just 5 points for coaches with more than 1,000 games coached.
Regular Season Loss: Minus 5.5 Points: minus 6.5 points for the loss plus 1 point for the experience. But it is minus 6.3 points for coaches with between 600 and 1,000 games coached and minus 6.5 points for coaches with more than 1,000 games coached.
Can you see what I think is the genius of this system? The more experienced coaches get experience points that obviously are not available to less experienced coaches. To partially or in some cases completely offset what would otherwise be an unfair advantage in the rating system, the more experienced coaches are expected to do somewhat better in winning and losing in order to achieve a net positive from their winning and losing toward their ratings. This is a primary mechanism used here that tends to even the playing field between coaches of widely differing amounts of experience, without being unfair to any type of coach. This whole project would have been largely a waste of time if I didn't have a good and fair way of varying the treatment of coaches with radically different amounts of experience, in this way.
Now here are the true net scores for playoff games:
Playoff Win: 28 points: 25 for the win and 3 for the experience.
Playoff Loss: Minus 29.5 points: minus 32.5 for the loss plus 3 for the experience.
PLAYOFF SUB RATING
Mechanically, the playoff sub rating is simply the rating you get when you factor in only the four playoffs-related factors. In the spreadsheet of the report, the Playoff Sub Ratings are just below the overall Ratings.
PLAYOFFS COACH SUB RATINGS
Two of the three sub ratings from 2008 are discontinued beginning 2009. Readers can now scan the raw data and get at least as much information as they could from the discontinued sub ratings. The only sub rating we are still publishing is the playoffs sub rating. (Who would have thought we'd key in on that one, laugh out loud.)
In the 2009 Ratings, George Karl is no longer at the very bottom of the playoffs sub ratings; he is ahead of Don Nelson thanks to Karl's Nuggets' 10-6 playoffs campaign in 2009. Golden State Warriors Coach Don Nelson is now dead last in the playoffs sub ratings. However, the deep hole that Karl dug in earlier years was so deep that the Nuggets' miraculous 2009 playoffs campaign was not enough to overall lift him very much in the playoffs sub rating. He is still showing up as a very poor playoffs coach, though Karl's rating is not as extremely poor as it was a year ago.
Karl has now won 78 playoff games and lost 93 of them. Prior to the 2009 playoffs, Karl had won just 68 playoff games and lost 87.
======= USAGE OF REAL COACH RATINGS ========
HOW TO INTERPRET DIFFERENCES IN RATINGS
We will use Phil Jackson versus George Karl from the 2009 Real Coach Ratings, published in early December, 2009. You can see that the best cautious rating system we can produce (the mone most in George Karl’s favor) and not be laughed out of the room shows that Los Angeles Lakers Coach Phil Jackson has a rating about ten times that of Denver Nuggets Coach George Karl.
You can interpret this in either of two ways. The first way to look at this is similar to the way that the Real Team Ratings are interpreted: It is about ten times more likely that Phil Jackson is a better coach and will defeat George Karl in a playoff series than the other way around, assuming the raw talent and injury situations of their teams are about the same. Given equal teams, Phil Jackson is going to beat George Karl barring something rare.
The other way to interpret this is to think of the differential between the two ratings as an amount which translates into an actual real life coaching difference. Then you plug that difference in with the other differences that determine who wins a playoff series. If the coaching difference and/or the size of the coaching component is big enough, it could result in the lesser skilled team winning the series if they have the better coach.
Even though we are unable at this time to properly estimate the actual size of the coaching factor in a playoff series, we know it is NOT negligible, trivial, or even very small. Coaching may be a small rather than a "middle-sized" factor (we don't have exact knowledge of how big a factor it is yet) but if the players between the two teams are evenly matched, then even a small difference in the coaching could determine the series.
In any event, the difference between Phil Jackson and George Karl is so large that even if the coaching impact on playoff games is at the low end of the possible range, George Karl would still have to have a much better team to be able to defeat Phil Jackson in a playoff series.
The same applies to Phil Jackson versus Boston Celtics Coach Doc Rivers. We think right now that the 2010 Championship is about a toss-up between the Celtics and the Lakers. But were it not for the Lakers' coaching advantage over the Celtics, the Celtics would have to be favored to win the Ring in 2010 by maybe 4 games to 2, since the Celtics are clearly better than the Lakers in terms of raw skill and raw potential.
CERTAIN VETERAN PLAYERS CAN COACH THEMSELVES TO A LARGE EXTENT
Always keep in mind that older, more veteran teams can coach themselves to one extent or another, particularly if the roster is both highly skilled and highly experienced. It doesn't matter who comes up with the winning schemes and patterns; what matters is that someone does. Younger teams, however, always need a good coaching staff to make headway in the playoffs.
Quest for the Ring has gone on record claiming that the 2007-08 Champion Boston Celtics are a good example of a team that could coach itself well to a large extent.
COACH OBJECTIVE #1: TO AVOID BEING FIRED
Calculations indicate that the average Real Coach Rating is currently 639 and the median is about 200. So the objective of all rookie coaches must be to increase their starting rating of 200 toward the average of 639 as soon as they can do so. You can think of the range between 200 and 600 as "the proving ground" or even the "make it or break it range" for coaches. Most coaches who drop below zero instead of going up from 200 during their first 3-6 years will be bounced out of the NBA.
Coaches who have ratings below 200 for several years, and especially coaches who have ratings below zero should be fired unless the managers and owners involved are sure that the coach has not had competitive players to work with, or are sure that the coach is getting better at his job, or unless there is some other unusual mitigating factor.
Coaches, who maintain their jobs with Real Player Ratings below 200, and especially with Real Coach Ratings below zero, are frequently going to be men who have very cordial relations with the managers and owners. In other words, they are being kept on the payroll because the managers and/or the owners involved personally like the coach in question enough to brush aside any concerns about whether that coach is doing a good enough job for their team. These dubious coaches are given the benefit of the doubt, in other words, or sort of a free pass.
It is also true that some managers and owners live in fear that they might go from bad to worse if they exchange one coach for another. They simply do not have enough courage to strike out and try a rookie or a near-rookie coach, or to pick up a coach who has been fired by another team but who deserves a second chance.
The key is balance. On the one hand you don't want to be stuck out of caution or fear with a veteran coach who is simply not among the best coaches. On the other hand, you can't just strike out and pick any one who has never coached an NBA team before but seems like he might be a good coach. Rahther, you have to do a lot of homework and research. You have to spend a lot of time and make every effort to find that one in a hundred candidate who will actually become one of the better and maybe even one of the best NBA coaches.
NEVER EVER HIRE A COACH WITH A POOR PLAYOFFS RECORD IF YOU WANT TO WIN A RING
The Nuggets hired Karl despite the fact that he had a poor playoffs rating. When the Nuggets hired Karl, his playoffs record was 65-71. While coaching the Nuggets, Karl's playoffs record is 13-22. Karl's playoff record has gotten worse while he has coached the Nuggets, not better (despite 2009).
The Nuggets were wrong to hire Karl, and they are also wrong not to fire him unless he wins the NBA Championship within the next year or two. Which by the way, the Nuggets were in 2008, possibly were in 2009, and are again for 2010 talented enough to win a Championship if the coaching was top notch. Coaches with losing playoff records are fired by all truly serious NBA franchises these days. Karl had a losing playoffs record when he was hired and it has only gotten worse since.
Why did the Nuggets hire Karl? I can only speculate. The Nuggets either knew in advance they would never win the Quest with Karl and hired him anyway, or they figured incorrectly that Karl's playoff record was trumped by better aspects of Karl's record, or they decided that Karl's playoff record could be excused for rational reasons, or there was some other unknown, off the wall reason for hiring Mr. Karl..
Whatever the Nuggets' management thought, they thought wrongly. If you are a team owner or manager, you can not afford to take any risk or to make any benign assumptions or weak rationalizations when you choose a head coach. If a coach has a poor playoffs record, you have no choice but to not hire that coach if you are serious about winning the Quest. There are going to be coaches who are good enough to do well in the regular season, but not good enough to prevail in the playoffs. You should not be the goober who hires one of them, obviously. Let some other franchise/team get stuck with that type of coach.
I have to be blunt here to make sure I am understood. You should never, ever do what the Nuggets did if you are serious about winning the Quest. Your coach should have a good record for both regular season and playoffs. The playoff record is even more important than the regular season record.
Finally, before leaving this crucial subject, I am going to state that given the choice between on the one hand a younger coach who is considered to be a good or great up and coming coach, but who has no NBA playoff record at all, and not much of a regular season one, and on the other hand a long-term veteran coach who has a decent regular season record but a poor, losing playoffs record, you are better off choosing the young coach with no playoff record.
In point blank and clear summary, a coach with a bad playoffs record is one of the worst things you can do if you want to win the Quest.
======= CAUTIONS REGARDING REAL COACH RATINGS ========
THE WIDELY DIFFERENT AMOUNTS OF EXPERIENCE PROBLEM
There is one big hurdle (or notable shortcoming if you want to be negative) in the Real Coach Ratings, and we have largely but probably not completely solved the problem as of 2009. This problem originates in the huge discrepancies in the amount of experience between long-term veteran coaches and much younger coaches. To some extent this makes comparing NBA coaches like trying to compare apples and oranges rather than like trying to compare various apples.
In the 2008 User Guide, this was what I had to say about this issue when I tackled it for the first time:
2008 WORK ON THE EXPERIENCE DISCREPANCY PROBLEM
As I was working on this I often had a sinking feeling that trying to fairly compare coaches with more than 10 years of experience with those with less than 2 years experience would be in the end impossible. But I persevered and scrapped and fought my way to the goal line and got it done. I achieved all of the balancing that I needed to achieve. Specifically, for example, I kept the points given for experience within reason, while making sure that regular season and playoff losses were penalized to the full extent they should be.
You must keep in mind that any coach who has been fired for not winning enough in the regular season, for not winning enough in the playoffs, or for both, and has not been rehired by another team, is not on this list. We don't care about them. The whole idea in multi-billion dollar professional sports is to win more than you lose, and that most obviously and most definitely includes the coaches. So a 50/50 record in either the regular season or in the playoffs is not good enough long term, and coaches who are not better than .500 get fired and not rehired sooner or later, and those who have met that fate already are not on this list.
To reflect the reality that coaches who can not win more than they lose are sooner or later going to be fired, and will most likely never advance in the playoffs before they are fired, it is necessary to make sure that losses entail a bigger negative number than do wins entail a positive number. But we have to avoid getting carried away. So when I add in the amount given for experience, the apparent gap between the award for winning and the penalty for losing is shrunk down to a small amount.
2009 WORK ON THE EXPERIENCE DIFFERENTIAL PROBLEM
Notice that in 2008 I said “we have to avoid getting carried away” in the 2008 attempt to solve this problem. Well, it turns out I probably did get a little carried away. The heavily experienced coaches with a lot of losses were being hammered a little bit too much!
So the number of points subtracted for losses was very slightly reduced for 2009. Regular season losses are now minus 6.5 (instead of minus 7).
However, due to another consideration, playoff losses are slightly greater minuses in 2009 than they were in 2008.
Where we are right now is that we are in very good shape overall, but out of respect for conservatism we may still have a small problem left with the experience discrepancy problem. In a nutshell, we decided to take the risk that the problem is not completely solved so as to avoid being overly harsh toward certain long-term coaches. "First, do no harm..."
When all is said and done, everyone, including a bad coach, can possibly improve even after many years of not improving. This fact, which we didn’t allow for last year, is the biggest reason why we tweaked the way we did. Unfortunately, the price for this is the real possibility that the experience discrepancy problem is not completely solved.
SLIGHTLY DIFFERENT REWARDS AND PENALTIES BASED ON EXPERIENCE
Even after the tweaking, this feature of the system goes a long way toward solving the experience differential problem. Here is how it works:
In the case of all coaches who have coached fewer than 600 games (which is currently 17 out of 30 of them) since a full point is given for every regular season game for just the experience factor, and since the award for a regular season win is 5 points, and since the penalty for a regular season loss is minus 6.5 points, these younger, less experienced coaches do slightly better than break even just by achieving a 50/50 regular season record. When you combine the win or loss points with the experience point, a win earns a new coach a total of 6 points, while a loss earns him minus 5.5 points.
The new coaches are learning, so the system must be slightly easier on them. They can not be expected to know everything right now that they will know in a year or two or three or four. And if they learn the right things, than they might become the next Phil Jackson or Rick Adelman!
Coaches who have coached more than 600 games but fewer than 1,000 games must do a little better than .500 in the regular season to achieve a net positive toward their overall Real Coach Ratings. These coaches get 5.2 points for each regular season win and lose 6.3 points for each loss.
The long-term veteran coaches who have coached more than 1,000 games get no experience points at all. So they get 5 points for each regular season win and lose 6.5 points for each regular season loss.
For the playoffs, all coaches have the same gain or loss: 28 points for a playoff win, and minus 29.5 points for a playoff loss.
REMAINING EXPERIENCE DISCREPANCY PROBLEM
The worst of the long-term veteran coaches probably have ratings that are slightly higher than what they really should be. If a Coach has received some "lucky breaks" by not being fired after bad losing seasons, and/or after bad losses in the playoffs, and he has over the years now accumulated 1,000 or more regular season games and 100 or more playoff games, his rating will likely be in effect slightly distorted on the high side relative to the other coaches.
This is because the long-time veteran Coach, who could have been fired a long time ago but was not fired, will max out on the experience points, and he will also have a few winning seasons to go with the losing seasons. The sum of the maximum experience points (which is 680 for regular season experience plus three times the number of playoff games) plus any positive net from winning seasons will tend to more than offset all the losses from the year(s) he might have been fired, despite the heavy negatives that losses carry.
Another way of thinking about this issue is that assuming a long-term veteran Coach has a too high rating due to the above, keep in mind that Coach would not even be in the ratings had he actually been fired. Coaching a professional sports team is about the worst job in existence for job security, since the vast majority of coaches are involuntarily fired.
Yet another way of focusing on this problem is realizing that pro basketball coaches are fired or not fired based on different criteria.
We can not simply remove experience from the set of factors, since in every single career there is, the more experience you have, the better you tend to be. Moreover, even if we did reduce or remove the experience factor, the same problem would still be there in the case of coaches who probably should have been fired, but are not and then end up fortunately coaching very skilled teams in subsequent years, thus piling up wins with those teams.
In other words, we have no choice but to proceed as if all coaches face the same criteria as to whether they are fired or not, even though we know that some coaches are treated much more leniently than others
One other thing to keep in mind about long-term veteran coaches (the ones with more than 1,000 regular season games coached) is that once such a Coach gets older than 60, 65, and then maybe even 70 years old, that Coach's abilities will probably be less than they were when he were younger. Whereas almost all coaches with little experience are under the age of 55.
For example, Utah Jazz Coach Jerry Sloan is 68 years old on March 28, 2010, so it is possible that he is a little too old now for maximum effectiveness.
The bottom line is that there will be a small number of older, veteran coaches whose ratings are misleading on the high side. Unfortunately, we are unable to correct for this or to properly estimate the amount of the unavoidable distortion at this time.
So we advise you when looking at the ratings to make sure you give the benefit of the doubt to younger coaches who seem to have good potential. The coaches whose ratings are most likely distorted upwards would be, at the moment, in order of the most likely amount of distortion, George Karl, Don Nelson, and Larry Brown. It is plausible, for example, that young Miami Heat Coach Erik Spoelstra is as good or better a Coach right now as is Don Nelson.
PROBABLE DOWNSIDE DISTORTIONS
The flip side of the above distortion is also going to be a problem sometimes. If you have a younger coach who has just started out, and he has a bad team to start with, then his rating will be much lower than it will be in some years if he avoids getting fired and gets much better teams to work with in subsequent years.
However, it is possible that in most cases the worst teams get only the medium and poor coaches, that in other words the really good coaches never have to start out coaching a bad team, so that any downside distortions are small and mostly moot points.
Generally speaking, we are still working on a way to make the comparisons between long-time veterans and much younger, newer coaches more valid than they are in the current system. We hope of course to make a breakthrough or two for next October's Report.
BE CAREFUL REGARDING THE VERY LARGE TIME SCALE OF THESE RATINGS
Keep in mind that each coach is rated using information from every season that he has been a head coach in the NBA. It is very plausible that some of the coaches will currently be substantially better or substantially worse than their overall career ratings indicate.
But while I am on this subject, I want to warn you to not make the assumption that all or even most coaches get better as they accumulate more and more experience. There is no empirical evidence I know of to back that sweeping generalization up, and nor is it in my view obvious or even likely to be true most or much of the time. It is plausible that coaches do not really improve that much after roughly 5 or 6 years of experience. It is also plausible that some of the heaviest experience coaches have not completely updated their beliefs and coaching schemes to reflect the current ways of basketball. They may be hurting their teams a little or even a lot by persisting with strategies and tactics that used to work well years ago but are not working very well in the NBA in 2008.
IF YOU COMPLETELY DISTRUST THE RATINGS
Even if you distrust the ratings themselves, you can evaluate the raw data yourself because Quest for the Ring beginning in 2009 provides the entire spreadsheet on which the Ratings are calculated.
FUTURE CHANGES
Are the factors set in stone forever and ever? No, but adjustments will be few, far between, and minor in the coming years. Although this is not a perfect system, it is at the very least a very good system. And it is light years ahead of having no system at all with which to fairly compare coaches of radically differing amounts of professional basketball head coach experience.
Monday, December 7, 2009
Saturday, November 14, 2009
User Guide for Real Player Rating Reports, November 2009
REAL PLAYER RATINGS BY TEAM USER GUIDE
Updated November 14, 2009
Sections Updated:
--Strategically Using RPR (All of the Evaluation scales were improved; the game evaluation scale was extensively revised)
--Mechanics of Real Player Ratings and Real Player Production (The new factors and zero thresholds were introduced.)
--Defensive and Offensive Sub Ratings Section (The mechanics of calculation remain the same; the new calibration between key components of RPR was discussed and/or upgraded.)
==========INTRODUCTION SECTION==========
INTRODUCTION TO THE CONCEPT OF REAL PLAYER RATINGS
The Real Player Rating (RPR) is a very carefully constructed all inclusive performance measure. Most things of value that a basketball player can do are carefully recorded by official NBA scorekeepers who sit right along the edge of the court, mid-court, and who are trained to observe and record everything that happens in a game.
Since these days all of these counts are immediately input into continually updated public data bases online, such as at ESPN, it is possible to in real time combine everything together into an overall performance measure for each player that is intended to evaluate how valuable each player is toward winning games. This is what the RPR does.
Real Player Rating or RPR is everything tracked by scorekeepers that a player does, good and bad, added and subtracted (with negative things such as turnovers and missed shots being subtracted). Very carefully calibrated factors, or weights, are applied to the different elements.
The calibration, as you would expect, is done to reflect the different value toward winning games that different actions on the court have. These factors are subject to very small annual adjustments as knowledge about how games are won and lost is fine tuned.
Then, all of the good and bad combined together is divided by minutes, yielding RPR, which is really the rate per unit of time of the good minus the rate per unit of time of the bad. This is what we need to determine the overall quality or value of the player toward the objective of winning basketball games.
QUALITY (RPR) AND QUANTITY (RPP} SUMMARIZED
RPR reports show for each player the RPR (Real Player Rating) which tells you how good a player did (all the good things minus all the bad things) out on the court per unit of time. The RPP (Real Player Production) report tells you how much in total (the sum of the of the good things minus the sum of the bad things) a player did out on the court, without regard to playing time.
Many and maybe most sports watchers and an unknown but probably disturbingly large number of sports managers make the mistakes of exaggerating the importance of quantity and overlooking to some extent quality. These reports allow you to expand your horizons. These reports put quantity and quality side by side, which is extremely valuable, because both are roughly equally important in explaining accurately why and how the team is playing the way it is.
SIMPLICITY, RELIABILITY, TRANSPARENCY, AND FOCUS ONLY ON "WINNING POWER"
Like everything statistical we do at Quest, we have kept this process as simple and reliable as possible, while at the same time spending as much time as necessary on design, quality control and performance evaluation. Unlike some other practitioners, we avoid what you might call layered complexity, which leads to formulas which can not be understood without studying them and which high traffic sites will not show on any of their web pages for fear that the public will rebel against the statistic. At Quest, we think that our rating systems can be understood and evaluated by most high school graduates, and we keep everything out in the open through User Guides such as this one.
Basketball statistical gurus frequently forget that no matter how intricate their formulas are, they are very heavily manipulating process items such as assists and rebounds while most likely spending very little time on how these things fit together to produce wins and losses. We think that they are making the mistake, whether or not they are aware, of injecting value adjustments regarding how they think the game should be played and value adjustments about which playing styles are better than others.
Whereas, the primary objectives of the relative simplicity (small number of formulas, to be more precise) of the Quest RPR is to avoid all how the game should be played and how players should play value judgments. We don't care about the styles, only about the results. The RPR is concerned first and in fact exclusively with the impact each player has on the potential for winning games.
Quest thinks it makes more sense to minimize the manipulation of process items, and to focus much more on coming up with the best possible estimation of how the process items impact points for and points against in games, which in turn of course determines wins. Whereas other "advanced statistics" might give you more depth and flavor regarding how a particular player plays (his style) the Quest RPR is a way for the reader to, in a very quick and easy way, determine what the overall value of the player is with respect to producing wins or losses.
In other words, the foundation of RPR is and will always be measurement of a player's power to help win basketball games, whereas the foundation of other, more complicated statistics may include preferences about how the game should be played and about the style of players, with winning power measured less accurately as a result of those focuses.
IMPORTANCE OF PER UNIT OF TIME
Because it is per time, RPR is immediately in the running to be the best possible measure of the net quality of a basketball player, or simply "how good" the player is (on average) for each minute of playing time. All per game statistics are inferior to any reasonably good per unit of playing time measure. For example, points per minute (or per 40 minutes or any number of minutes) is a much better thing to look at than points per game.
REAL PLAYER RATING REPORTS CAN BE FOR THE WHOLE NBA, FOR A TEAM, FOR A GAME, OR FOR A CAREER
With a Real Player Ratings Report for the entire NBA, you can see very rapidly who the best players in the NBA have been during the course of the season.
With a Real Player Ratings Report for a Team for the Regular Season, you can see very rapidly who the best players on the team have been during the course of the season. You can use this information to investigate the possibility that the coach is not perfect. Well, we know that no coach is perfect. So really, with the benefit of 20/20 hindsight, we can investigate and determine what mistakes the coach has apparently made with regard to rotations and playing times. Furthermore, by using the Ratings, basketball knowledge, a little creativity, and logical deduction, we can also investigate and perhaps determine whether the coach has made incorrect decisions regarding which strategies and plays are best for his team's offense and defense.
Real Player Ratings for games are a major part of Reports called Ultimate Game Breakdowns.
Real Player Ratings for a player's career, year by year and in total, are obviously very valuable looks at how the player changed over the years. Of course, most players get better from how they started in their rookie years.
[End of Introduction Section.]
============================================
CAUTIONS SECTION
To be completely honest and clear, although it is the best possible overall real life measure, RPR is still not a perfect or absolute, "final word" measure on any player. In general, you must remember that all performance measures including this one for the NBA are relative rather than absolute measures. The ratings are relative to the team context. Players do not exist in a vacuum, especially in basketball.
Several specific cautions will now be described.
RPRs ARE RELATIVE TO TEAMS, AND ARE SUBJECT TO THE CROWDING OUT EFFECT
Because basketball is a team game, and more so than most other sports, players who are on really good teams might have their own performances "crowded out" to some extent by just as good players and especially by even better players. So paradoxically, ratings of players of all ratings levels who are on better teams will generally have slightly lower ratings than they would have if they were on a not as good team. Conversely, players (at all ratings levels) who are on bad teams will have slightly higher ratings than they would have if they were on a better team. Numerically, a player on the best NBA team could easily have a RPR that is 15-20% less than what it would be if he was a player on the worst NBA team.
Always remember this important point, which we restate for emphasis. If a good player is on a good team where there are a number of players as good and even better than he is, than his RPR will likely be lower than it would be if he were on a not as good team.
Position in the team context can impact RPR as well. If a good player plays a certain position for which his team has an even better player, then it's probable that the better player will crowd out the lesser player to one extent or another, so that the lesser player's RPR will be lower than what it would be if he were the best player at the position on the team. Conversely, the best player at a position on a bad team can have a RPR which is higher than what it would be on many other teams.
ACTUAL RPR DIFFERENCES BETWEEN TEAMS ARE GREATER THAN APPARENT DIFFERENCES
An important implication of crowding out and relativity is that the average RPR among the best five, six, or seven players of the best teams in most cases will understate the real "potential RPR" of those players, where potential RPR is RPR with the least possible crowding out. In other words, the potential RPRs of players on the best teams is higher than their actual RPRs. Conversely, the long-run, true potential RPRs of the apparently better players on bad teams is actually lower than their actual RPRs.
This plays out at the team level in a very important way. Always remember this: the actual underlying gap in the real quality of the players between good and bad teams is greater than the actual RPRs are indicating. The true RPR differential between the best and the worst NBA team could easily be 20-30% greater than the apparent differential. In other words, team RPR averages understate real quality differences between teams.
PLAYERS NEED THE BALL FOR HIGHER RPRs
Players need not only playing time but possession of the ball in order to produce many of the things that count in the rating. So if, for whatever reason, a player does not get the ball as often as he would on a different team, or with a different coach, or with whatever other circumstances you can dream of, then his RPR will be lower than what it could or would be.
DO NOT FORGET WHAT THE RATINGS YOU ARE LOOKING AT ARE MEASURING
Many ratings that you see on Quest are only for the current season. It has recently been discovered that many player's ratings often change up or down by 10% from year to year, and can change by about 15% up or down without too much trouble from one year to the next. Moreover, over the course of a player's entire career, RPR ratings by year can and often do vary by 50% or even more when you compare the highest year or two to the lowest year or two. Although there are a fairly good number of exceptions, many NBA players have much lower RPRs in their first year or two in the NBA than they will eventually average out to.
INJURIES AND RECOVERIES FROM INJURIES
Players often play with minor injuries. They also often start playing again before they are 100% recovered from an injury. They sometimes even postpone surgery that has become necessary due to injury until the off-season, and play with some type of impairment in the meantime. In all of these situations, RPR will be lower than it would be were the player not dealing with any injury.
MAGNITUDE OF THE ADJUSTMENT FOR HIDDEN DEFENDING
Those who think defense in basketball is much more important than offense will consider the magnitude of the defensive adjustment to be inadequate. They will contend that defensive specialists who are poor offensive players should have a higher rating. While we realized that we needed to adjust the ratings for defending not tracked by NBA scorekeepers, and while we put in a huge effort to come up with a valid adjustment system, we continue to believe that players who are great defensive specialists but poor or undeveloped offensive players should in most cases rank no higher than the Major Role Player/Good Enough to Start level, which is the level just below the Solid Starter level. In a few rare cases, defensive specialists who have decent offensive games will be ranked as Solid Starters.
AVOID BEING CONFUSED BETWEEN RPR AND RPP AND DO NOT MINIMIZE THE IMPORTANCE OF RPP
Do not forget that RPR is a per time measure. RPP and not RPR measures total impact of a player. RPR measures how valuable a player has been toward winning basketball games, per unit of time.
Do not make the mistake of ignoring the importance of RPP, now improved to TRPP. Players with the highest TRPP are showing they have the stamina, knowledge, and trust of the coaching staff to be able to get all the playing time needed to produce that. So even if their RPRs are a little lower than you might expect, players with the highest TRPPs should still be considered as extremely important and valuable players.
Having said that, one of the most important objectives for any top Coach must be to make sure that his highest RPR players are also found at or close to the top of the TRPP list.
THE CLASSIFICATION SCHEME IS RELATIVE TOO
The classification scheme, like the ratings, is relative. A role player on a really good team might be a solid starter on a bad team. A star on a bad team might be just a major role player on a really good team. And so on and so forth. A player is a star, a role player, or whatever only in the contexts of the particular season and the particular team involved. If he was on a different team, or if it was a different year, his classification could easily be different.
So to conclude the Cautions section of this guide, don't think of RPR as the ultimate gospel or bible on how good players are. But do think of it as an extremely accurate and reliable summary of how good the players actually have been in real life in the specific time (season or playoffs) and place (team) involved.
[end of cautions section]
===========STRATEGICALLY USING RPR SECTION============
RELATIVITY ADJUSTMENT FOR PROJECTED RPR FOR PLAYERS CHANGING TEAMS
When you are trying to judge how good a player might be if he were on another team, you need to, due to the relativity factor discussed previously, adjust the expected RPR upwards if the player is moving to a lower quality team and to adjust the expected RPR downwards if the player is moving to a higher quality team. The maximum such adjustment necessary is believed to be about 15%, with that full amount applied only when the player is moving from one of the worst teams to one of the very best teams, or vice versa.
GREAT VARIABILITY OF PLAYER RPRs FOR INDIVIDUAL GAMES
Not as many breakdowns of individual game ratings are going to closely track the overall average for the roster as you might think. This is because one of the interesting things about basketball that makes it different from most other sports is that "how good" a player is from game to game varies radically. The best players sometimes have terrible games where they do almost nothing, while players who normally do not do much can every once in a while have outstanding games, at least if you measure it per minute on the court anyway. If you just looked at actual production, and never at a reserve player's Real Player Rating, you would hardly notice any of his unusually outstanding games, since players who normally do not do much will normally not have much playing time.
INTERACTIONS BETWEEN PLAYING TIMES, RPRs, TRPPs, AND THE NEEDS OF TEAMS
There are certain things that only certain players can do very well, and if those things are crucial for the team, than those players will have to play more minutes than they might otherwise play. The extra minutes might tend to reduce the player's Real Player Rating, while his total production will rise with the additional minutes. So to fairly and completely evaluate any player, you must always look at both the Real Player Rating (RPR) and the Real Player Production (RPP).
Furthermore, it is strongly suspected that, in order to compete in the playoffs, a team must have as many players of as high a quality (RPR) as possible, while at the same time having at least one or two players whose actual production is among the highest in the NBA regardless of exactly how high the RPRs happen to be. (All high RPP players will be relatively high RPR players; some will be higher than others.) Specifically for example, LeBron James' actual massive amount of production is most likely just as important to the Cleveland Cavaliers as is his RPR or, in other words, as is his rate of production. Similarly, Kobe Bryant's quantity is probably at least as important to the Lakers as is his quality.
Whereas, teams such as the Denver Nuggets, who have instructed a possible huge producer, Carmelo Anthony, to "not worry about scoring," may have made a fatal mistake relative to the playoffs, because teams with no extremely high rate producers may be generally doomed to lose quickly in the playoffs even if they have an unusually large number of high quality players as shown by RPR. This is because extremely high RPP players can by themselves "dominate a game" to some extent, meaning they can by themselves possibly win the game for their team, without worrying about complications that come in to play if you need to coordinate several high RPR but ultimately and theoretically limited RPP players.
Players who over the course of a season appear to rank higher in RPR (quality) but lower in RPP (quantity) may not be getting enough playing time. Players who over the course of a season appear to rank lower in RPR (quality) but higher in RPP (quantity) may be getting too much playing time. But as alluded to earlier, you must not automatically conclude this, because some skills are needed out on the court most of the time, but yet may be available only from a small number players on the roster. Such players may have to get more playing time due to that critical skill in short supply, even if their overall quality does not seem to justify all of that playing time.
A relatively common reason for unusual playing time will be players who are either truly outstanding defenders (who get extra playing time) or truly bad defenders (who get their playing time reduced).
Another common reason for extra playing time will be if a team has a point guard who has many more turnovers than the average point guard has. Because the point guard is so important, a good coach has to play his best guard who can make plays at the position for a full set of minutes every game, and he or she must do so almost regardless of how many turnovers that player makes. If you take out your designated point guard due to "too many turnovers," it may end up sort of like cutting your foot off because you have a bad case of athletes foot!
EVALUATION OF REAL PLAYER RATINGS
EVALUATION SCALE FOR A REGULAR SEASON
Remember: Meaningful regular season ratings are not possible until about Jan. 20 of each year. Ratings become even more valuable after the all star break.
Perfect for all Practical Purposes / Major Historic Super Star 1.100 and more
Historic Super Star 1.000 1.099
Super Star 0.910 0.999
A Star Player / A Well Above Normal Starter 0.830 0.909
Very Good Player: A Solid Starter 0.760 0.829
Major Role Player / Good Enough to Start 0.700 0.759
Good Role Player / Often a Good 6th Man 0.650 0.699
Satisfactory Role Player 0.590 0.649
Marginal Role Player 0.530 0.589
Poor Player 0.470 0.529
Very Poor Player 0.400 0.469
Extremely Poor Player and less 0.399
NOTE ABOUT LOW REGULAR SEASON RATINGS
Players rated below about .550 often get playing time based largely on factors outside of RPR, but valued by coaches and other players, such as:
--Great energy and hustle
--Toughness, such as diving after loose balls and taking charges
--Leadership and/or knowledge, especially in the case of veterans
--Perceived potential for future improvement in terms of real basketball production, especially in the case of young players
--See also the User Guide section called "Cautions"
EVALUATION SCALE FOR SINGLE GAMES
Remember: Single game ratings do NOT include any adjustment for hidden defending
EVALUATION SCALE FOR BASIC REAL PLAYER RATINGS FOR A SINGLE GAME
Perfect for all Practical Purposes / Major Historic Super Star 1.100 and more
Historic Super Star 1.000 1.099
Super Star 0.900 0.999
A Star Player / A Well Above Normal Starter 0.800 0.899
Very Good Player: A Solid Starter 0.700 0.799
Major Role Player / Good Enough to Start 0.600 0.699
Good Role Player / Often a Good 6th Man 0.500 0.599
Satisfactory Role Player 0.425 0.499
Marginal Role Player 0.350 0.424
Poor Player 0.275 0.349
Very Poor Player 0.200 0.274
Extremely Poor Player .199 and less
NOTE ABOUT LOW SINGLE GAME RATINGS
Players rated below about .400 often get playing time based largely on factors outside of RPR, but valued by coaches and other players, such as:
--Great energy and hustle
--Toughness, such as diving after loose balls and taking charges
--Leadership and/or knowledge, especially in the case of veterans
--Perceived potential for future improvement in terms of real basketball production, especially in the case of young players
--See also the User Guide section called "Cautions"
EVALUATION SCALE FOR A CAREER (OF A PLAYER)
Remember: Many players have lower ratings in their first one to three years than they will have ultimately.
Perfect for all Practical Purposes / Major Historic Super Star 1.000 and more
Historic Super Star 0.940 0.999
Super Star 0.870 0.939
A Star Player / A Well Above Normal Starter 0.800 0.869
A Very Good Player: A Solid Starter 0.750 0.799
Major Role Player / Good Enough to Start 0.700 0.749
Good Role Player / Often a Good 6th Man 0.650 0.699
Satisfactory Role Player 0.600 0.649
Marginal Role Player 0.540 0.599
Poor Player 0.480 0.539
Very Poor Player 0.420 0.479
Extremely Poor Player 0.419 and Less
NOTE ABOUT LOW CAREER RATINGS
Players rated below about .580 often get playing time based largely on factors outside of RPR, but valued by coaches and other players, such as:
--Great energy and hustle
--Toughness, such as diving after loose balls and taking charges
--Leadership and/or knowledge, especially in the case of veterans
--Perceived potential for future improvement in terms of real basketball production, especially in the case of young players
--See also the User Guide section called "Cautions"
[End of the Strategic Use of Ratings Section]
==========MECHANICS OF REAL PLAYER RATINGS AND REAL PLAYER PRODUCTION==========
MINIMUM PLAYING TIME RULES
Due to the minimum sample size requirement for the adjustment for hidden defending, regular season ratings for NBA players can not be meaningfully done until at least mid January. Generally, we need at least 3 players to have played 1,500 minutes or more before we can or will rate that team's players.
Also, only players who played at least 10% of the minutes of whoever has played the most minutes on the team are included in intra-regular-season team RPR reports. Any player who has played for less than 10% of the minutes of the player who has played the most minutes is not included, since he didn't play for long enough to be fairly or reasonably compared with the other players. However, in the interests of completeness, all players who have played at least 300 minutes are included in the end of season team RPR report, even though a small number of players will have played for less than 10% of the time of the highest minutes player.
Furthermore, as described extensively in the adjustment for defending section of this Guide, only players who have played at least 300 minutes can have a defensive rating, or an overall RPR given to him. Both the 10% and the 300 minutes minimums must be met for a player to be rated.
REAL PLAYER PRODUCTION
Of course, looking at actual production (everything positive added together and everything negative subtracted out) is something that is extremely important too. The total production (everything good and everything bad combined together) is simply called Real Player Production or RPP.
BASIC VERSUS TOTAL REAL PLAYER PRODUCTION
Basic RPP does not include any estimate of how much value from hidden defending was done by the player. Starting from June 2009, there is an estimate made for the value of hidden defending of each player, calculated from the following formula:
Hidden Defending Production = Total Scored Defensive Production * (Hidden Defending Rating / Total Scored Defensive Rating)
The validity of this adjustment is somewhat less than the high validity of the defending adjustments for RPR in general. Therefore, the user is advised to not go overboard in using the results.
Then of course Total Real Player Production is Basic Real Player Production plus Hidden Defending Production. Note: At this time, RPP still refers to basic RPP, and so TRPP is the adjusted version.
SOURCE OF TRACKED BASKETBALL COUNTS
The sources for the raw counts of scores, rebounds, steals, turnovers, and so forth are ESPN.com and NBA.com. Other sites used as important data sources are Basketball-reference.com, Knickerblogger.net, and USAToday.com.
NOTES ON SOME OF THE TECHNOLOGIES USED
Microsoft Excel is extensively used to accurately produce RPR reports. Hundreds of Internet sites have been used to one extent or another in the development and in the continuing production of RPR and related reports.
THE BASIC FORMULA
For 2009-10, the RPR formula has been very carefully and accurately tweaked again and is set to be as follows:
POSITIVE FACTORS
Points 1.00 (at par)
Number of 3-Pt FGs Made 1.00
Number of 2-Pt FGs Made 0.40
Number of FTs Made 0 (no "bonus for a made free throw; just the point itself goes into RPR)
Assists 2.15
Offensive Rebounds 1.43
Defensive Rebounds 1.31
Blocks 1.80
Steals 2.30
NEGATIVE FACTORS
3-Pt FGs Missed -1.00
2-Pt FGs Missed -1.03
FTs Missed -1.3256
Turnovers -1.95
Personal Fouls -1.00
ACTUAL COMBINED AWARD OR PENALTY BY TYPE OF SHOT
3-Pointer Made 4.00
2-Pointer Made 2.40
Free Throw Made 1.00
3-Pointer Missed -1.00
2-Pointer Missed -1.03
Free Throw Missed -1.3256
ZERO POINTS: PERCENTAGES BELOW WHICH THERE IS A NEGATIVE NET RESULT
3-Pointer 0 score % 0.200
2-Pointer 0 score % 0.300
1-Pointer 0 score % 0.570
This means that if a player has a lower percentage than any of the three above, then his RPR would be lower rather than higher as a result of his shooting that type of shot.
ASSISTS VERSUS TURNOVERS ZERO POINT
Assist/Turnover Ratio That Yields 0 Net Points: 0.908
Asset/turnover rations greater than .908 are positive with respect to RPR. This also means that any player who has an assist/turnover ration of less than .908 is losing RPR rating when assists and turnovers are considered. He would have to either increase assisting or reduce turnovers to turn the combined effect from assists and turnovers positive.
DEFENDING RATING
A quality of defending rating of between 0 and .307 is added to base or unadjusted RPR. In most cases, the defending rating is between 0.070 and .240. See the Hidden Defending Adjustment to Real Player Ratings below below for a very detailed explanation of how we determine player defensive ratings and how we combine them with base RPR.
[End of Mechanics of Real Player Ratings and Real Player Production Section.]
======== DEFENSIVE AND OFFENSIVE SUB RATINGS ======================
DEFENDING SUB RATING
THE HIDDEN DEFENDING ADJUSTMENT (HDA) TO REAL PLAYER RATINGS
As of January 8, 2009, The Quest is proud to announce to you that the second major improvement to Real Player Ratings (RPR) in less than half a year is now fully up and running. The first major improvement were some needed changes in the factors used for RPR. The second major improvement (series of improvements, actually) is so far as I am aware the first ever effort to rate the defensive efforts of players that are hidden unless you watch all that player's games, because they are not scored or tracked by scorekeepers.
As of June 8, 2009, the mechanics of the HDA were slightly changed to increase accuracy; see the outlier adjsutment below for details. As if November 14, 2009, the HDA was upgraded, on the average, from about 40% of the overall defensive subrating of players to about 45%. Furthermore, as of November 14, 2009, the overall defensive subrating was recalibrated so that it would now be about 45% of the overall RPR, versus about 42% in 2008-09. The offensive subrating iw recalibrated so that it will now on average constitute about 55% of the overall RPR, versus about 58% in 2008-09.
Obviously, some valuable things that basketball players do are never counted by scorekeepers. Many of these uncounted things are defensive, insofar as they prevent scores, or reduce the scoring opportunities of the opponent. These things would include chasing down loose balls, taking charges, and good or great man to man defending. Man to man defending that is good enough to prevent what would have been a score from actually being a score is the most common and important basketball action which can not be and is not tracked by NBA scorekeepers.
Man to man defending however, although the most important, is not by any means the only defensive element that can not be tracked or scored. Broadly, what is missed or hidden is all the things that the player does to make the possessions of the opposing teams worthless other than what is already counted, which would be rebounds, steals, blocks, and personal fouls. These untracked or hidden actions would include:
SOME BASKETBALL FACTORS ESTIMATED BY THE HIDDEN DEFENDING ADJUSTMENT TO RPR
--effective man to man defending
--effective rotation / switching on defense, especially off screens and picks
--effective pick and roll defense
--effective defensive recognition
--quickness of defensive reaction
--energy and hustle on defense
--effective taking of charges (causing a driving offensive player to be called for an offensive foul)
--effective hustling after loose balls
These things would be counted by scorekeepers if it were possible. But, for example, there is no way to know exactly how many shots a good (or any kind of) defender has changed from being a score to a miss.
Quest for the Ring has developed a statistically valid way to accurately estimate the untracked or hidden aspects of defending. This is described in complete detail in the latter sections of this Guide.
HDA IS AN UPGRADE TO DEFENSIVE EFFICIENCY RATINGS OF PLAYERS SEEN ON OTHER SITES
There are a small number of sites that show you each player's "defensive efficiency," which is number of points allowed per 100 possessions. This sounds nice, but it actaully is not all that valuable. The Hidden Defending Adjustment of RPR is an upgrade for this.
Probably the most important improvement is that in HDA, players' defending is standardized for team defending. With the defensive efficiency on certain other sites, players who are on good defensive teams have elevated ratings simply because they are on those teams. But obviously, many of the players on a good defensive team are producing that good defense, not just any one of them. The Hidden Defensive Adjustment corrects for this quality of team defense bias, which enables players on different teams to be fairly compared with respect to hidden defending.
THE HIDDEN DEFENDING ADJUSTMENT EXPLAINED
It took almost two years of hoping, searching for things, planning, and then developing, but finally the breakthrough was achieved in the objective of correct evaluation of defending. Now that the breakthrough has come, I am now more certain than ever that RPR is the best overall rating system in existence, and that it is now roughly as good as it will ever or can ever be.
HDA is a statistically valid way to rate the hidden defending of players, that is, what they do to prevent scores other than rebounding, blocks, steals, and fouls, which were always included in RPR. This would include man to man defending, zone defending, pick and roll defending, defensive recognition, and defensive rotation.
Although the technique used had to be indirect and subject to a very small amount of statistical error, it validly awards the better defenders with bigger RPR bonuses. It has been validated by comparing results obtained with the player defensive efficiency ratings shown on three different "advanced basketball statistics" web sites. HDA results were shown to be highly correlated with those efficiency ratings.
Where there are small differences, HDA is better, because of the correction for team defense bias, because HDA uses simple, bedrock statistical theory rather than involved formulas involving assumptions, and for other lessor reasons.
HOW TO REVEAL HIDDEN DEFENDING IN FOUR STEPS
STEP ONE: CALCUATION OF RAW HIDDEN DEFENDING ADJUSTMENTS
Unlike most "advanced statistics" that are published on the internet or in print, we give you all the details about how we do ours, so that you can evaluate the evaluations, so to speak. The following is specifically what we are doing to be able to accurately and fairly compare players' defending:
Where do we start to discover what is hidden? We keep it as simple and yet as accurate as possible. We use the most official and therefore presumably the most reliable data as the building blocks for rating the defense of NBA players. We start with the player minutes and points scored by the other team while the player was on the court that are shown in the plus/minus statistical section at NBA.com.
There are no value judgments made regarding a player's defending style, or regarding a team's defending style for that matter. We don't care about style. Using points allowed per minute is looking at results, nothing more and nothing less.
After simply dividing points allowed by minutes on the court, we adjust (we standardize, to be more precise) that rate for the pace of the team and for the quality of the team's defense. The two adjustments are needed so that the ratings of players who are on different teams can be fairly compared. For example, it would be grossly unfair to compare the rate of points allowed of a player on a fast paced, poor defense quality team to a player on a slow paced, high defense quality team.
Players who are on teams with faster paces give up more points per minute through no fault of their own. Similarly, players who are on teams with less efficient defenses give up more points per minute, regardless of how well they defend, everything else held constant. You can not fairly compare players on two or more teams with different paces and different team defense qualities unless you standardize, or in other words control for those differences for all NBA players.
USE OF BASIC STATISTICAL SAMPLING THEORY
What we are doing is using an indirect and inexact yet accurate and statistically valid way to discover who the better defenders are. No two players are out on the court for all the exact same minutes. So although for every player, what the other players out on the court do defensively while they are out on the court is a very large factor determining what that player's points per minute allowed will be, when you look at many, many hundreds of minutes, what the individual player does, or does not do defensively, as the case may be, will eventually show up in that particular player's points allowed per minute statistic.
In other words, what any individual player does defensively has to sooner or later show itself in a differentiation from other players of his points allowed per minute. As the number of minutes rise above 500, and then 1,000 and then, for many players, above 2,000 and even 3,000 for a regular season, what a particular player does or does not do defensively becomes more and more exactly shown by the points allowed per minute number. This is very basic statistical sampling theory in operation. Statistical sampling theory is the easy to understand bedrock theory of statistics.
Due to the necessity of a large sample of minutes, we will not do defending estimates for any player who has played for fewer than 300 minutes. Quality of defending estimates will be slightly less accurate for players who have only played between 301 and about 600 minutes than they will be for players who have played for more than 600 minutes. We believe that the estimates are going to be extremely accurate for all players who have played 750 minutes or more. The idea is relatively simple: as the number of hundreds of minutes played goes up, the accuracy of this system improves, to the point where it gives you the same information you would have if you knew exactly how many possessions of the other team each player ruined with his defending.
For your information, after adjustments for pace and team defense quality, all players allow between 1.88 and 2.28 points per minute; most allow between 1.98 and 2.18. The overall NBA average is about 2.08 points per minute allowed.
STEP TWO: CONVERSION OF RAW HIDDEN DEFENDING ADJUSTMENT TO UNADJUSTED HIDDEN DEFENDING RATING
We need to translate the adjusted or filtered points allowed per minute into numerical terms that are the most useful with respect to RPR. So with a very carefully designed translation scale, we amplify the very small differences in different player's points allowed per minute numbers into much larger different hidden defending ratings for each player. The scale is approximately symmetrically (two sided) inverse logarithmic with respect to the average point.
STEP THREE: APPLICATION OF OUTLIER RULES IN VERY LIMITED NUMBER OF EXTREME CASES
An outlier rule is not the best way to produce a valid and useful measure. However, if it is the only technique available to correct for bias, using it is certainly better than not using it.
Although the HDRs emerging from Step 2 are valid and valuable, there are apparently two kinds of players whose HDRs are biased upwards. These players are relatively rare; we are only talking about fewer than two dozen of the close to 400 NBA players who play for 300 minutes or more during a season.
In general the problem is that a tiny number of forwards / centers make most of their defensive efforts in the tracked actions, such as defensive rebounding, and a very small number of guards, at the opposite extreme, make most of their defensive efforts in the hidden defending actions.
The problem is that the HDA scaling presupposes that there is some correlation between hidden defending and scored or tracked defending. For at least 90% of the players, there is a strong correlation. But where a player's defending is mostly tracked and very little hidden, or vice versa, the HDR rating is biased upward slightly. In both cases, the player would be given a higher HDA than he in real life deserves were it not for the outlier rules.
So the following are the two outliers that require minor adjustments. Of the two kinds of players who have been flagged for adjustment, the first one below, the guards, are much more common in real life.
1. Guards who do not rebound, steal, or block much, but who are great at hidden defending actions, sometimes end up with excessive HDR ratings, ones that they can not possibly be due.
An outlier rule is applied to prevent this bias from showing up in the final RPR of such guards. No player can have an HDR rating higher than double what his Scored Defensive Rating (SDR). The SDR is defensive rebounds, steals, and blocks minus personal fouls factored according to RPR. However, no player's HDR rating will be adjusted downward by more than .100 under this rule, so as to prevent excessive downward adjustment for a handful of guards who are truly good man to man defenders but who are not at all good at rebounding, stealing, and blocking.
2. At the opposite extreme, you have forwards and centers who produce an extreme number of defensive rebounds, steals, and blocks. Obviously, these players will tend to have high raw HDRs by virtue of those things alone. It is both logical and observable in games that such players can not get a very large number of rebounds/steals/blocks, and defensive stops via great man to man defending, at the same time. Technically, a great man to man defend on a possession is essentially an uncalled block. And forwards and centers who get a very large number of rebounds can not do that and get in a lot of good or great man to man defending in the same game.
Therefore, for such players, HDR would be biased upward; there would be some double counting involved.
An outlier rule is applied to prevent most of this bias from showing up in the final RPR. No player can have an HDA rating and a Scored Defensive Rating (SDR) combined higher than .600, which in turn is approximately 1.5 times the average Scored Offensive Rating for NBA players. (The average scored offensive rating is roughly .400.) For example, if an outstanding rebounder has a SDR of .400, his HDR must be reduced to .200 if it exceeds .200.
Remember, less than two dozen NBA players are subject to either of the outlier rules. The huge majority of NBA players are within the scaling parameters for HDR and so there is very low bias for them.
STEP FOUR: CALCULATION OF REAL PLAYER RATING (ADJUSTED FOR HIDDEN DEFENDING)
The final step is to simply add the hidden defending rating to the Base RPR to yield RPR.
USE OF HIDDEN DEFENDING RATING
We now have added in a reasonably good estimate of the value of actions of players that are not even kept track of by scorekeepers! Technically, you could call the final result "Ajusted RPR," but we are trying to avoid that terminology because of how important we think it is to include the hidden defending in the performance measure.
SIZE OF THE DEFENDING ADJUSTMENTS
Base regular season RPR's for most NBA players range between .400 and 1.000. The total range of possible defending adjustments to the base RPRs is from 0 to .307. In most cases, however, the adjustment will be between 0.070 and .240.
THE DEFENDING SUB RATING: PUTTING THE HIDDEN AND THE UNHIDDEN TOGETHER
Aside from the Hidden Defending Rating we can find out how well each player does in terms of unhidden or scored defending, can't we? Of course we can.
Unhidden or tracked defending, is defensive rebounding plus steals plus blocks minus personal fouls, calibrated according to the usual RPR factors. If we extract the combination of those four out of the same counts that underlie the RPR as a whole, and use the usual factors, we get what we are going to call the Scored Defending Production. This could also be thought of as Tracked Defending Production if you prefer. Then if we divide this by minutes, we have a Scored (or Tracked) Defending Rating.
Finally, if we combine Hidden Defending Rating (HDR) with Scored Defending Rating (SDR) we can have an Overall Defending Rating (ODR).
Obviously, the HDR scaling is designed to coordinate correctly with both SDR and with RPR as a whole. All of the coordinations reflect the latest undertanding of how basketball games are won and lost. The HDR constitutes about 45% of ODR while SDR constitutes the other 55%. In other words, the value of hidden defending is perceived to be about 45% of the overall value of defending, while the value of scored (unhidden) defending is perceived to be about 55% of the overall value of defending.
There appear to be many coaches and not a few hardcore basketball fans who believe that hidden defending is actually more important than scored defending, but I am never going to agree with that. I think that although hidden defending is important, and plausibly almost as important as tracked defending, that it can not be more than this. Hidden defending is like a quicksand, in that there seems to be a tendency for a substantial minority of basketball people to get carried away with estimating the importance of it, and then become more and more trapped by their error in terms of how they look at basketball or in terms of how they coach their team if they are coaching.
FORWARDS AND CENTERS WILL GENERALLY HAVE SUBSTANTIALLY HIGHER DEFENDING RATINGS
Due to having primary responsibility for defense of the paint and for rebounding, centers and forwards are going to inevitably have higher defensive ratings than will guards. Along with much greater opportunity for rebounds and blocks, centers and forwards also have more opportunity for such hidden defending actions as good man to man defending and correct rotations than do guards. Guards out on the perimeter generally should not and do not man to man defend as closely as do interior defenders, due to the well known guideline that it is quite foolish to foul a jump shooter outside of the paint.
THE OFFENSIVE SUB RATING
The Offensive Sub Rating is all tracked actions other than the defensive ones (defensive rebounding, steals, blocks, and personal fouls) combined together using the RPR weights, divided by minutes. In other words, it is Total Offensive Production divided by minutes. For the list of all tracked actions and the weight factors assigned to each, see the secion titled "The Formula" above.
THE BEST GUARDS WILL HAVE THE HIGHEST OFFENSIVE SUB RATINGS
The very best guards in basketball are ones who, although they are not afraid to drive to the hoop from time to time, are able to make outside shots at a good rate. Also, guards in general, and especially point guards, are usually primarily responsible for making assists. These two are among the several reasons why the better guards in pro basketball will have the highest offensive sub ratings in the League.
On the other hand, some of the most valuable players in the NBA are centers and forwards who are great defenders and efficient inside scorers at the same time. Even more unusual and probably for that reason more valuable is a forward who is (a) a great inside defender (b) a great inside scorer and (c) someone who can hit jump shots, perhaps even including threes, from outside the paint. Lamar Odom is an example of this kind of extremely valuable player.
Some of these big men will have offensive sub ratings that exceed those of the lessor skilled shooting guards and even those of some of the less skilled point guards.
[End of Defensive and Offensive Sub Ratings Section.]
======== SUMMARY OF PRIMARY FORMULAS SECTION =================
Real Player Production or RPR = (All tracked or scored actions weighted according to best available analysis of importance / minutes) + Hidden Defending Rating
Real Player Production or RPP = Total Offensive Production + Total Defensive Production. (All tracked or scored actions weighted according to best available analysis of importance.)
Offensive Sub Rating = Total Scored or Tracked Offensive Production / Minutes
Defensive Sub Rating = Total Scored or Tracked Defensive Production + Hidden Defending Rating
Hidden Defending Rating = Raw Hidden Defending Adjustment amplified via an approximtely reverse logarithmic scale symmetric as to the average. The objective of this is to correctly show value toward winning games via amplification of seemingly minute differences in points allowed per minute.
Raw Hidden Defending Adjustment = Points Allowed Per Minute * Pace Adjustment Factor * Team Defense Adjustment
Updated November 14, 2009
Sections Updated:
--Strategically Using RPR (All of the Evaluation scales were improved; the game evaluation scale was extensively revised)
--Mechanics of Real Player Ratings and Real Player Production (The new factors and zero thresholds were introduced.)
--Defensive and Offensive Sub Ratings Section (The mechanics of calculation remain the same; the new calibration between key components of RPR was discussed and/or upgraded.)
==========INTRODUCTION SECTION==========
INTRODUCTION TO THE CONCEPT OF REAL PLAYER RATINGS
The Real Player Rating (RPR) is a very carefully constructed all inclusive performance measure. Most things of value that a basketball player can do are carefully recorded by official NBA scorekeepers who sit right along the edge of the court, mid-court, and who are trained to observe and record everything that happens in a game.
Since these days all of these counts are immediately input into continually updated public data bases online, such as at ESPN, it is possible to in real time combine everything together into an overall performance measure for each player that is intended to evaluate how valuable each player is toward winning games. This is what the RPR does.
Real Player Rating or RPR is everything tracked by scorekeepers that a player does, good and bad, added and subtracted (with negative things such as turnovers and missed shots being subtracted). Very carefully calibrated factors, or weights, are applied to the different elements.
The calibration, as you would expect, is done to reflect the different value toward winning games that different actions on the court have. These factors are subject to very small annual adjustments as knowledge about how games are won and lost is fine tuned.
Then, all of the good and bad combined together is divided by minutes, yielding RPR, which is really the rate per unit of time of the good minus the rate per unit of time of the bad. This is what we need to determine the overall quality or value of the player toward the objective of winning basketball games.
QUALITY (RPR) AND QUANTITY (RPP} SUMMARIZED
RPR reports show for each player the RPR (Real Player Rating) which tells you how good a player did (all the good things minus all the bad things) out on the court per unit of time. The RPP (Real Player Production) report tells you how much in total (the sum of the of the good things minus the sum of the bad things) a player did out on the court, without regard to playing time.
Many and maybe most sports watchers and an unknown but probably disturbingly large number of sports managers make the mistakes of exaggerating the importance of quantity and overlooking to some extent quality. These reports allow you to expand your horizons. These reports put quantity and quality side by side, which is extremely valuable, because both are roughly equally important in explaining accurately why and how the team is playing the way it is.
SIMPLICITY, RELIABILITY, TRANSPARENCY, AND FOCUS ONLY ON "WINNING POWER"
Like everything statistical we do at Quest, we have kept this process as simple and reliable as possible, while at the same time spending as much time as necessary on design, quality control and performance evaluation. Unlike some other practitioners, we avoid what you might call layered complexity, which leads to formulas which can not be understood without studying them and which high traffic sites will not show on any of their web pages for fear that the public will rebel against the statistic. At Quest, we think that our rating systems can be understood and evaluated by most high school graduates, and we keep everything out in the open through User Guides such as this one.
Basketball statistical gurus frequently forget that no matter how intricate their formulas are, they are very heavily manipulating process items such as assists and rebounds while most likely spending very little time on how these things fit together to produce wins and losses. We think that they are making the mistake, whether or not they are aware, of injecting value adjustments regarding how they think the game should be played and value adjustments about which playing styles are better than others.
Whereas, the primary objectives of the relative simplicity (small number of formulas, to be more precise) of the Quest RPR is to avoid all how the game should be played and how players should play value judgments. We don't care about the styles, only about the results. The RPR is concerned first and in fact exclusively with the impact each player has on the potential for winning games.
Quest thinks it makes more sense to minimize the manipulation of process items, and to focus much more on coming up with the best possible estimation of how the process items impact points for and points against in games, which in turn of course determines wins. Whereas other "advanced statistics" might give you more depth and flavor regarding how a particular player plays (his style) the Quest RPR is a way for the reader to, in a very quick and easy way, determine what the overall value of the player is with respect to producing wins or losses.
In other words, the foundation of RPR is and will always be measurement of a player's power to help win basketball games, whereas the foundation of other, more complicated statistics may include preferences about how the game should be played and about the style of players, with winning power measured less accurately as a result of those focuses.
IMPORTANCE OF PER UNIT OF TIME
Because it is per time, RPR is immediately in the running to be the best possible measure of the net quality of a basketball player, or simply "how good" the player is (on average) for each minute of playing time. All per game statistics are inferior to any reasonably good per unit of playing time measure. For example, points per minute (or per 40 minutes or any number of minutes) is a much better thing to look at than points per game.
REAL PLAYER RATING REPORTS CAN BE FOR THE WHOLE NBA, FOR A TEAM, FOR A GAME, OR FOR A CAREER
With a Real Player Ratings Report for the entire NBA, you can see very rapidly who the best players in the NBA have been during the course of the season.
With a Real Player Ratings Report for a Team for the Regular Season, you can see very rapidly who the best players on the team have been during the course of the season. You can use this information to investigate the possibility that the coach is not perfect. Well, we know that no coach is perfect. So really, with the benefit of 20/20 hindsight, we can investigate and determine what mistakes the coach has apparently made with regard to rotations and playing times. Furthermore, by using the Ratings, basketball knowledge, a little creativity, and logical deduction, we can also investigate and perhaps determine whether the coach has made incorrect decisions regarding which strategies and plays are best for his team's offense and defense.
Real Player Ratings for games are a major part of Reports called Ultimate Game Breakdowns.
Real Player Ratings for a player's career, year by year and in total, are obviously very valuable looks at how the player changed over the years. Of course, most players get better from how they started in their rookie years.
[End of Introduction Section.]
============================================
CAUTIONS SECTION
To be completely honest and clear, although it is the best possible overall real life measure, RPR is still not a perfect or absolute, "final word" measure on any player. In general, you must remember that all performance measures including this one for the NBA are relative rather than absolute measures. The ratings are relative to the team context. Players do not exist in a vacuum, especially in basketball.
Several specific cautions will now be described.
RPRs ARE RELATIVE TO TEAMS, AND ARE SUBJECT TO THE CROWDING OUT EFFECT
Because basketball is a team game, and more so than most other sports, players who are on really good teams might have their own performances "crowded out" to some extent by just as good players and especially by even better players. So paradoxically, ratings of players of all ratings levels who are on better teams will generally have slightly lower ratings than they would have if they were on a not as good team. Conversely, players (at all ratings levels) who are on bad teams will have slightly higher ratings than they would have if they were on a better team. Numerically, a player on the best NBA team could easily have a RPR that is 15-20% less than what it would be if he was a player on the worst NBA team.
Always remember this important point, which we restate for emphasis. If a good player is on a good team where there are a number of players as good and even better than he is, than his RPR will likely be lower than it would be if he were on a not as good team.
Position in the team context can impact RPR as well. If a good player plays a certain position for which his team has an even better player, then it's probable that the better player will crowd out the lesser player to one extent or another, so that the lesser player's RPR will be lower than what it would be if he were the best player at the position on the team. Conversely, the best player at a position on a bad team can have a RPR which is higher than what it would be on many other teams.
ACTUAL RPR DIFFERENCES BETWEEN TEAMS ARE GREATER THAN APPARENT DIFFERENCES
An important implication of crowding out and relativity is that the average RPR among the best five, six, or seven players of the best teams in most cases will understate the real "potential RPR" of those players, where potential RPR is RPR with the least possible crowding out. In other words, the potential RPRs of players on the best teams is higher than their actual RPRs. Conversely, the long-run, true potential RPRs of the apparently better players on bad teams is actually lower than their actual RPRs.
This plays out at the team level in a very important way. Always remember this: the actual underlying gap in the real quality of the players between good and bad teams is greater than the actual RPRs are indicating. The true RPR differential between the best and the worst NBA team could easily be 20-30% greater than the apparent differential. In other words, team RPR averages understate real quality differences between teams.
PLAYERS NEED THE BALL FOR HIGHER RPRs
Players need not only playing time but possession of the ball in order to produce many of the things that count in the rating. So if, for whatever reason, a player does not get the ball as often as he would on a different team, or with a different coach, or with whatever other circumstances you can dream of, then his RPR will be lower than what it could or would be.
DO NOT FORGET WHAT THE RATINGS YOU ARE LOOKING AT ARE MEASURING
Many ratings that you see on Quest are only for the current season. It has recently been discovered that many player's ratings often change up or down by 10% from year to year, and can change by about 15% up or down without too much trouble from one year to the next. Moreover, over the course of a player's entire career, RPR ratings by year can and often do vary by 50% or even more when you compare the highest year or two to the lowest year or two. Although there are a fairly good number of exceptions, many NBA players have much lower RPRs in their first year or two in the NBA than they will eventually average out to.
INJURIES AND RECOVERIES FROM INJURIES
Players often play with minor injuries. They also often start playing again before they are 100% recovered from an injury. They sometimes even postpone surgery that has become necessary due to injury until the off-season, and play with some type of impairment in the meantime. In all of these situations, RPR will be lower than it would be were the player not dealing with any injury.
MAGNITUDE OF THE ADJUSTMENT FOR HIDDEN DEFENDING
Those who think defense in basketball is much more important than offense will consider the magnitude of the defensive adjustment to be inadequate. They will contend that defensive specialists who are poor offensive players should have a higher rating. While we realized that we needed to adjust the ratings for defending not tracked by NBA scorekeepers, and while we put in a huge effort to come up with a valid adjustment system, we continue to believe that players who are great defensive specialists but poor or undeveloped offensive players should in most cases rank no higher than the Major Role Player/Good Enough to Start level, which is the level just below the Solid Starter level. In a few rare cases, defensive specialists who have decent offensive games will be ranked as Solid Starters.
AVOID BEING CONFUSED BETWEEN RPR AND RPP AND DO NOT MINIMIZE THE IMPORTANCE OF RPP
Do not forget that RPR is a per time measure. RPP and not RPR measures total impact of a player. RPR measures how valuable a player has been toward winning basketball games, per unit of time.
Do not make the mistake of ignoring the importance of RPP, now improved to TRPP. Players with the highest TRPP are showing they have the stamina, knowledge, and trust of the coaching staff to be able to get all the playing time needed to produce that. So even if their RPRs are a little lower than you might expect, players with the highest TRPPs should still be considered as extremely important and valuable players.
Having said that, one of the most important objectives for any top Coach must be to make sure that his highest RPR players are also found at or close to the top of the TRPP list.
THE CLASSIFICATION SCHEME IS RELATIVE TOO
The classification scheme, like the ratings, is relative. A role player on a really good team might be a solid starter on a bad team. A star on a bad team might be just a major role player on a really good team. And so on and so forth. A player is a star, a role player, or whatever only in the contexts of the particular season and the particular team involved. If he was on a different team, or if it was a different year, his classification could easily be different.
So to conclude the Cautions section of this guide, don't think of RPR as the ultimate gospel or bible on how good players are. But do think of it as an extremely accurate and reliable summary of how good the players actually have been in real life in the specific time (season or playoffs) and place (team) involved.
[end of cautions section]
===========STRATEGICALLY USING RPR SECTION============
RELATIVITY ADJUSTMENT FOR PROJECTED RPR FOR PLAYERS CHANGING TEAMS
When you are trying to judge how good a player might be if he were on another team, you need to, due to the relativity factor discussed previously, adjust the expected RPR upwards if the player is moving to a lower quality team and to adjust the expected RPR downwards if the player is moving to a higher quality team. The maximum such adjustment necessary is believed to be about 15%, with that full amount applied only when the player is moving from one of the worst teams to one of the very best teams, or vice versa.
GREAT VARIABILITY OF PLAYER RPRs FOR INDIVIDUAL GAMES
Not as many breakdowns of individual game ratings are going to closely track the overall average for the roster as you might think. This is because one of the interesting things about basketball that makes it different from most other sports is that "how good" a player is from game to game varies radically. The best players sometimes have terrible games where they do almost nothing, while players who normally do not do much can every once in a while have outstanding games, at least if you measure it per minute on the court anyway. If you just looked at actual production, and never at a reserve player's Real Player Rating, you would hardly notice any of his unusually outstanding games, since players who normally do not do much will normally not have much playing time.
INTERACTIONS BETWEEN PLAYING TIMES, RPRs, TRPPs, AND THE NEEDS OF TEAMS
There are certain things that only certain players can do very well, and if those things are crucial for the team, than those players will have to play more minutes than they might otherwise play. The extra minutes might tend to reduce the player's Real Player Rating, while his total production will rise with the additional minutes. So to fairly and completely evaluate any player, you must always look at both the Real Player Rating (RPR) and the Real Player Production (RPP).
Furthermore, it is strongly suspected that, in order to compete in the playoffs, a team must have as many players of as high a quality (RPR) as possible, while at the same time having at least one or two players whose actual production is among the highest in the NBA regardless of exactly how high the RPRs happen to be. (All high RPP players will be relatively high RPR players; some will be higher than others.) Specifically for example, LeBron James' actual massive amount of production is most likely just as important to the Cleveland Cavaliers as is his RPR or, in other words, as is his rate of production. Similarly, Kobe Bryant's quantity is probably at least as important to the Lakers as is his quality.
Whereas, teams such as the Denver Nuggets, who have instructed a possible huge producer, Carmelo Anthony, to "not worry about scoring," may have made a fatal mistake relative to the playoffs, because teams with no extremely high rate producers may be generally doomed to lose quickly in the playoffs even if they have an unusually large number of high quality players as shown by RPR. This is because extremely high RPP players can by themselves "dominate a game" to some extent, meaning they can by themselves possibly win the game for their team, without worrying about complications that come in to play if you need to coordinate several high RPR but ultimately and theoretically limited RPP players.
Players who over the course of a season appear to rank higher in RPR (quality) but lower in RPP (quantity) may not be getting enough playing time. Players who over the course of a season appear to rank lower in RPR (quality) but higher in RPP (quantity) may be getting too much playing time. But as alluded to earlier, you must not automatically conclude this, because some skills are needed out on the court most of the time, but yet may be available only from a small number players on the roster. Such players may have to get more playing time due to that critical skill in short supply, even if their overall quality does not seem to justify all of that playing time.
A relatively common reason for unusual playing time will be players who are either truly outstanding defenders (who get extra playing time) or truly bad defenders (who get their playing time reduced).
Another common reason for extra playing time will be if a team has a point guard who has many more turnovers than the average point guard has. Because the point guard is so important, a good coach has to play his best guard who can make plays at the position for a full set of minutes every game, and he or she must do so almost regardless of how many turnovers that player makes. If you take out your designated point guard due to "too many turnovers," it may end up sort of like cutting your foot off because you have a bad case of athletes foot!
EVALUATION OF REAL PLAYER RATINGS
EVALUATION SCALE FOR A REGULAR SEASON
Remember: Meaningful regular season ratings are not possible until about Jan. 20 of each year. Ratings become even more valuable after the all star break.
Perfect for all Practical Purposes / Major Historic Super Star 1.100 and more
Historic Super Star 1.000 1.099
Super Star 0.910 0.999
A Star Player / A Well Above Normal Starter 0.830 0.909
Very Good Player: A Solid Starter 0.760 0.829
Major Role Player / Good Enough to Start 0.700 0.759
Good Role Player / Often a Good 6th Man 0.650 0.699
Satisfactory Role Player 0.590 0.649
Marginal Role Player 0.530 0.589
Poor Player 0.470 0.529
Very Poor Player 0.400 0.469
Extremely Poor Player and less 0.399
NOTE ABOUT LOW REGULAR SEASON RATINGS
Players rated below about .550 often get playing time based largely on factors outside of RPR, but valued by coaches and other players, such as:
--Great energy and hustle
--Toughness, such as diving after loose balls and taking charges
--Leadership and/or knowledge, especially in the case of veterans
--Perceived potential for future improvement in terms of real basketball production, especially in the case of young players
--See also the User Guide section called "Cautions"
EVALUATION SCALE FOR SINGLE GAMES
Remember: Single game ratings do NOT include any adjustment for hidden defending
EVALUATION SCALE FOR BASIC REAL PLAYER RATINGS FOR A SINGLE GAME
Perfect for all Practical Purposes / Major Historic Super Star 1.100 and more
Historic Super Star 1.000 1.099
Super Star 0.900 0.999
A Star Player / A Well Above Normal Starter 0.800 0.899
Very Good Player: A Solid Starter 0.700 0.799
Major Role Player / Good Enough to Start 0.600 0.699
Good Role Player / Often a Good 6th Man 0.500 0.599
Satisfactory Role Player 0.425 0.499
Marginal Role Player 0.350 0.424
Poor Player 0.275 0.349
Very Poor Player 0.200 0.274
Extremely Poor Player .199 and less
NOTE ABOUT LOW SINGLE GAME RATINGS
Players rated below about .400 often get playing time based largely on factors outside of RPR, but valued by coaches and other players, such as:
--Great energy and hustle
--Toughness, such as diving after loose balls and taking charges
--Leadership and/or knowledge, especially in the case of veterans
--Perceived potential for future improvement in terms of real basketball production, especially in the case of young players
--See also the User Guide section called "Cautions"
EVALUATION SCALE FOR A CAREER (OF A PLAYER)
Remember: Many players have lower ratings in their first one to three years than they will have ultimately.
Perfect for all Practical Purposes / Major Historic Super Star 1.000 and more
Historic Super Star 0.940 0.999
Super Star 0.870 0.939
A Star Player / A Well Above Normal Starter 0.800 0.869
A Very Good Player: A Solid Starter 0.750 0.799
Major Role Player / Good Enough to Start 0.700 0.749
Good Role Player / Often a Good 6th Man 0.650 0.699
Satisfactory Role Player 0.600 0.649
Marginal Role Player 0.540 0.599
Poor Player 0.480 0.539
Very Poor Player 0.420 0.479
Extremely Poor Player 0.419 and Less
NOTE ABOUT LOW CAREER RATINGS
Players rated below about .580 often get playing time based largely on factors outside of RPR, but valued by coaches and other players, such as:
--Great energy and hustle
--Toughness, such as diving after loose balls and taking charges
--Leadership and/or knowledge, especially in the case of veterans
--Perceived potential for future improvement in terms of real basketball production, especially in the case of young players
--See also the User Guide section called "Cautions"
[End of the Strategic Use of Ratings Section]
==========MECHANICS OF REAL PLAYER RATINGS AND REAL PLAYER PRODUCTION==========
MINIMUM PLAYING TIME RULES
Due to the minimum sample size requirement for the adjustment for hidden defending, regular season ratings for NBA players can not be meaningfully done until at least mid January. Generally, we need at least 3 players to have played 1,500 minutes or more before we can or will rate that team's players.
Also, only players who played at least 10% of the minutes of whoever has played the most minutes on the team are included in intra-regular-season team RPR reports. Any player who has played for less than 10% of the minutes of the player who has played the most minutes is not included, since he didn't play for long enough to be fairly or reasonably compared with the other players. However, in the interests of completeness, all players who have played at least 300 minutes are included in the end of season team RPR report, even though a small number of players will have played for less than 10% of the time of the highest minutes player.
Furthermore, as described extensively in the adjustment for defending section of this Guide, only players who have played at least 300 minutes can have a defensive rating, or an overall RPR given to him. Both the 10% and the 300 minutes minimums must be met for a player to be rated.
REAL PLAYER PRODUCTION
Of course, looking at actual production (everything positive added together and everything negative subtracted out) is something that is extremely important too. The total production (everything good and everything bad combined together) is simply called Real Player Production or RPP.
BASIC VERSUS TOTAL REAL PLAYER PRODUCTION
Basic RPP does not include any estimate of how much value from hidden defending was done by the player. Starting from June 2009, there is an estimate made for the value of hidden defending of each player, calculated from the following formula:
Hidden Defending Production = Total Scored Defensive Production * (Hidden Defending Rating / Total Scored Defensive Rating)
The validity of this adjustment is somewhat less than the high validity of the defending adjustments for RPR in general. Therefore, the user is advised to not go overboard in using the results.
Then of course Total Real Player Production is Basic Real Player Production plus Hidden Defending Production. Note: At this time, RPP still refers to basic RPP, and so TRPP is the adjusted version.
SOURCE OF TRACKED BASKETBALL COUNTS
The sources for the raw counts of scores, rebounds, steals, turnovers, and so forth are ESPN.com and NBA.com. Other sites used as important data sources are Basketball-reference.com, Knickerblogger.net, and USAToday.com.
NOTES ON SOME OF THE TECHNOLOGIES USED
Microsoft Excel is extensively used to accurately produce RPR reports. Hundreds of Internet sites have been used to one extent or another in the development and in the continuing production of RPR and related reports.
THE BASIC FORMULA
For 2009-10, the RPR formula has been very carefully and accurately tweaked again and is set to be as follows:
POSITIVE FACTORS
Points 1.00 (at par)
Number of 3-Pt FGs Made 1.00
Number of 2-Pt FGs Made 0.40
Number of FTs Made 0 (no "bonus for a made free throw; just the point itself goes into RPR)
Assists 2.15
Offensive Rebounds 1.43
Defensive Rebounds 1.31
Blocks 1.80
Steals 2.30
NEGATIVE FACTORS
3-Pt FGs Missed -1.00
2-Pt FGs Missed -1.03
FTs Missed -1.3256
Turnovers -1.95
Personal Fouls -1.00
ACTUAL COMBINED AWARD OR PENALTY BY TYPE OF SHOT
3-Pointer Made 4.00
2-Pointer Made 2.40
Free Throw Made 1.00
3-Pointer Missed -1.00
2-Pointer Missed -1.03
Free Throw Missed -1.3256
ZERO POINTS: PERCENTAGES BELOW WHICH THERE IS A NEGATIVE NET RESULT
3-Pointer 0 score % 0.200
2-Pointer 0 score % 0.300
1-Pointer 0 score % 0.570
This means that if a player has a lower percentage than any of the three above, then his RPR would be lower rather than higher as a result of his shooting that type of shot.
ASSISTS VERSUS TURNOVERS ZERO POINT
Assist/Turnover Ratio That Yields 0 Net Points: 0.908
Asset/turnover rations greater than .908 are positive with respect to RPR. This also means that any player who has an assist/turnover ration of less than .908 is losing RPR rating when assists and turnovers are considered. He would have to either increase assisting or reduce turnovers to turn the combined effect from assists and turnovers positive.
DEFENDING RATING
A quality of defending rating of between 0 and .307 is added to base or unadjusted RPR. In most cases, the defending rating is between 0.070 and .240. See the Hidden Defending Adjustment to Real Player Ratings below below for a very detailed explanation of how we determine player defensive ratings and how we combine them with base RPR.
[End of Mechanics of Real Player Ratings and Real Player Production Section.]
======== DEFENSIVE AND OFFENSIVE SUB RATINGS ======================
DEFENDING SUB RATING
THE HIDDEN DEFENDING ADJUSTMENT (HDA) TO REAL PLAYER RATINGS
As of January 8, 2009, The Quest is proud to announce to you that the second major improvement to Real Player Ratings (RPR) in less than half a year is now fully up and running. The first major improvement were some needed changes in the factors used for RPR. The second major improvement (series of improvements, actually) is so far as I am aware the first ever effort to rate the defensive efforts of players that are hidden unless you watch all that player's games, because they are not scored or tracked by scorekeepers.
As of June 8, 2009, the mechanics of the HDA were slightly changed to increase accuracy; see the outlier adjsutment below for details. As if November 14, 2009, the HDA was upgraded, on the average, from about 40% of the overall defensive subrating of players to about 45%. Furthermore, as of November 14, 2009, the overall defensive subrating was recalibrated so that it would now be about 45% of the overall RPR, versus about 42% in 2008-09. The offensive subrating iw recalibrated so that it will now on average constitute about 55% of the overall RPR, versus about 58% in 2008-09.
Obviously, some valuable things that basketball players do are never counted by scorekeepers. Many of these uncounted things are defensive, insofar as they prevent scores, or reduce the scoring opportunities of the opponent. These things would include chasing down loose balls, taking charges, and good or great man to man defending. Man to man defending that is good enough to prevent what would have been a score from actually being a score is the most common and important basketball action which can not be and is not tracked by NBA scorekeepers.
Man to man defending however, although the most important, is not by any means the only defensive element that can not be tracked or scored. Broadly, what is missed or hidden is all the things that the player does to make the possessions of the opposing teams worthless other than what is already counted, which would be rebounds, steals, blocks, and personal fouls. These untracked or hidden actions would include:
SOME BASKETBALL FACTORS ESTIMATED BY THE HIDDEN DEFENDING ADJUSTMENT TO RPR
--effective man to man defending
--effective rotation / switching on defense, especially off screens and picks
--effective pick and roll defense
--effective defensive recognition
--quickness of defensive reaction
--energy and hustle on defense
--effective taking of charges (causing a driving offensive player to be called for an offensive foul)
--effective hustling after loose balls
These things would be counted by scorekeepers if it were possible. But, for example, there is no way to know exactly how many shots a good (or any kind of) defender has changed from being a score to a miss.
Quest for the Ring has developed a statistically valid way to accurately estimate the untracked or hidden aspects of defending. This is described in complete detail in the latter sections of this Guide.
HDA IS AN UPGRADE TO DEFENSIVE EFFICIENCY RATINGS OF PLAYERS SEEN ON OTHER SITES
There are a small number of sites that show you each player's "defensive efficiency," which is number of points allowed per 100 possessions. This sounds nice, but it actaully is not all that valuable. The Hidden Defending Adjustment of RPR is an upgrade for this.
Probably the most important improvement is that in HDA, players' defending is standardized for team defending. With the defensive efficiency on certain other sites, players who are on good defensive teams have elevated ratings simply because they are on those teams. But obviously, many of the players on a good defensive team are producing that good defense, not just any one of them. The Hidden Defensive Adjustment corrects for this quality of team defense bias, which enables players on different teams to be fairly compared with respect to hidden defending.
THE HIDDEN DEFENDING ADJUSTMENT EXPLAINED
It took almost two years of hoping, searching for things, planning, and then developing, but finally the breakthrough was achieved in the objective of correct evaluation of defending. Now that the breakthrough has come, I am now more certain than ever that RPR is the best overall rating system in existence, and that it is now roughly as good as it will ever or can ever be.
HDA is a statistically valid way to rate the hidden defending of players, that is, what they do to prevent scores other than rebounding, blocks, steals, and fouls, which were always included in RPR. This would include man to man defending, zone defending, pick and roll defending, defensive recognition, and defensive rotation.
Although the technique used had to be indirect and subject to a very small amount of statistical error, it validly awards the better defenders with bigger RPR bonuses. It has been validated by comparing results obtained with the player defensive efficiency ratings shown on three different "advanced basketball statistics" web sites. HDA results were shown to be highly correlated with those efficiency ratings.
Where there are small differences, HDA is better, because of the correction for team defense bias, because HDA uses simple, bedrock statistical theory rather than involved formulas involving assumptions, and for other lessor reasons.
HOW TO REVEAL HIDDEN DEFENDING IN FOUR STEPS
STEP ONE: CALCUATION OF RAW HIDDEN DEFENDING ADJUSTMENTS
Unlike most "advanced statistics" that are published on the internet or in print, we give you all the details about how we do ours, so that you can evaluate the evaluations, so to speak. The following is specifically what we are doing to be able to accurately and fairly compare players' defending:
Where do we start to discover what is hidden? We keep it as simple and yet as accurate as possible. We use the most official and therefore presumably the most reliable data as the building blocks for rating the defense of NBA players. We start with the player minutes and points scored by the other team while the player was on the court that are shown in the plus/minus statistical section at NBA.com.
There are no value judgments made regarding a player's defending style, or regarding a team's defending style for that matter. We don't care about style. Using points allowed per minute is looking at results, nothing more and nothing less.
After simply dividing points allowed by minutes on the court, we adjust (we standardize, to be more precise) that rate for the pace of the team and for the quality of the team's defense. The two adjustments are needed so that the ratings of players who are on different teams can be fairly compared. For example, it would be grossly unfair to compare the rate of points allowed of a player on a fast paced, poor defense quality team to a player on a slow paced, high defense quality team.
Players who are on teams with faster paces give up more points per minute through no fault of their own. Similarly, players who are on teams with less efficient defenses give up more points per minute, regardless of how well they defend, everything else held constant. You can not fairly compare players on two or more teams with different paces and different team defense qualities unless you standardize, or in other words control for those differences for all NBA players.
USE OF BASIC STATISTICAL SAMPLING THEORY
What we are doing is using an indirect and inexact yet accurate and statistically valid way to discover who the better defenders are. No two players are out on the court for all the exact same minutes. So although for every player, what the other players out on the court do defensively while they are out on the court is a very large factor determining what that player's points per minute allowed will be, when you look at many, many hundreds of minutes, what the individual player does, or does not do defensively, as the case may be, will eventually show up in that particular player's points allowed per minute statistic.
In other words, what any individual player does defensively has to sooner or later show itself in a differentiation from other players of his points allowed per minute. As the number of minutes rise above 500, and then 1,000 and then, for many players, above 2,000 and even 3,000 for a regular season, what a particular player does or does not do defensively becomes more and more exactly shown by the points allowed per minute number. This is very basic statistical sampling theory in operation. Statistical sampling theory is the easy to understand bedrock theory of statistics.
Due to the necessity of a large sample of minutes, we will not do defending estimates for any player who has played for fewer than 300 minutes. Quality of defending estimates will be slightly less accurate for players who have only played between 301 and about 600 minutes than they will be for players who have played for more than 600 minutes. We believe that the estimates are going to be extremely accurate for all players who have played 750 minutes or more. The idea is relatively simple: as the number of hundreds of minutes played goes up, the accuracy of this system improves, to the point where it gives you the same information you would have if you knew exactly how many possessions of the other team each player ruined with his defending.
For your information, after adjustments for pace and team defense quality, all players allow between 1.88 and 2.28 points per minute; most allow between 1.98 and 2.18. The overall NBA average is about 2.08 points per minute allowed.
STEP TWO: CONVERSION OF RAW HIDDEN DEFENDING ADJUSTMENT TO UNADJUSTED HIDDEN DEFENDING RATING
We need to translate the adjusted or filtered points allowed per minute into numerical terms that are the most useful with respect to RPR. So with a very carefully designed translation scale, we amplify the very small differences in different player's points allowed per minute numbers into much larger different hidden defending ratings for each player. The scale is approximately symmetrically (two sided) inverse logarithmic with respect to the average point.
STEP THREE: APPLICATION OF OUTLIER RULES IN VERY LIMITED NUMBER OF EXTREME CASES
An outlier rule is not the best way to produce a valid and useful measure. However, if it is the only technique available to correct for bias, using it is certainly better than not using it.
Although the HDRs emerging from Step 2 are valid and valuable, there are apparently two kinds of players whose HDRs are biased upwards. These players are relatively rare; we are only talking about fewer than two dozen of the close to 400 NBA players who play for 300 minutes or more during a season.
In general the problem is that a tiny number of forwards / centers make most of their defensive efforts in the tracked actions, such as defensive rebounding, and a very small number of guards, at the opposite extreme, make most of their defensive efforts in the hidden defending actions.
The problem is that the HDA scaling presupposes that there is some correlation between hidden defending and scored or tracked defending. For at least 90% of the players, there is a strong correlation. But where a player's defending is mostly tracked and very little hidden, or vice versa, the HDR rating is biased upward slightly. In both cases, the player would be given a higher HDA than he in real life deserves were it not for the outlier rules.
So the following are the two outliers that require minor adjustments. Of the two kinds of players who have been flagged for adjustment, the first one below, the guards, are much more common in real life.
1. Guards who do not rebound, steal, or block much, but who are great at hidden defending actions, sometimes end up with excessive HDR ratings, ones that they can not possibly be due.
An outlier rule is applied to prevent this bias from showing up in the final RPR of such guards. No player can have an HDR rating higher than double what his Scored Defensive Rating (SDR). The SDR is defensive rebounds, steals, and blocks minus personal fouls factored according to RPR. However, no player's HDR rating will be adjusted downward by more than .100 under this rule, so as to prevent excessive downward adjustment for a handful of guards who are truly good man to man defenders but who are not at all good at rebounding, stealing, and blocking.
2. At the opposite extreme, you have forwards and centers who produce an extreme number of defensive rebounds, steals, and blocks. Obviously, these players will tend to have high raw HDRs by virtue of those things alone. It is both logical and observable in games that such players can not get a very large number of rebounds/steals/blocks, and defensive stops via great man to man defending, at the same time. Technically, a great man to man defend on a possession is essentially an uncalled block. And forwards and centers who get a very large number of rebounds can not do that and get in a lot of good or great man to man defending in the same game.
Therefore, for such players, HDR would be biased upward; there would be some double counting involved.
An outlier rule is applied to prevent most of this bias from showing up in the final RPR. No player can have an HDA rating and a Scored Defensive Rating (SDR) combined higher than .600, which in turn is approximately 1.5 times the average Scored Offensive Rating for NBA players. (The average scored offensive rating is roughly .400.) For example, if an outstanding rebounder has a SDR of .400, his HDR must be reduced to .200 if it exceeds .200.
Remember, less than two dozen NBA players are subject to either of the outlier rules. The huge majority of NBA players are within the scaling parameters for HDR and so there is very low bias for them.
STEP FOUR: CALCULATION OF REAL PLAYER RATING (ADJUSTED FOR HIDDEN DEFENDING)
The final step is to simply add the hidden defending rating to the Base RPR to yield RPR.
USE OF HIDDEN DEFENDING RATING
We now have added in a reasonably good estimate of the value of actions of players that are not even kept track of by scorekeepers! Technically, you could call the final result "Ajusted RPR," but we are trying to avoid that terminology because of how important we think it is to include the hidden defending in the performance measure.
SIZE OF THE DEFENDING ADJUSTMENTS
Base regular season RPR's for most NBA players range between .400 and 1.000. The total range of possible defending adjustments to the base RPRs is from 0 to .307. In most cases, however, the adjustment will be between 0.070 and .240.
THE DEFENDING SUB RATING: PUTTING THE HIDDEN AND THE UNHIDDEN TOGETHER
Aside from the Hidden Defending Rating we can find out how well each player does in terms of unhidden or scored defending, can't we? Of course we can.
Unhidden or tracked defending, is defensive rebounding plus steals plus blocks minus personal fouls, calibrated according to the usual RPR factors. If we extract the combination of those four out of the same counts that underlie the RPR as a whole, and use the usual factors, we get what we are going to call the Scored Defending Production. This could also be thought of as Tracked Defending Production if you prefer. Then if we divide this by minutes, we have a Scored (or Tracked) Defending Rating.
Finally, if we combine Hidden Defending Rating (HDR) with Scored Defending Rating (SDR) we can have an Overall Defending Rating (ODR).
Obviously, the HDR scaling is designed to coordinate correctly with both SDR and with RPR as a whole. All of the coordinations reflect the latest undertanding of how basketball games are won and lost. The HDR constitutes about 45% of ODR while SDR constitutes the other 55%. In other words, the value of hidden defending is perceived to be about 45% of the overall value of defending, while the value of scored (unhidden) defending is perceived to be about 55% of the overall value of defending.
There appear to be many coaches and not a few hardcore basketball fans who believe that hidden defending is actually more important than scored defending, but I am never going to agree with that. I think that although hidden defending is important, and plausibly almost as important as tracked defending, that it can not be more than this. Hidden defending is like a quicksand, in that there seems to be a tendency for a substantial minority of basketball people to get carried away with estimating the importance of it, and then become more and more trapped by their error in terms of how they look at basketball or in terms of how they coach their team if they are coaching.
FORWARDS AND CENTERS WILL GENERALLY HAVE SUBSTANTIALLY HIGHER DEFENDING RATINGS
Due to having primary responsibility for defense of the paint and for rebounding, centers and forwards are going to inevitably have higher defensive ratings than will guards. Along with much greater opportunity for rebounds and blocks, centers and forwards also have more opportunity for such hidden defending actions as good man to man defending and correct rotations than do guards. Guards out on the perimeter generally should not and do not man to man defend as closely as do interior defenders, due to the well known guideline that it is quite foolish to foul a jump shooter outside of the paint.
THE OFFENSIVE SUB RATING
The Offensive Sub Rating is all tracked actions other than the defensive ones (defensive rebounding, steals, blocks, and personal fouls) combined together using the RPR weights, divided by minutes. In other words, it is Total Offensive Production divided by minutes. For the list of all tracked actions and the weight factors assigned to each, see the secion titled "The Formula" above.
THE BEST GUARDS WILL HAVE THE HIGHEST OFFENSIVE SUB RATINGS
The very best guards in basketball are ones who, although they are not afraid to drive to the hoop from time to time, are able to make outside shots at a good rate. Also, guards in general, and especially point guards, are usually primarily responsible for making assists. These two are among the several reasons why the better guards in pro basketball will have the highest offensive sub ratings in the League.
On the other hand, some of the most valuable players in the NBA are centers and forwards who are great defenders and efficient inside scorers at the same time. Even more unusual and probably for that reason more valuable is a forward who is (a) a great inside defender (b) a great inside scorer and (c) someone who can hit jump shots, perhaps even including threes, from outside the paint. Lamar Odom is an example of this kind of extremely valuable player.
Some of these big men will have offensive sub ratings that exceed those of the lessor skilled shooting guards and even those of some of the less skilled point guards.
[End of Defensive and Offensive Sub Ratings Section.]
======== SUMMARY OF PRIMARY FORMULAS SECTION =================
Real Player Production or RPR = (All tracked or scored actions weighted according to best available analysis of importance / minutes) + Hidden Defending Rating
Real Player Production or RPP = Total Offensive Production + Total Defensive Production. (All tracked or scored actions weighted according to best available analysis of importance.)
Offensive Sub Rating = Total Scored or Tracked Offensive Production / Minutes
Defensive Sub Rating = Total Scored or Tracked Defensive Production + Hidden Defending Rating
Hidden Defending Rating = Raw Hidden Defending Adjustment amplified via an approximtely reverse logarithmic scale symmetric as to the average. The objective of this is to correctly show value toward winning games via amplification of seemingly minute differences in points allowed per minute.
Raw Hidden Defending Adjustment = Points Allowed Per Minute * Pace Adjustment Factor * Team Defense Adjustment
Friday, May 29, 2009
User Guide for the Real Player Ratings Interactive Tool
The Quest for the Ring Toolbox is the only known place on the internet where anyone can rate players by entering game or season performance measurements.
TO USE THE REAL PLAYER RATING TOOL
1. Click "Add" A form will pop up allowing you to enter player numbers. DO NOT enter numbers in the view mode, which is the mode that you see when you arrive on this page. You must click Add and use the pop-up form.
2. Enter the performance measures shown for the player.
3. Click "submit"
4. Refresh this page. The Real Player Rating and the Real Player Producton for the player should now appear, along with all of the numbers you entered.
5. Repeat 1 through 4 for more players or more teams.
USE THE TOOL FOR ANY TIME FRAME YOU NEED
Provided you have the correct statistics, you can look at a player's performance for an individual game, for his or her entire career, or for anything in between, such as a season.
USE OF THE TOOL TO COMPARE TEAMS
You can also use the tool to rate and compare entire teams, simply by using the combined measures for all the players. Suppose you have two teams in a League that were considered extremely close, and they play in the Champiionship, and the Championship is decided in overtime. In such a case you might not be convinced that the team that won the Championship was really the better team. To investigate, you could compare the team RPRs of the two teams to try to get at which was really and truly the better team.
One interesting idea for Team RPR is to use combined team RPR (the sum of the player RPRs) to compare the same team from one year to another, which would go a long way towards answering a question that everyone asks all the time but that often no one ever has a very good answer for: which team was better: last year's or this year's?
CUSTOMIZED RATING
To request a custom rating scheme different from the one used in RPR, you can e-mail your request to questforthering at gmail.com.
BASE VERSUS ADJUSTED REAL PLAYER RATINGS
The tool here technically shows Base Real Player Rating (BRPR). This is already an extremely useful measure. But when Quest for the Ring reports team or NBA Real Player Ratings (RPRs) it reports them adjusted for "hidden defending".
On the other hand, Quest uses BRPR for Ultimate Game Breakdowns for individual games. It appears to be impossible to validly quantify hidden defending for individual games. With sample sizes of more than 300 minutes and especially for samples sizes in excess of 600 minutes, it is possible, under basic Statistical Smapling Theory, to come up with statistically valid estimations of hidden defending of players. Quest for the Ring has accomplished this as of very early 2009, and RPRs for teams and for the NBA as a whole are reported out only after the hidden defending adjustment has been included. Quest for the Ring is the only known site which as achieved this capability.
The root issue is that only defensive actions tracked by scorekeepers are counted in Base RPR, which is calculated in the above tool. Hidden defending is not included in the tool simply because it is impossible to quantify validly.
Such defending includes such things as man to man defending, defensive recognition, defensive rotation, pick and roll defending, defensive intensity, and defensive aggressiveness.
AN APPROXIMATE WAY TO ADD HIDDEN DEFENDING TO BASE REAL PLAYER RATINGS
There is an inexact but valid thing you can do to add the value of hidden defending into base RPR. Experience has taught that the value of hidden defending has a maximum theoretical range of 0 to .280 expressed in terms of the Real Player Rating. The vast majority of players will fall into the range of .040 at the low end to .240 at the high end. At the most, about 3% of players will be below .040 and another 3% will be above .240.
So what you can do is simply use your best judgement about how good a player is at defending not counting defensive rebounds, steals, blocks, and personal fouls, which are already counted in base RPR.
Theoretically, a player who never changes any shots from makes to misses would have a hidden defending rating of as low as .000. But even the "worst defensive players" in terms of "made them miss" defending, via untracked actions, will generally have hidden defending ratings of between about .040 and .060. Exactly in the middle players in terms of hidden defending will have hidden defending ratings of between .130 to .150. And the best defensive players in terms of hidden d3fending will generally have hidden defending ratings of between .220 and .240, although the absolute best such players can theoretically deserve a rating of up to .280.
AVOID BIAS WHEN ESTIMATING HIDDEN DEFENDING
I can not stress enough that when you add an amount for hidden defending to the base Real Player Rating, you are evaluating players based on defending actions other than defensive rebounding, steals, blocks, and personal fouls. Specifically, for example, when you estimate how good a player's hidden defending is, do not be biased either for or against players who make a lot of defensive rebounds.
Ironically though, players who make a large number of defensive rebounds and blocks often have lower hidden defending ratings than do "defensive specialists" who do not make a truly large number of defensive rebounds and blocks. This makes sense insofar as that it is not automatic or all that easy for players to be extremely good at rebounding and blocking and for example man to man defending at the same time. To some extent, with defending it is an either/or proposition. Great defenders can be either great rebounders and blockers, or alternatively they can be great man to man defenders and defensive recognizers and rotators.
Of course, there can be other combinations. For example, there will also be players who are average in rebounding and a little above average in man to man defending. It's just that it would be rare for a player to be an outstanding rebounder, blocker, and man to man defender all at the same time.
AVOID ALL OTHER BIASES WHEN ESTIMATING HIDDEN DEFENDING
And obviously, you should avoid bias for or against good offensive players. Quite honestly, how well or how bad a spcific player is on offense has almost nothing to do with how well or bad that player is on defense, allthough broadly speaking across the whole universe of players there is some degree of correlation.
REAL PLAYER RATING USER GUIDE
You need to transfer to another Guide if you want information on how the RPR is calculated in the above tool! To see all the details about how and why the Real Player Rating works, see the Real Player Rating User Guide. This Guide will, among other things, show you the exact factors involved in translating all of the basketball actions into ratings.
TO USE THE REAL PLAYER RATING TOOL
1. Click "Add" A form will pop up allowing you to enter player numbers. DO NOT enter numbers in the view mode, which is the mode that you see when you arrive on this page. You must click Add and use the pop-up form.
2. Enter the performance measures shown for the player.
3. Click "submit"
4. Refresh this page. The Real Player Rating and the Real Player Producton for the player should now appear, along with all of the numbers you entered.
5. Repeat 1 through 4 for more players or more teams.
USE THE TOOL FOR ANY TIME FRAME YOU NEED
Provided you have the correct statistics, you can look at a player's performance for an individual game, for his or her entire career, or for anything in between, such as a season.
USE OF THE TOOL TO COMPARE TEAMS
You can also use the tool to rate and compare entire teams, simply by using the combined measures for all the players. Suppose you have two teams in a League that were considered extremely close, and they play in the Champiionship, and the Championship is decided in overtime. In such a case you might not be convinced that the team that won the Championship was really the better team. To investigate, you could compare the team RPRs of the two teams to try to get at which was really and truly the better team.
One interesting idea for Team RPR is to use combined team RPR (the sum of the player RPRs) to compare the same team from one year to another, which would go a long way towards answering a question that everyone asks all the time but that often no one ever has a very good answer for: which team was better: last year's or this year's?
CUSTOMIZED RATING
To request a custom rating scheme different from the one used in RPR, you can e-mail your request to questforthering at gmail.com.
BASE VERSUS ADJUSTED REAL PLAYER RATINGS
The tool here technically shows Base Real Player Rating (BRPR). This is already an extremely useful measure. But when Quest for the Ring reports team or NBA Real Player Ratings (RPRs) it reports them adjusted for "hidden defending".
On the other hand, Quest uses BRPR for Ultimate Game Breakdowns for individual games. It appears to be impossible to validly quantify hidden defending for individual games. With sample sizes of more than 300 minutes and especially for samples sizes in excess of 600 minutes, it is possible, under basic Statistical Smapling Theory, to come up with statistically valid estimations of hidden defending of players. Quest for the Ring has accomplished this as of very early 2009, and RPRs for teams and for the NBA as a whole are reported out only after the hidden defending adjustment has been included. Quest for the Ring is the only known site which as achieved this capability.
The root issue is that only defensive actions tracked by scorekeepers are counted in Base RPR, which is calculated in the above tool. Hidden defending is not included in the tool simply because it is impossible to quantify validly.
Such defending includes such things as man to man defending, defensive recognition, defensive rotation, pick and roll defending, defensive intensity, and defensive aggressiveness.
AN APPROXIMATE WAY TO ADD HIDDEN DEFENDING TO BASE REAL PLAYER RATINGS
There is an inexact but valid thing you can do to add the value of hidden defending into base RPR. Experience has taught that the value of hidden defending has a maximum theoretical range of 0 to .280 expressed in terms of the Real Player Rating. The vast majority of players will fall into the range of .040 at the low end to .240 at the high end. At the most, about 3% of players will be below .040 and another 3% will be above .240.
So what you can do is simply use your best judgement about how good a player is at defending not counting defensive rebounds, steals, blocks, and personal fouls, which are already counted in base RPR.
Theoretically, a player who never changes any shots from makes to misses would have a hidden defending rating of as low as .000. But even the "worst defensive players" in terms of "made them miss" defending, via untracked actions, will generally have hidden defending ratings of between about .040 and .060. Exactly in the middle players in terms of hidden defending will have hidden defending ratings of between .130 to .150. And the best defensive players in terms of hidden d3fending will generally have hidden defending ratings of between .220 and .240, although the absolute best such players can theoretically deserve a rating of up to .280.
AVOID BIAS WHEN ESTIMATING HIDDEN DEFENDING
I can not stress enough that when you add an amount for hidden defending to the base Real Player Rating, you are evaluating players based on defending actions other than defensive rebounding, steals, blocks, and personal fouls. Specifically, for example, when you estimate how good a player's hidden defending is, do not be biased either for or against players who make a lot of defensive rebounds.
Ironically though, players who make a large number of defensive rebounds and blocks often have lower hidden defending ratings than do "defensive specialists" who do not make a truly large number of defensive rebounds and blocks. This makes sense insofar as that it is not automatic or all that easy for players to be extremely good at rebounding and blocking and for example man to man defending at the same time. To some extent, with defending it is an either/or proposition. Great defenders can be either great rebounders and blockers, or alternatively they can be great man to man defenders and defensive recognizers and rotators.
Of course, there can be other combinations. For example, there will also be players who are average in rebounding and a little above average in man to man defending. It's just that it would be rare for a player to be an outstanding rebounder, blocker, and man to man defender all at the same time.
AVOID ALL OTHER BIASES WHEN ESTIMATING HIDDEN DEFENDING
And obviously, you should avoid bias for or against good offensive players. Quite honestly, how well or how bad a spcific player is on offense has almost nothing to do with how well or bad that player is on defense, allthough broadly speaking across the whole universe of players there is some degree of correlation.
REAL PLAYER RATING USER GUIDE
You need to transfer to another Guide if you want information on how the RPR is calculated in the above tool! To see all the details about how and why the Real Player Rating works, see the Real Player Rating User Guide. This Guide will, among other things, show you the exact factors involved in translating all of the basketball actions into ratings.
Monday, May 25, 2009
User Guide for Real Game Ratings of Ultimate Game Breakdowns
Real Game Ratings will be a set of team performance measures in games that, quite simply, allow the user to be at a higher level of knowledge and appreciation about basketball than those who are limited to traditional box scores and statistics.
Some of these ratings have been developed by statistical gurus over the last 20 years or so. Some of them have been developed by Quest and have never been seen before. Although these measures are not rocket science, Quest is indebted to "those who have gone before" in developing sophisticated ways of looking at basketball games and players.
Definitely for Quest and hopefully for most of the statistical experts who have blazed the trail, the objective has to be to reveal how basketball games are won.
ADVANCED MEASURES FOR TEAMS IN GAMES
POSSESSIONS
The number of possessions is the foundation needed for several extremely important performance measures. Several statistical gurus have developed formulas for calculating the number of possessions a team had using box score numbers. The results of these formulas are extremely similar. Quest uses the following formula. Though relatively simple, it yields almost exactly the same number of possessions as do more complicated formulas.
Possessions = Field Goals Attempted + Turnovers + (.44 * Free Throws Attempted) - Offensive Rebounds
EFFICIENCY
Efficiency is the single most important "advanced" performance measure. Anytime you are in a hurry, you can simply look at efficiency to evaluate how well a team played either on offense or defense. Efficiency is:
Efficiency: Points / Possessions
Quest has already been reporting team offensive and defensive efficiency separately and as part of the Real Team Ratings. We will now be including this crucial measure in Ultimate Game Breakdowns, which as explained in the 2009 Site News Update in the User Guide will be mostly for playoff games in the future. In other words, Quest will become virtually the only source on the Internet for team offensive and team defensive efficiency in NBA playoff games.
OFFENSIVE REBOUND PERCENTAGE
Most everyone knows that offensive rebounding is very important toward winning games, especially close games. On the other hand, offensive rebounding is less important for the task of looking at a basketball offense in isolation and evaluating how good it is, and how good the guards are in that offense.
Quest will report this in the Ultimate Game Breakdowns for NBA playoff games. We will be virtually the only known source for this information. The formula is:
Offensive Rebound Percentage = Offensive Rebounds / (Offensive Rebounds + Opponent's Defensive Rebounds)
As you can see, this tells you how many of all of the available rebounds were snagged by the offensive squad.
TURNOVER PERCENTAGE
Turnovers are very, very important in determining which team wins the game, especially in close games. Turnovers are interwoven into the only at Quest offensive quality and power measures.
Quest will report this in the Ultimate Game Breakdowns for NBA playoff games We will be virtually the only known source for this information. The formula is:
Turnover Percentage = Turnovers / Possessions
GETTING TO THE LINE
When a team is playing a good defending team, a rough defending team such as the 2009 Denver Nuggets, and/or a team with very tall centers and power forwards, there is a tendency to settle for more outside jump shots than is wise. Basketball players are human, and given the choice between scoring without taking abuse in the paint and scoring with abuse, they will choose the former.
While it is not true that you can win games simply by excessively over weighting driving into the paint in hopes of dunks, layups, and fouls, it is true that you have to maintain some kind of balance between so doing and between shooting from outside the paint. The main reason the balance is important is that it is much more difficult to defend a team that mixes up well drives in the paint with outside shooting.
One complication involved in determining how much a team should take it to the rim is how closely the referees are calling a game. If the referees are calling the game loosely, if in other words the refs are "letting them play," the defenders have an unusual advantage in the paint, and the offense will be penalized if it drives into the paint too much. If the referees are calling a game tightly, than the offense in many cases will have the advantage in the paint, so obviously the coach should have the offense drive into the paint much more in that case. Keep in mind though that the referees may change how tightly they are calling the game as the game goes along.
Aside from the factor discussed in the previous paragraph, other factors that determine exactly what the balance should be between drives into the paint and outside shots is relatively complicated, and is beyond the scope of this User Guide. But this very, very important subject will be the subject of future Quest reports.
Quest will report the extent to which each team "got to the line" in Ultimate Game Breakdowns for NBA playoff games. We will be virtually the only known source for this information. The measure will be called Getting to the Line:
Getting to the Line = Free Throws Attempted / Field Goals Attempted
As you can see, this is the ratio of free throws to field goals attempted.
EFFECTIVE FIELD GOAL PERCENTAGE
This is simply a juiced up version of shooting accuracy. Basic shooting accuracy, as reported in box scores as field goals made / field goals attempted, is not a very good measure, because two-point and three-point scores are combined in together as if they are the same thing. Effective field goal percentage adjusts basic shooting percentage so that it reflects the extra value of 3-point scores. So this is where the crucial 3-point shooters are given credit for their contributions toward winning the game.
Obviously, this is one of the most important measures for deciding who wins basketball games, and at the player level, for determining who the most valuable offensive players really are. Defensive Effective Field Goal Percentage is just as important for evaluating team defense as is the flip side.
Effective field goal percentage is a crucial part of efficiency which, as explained above, is the most crucial measure of all for determining who is going to win the basketball game.
Quest will be reporting the Effective Field Goal Percentage for teams in Ultimate Game Breakdowns for NBA playoff games. We will be virtually the only known source for this information. The formula is:
Effective Field Goal Percentage = (Field Goals Made + (0.5 * 3-Point Field Goals Made)) / Field Goals Attempted
ASSIST / TURNOVER RATIO
This is number of assists divided by number of turnovers. Point guards have surprisingly different turnover rates. The ones with the lowest turnover rates are obviously the best for efficiency per se, but for overall effectiveness, you need to look at this ratio. A high turnover point guard can nevertheless be a very good point guard if he makes a truly large number of assists. In general, for every turnover a point guard suffers, the more assists he needs to make up for it.
EVALUATION SCALE FOR ASSIST / TURNOVER RATIO
4.00 and More: Ultra Careful Point Guard, arguably too careful
3.50 to 3.99: Extremely Careful Point Guard, possibly too careful
3.00 to 3.49: Very Careful Point Guard
2.60 to 2.99: Careful Point Guard
2.20 to 2.59: Medium Point Guard
1.90 to 2.19: Slightly Careless Point Guard
1.60 to 1.89: Careless Point Guard
1.40 to 1.59: Very Careless Point Guard
1.20 to 1.39: Extremely Careless Point Guard
1.19 and Less: Ultra Careless Point Guard
Unfortunately, it seems that the assist / turnover ratio by itself is not extremely useful for either evaluating point guards or even at the team level for determining how good an offense really is. The problem seems to be that some point guards "need" more turnovers to produce a lot of assists than do others. Some not very careful point guards can more than make up for turnovers by making assists that are more impressive and important than the assists made by careful point guards.
On the other hand, very careless and worse point guards are not going to be able to fully make up for all their turnovers no matter what they do. Assist / turnover ratios below 1.60 would signal point guards who are simply making too many turnovers to have any chance of being truly effective playmakers. Keep in mind though that young point guards will often have higher or much higher ratios than they will have later on.
So although by itself the ratio is not a greatly important thing, when used in conjunction with other offensive indicators, as Quest does, the assist/turnover ratio becomes much more useful.
Quest will be reporting the Assists/Turnovers ratio for NBA playoff games and for a limited number of regular season games. This will be one of the only sources for this, although of course it is easy to make a rough calculation of this in your head simply by looking at a box score.
QUEST FOR THE RING ORIGINAL SYSTEM FOR RATING THE QUALITY AND POWER OF BASKETBALL OFFENSES
Quest as of June 2009 is officially introducing high level performance measures found no where else on the Internet. Most of these are focused on the offense. But obviously, if you look at how an opponent did in these things, you can evaluate a defense using them. Very intelligent basketball fans, offensive basketball coaches, shooting guards, and especially point guards will be able to make the most use of these new measures.
PLAYMAKING IDENTITY
The Quest discussed during many reports in the first 18 months of the site a concept called "playmaking identity". This is basically to what extent a team's offense is organized for maximum effectiveness. The more a team's offense is directed by the guards in general and especially by the point guards, the more effective it will be. Here are some of the reasons for that:
1. Point Guards bring up the ball. For that and for traditional reasons, point guards are supposed to be able to direct, or in other words to organize the offense to some extent. In theory, the more organized the offense, the more effective it will be, mainly because the more organized it is, the more the plays are repetitive, and the more repetitive and practiced the plays, the easier it is to score.
2. Guards in general and especially point guards are directly responsible for running specific plays called by coaches, both plays in general for every game, and specific plays called in specific situations, especially off time outs and in critical late game situations.
3. Point Guards are supposed to be able to read defenses and to be able to evaluate how well defenders are playing in a particular game. They are supposed to be able to use this knowledge to adjust their offense so as to avoid the good defending and attack the bad defending.
Quest is now formalizing the concept. The definition of playmaking identity will be:
Playmaking Identity = ((2 * Point Guard Assists) + Shooting Guard Assists) / Total Assists
As you can see, this is an adjusted version of percentage of assists by guards. It's adjusted because the point guard assists are double weighted while the shooting guard assists are single weighted. In terms of ultimately rating how good the offense is, point guard assists are the most important, shooting guard assists are of medium importance, and assists by forwards and centers are less important.
Assists by forwards and centers are left out of playmaking identity, which is part of the main point of this new measure, because assists by them, while better than no assists at all, are not very reflective of a quality, organized, and efficient offense.
On the other hand, total assists and the assist/turnover ratio, which would include assists by forwards and centers, are very important, as you will see shortly.
Quest for the Ring will be reporting Playmaking Identity for most NBA playoff games and for carefully selected regular season games. This measure has been created here and will definitely be available only here.
PLAYMAKING QUALITY
Playmaking Quality is an extremely important measure developed by Quest. Not only has this particular measure never been seen before, there has never been a measure that gets at how "good" an offense really is as well as this one does.
The idea, like many of the most useful ideas, is relatively simple actually. The theory is that the two most important things in a basketball offense is how well organized it is, as reflected by Playmaking Identity, and how well it scores, as measured by Effective Field Goal Percentage. So the formula is:
Playmaking Quality = Playmaking Identity * Effective Field Goal Percentage
A way to look at this is that it is effective or real shooting adjusted by to what extent the shooting was organized or not. In theory, the more organized the shooting, the more inevitable it was in the game (and the less by chance it was). So this would be an indicator that you can get from every game as to how good the team's offense really is.
The higher the Playmaking Quality as measured by more and more games, the more wins from offense you can expect for that team over the course of a season. Also, the higher the Playmaking Quality, the lessor the chance that even very good defending opposing teams can win with defense alone.
Quest believes that Playmaking Quality may prove to have one of the most high correlations with winning playoff games and Championships of all measures in existence. Why? For one thing, and to reemphasize, Playmaking Quality measures the extent to which an offense is invulnerable to losing to a quality defense.
Quest for the Ring will be reporting Playmaking Quality for most NBA playoff games and for carefully selected regular season games. This measure has been created here and will definitely be available only here.
PLAYMAKING POWER
While Playmaking Quality alone may be enough to ultimately explain why NBA playoff games are won and lost, Quest is introducing another one that may possibly be slightly more important still: Playmaking Power. This is Playmaking Quality multiplied by the team Assists / Turnover ratio. The formula is:
Playmaking Power: Playmaking Quality * (Assists / Turnovers)
Think of this as the ultimate summary measure of the quality of the offense of a team, with everything including the kitchen sink thrown in. In general, we are taking the best offensive quality measure possible (Playmaking Quality) and multiplying by the effective quantity of that offense, as shown by assists / turnovers. Gross quantity of the offense in this framework would be assists. Net or effective quantity would be assists / turnovers, since the more turnovers there are, the less valuable the assists actually are.
Quest for the Ring will be reporting Playmaking Power for most NBA playoff games and for carefully selected regular season games. This measure has been created here and will definitely be available only here.
PRODUCTION OUTLOOK
Ultimate Game Breakdowns for NBA playoff games and for a small number of regular season games will from now on consist of Real Player Ratings and of the Real Game Ratings, the latter as explained in this User Guide, and the former explained in a separate User Guide.
Unfortunately, we do not have the resources at this time to produce all of this for a substantial number of regular season games, let alone for all regular season games. We will at the least produce this for all NBA Championship games, for all Conference Finals games, and for all Conference semifinals games. To the extent possible, we will produce this for Conference quarterfinals, also known as the first round of the NBA playoffs.
Also due to limitations currently existing, Ultimate Game Breakdowns for NBA playoff games, as detailed, will not be available for days, weeks, or possibly even months following those games. We will, however, be able to make sure that all the Breakdowns for a given year's playoff games are completed at the latest by the end of the year in which those games were played. And we will do everything possible to get the Ultimate Game Breakdowns for the Championship out quickly.
If someday we can find qualified individuals to join the Quest Performance Measure Division (so to speak) then we will be able to do more Breakdowns and we will be able to get the Breakdowns done more quickly.
You Can Post Your Response to Anything on Quest Here
Some of these ratings have been developed by statistical gurus over the last 20 years or so. Some of them have been developed by Quest and have never been seen before. Although these measures are not rocket science, Quest is indebted to "those who have gone before" in developing sophisticated ways of looking at basketball games and players.
Definitely for Quest and hopefully for most of the statistical experts who have blazed the trail, the objective has to be to reveal how basketball games are won.
ADVANCED MEASURES FOR TEAMS IN GAMES
POSSESSIONS
The number of possessions is the foundation needed for several extremely important performance measures. Several statistical gurus have developed formulas for calculating the number of possessions a team had using box score numbers. The results of these formulas are extremely similar. Quest uses the following formula. Though relatively simple, it yields almost exactly the same number of possessions as do more complicated formulas.
Possessions = Field Goals Attempted + Turnovers + (.44 * Free Throws Attempted) - Offensive Rebounds
EFFICIENCY
Efficiency is the single most important "advanced" performance measure. Anytime you are in a hurry, you can simply look at efficiency to evaluate how well a team played either on offense or defense. Efficiency is:
Efficiency: Points / Possessions
Quest has already been reporting team offensive and defensive efficiency separately and as part of the Real Team Ratings. We will now be including this crucial measure in Ultimate Game Breakdowns, which as explained in the 2009 Site News Update in the User Guide will be mostly for playoff games in the future. In other words, Quest will become virtually the only source on the Internet for team offensive and team defensive efficiency in NBA playoff games.
OFFENSIVE REBOUND PERCENTAGE
Most everyone knows that offensive rebounding is very important toward winning games, especially close games. On the other hand, offensive rebounding is less important for the task of looking at a basketball offense in isolation and evaluating how good it is, and how good the guards are in that offense.
Quest will report this in the Ultimate Game Breakdowns for NBA playoff games. We will be virtually the only known source for this information. The formula is:
Offensive Rebound Percentage = Offensive Rebounds / (Offensive Rebounds + Opponent's Defensive Rebounds)
As you can see, this tells you how many of all of the available rebounds were snagged by the offensive squad.
TURNOVER PERCENTAGE
Turnovers are very, very important in determining which team wins the game, especially in close games. Turnovers are interwoven into the only at Quest offensive quality and power measures.
Quest will report this in the Ultimate Game Breakdowns for NBA playoff games We will be virtually the only known source for this information. The formula is:
Turnover Percentage = Turnovers / Possessions
GETTING TO THE LINE
When a team is playing a good defending team, a rough defending team such as the 2009 Denver Nuggets, and/or a team with very tall centers and power forwards, there is a tendency to settle for more outside jump shots than is wise. Basketball players are human, and given the choice between scoring without taking abuse in the paint and scoring with abuse, they will choose the former.
While it is not true that you can win games simply by excessively over weighting driving into the paint in hopes of dunks, layups, and fouls, it is true that you have to maintain some kind of balance between so doing and between shooting from outside the paint. The main reason the balance is important is that it is much more difficult to defend a team that mixes up well drives in the paint with outside shooting.
One complication involved in determining how much a team should take it to the rim is how closely the referees are calling a game. If the referees are calling the game loosely, if in other words the refs are "letting them play," the defenders have an unusual advantage in the paint, and the offense will be penalized if it drives into the paint too much. If the referees are calling a game tightly, than the offense in many cases will have the advantage in the paint, so obviously the coach should have the offense drive into the paint much more in that case. Keep in mind though that the referees may change how tightly they are calling the game as the game goes along.
Aside from the factor discussed in the previous paragraph, other factors that determine exactly what the balance should be between drives into the paint and outside shots is relatively complicated, and is beyond the scope of this User Guide. But this very, very important subject will be the subject of future Quest reports.
Quest will report the extent to which each team "got to the line" in Ultimate Game Breakdowns for NBA playoff games. We will be virtually the only known source for this information. The measure will be called Getting to the Line:
Getting to the Line = Free Throws Attempted / Field Goals Attempted
As you can see, this is the ratio of free throws to field goals attempted.
EFFECTIVE FIELD GOAL PERCENTAGE
This is simply a juiced up version of shooting accuracy. Basic shooting accuracy, as reported in box scores as field goals made / field goals attempted, is not a very good measure, because two-point and three-point scores are combined in together as if they are the same thing. Effective field goal percentage adjusts basic shooting percentage so that it reflects the extra value of 3-point scores. So this is where the crucial 3-point shooters are given credit for their contributions toward winning the game.
Obviously, this is one of the most important measures for deciding who wins basketball games, and at the player level, for determining who the most valuable offensive players really are. Defensive Effective Field Goal Percentage is just as important for evaluating team defense as is the flip side.
Effective field goal percentage is a crucial part of efficiency which, as explained above, is the most crucial measure of all for determining who is going to win the basketball game.
Quest will be reporting the Effective Field Goal Percentage for teams in Ultimate Game Breakdowns for NBA playoff games. We will be virtually the only known source for this information. The formula is:
Effective Field Goal Percentage = (Field Goals Made + (0.5 * 3-Point Field Goals Made)) / Field Goals Attempted
ASSIST / TURNOVER RATIO
This is number of assists divided by number of turnovers. Point guards have surprisingly different turnover rates. The ones with the lowest turnover rates are obviously the best for efficiency per se, but for overall effectiveness, you need to look at this ratio. A high turnover point guard can nevertheless be a very good point guard if he makes a truly large number of assists. In general, for every turnover a point guard suffers, the more assists he needs to make up for it.
EVALUATION SCALE FOR ASSIST / TURNOVER RATIO
4.00 and More: Ultra Careful Point Guard, arguably too careful
3.50 to 3.99: Extremely Careful Point Guard, possibly too careful
3.00 to 3.49: Very Careful Point Guard
2.60 to 2.99: Careful Point Guard
2.20 to 2.59: Medium Point Guard
1.90 to 2.19: Slightly Careless Point Guard
1.60 to 1.89: Careless Point Guard
1.40 to 1.59: Very Careless Point Guard
1.20 to 1.39: Extremely Careless Point Guard
1.19 and Less: Ultra Careless Point Guard
Unfortunately, it seems that the assist / turnover ratio by itself is not extremely useful for either evaluating point guards or even at the team level for determining how good an offense really is. The problem seems to be that some point guards "need" more turnovers to produce a lot of assists than do others. Some not very careful point guards can more than make up for turnovers by making assists that are more impressive and important than the assists made by careful point guards.
On the other hand, very careless and worse point guards are not going to be able to fully make up for all their turnovers no matter what they do. Assist / turnover ratios below 1.60 would signal point guards who are simply making too many turnovers to have any chance of being truly effective playmakers. Keep in mind though that young point guards will often have higher or much higher ratios than they will have later on.
So although by itself the ratio is not a greatly important thing, when used in conjunction with other offensive indicators, as Quest does, the assist/turnover ratio becomes much more useful.
Quest will be reporting the Assists/Turnovers ratio for NBA playoff games and for a limited number of regular season games. This will be one of the only sources for this, although of course it is easy to make a rough calculation of this in your head simply by looking at a box score.
QUEST FOR THE RING ORIGINAL SYSTEM FOR RATING THE QUALITY AND POWER OF BASKETBALL OFFENSES
Quest as of June 2009 is officially introducing high level performance measures found no where else on the Internet. Most of these are focused on the offense. But obviously, if you look at how an opponent did in these things, you can evaluate a defense using them. Very intelligent basketball fans, offensive basketball coaches, shooting guards, and especially point guards will be able to make the most use of these new measures.
PLAYMAKING IDENTITY
The Quest discussed during many reports in the first 18 months of the site a concept called "playmaking identity". This is basically to what extent a team's offense is organized for maximum effectiveness. The more a team's offense is directed by the guards in general and especially by the point guards, the more effective it will be. Here are some of the reasons for that:
1. Point Guards bring up the ball. For that and for traditional reasons, point guards are supposed to be able to direct, or in other words to organize the offense to some extent. In theory, the more organized the offense, the more effective it will be, mainly because the more organized it is, the more the plays are repetitive, and the more repetitive and practiced the plays, the easier it is to score.
2. Guards in general and especially point guards are directly responsible for running specific plays called by coaches, both plays in general for every game, and specific plays called in specific situations, especially off time outs and in critical late game situations.
3. Point Guards are supposed to be able to read defenses and to be able to evaluate how well defenders are playing in a particular game. They are supposed to be able to use this knowledge to adjust their offense so as to avoid the good defending and attack the bad defending.
Quest is now formalizing the concept. The definition of playmaking identity will be:
Playmaking Identity = ((2 * Point Guard Assists) + Shooting Guard Assists) / Total Assists
As you can see, this is an adjusted version of percentage of assists by guards. It's adjusted because the point guard assists are double weighted while the shooting guard assists are single weighted. In terms of ultimately rating how good the offense is, point guard assists are the most important, shooting guard assists are of medium importance, and assists by forwards and centers are less important.
Assists by forwards and centers are left out of playmaking identity, which is part of the main point of this new measure, because assists by them, while better than no assists at all, are not very reflective of a quality, organized, and efficient offense.
On the other hand, total assists and the assist/turnover ratio, which would include assists by forwards and centers, are very important, as you will see shortly.
Quest for the Ring will be reporting Playmaking Identity for most NBA playoff games and for carefully selected regular season games. This measure has been created here and will definitely be available only here.
PLAYMAKING QUALITY
Playmaking Quality is an extremely important measure developed by Quest. Not only has this particular measure never been seen before, there has never been a measure that gets at how "good" an offense really is as well as this one does.
The idea, like many of the most useful ideas, is relatively simple actually. The theory is that the two most important things in a basketball offense is how well organized it is, as reflected by Playmaking Identity, and how well it scores, as measured by Effective Field Goal Percentage. So the formula is:
Playmaking Quality = Playmaking Identity * Effective Field Goal Percentage
A way to look at this is that it is effective or real shooting adjusted by to what extent the shooting was organized or not. In theory, the more organized the shooting, the more inevitable it was in the game (and the less by chance it was). So this would be an indicator that you can get from every game as to how good the team's offense really is.
The higher the Playmaking Quality as measured by more and more games, the more wins from offense you can expect for that team over the course of a season. Also, the higher the Playmaking Quality, the lessor the chance that even very good defending opposing teams can win with defense alone.
Quest believes that Playmaking Quality may prove to have one of the most high correlations with winning playoff games and Championships of all measures in existence. Why? For one thing, and to reemphasize, Playmaking Quality measures the extent to which an offense is invulnerable to losing to a quality defense.
Quest for the Ring will be reporting Playmaking Quality for most NBA playoff games and for carefully selected regular season games. This measure has been created here and will definitely be available only here.
PLAYMAKING POWER
While Playmaking Quality alone may be enough to ultimately explain why NBA playoff games are won and lost, Quest is introducing another one that may possibly be slightly more important still: Playmaking Power. This is Playmaking Quality multiplied by the team Assists / Turnover ratio. The formula is:
Playmaking Power: Playmaking Quality * (Assists / Turnovers)
Think of this as the ultimate summary measure of the quality of the offense of a team, with everything including the kitchen sink thrown in. In general, we are taking the best offensive quality measure possible (Playmaking Quality) and multiplying by the effective quantity of that offense, as shown by assists / turnovers. Gross quantity of the offense in this framework would be assists. Net or effective quantity would be assists / turnovers, since the more turnovers there are, the less valuable the assists actually are.
Quest for the Ring will be reporting Playmaking Power for most NBA playoff games and for carefully selected regular season games. This measure has been created here and will definitely be available only here.
PRODUCTION OUTLOOK
Ultimate Game Breakdowns for NBA playoff games and for a small number of regular season games will from now on consist of Real Player Ratings and of the Real Game Ratings, the latter as explained in this User Guide, and the former explained in a separate User Guide.
Unfortunately, we do not have the resources at this time to produce all of this for a substantial number of regular season games, let alone for all regular season games. We will at the least produce this for all NBA Championship games, for all Conference Finals games, and for all Conference semifinals games. To the extent possible, we will produce this for Conference quarterfinals, also known as the first round of the NBA playoffs.
Also due to limitations currently existing, Ultimate Game Breakdowns for NBA playoff games, as detailed, will not be available for days, weeks, or possibly even months following those games. We will, however, be able to make sure that all the Breakdowns for a given year's playoff games are completed at the latest by the end of the year in which those games were played. And we will do everything possible to get the Ultimate Game Breakdowns for the Championship out quickly.
If someday we can find qualified individuals to join the Quest Performance Measure Division (so to speak) then we will be able to do more Breakdowns and we will be able to get the Breakdowns done more quickly.
You Can Post Your Response to Anything on Quest Here
Thursday, March 5, 2009
User Guide for Real Team Ratings Reports: Updated as of May 5 2009
USER GUIDE FOR REAL TEAM RATINGS REPORTS
UPDATED AS OF MAY 5 2009
The only update done on May 5 was a new standardization adjustment for the wins and losses against playoff teams factor, which is described in detail in the appropriate section below.
The Real Team Ratings(RTR)are intended to rate teams with respect to how much success they are most likely to have in the playoffs. They are not intended to be simply a summary of how well teams have done. The ratings are calculated for all teams, even though 14 of the 30 NBA teams do not qualify for the playoffs. Even though they will not be playing any playoff games, the ratings for the lower teams nevertheless give an accurate measure of how well those teams would most likely do if they were in the playoffs. So for those lottery teams, RTR is an interesting hypothetical.
As of 2009 the RTR rating system was improved. It was improved to make absolutely certain that you can predict the outcome of the playoffs in advance as accurately as possible. All crucial factors are now included and weighted very carefully. The biggest and most important improvement for 2009 and beyond was the introduction of points for wins over and points subtracted for losses to the top 16 teams (which would be the playoff teams themselves.)
Broadly speaking, the system is a combination of net efficiency (net points per 100 possessions), net points per game, the defensive overweight adjustment, the wins and losses against playoff bound teams, and the relatively small pace overweight adjustment. Each of these is now described in detail.
THE FIVE FACTORS USED FOR REAL TEAM RATINGS
1.NET EFFICIENCY
Offensive Efficiency minus Defensive Efficiency equals Net Efficiency. A weight of 2.0 is applied to Net Efficiency.
1a. Offensive Efficiency: Points Scored per 100 Possessions
1b. Defensive Efficiency: Points Allowed per 100 Possessions
Note that net efficiency, or net points gained or lossed per each 100 possessions, is double weighted, whereas the simple net points per game is single weighted. In the original version of the Real Team Ratings, the efficiency was single weighted and the points per game was not a part of the system.
2.NET POINTS PER GAME
Points Per Game Scored Minus Points Per Game Allowed. A weight of 1.0 is applied to Net Points Per Game. In other words, the straight up points scored per game minus points allowed per game is input into the formula.
3.DEFENSIVE OVERWEIGHT ADJUSTMENT
The teams are sorted by defensive efficiency. Then, using a range from 5.8 to -5.8, points are assigned, in equal increments of .4, to each team in order of how it ranks according to defensive efficiency. Specifically, the team with the best defensive efficiency (fewest points allowed per 100 possessions) is given 5.8 points, the 2nd most defensively efficient team gets 5.4 points, the third most defensively efficient team gets 5.0 points, and so on, until the least defensively efficient team gets minus 5.8 points.
It is well known that, for the playoffs, how well a team can defend is generally somewhat more important than during the regular season. This factor answers the need to overweight defending in order to get accurate playoff projections. The adjustment gives an increase or a decrease in every team's rating in accordance with how each team ranks in defensive efficiency in the NBA.
The amount of the adjustment is carefully calibrated to be sufficient without being excessive. Since almost all teams ramp up their defense in the playoffs, you have to be careful here to avoid getting carried away and putting in adjustments that are too large.
4.WINS OVER AND LOSSES TO PLAYOFF TEAMS
Each team's win-loss record is accessed for games it played against the top 16 teams and, separately, each team's win-loss record is accessed for games it played against the top 10 teams. These two records are added together, which has the effect of double weighting wins and losses versus top 10 teams, while leaving wins and losses versus the 11th through the 16th best teams single weighted. In other words, the sum of the wins versus the top 10 teams is added to the sum of the wins versus the top 16 teams, and the sum of these two sums is added to the overall Real Team Rating formula (with weight of simply 1.0). Losses in games against the top 10 and against the top 16 are subtracted from the overall Real Team Rating in the same way that the wins are added.
In late April 2009, it was realized that since the teams play a different total number of games against the top 10 and the top 16, that a distortion would be introduced into the RTR as designed at the time. Therefore, a new adjustment of the wins/losses against playoff teams was rolled out. The purpose of the adjustment is to standardize the wins/losses against playoff teams records so that the differing number of such games played by different teams is no longer a distorting factor.
Were this adjustment not done, the impact of wins and losses against playoff teams on the overall RTR would be greater for teams that played more games against other playoff teams, which is a significant distortion.
The adjustment works as follows. Whichever team played the most games versus top 10 and top 16 teams (combined) is ascertained, and then that number of games is used as the standarization base. The unadjusted wins percentage for each team is then multiplied by the standardization base to yield the best estimate for how many wins each team would have won against the top 10 and the top 16 had they played as many games (the standard number of games) as the team that played the most such games.
The best estimate for standarized losses is then of course the number of standard games minus the standardized wins. Both the standardized wins and the standardized losses are projected to two decimal points (two numbers beyond the decimal points.)
This factor, wins over and losses to playoff teams, is the key improvement from the early versions of RTR, and helps to clearly establish Real Team Ratings as the most accurate playoff predictor possible. By counting in the overall formula actual wins and losses in games between the likely playoff teams, you have gone in a straight line directly to evidence for the question we are out to answer: how good are the teams really going to be in the playoffs, according to everything known now?
5.PACE OVERWEIGHT ADJUSTMENT
The teams are sorted by pace. Pace for each team is the average number of possessions per game for that team's games. Then, using a range from 2.9 to -2.9, points are assigned, in equal increments of .2, to each team in order of how it ranks according to pace. Specifically, the team with the slowest pace (fewest possessions per game) is given 2.9 points, the 2nd slowest pace team gets 2.7 points, the third slowest pace team gets 2.5 points, and so on, until the fastest pace team gets minus 2.9 points.
The reason for the pace adjustment is that there is a mild but definite correlation between slower pace and winning playoff series. It is a little more difficult, on average, for fast pace teams to win playoff series than it is for slow pace teams to win them. Therefore, a small adjustment called the pace overweight adjustment is factored in to RTR.
Why exactly do slower paced teams have a slightly easier job winning playoff series? Consider an example. For example, consider the Denver Nuggets. They are one of the fastest paced teams in the NBA during the regular season. If you just look at the efficiency measures, the Nuggets might appear to be almost identical to another, much slower team. But these two teams would be very different when you look at efficiency and pace together. In theory, slower paced teams can more reliably reproduce their nice regular season net efficiency in the playoffs than can faster paced teams, mostly because the playoffs feature a higher defensive intensity and aggressiveness, which automatically slows down the pace.
Suppose that in the playoffs, the fast paced Nuggets and a slow paced team play. Each team had almost exactly the same offensive, defensive, and net efficiency numbers during the regular season. By playing extra hard on defense, the slow pace team can automatically slow down the game to some degree, which will disrupt the offensive (and possibly the defensive) efficiency of the Nuggets, the team that was fast pace in the regular season. In other words, there will be fewer possessions for the fast pace team in the playoff games than it typically had in the regular season. This in turn means that the fast pace team will be disrupted from what they did during the regular season to one extent or another.
This means that for the fast pace team, both the offensive and the defensive efficiency could change in the playoffs from what it was in the regular season, due to all of the changes forced on the fast pace team by the change of pace. Both the offensive and the defensive efficiency might change, and each change could be either for the better or for the worse, but by far the most likely changes would be that the offense would be substantially less efficient, while the defense would not be changed much. A much less efficient offense, but about the same defense, is exactly what we have seen from the Nuggets in their numerous playoff series losses in recent years.
In extreme cases, such as the fastest pace team being slowed down dramatically in the playoffs by an extremely slow team, the pace adjustment may be inadequate, so that there may still be some forecast error even after everything we have done.
The bottom line is that in all known cases, faster paced teams do not do as well in the playoffs as they do in the regular season, all other things equal. If a fast paced team wants to win in the playoffs, it would be wise to do some things better in the playoffs than they did those things in the regular season, in order to compensate for being forced to operate at a slower pace.
CALCULATION OF RTR: THE FORMULA
The easiest way to describe the final calculation of RTR is to give you the formula.
RTR = 2 X Net Efficiency + Net Points per Game + Defense Overweight Adjustment + Number of Wins Over Top 10 Teams + Number of Wins Over Top 16 Teams - Number of Losses To Top 10 Teams - Number of Losses To Top 16 Teams + Pace Overweight Adjustment
OTHER FACTORS THAT COULD SWING A SERIES ARE UNKNOWABLE
To get even better accuracy than RTR, you would have to know what the injury situation is going to be in late April and throughout May and June. And you would have to start guessing how unmeasurable factors such as currently unknown coaching strategies and possible injury problems will affect individual playoff series.
IT IS NOT SAFE TO USE THE RTR TO TRY TO WIN MONEY BETTING ON GAMES OR SERIES
As of 2009 you can no longer use the actual ratings as a starting point for estimations of final score margins of games between teams, which is no real loss, because you could not do that very accurately with the previous system anyway. Remember, you should never bet money on the outcome of games or even on the outcome of playoff series, because there are always variables that neither you nor anyone else will be aware of that will go into determining the outcome of each game.
ADJUSTING BASE RTR FOR FACTORS NOT INCLUDED IN BASE RTR THAT CAN BE MAJOR FOR DETERMINING WHO WINS PLAYOFF SERIES
Of all the popular American sports Leagues, the NBA is the one where the better team is most likely to avoid being upset in the playoffs. Therefore, the RTR system can be used to gain knowledge of which team is most likely to win playoff series. It can also be used to determine whether how good various players played led to an upset or not, and to get a general idea of how much better or worse than expected teams played in playoff series.
In order to do these things, however, it is necessary to modify base RTR for at least two factors that can not be directly embedded in it: home court advantage and players unavailable or playing badly due to injuries.
Even just to start with, due to a small amount of unavoidable statistical error remaining in the base RTR, there has to be a 7-12 points difference between teams before you can start to have any big confidence that one team will defeat another in the playoffs. More importantly, there are also factors unknown until close to or exactly when the series is actually being played, especially what the injury situation for each team is.
Another factor that becomes a big one when two teams with very similar ratings are playing is home court advantage. Home court advantage is estimated to be worth between 5 and 7 points.
Even if it were not for the injury wild card factor, use of RTR to predict playoff series prior to March 1 would not be very useful, due to the need for a fairly high percentage of the season to be over before the projections in base RTR are statistically meaningful, and due to the fact that the ratings are not relative to time, but rather expand with time.
MAJOR FACTORS DETERMINING WHO WINS PLAYOFF SERIES NOT BUILT IN TO THE BASIC RTR
1.HOME COURT ADVANTAGE
The home court advantage in RTR terms is believed to be 5 to 7 points
2. PLAYERS UNAVAILABLE (OR PLAYING POORLY) DUE TO INJURES
The impact relative to RTR is believed to be mathematically anything from almost 0 to 100 points for each injured and unavailable player who played during the regular season, depending mostly on how good the injured player is and depending on to what extent other players are able to step up and replace the injured player or players. Although the mathematical range of impact is 0 to 100, the realistically true and common relevant range covering the loss of starters and key reserves, is about 10 to 60 RTR points.
Players who were injured the entire season are irrelevant, except of course they are relevant in the hypothetical sense of how the season could have been different. Players who were injured relatively early in the regular season, in November or December, are only slightly relevant, and the loss of them would be a much smaller number of reduced RTR points than when the loss is later. Players who were injured late in the season, from mid-February to mid-April, have the most relevancy to whether playoff series can be won or lost, and the RTR reduction for them is much higher.
INJURY ADJUSTMENT TO RTR DIFFERENTIALS (DEVELOPMENTAL BUT HAS PASSED INITAL REVIEWS)
This adjustment is under development and review. However, it has passed initial approvals, which means it is very likely to be officially adopted in the near future.
The base or starting point is the quality of the player, as shown by the Real Player Ratings including the defending adjustment.
BASE FOR THE DOWNWARD ADJUSTMENT TO RTR DUE TO INJURY/UNAVAILABLILTY OF PLAYERS
Almost Perfect Player: 125
Historical Superstar: 113
Superstar: 98
Star: 81
Outstanding/Solid Starter: 64
Major Role Player: 49
Role Player: 36
Minor Role Player: 25
Very Minor Role Player: 16
Poor Player: 9
Very Poor Player: 4
Extremely Poor Player: 1
Remember, the base is just the hypothetical maximum possible impact, and is not a relevant real world outlook by itself. The RTR impact and the real world impact of the loss is determined by many variables, the most important of which are when the player was lost, and to what extent the other players can make up for the loss of that player.
DOWNWARD ADJUSTMENT FACTORS TO THE RTR INJURY ADJUSTMENT BASE
FACTOR FOR TIME IN THE SEASON WHEN THE PLAYER WAS LOST
November .1
December .25
January .45
February .70
March .95
April 1.0
FACTOR FOR TO WHAT EXTENT OTHER PLAYERS CAN MAKE UP FOR THE UNAVAILABLE PLAYER
We currently have to describe this adjustment generically:
(1)The player or players getting the lost minutes are close in quality (no more than one Real Player Rating category lower) and also close in terms of basketball position played, to the lost player: .4 to .6
(2)The player or players getting the lost minutes are close in quality, but are not close in terms of the position(s) played: .55 to .75
(3)The player or players getting the lost minutes are not close in quality, but are close in terms of the position played: .7 to .9
FACTOR FOR IMPORTANCE OF PLAYER TO THE TEAM (SHOWN BY MINUTES PER GAME)
32 mpg and more: 1.0
28 to 31.1: .9
24 to 27.9: .8
20 to 23.9: .7
16 to 19.9: .6
12 to 15.9: .5
8 to 11.9: .4
4 to 7.9: .3
Less than 4: .2
AN EXAMPLE OF HOW BASE RTR IS ADJUSTED FOR INJURIES: THE 2009 BOSTON CELTICS
Ok, now lets consider an example to see how all of this works. We have this year Kevin Garnett, one of the very best players in the NBA, not available to the Celtics due to injury. Garnett is in the almost perfect player range, so you start with 125 points as the unadjusted impact of the Garnett loss to the Celtics' Real Team Rating.
Now we need to breakdown that 125 gross number. First, we need the when the player was lost factor. Although Garnett came back for a few games in March, he was essentially lost for the season in February, so the factor to use is .70.
Next we need the stepping up of other players factor. Rajon Rondo has responded by being even better than he was in the regular season, which already was at the superstar level. And Rondo overall is playing as good as Garnett would be if he were playing, at the almost perfect level to be exact. So this turn of events would put the Celtics in the (2) situation above, so the factor to use would be between .55 to .75. Based on how extremely well Rondo is playing, and based on the overall resiliency of the Celtics, we will use .60 as the factor.
The third and final factor to use on the gross injury base RTR adjustment is how important the player is to the team, as shown by minutes per game. Garnett was about 31 minutes a game this season, so the factor to use is .9.
Now we can calculate by how many points we should downgrade the Celtics' RTR due to the loss of Kevin Garnett:
125 *.60 *.70 *.90 = 47.3
LEON POWE
Now we need to figure out how much the Celtics have lost due to Leon Powe not being available. Because the Celtics are also without Leon Powe, a very underrated 17 minutes a game forward, whose 2008-09 quality level was major role player.
Since he is a major role player, we start with 49 points. He was lost at the very end of the season, so the when lost factor is 1.0. Due to the simultaneous loss of Garnett and Powe, the Celtics are severly depleted upfront, which means you can positionally replace Garnett but not both Garnett and Powe at the same time, which in turn translates into a factor of about .75 for the to what extent other players can replace Powe factor. Finally, Powe played about 17 minutes a game this season, which means you should use a factor of .6 for importance of the player to the team.
Now we can calculate about how many RTR points the Celtics have lost due to Leon Powe not being available:
49 * 1.0 * .75 * .60 = 22.1
The two losses cost the Celtics roughly 47.3 + 22.1 = 69.4 points.
The Celtics' 2008-09 RTR unadjusted for injuries was 54.4. The Chicago Bulls' RTR was -24.02. So the difference between the two teams started out as 78.42. But the loss of Garnett and Powe cost the Celtics roughly 69.4 RTR points, leaving them with only a measly 9 points advantage over the Bulls, or 15 points once you add on the 6 points for home court advantage.
According to the Guidelines for Interpreting Differences in Real Team Ratings (which is reprinted from the overall User Guide below) a 15 points difference means there was somewhere between a 72% and an 80% that the Celtics were going to win the series.
So in real life, the series played out exactly as the RTR predicted that it would, at least once you have adjusted the RTR for players not available!
So what has to be done to figure out the real differences between any two playoff teams is to start with RTR, and then use the under development but destined to become officially approved Injury Adjustments. So we will have RTR-IA to use to predict playoff series, and to judge whether a series has been an upset or not, and also to judge just how heroic players such as Rajon Rondo are. (Note: always remember to add in about 6 points for the home court advantage, too.)
Bottom line for the Celtics: Rajon Rondo has been out there making up for both Kevin Garnett and Leon Powe. He has been a hero, not only for that, but also because he was already a superstar in the regular season, and then he had to get even better than being a superstar.
How would you like someone to come up to you and say to you: "You are a superstar, but that is not good enouogh right now. Now you have to be better than a superstar, you have to be almost perfect." You might be thinking to yourself: "Yeah right, I'm going to be able to suddenly be even better than I already am. I'll try, but I doubt it's going to happen; I'm not a miracle worker".
Whereas Rondo said: "No problem, I'm down with that." And then he proceeded to go out there and be almost perfect, as if that is as easy as waking up in the morning.
So Rajon Rondo is no doubt one of the heroes of the 2009 playoffs. And he has a very good chance of being the first ever Real Playoffs MVP here at Quest, which will be an honor given out to the player who did the most to help his team, regardless of exactly how many games and series the team won.
It's a good thing he already has a ring, because if he didn't fully earn it last year, he sure as hell earned one this year.
But the Celtics will need a miracle to defeat the Magic despite this.
GUIDE TO INTERPRETATION OF DIFFERENCES BETWEEN TEAMS' REAL TEAM RATINGS
In the numeric interpretaton guide that follows, the word "roughly" is repeatedly used in front of the probability numbers, as a reminder about the small amount of unavoidable statistical error, and to emphasize that unknown factors, especially injuries, will in some cases result in substantially different actual probabilities.
The probability percentages are based on the historical results in the NBA:
REMEMBER:
In order to realistically compare teams using RTR, particularly if your objective is to know who is most likely to win playoff series, you MUST add 5 to 7 points to the RTR of the team that has home court advantage (HCA), and you MUST use the Injury Adjustment (IA) for all injured players on all teams evaluated. The injury adjustment is under development, is in beta currently, and is described extensively above. In other words, for any and all teams being evaluated, you MUST use RTR + HCA + IA
The following guidelines assume that the HCA and all injury adjustments have been correctly done.
DIFFERENCE IN RATINGS IS BETWEEN 0 AND 5.9
The series is a complete toss-up, when statistical error is considered. There is a strong possibility of a 7 game series. The higher team has a 50% to 57% chance of winning, depending on what exactly the difference is. These probabilities are too low for anyone to have any confidence in using adjusted RTR to say who will win. All series of this type are decided quite simply by who plays better, by who coaches better, or both.
DIFFERENCE IN RATINGS IS BETWEEN 6.0 AND 11.9
The series can easily go either way, although the higher team has a slight edge, and has between a 57% to 66% chance of winning, depending on where in the range the difference is. There is a very substantial chance of a 7-game series. If the lower team wins, it is nothing more than a very minor upset. Either slight differences in the quality of coaching, certain players playing a little better or a little worse than they did in the regular season, or both, could be responsible for an upset at this level.
DIFFERENCE IN RATINGS IS BETWEEN 12.0 AND 17.9
The series can go either way, and this type of difference gives a substantial chance for a 7-game series. But the higher team has a clear edge. The higher team has between a 66% and a 76% probability of winning, depending on where in the range the difference is. If the lower team wins, it is an upset, though just a small upset. Either slight differences in the quality of coaching, certain players playing a little better or a little worse than they did in the regular season, or both, could be responsible for an upset at this level.
DIFFERENCE IN RATINGS IS BETWEEN 18.0 AND 23.9
The higher team has roughly between a 76% to 84% probability of winning, depending on where in the range the difference is. There is still a chance, but a relatively small one, for a 7-game series. If the lower team wins, it is clearly an upset, though generally not a major upset. Either coaches, certain players, or both could be responsible for an upset at this level.
DIFFERENCE IN RATINGS IS BETWEEN 24.0 AND 29.9
The higher team has roughly between an 84% to a 91% probability of winning, depending on where in the range the difference is. In this kind of series, often the only way the lower team can win the series is by extending the series out to 7 games and then somehow winning the 7th game, thus taking the series 4 games to 3. However, it is not uncommon, assuming there is an upset in this type of series, for the lower team to so severly disrupt the favored team that the lower team upsets the higher, favored team 4 games to 2. Whichever way it does it, if the lower team does win coming in down by this amount, it should be considered a major upset. In many such cases, the coaching would have to be very wrong and/or negligent.
DIFFERENCE IN RATINGS IS BETWEEN 30.0 AND 35.9
The higher team has roughly between a 91% and a 95.5% probability of winning. depending on where in the range the difference is. In this kind of series, often the only way the lower team can win the series is by taking the series 7 games and winning the 7th game, thus taking the series 4 games to 3. However, there have been a tiny number of series where a team with this amount of a RTR deficit has won the series by so severly disrupting the favored team that it is able to win the series 4 games to 2. In the vast majority of such cases, the coaching for the higher team was severely wrong and/or negligent. Whether accomplished in 6 games or 7, the lower team winning despite being this far behind in RTR is extremely rare, and would be considered a very shocking, historical upset.
DIFFERENCE IN RATINGS IS BETWEEN 36.0 AND 41.9
The higher team has roughly between a 95.5% and a 98% probability of winning, depending on where in the range the difference is. Obviously, an upset would be extremely rare, shocking, and historical. It would in most cases be caused substantially by incompetent and/or severely negligent coaching. With this amount of difference, any upset would almost certainly have to be with the series going all seven games.
DIFFERENCE IN RATINGS IS 42.0 AND 47.9
The higher team has a roughly between a 98% and a 99.5% probability of winning the series. Obviously, an upset would be extremely rare, shocking, and historical. It would in most cases be caused substantially by incompetent and/or severely negligent coaching. With this amount of difference, any upset would almost certainly have to be with the series going all seven games.
DEFFERENCE IN RATINGS IS 48.0 OR MORE
It is close to a 100% certainty that the higher team will win the series. Obviously, an upset would be extremely rare, shocking, and historical. It would in the vast majority of cases be caused substantially by incompetent and/or severely negligent coaching. With this amount of difference, any upset would almost certainly have to be with the series going all seven games.
UPDATED AS OF MAY 5 2009
The only update done on May 5 was a new standardization adjustment for the wins and losses against playoff teams factor, which is described in detail in the appropriate section below.
The Real Team Ratings(RTR)are intended to rate teams with respect to how much success they are most likely to have in the playoffs. They are not intended to be simply a summary of how well teams have done. The ratings are calculated for all teams, even though 14 of the 30 NBA teams do not qualify for the playoffs. Even though they will not be playing any playoff games, the ratings for the lower teams nevertheless give an accurate measure of how well those teams would most likely do if they were in the playoffs. So for those lottery teams, RTR is an interesting hypothetical.
As of 2009 the RTR rating system was improved. It was improved to make absolutely certain that you can predict the outcome of the playoffs in advance as accurately as possible. All crucial factors are now included and weighted very carefully. The biggest and most important improvement for 2009 and beyond was the introduction of points for wins over and points subtracted for losses to the top 16 teams (which would be the playoff teams themselves.)
Broadly speaking, the system is a combination of net efficiency (net points per 100 possessions), net points per game, the defensive overweight adjustment, the wins and losses against playoff bound teams, and the relatively small pace overweight adjustment. Each of these is now described in detail.
THE FIVE FACTORS USED FOR REAL TEAM RATINGS
1.NET EFFICIENCY
Offensive Efficiency minus Defensive Efficiency equals Net Efficiency. A weight of 2.0 is applied to Net Efficiency.
1a. Offensive Efficiency: Points Scored per 100 Possessions
1b. Defensive Efficiency: Points Allowed per 100 Possessions
Note that net efficiency, or net points gained or lossed per each 100 possessions, is double weighted, whereas the simple net points per game is single weighted. In the original version of the Real Team Ratings, the efficiency was single weighted and the points per game was not a part of the system.
2.NET POINTS PER GAME
Points Per Game Scored Minus Points Per Game Allowed. A weight of 1.0 is applied to Net Points Per Game. In other words, the straight up points scored per game minus points allowed per game is input into the formula.
3.DEFENSIVE OVERWEIGHT ADJUSTMENT
The teams are sorted by defensive efficiency. Then, using a range from 5.8 to -5.8, points are assigned, in equal increments of .4, to each team in order of how it ranks according to defensive efficiency. Specifically, the team with the best defensive efficiency (fewest points allowed per 100 possessions) is given 5.8 points, the 2nd most defensively efficient team gets 5.4 points, the third most defensively efficient team gets 5.0 points, and so on, until the least defensively efficient team gets minus 5.8 points.
It is well known that, for the playoffs, how well a team can defend is generally somewhat more important than during the regular season. This factor answers the need to overweight defending in order to get accurate playoff projections. The adjustment gives an increase or a decrease in every team's rating in accordance with how each team ranks in defensive efficiency in the NBA.
The amount of the adjustment is carefully calibrated to be sufficient without being excessive. Since almost all teams ramp up their defense in the playoffs, you have to be careful here to avoid getting carried away and putting in adjustments that are too large.
4.WINS OVER AND LOSSES TO PLAYOFF TEAMS
Each team's win-loss record is accessed for games it played against the top 16 teams and, separately, each team's win-loss record is accessed for games it played against the top 10 teams. These two records are added together, which has the effect of double weighting wins and losses versus top 10 teams, while leaving wins and losses versus the 11th through the 16th best teams single weighted. In other words, the sum of the wins versus the top 10 teams is added to the sum of the wins versus the top 16 teams, and the sum of these two sums is added to the overall Real Team Rating formula (with weight of simply 1.0). Losses in games against the top 10 and against the top 16 are subtracted from the overall Real Team Rating in the same way that the wins are added.
In late April 2009, it was realized that since the teams play a different total number of games against the top 10 and the top 16, that a distortion would be introduced into the RTR as designed at the time. Therefore, a new adjustment of the wins/losses against playoff teams was rolled out. The purpose of the adjustment is to standardize the wins/losses against playoff teams records so that the differing number of such games played by different teams is no longer a distorting factor.
Were this adjustment not done, the impact of wins and losses against playoff teams on the overall RTR would be greater for teams that played more games against other playoff teams, which is a significant distortion.
The adjustment works as follows. Whichever team played the most games versus top 10 and top 16 teams (combined) is ascertained, and then that number of games is used as the standarization base. The unadjusted wins percentage for each team is then multiplied by the standardization base to yield the best estimate for how many wins each team would have won against the top 10 and the top 16 had they played as many games (the standard number of games) as the team that played the most such games.
The best estimate for standarized losses is then of course the number of standard games minus the standardized wins. Both the standardized wins and the standardized losses are projected to two decimal points (two numbers beyond the decimal points.)
This factor, wins over and losses to playoff teams, is the key improvement from the early versions of RTR, and helps to clearly establish Real Team Ratings as the most accurate playoff predictor possible. By counting in the overall formula actual wins and losses in games between the likely playoff teams, you have gone in a straight line directly to evidence for the question we are out to answer: how good are the teams really going to be in the playoffs, according to everything known now?
5.PACE OVERWEIGHT ADJUSTMENT
The teams are sorted by pace. Pace for each team is the average number of possessions per game for that team's games. Then, using a range from 2.9 to -2.9, points are assigned, in equal increments of .2, to each team in order of how it ranks according to pace. Specifically, the team with the slowest pace (fewest possessions per game) is given 2.9 points, the 2nd slowest pace team gets 2.7 points, the third slowest pace team gets 2.5 points, and so on, until the fastest pace team gets minus 2.9 points.
The reason for the pace adjustment is that there is a mild but definite correlation between slower pace and winning playoff series. It is a little more difficult, on average, for fast pace teams to win playoff series than it is for slow pace teams to win them. Therefore, a small adjustment called the pace overweight adjustment is factored in to RTR.
Why exactly do slower paced teams have a slightly easier job winning playoff series? Consider an example. For example, consider the Denver Nuggets. They are one of the fastest paced teams in the NBA during the regular season. If you just look at the efficiency measures, the Nuggets might appear to be almost identical to another, much slower team. But these two teams would be very different when you look at efficiency and pace together. In theory, slower paced teams can more reliably reproduce their nice regular season net efficiency in the playoffs than can faster paced teams, mostly because the playoffs feature a higher defensive intensity and aggressiveness, which automatically slows down the pace.
Suppose that in the playoffs, the fast paced Nuggets and a slow paced team play. Each team had almost exactly the same offensive, defensive, and net efficiency numbers during the regular season. By playing extra hard on defense, the slow pace team can automatically slow down the game to some degree, which will disrupt the offensive (and possibly the defensive) efficiency of the Nuggets, the team that was fast pace in the regular season. In other words, there will be fewer possessions for the fast pace team in the playoff games than it typically had in the regular season. This in turn means that the fast pace team will be disrupted from what they did during the regular season to one extent or another.
This means that for the fast pace team, both the offensive and the defensive efficiency could change in the playoffs from what it was in the regular season, due to all of the changes forced on the fast pace team by the change of pace. Both the offensive and the defensive efficiency might change, and each change could be either for the better or for the worse, but by far the most likely changes would be that the offense would be substantially less efficient, while the defense would not be changed much. A much less efficient offense, but about the same defense, is exactly what we have seen from the Nuggets in their numerous playoff series losses in recent years.
In extreme cases, such as the fastest pace team being slowed down dramatically in the playoffs by an extremely slow team, the pace adjustment may be inadequate, so that there may still be some forecast error even after everything we have done.
The bottom line is that in all known cases, faster paced teams do not do as well in the playoffs as they do in the regular season, all other things equal. If a fast paced team wants to win in the playoffs, it would be wise to do some things better in the playoffs than they did those things in the regular season, in order to compensate for being forced to operate at a slower pace.
CALCULATION OF RTR: THE FORMULA
The easiest way to describe the final calculation of RTR is to give you the formula.
RTR = 2 X Net Efficiency + Net Points per Game + Defense Overweight Adjustment + Number of Wins Over Top 10 Teams + Number of Wins Over Top 16 Teams - Number of Losses To Top 10 Teams - Number of Losses To Top 16 Teams + Pace Overweight Adjustment
OTHER FACTORS THAT COULD SWING A SERIES ARE UNKNOWABLE
To get even better accuracy than RTR, you would have to know what the injury situation is going to be in late April and throughout May and June. And you would have to start guessing how unmeasurable factors such as currently unknown coaching strategies and possible injury problems will affect individual playoff series.
IT IS NOT SAFE TO USE THE RTR TO TRY TO WIN MONEY BETTING ON GAMES OR SERIES
As of 2009 you can no longer use the actual ratings as a starting point for estimations of final score margins of games between teams, which is no real loss, because you could not do that very accurately with the previous system anyway. Remember, you should never bet money on the outcome of games or even on the outcome of playoff series, because there are always variables that neither you nor anyone else will be aware of that will go into determining the outcome of each game.
ADJUSTING BASE RTR FOR FACTORS NOT INCLUDED IN BASE RTR THAT CAN BE MAJOR FOR DETERMINING WHO WINS PLAYOFF SERIES
Of all the popular American sports Leagues, the NBA is the one where the better team is most likely to avoid being upset in the playoffs. Therefore, the RTR system can be used to gain knowledge of which team is most likely to win playoff series. It can also be used to determine whether how good various players played led to an upset or not, and to get a general idea of how much better or worse than expected teams played in playoff series.
In order to do these things, however, it is necessary to modify base RTR for at least two factors that can not be directly embedded in it: home court advantage and players unavailable or playing badly due to injuries.
Even just to start with, due to a small amount of unavoidable statistical error remaining in the base RTR, there has to be a 7-12 points difference between teams before you can start to have any big confidence that one team will defeat another in the playoffs. More importantly, there are also factors unknown until close to or exactly when the series is actually being played, especially what the injury situation for each team is.
Another factor that becomes a big one when two teams with very similar ratings are playing is home court advantage. Home court advantage is estimated to be worth between 5 and 7 points.
Even if it were not for the injury wild card factor, use of RTR to predict playoff series prior to March 1 would not be very useful, due to the need for a fairly high percentage of the season to be over before the projections in base RTR are statistically meaningful, and due to the fact that the ratings are not relative to time, but rather expand with time.
MAJOR FACTORS DETERMINING WHO WINS PLAYOFF SERIES NOT BUILT IN TO THE BASIC RTR
1.HOME COURT ADVANTAGE
The home court advantage in RTR terms is believed to be 5 to 7 points
2. PLAYERS UNAVAILABLE (OR PLAYING POORLY) DUE TO INJURES
The impact relative to RTR is believed to be mathematically anything from almost 0 to 100 points for each injured and unavailable player who played during the regular season, depending mostly on how good the injured player is and depending on to what extent other players are able to step up and replace the injured player or players. Although the mathematical range of impact is 0 to 100, the realistically true and common relevant range covering the loss of starters and key reserves, is about 10 to 60 RTR points.
Players who were injured the entire season are irrelevant, except of course they are relevant in the hypothetical sense of how the season could have been different. Players who were injured relatively early in the regular season, in November or December, are only slightly relevant, and the loss of them would be a much smaller number of reduced RTR points than when the loss is later. Players who were injured late in the season, from mid-February to mid-April, have the most relevancy to whether playoff series can be won or lost, and the RTR reduction for them is much higher.
INJURY ADJUSTMENT TO RTR DIFFERENTIALS (DEVELOPMENTAL BUT HAS PASSED INITAL REVIEWS)
This adjustment is under development and review. However, it has passed initial approvals, which means it is very likely to be officially adopted in the near future.
The base or starting point is the quality of the player, as shown by the Real Player Ratings including the defending adjustment.
BASE FOR THE DOWNWARD ADJUSTMENT TO RTR DUE TO INJURY/UNAVAILABLILTY OF PLAYERS
Almost Perfect Player: 125
Historical Superstar: 113
Superstar: 98
Star: 81
Outstanding/Solid Starter: 64
Major Role Player: 49
Role Player: 36
Minor Role Player: 25
Very Minor Role Player: 16
Poor Player: 9
Very Poor Player: 4
Extremely Poor Player: 1
Remember, the base is just the hypothetical maximum possible impact, and is not a relevant real world outlook by itself. The RTR impact and the real world impact of the loss is determined by many variables, the most important of which are when the player was lost, and to what extent the other players can make up for the loss of that player.
DOWNWARD ADJUSTMENT FACTORS TO THE RTR INJURY ADJUSTMENT BASE
FACTOR FOR TIME IN THE SEASON WHEN THE PLAYER WAS LOST
November .1
December .25
January .45
February .70
March .95
April 1.0
FACTOR FOR TO WHAT EXTENT OTHER PLAYERS CAN MAKE UP FOR THE UNAVAILABLE PLAYER
We currently have to describe this adjustment generically:
(1)The player or players getting the lost minutes are close in quality (no more than one Real Player Rating category lower) and also close in terms of basketball position played, to the lost player: .4 to .6
(2)The player or players getting the lost minutes are close in quality, but are not close in terms of the position(s) played: .55 to .75
(3)The player or players getting the lost minutes are not close in quality, but are close in terms of the position played: .7 to .9
FACTOR FOR IMPORTANCE OF PLAYER TO THE TEAM (SHOWN BY MINUTES PER GAME)
32 mpg and more: 1.0
28 to 31.1: .9
24 to 27.9: .8
20 to 23.9: .7
16 to 19.9: .6
12 to 15.9: .5
8 to 11.9: .4
4 to 7.9: .3
Less than 4: .2
AN EXAMPLE OF HOW BASE RTR IS ADJUSTED FOR INJURIES: THE 2009 BOSTON CELTICS
Ok, now lets consider an example to see how all of this works. We have this year Kevin Garnett, one of the very best players in the NBA, not available to the Celtics due to injury. Garnett is in the almost perfect player range, so you start with 125 points as the unadjusted impact of the Garnett loss to the Celtics' Real Team Rating.
Now we need to breakdown that 125 gross number. First, we need the when the player was lost factor. Although Garnett came back for a few games in March, he was essentially lost for the season in February, so the factor to use is .70.
Next we need the stepping up of other players factor. Rajon Rondo has responded by being even better than he was in the regular season, which already was at the superstar level. And Rondo overall is playing as good as Garnett would be if he were playing, at the almost perfect level to be exact. So this turn of events would put the Celtics in the (2) situation above, so the factor to use would be between .55 to .75. Based on how extremely well Rondo is playing, and based on the overall resiliency of the Celtics, we will use .60 as the factor.
The third and final factor to use on the gross injury base RTR adjustment is how important the player is to the team, as shown by minutes per game. Garnett was about 31 minutes a game this season, so the factor to use is .9.
Now we can calculate by how many points we should downgrade the Celtics' RTR due to the loss of Kevin Garnett:
125 *.60 *.70 *.90 = 47.3
LEON POWE
Now we need to figure out how much the Celtics have lost due to Leon Powe not being available. Because the Celtics are also without Leon Powe, a very underrated 17 minutes a game forward, whose 2008-09 quality level was major role player.
Since he is a major role player, we start with 49 points. He was lost at the very end of the season, so the when lost factor is 1.0. Due to the simultaneous loss of Garnett and Powe, the Celtics are severly depleted upfront, which means you can positionally replace Garnett but not both Garnett and Powe at the same time, which in turn translates into a factor of about .75 for the to what extent other players can replace Powe factor. Finally, Powe played about 17 minutes a game this season, which means you should use a factor of .6 for importance of the player to the team.
Now we can calculate about how many RTR points the Celtics have lost due to Leon Powe not being available:
49 * 1.0 * .75 * .60 = 22.1
The two losses cost the Celtics roughly 47.3 + 22.1 = 69.4 points.
The Celtics' 2008-09 RTR unadjusted for injuries was 54.4. The Chicago Bulls' RTR was -24.02. So the difference between the two teams started out as 78.42. But the loss of Garnett and Powe cost the Celtics roughly 69.4 RTR points, leaving them with only a measly 9 points advantage over the Bulls, or 15 points once you add on the 6 points for home court advantage.
According to the Guidelines for Interpreting Differences in Real Team Ratings (which is reprinted from the overall User Guide below) a 15 points difference means there was somewhere between a 72% and an 80% that the Celtics were going to win the series.
So in real life, the series played out exactly as the RTR predicted that it would, at least once you have adjusted the RTR for players not available!
So what has to be done to figure out the real differences between any two playoff teams is to start with RTR, and then use the under development but destined to become officially approved Injury Adjustments. So we will have RTR-IA to use to predict playoff series, and to judge whether a series has been an upset or not, and also to judge just how heroic players such as Rajon Rondo are. (Note: always remember to add in about 6 points for the home court advantage, too.)
Bottom line for the Celtics: Rajon Rondo has been out there making up for both Kevin Garnett and Leon Powe. He has been a hero, not only for that, but also because he was already a superstar in the regular season, and then he had to get even better than being a superstar.
How would you like someone to come up to you and say to you: "You are a superstar, but that is not good enouogh right now. Now you have to be better than a superstar, you have to be almost perfect." You might be thinking to yourself: "Yeah right, I'm going to be able to suddenly be even better than I already am. I'll try, but I doubt it's going to happen; I'm not a miracle worker".
Whereas Rondo said: "No problem, I'm down with that." And then he proceeded to go out there and be almost perfect, as if that is as easy as waking up in the morning.
So Rajon Rondo is no doubt one of the heroes of the 2009 playoffs. And he has a very good chance of being the first ever Real Playoffs MVP here at Quest, which will be an honor given out to the player who did the most to help his team, regardless of exactly how many games and series the team won.
It's a good thing he already has a ring, because if he didn't fully earn it last year, he sure as hell earned one this year.
But the Celtics will need a miracle to defeat the Magic despite this.
GUIDE TO INTERPRETATION OF DIFFERENCES BETWEEN TEAMS' REAL TEAM RATINGS
In the numeric interpretaton guide that follows, the word "roughly" is repeatedly used in front of the probability numbers, as a reminder about the small amount of unavoidable statistical error, and to emphasize that unknown factors, especially injuries, will in some cases result in substantially different actual probabilities.
The probability percentages are based on the historical results in the NBA:
REMEMBER:
In order to realistically compare teams using RTR, particularly if your objective is to know who is most likely to win playoff series, you MUST add 5 to 7 points to the RTR of the team that has home court advantage (HCA), and you MUST use the Injury Adjustment (IA) for all injured players on all teams evaluated. The injury adjustment is under development, is in beta currently, and is described extensively above. In other words, for any and all teams being evaluated, you MUST use RTR + HCA + IA
The following guidelines assume that the HCA and all injury adjustments have been correctly done.
DIFFERENCE IN RATINGS IS BETWEEN 0 AND 5.9
The series is a complete toss-up, when statistical error is considered. There is a strong possibility of a 7 game series. The higher team has a 50% to 57% chance of winning, depending on what exactly the difference is. These probabilities are too low for anyone to have any confidence in using adjusted RTR to say who will win. All series of this type are decided quite simply by who plays better, by who coaches better, or both.
DIFFERENCE IN RATINGS IS BETWEEN 6.0 AND 11.9
The series can easily go either way, although the higher team has a slight edge, and has between a 57% to 66% chance of winning, depending on where in the range the difference is. There is a very substantial chance of a 7-game series. If the lower team wins, it is nothing more than a very minor upset. Either slight differences in the quality of coaching, certain players playing a little better or a little worse than they did in the regular season, or both, could be responsible for an upset at this level.
DIFFERENCE IN RATINGS IS BETWEEN 12.0 AND 17.9
The series can go either way, and this type of difference gives a substantial chance for a 7-game series. But the higher team has a clear edge. The higher team has between a 66% and a 76% probability of winning, depending on where in the range the difference is. If the lower team wins, it is an upset, though just a small upset. Either slight differences in the quality of coaching, certain players playing a little better or a little worse than they did in the regular season, or both, could be responsible for an upset at this level.
DIFFERENCE IN RATINGS IS BETWEEN 18.0 AND 23.9
The higher team has roughly between a 76% to 84% probability of winning, depending on where in the range the difference is. There is still a chance, but a relatively small one, for a 7-game series. If the lower team wins, it is clearly an upset, though generally not a major upset. Either coaches, certain players, or both could be responsible for an upset at this level.
DIFFERENCE IN RATINGS IS BETWEEN 24.0 AND 29.9
The higher team has roughly between an 84% to a 91% probability of winning, depending on where in the range the difference is. In this kind of series, often the only way the lower team can win the series is by extending the series out to 7 games and then somehow winning the 7th game, thus taking the series 4 games to 3. However, it is not uncommon, assuming there is an upset in this type of series, for the lower team to so severly disrupt the favored team that the lower team upsets the higher, favored team 4 games to 2. Whichever way it does it, if the lower team does win coming in down by this amount, it should be considered a major upset. In many such cases, the coaching would have to be very wrong and/or negligent.
DIFFERENCE IN RATINGS IS BETWEEN 30.0 AND 35.9
The higher team has roughly between a 91% and a 95.5% probability of winning. depending on where in the range the difference is. In this kind of series, often the only way the lower team can win the series is by taking the series 7 games and winning the 7th game, thus taking the series 4 games to 3. However, there have been a tiny number of series where a team with this amount of a RTR deficit has won the series by so severly disrupting the favored team that it is able to win the series 4 games to 2. In the vast majority of such cases, the coaching for the higher team was severely wrong and/or negligent. Whether accomplished in 6 games or 7, the lower team winning despite being this far behind in RTR is extremely rare, and would be considered a very shocking, historical upset.
DIFFERENCE IN RATINGS IS BETWEEN 36.0 AND 41.9
The higher team has roughly between a 95.5% and a 98% probability of winning, depending on where in the range the difference is. Obviously, an upset would be extremely rare, shocking, and historical. It would in most cases be caused substantially by incompetent and/or severely negligent coaching. With this amount of difference, any upset would almost certainly have to be with the series going all seven games.
DIFFERENCE IN RATINGS IS 42.0 AND 47.9
The higher team has a roughly between a 98% and a 99.5% probability of winning the series. Obviously, an upset would be extremely rare, shocking, and historical. It would in most cases be caused substantially by incompetent and/or severely negligent coaching. With this amount of difference, any upset would almost certainly have to be with the series going all seven games.
DEFFERENCE IN RATINGS IS 48.0 OR MORE
It is close to a 100% certainty that the higher team will win the series. Obviously, an upset would be extremely rare, shocking, and historical. It would in the vast majority of cases be caused substantially by incompetent and/or severely negligent coaching. With this amount of difference, any upset would almost certainly have to be with the series going all seven games.
Thursday, February 26, 2009
User Guide for Real Player Rating Reports for the NBA, for NBA Teams, and for Games
NOTICE: THIS USER GUIDE HAS BEEN UPDATED; THE NEW VERSION IS HERE. This version apples for ratings produced from late February until the end of May 2009.
REAL PLAYER RATINGS BY TEAM USER GUIDE
Updated Feb. 25, 2009
INTRODUCTION TO THE CONCEPT OF REAL PLAYER RATINGS
The Real Player Rating (RPR) is a very carefully constructed all inclusive performance measure. Everything of value that a basketball player can do is recorded by official NBA scorekeepers who sit right along the edge of the court, mid-court, and who are trained to observe and record everything that happens in a game.
Since these days all of these counts are immediately input into continually updated public data bases online, such as at ESPN, it is theoretically possible to combine everything together into an overall performance measure for each player. This is what the RPR does.
Real Player Rating or RPR is everything tracked by scorekeepers that a player does, good and bad, added and subtracted (with negative things such as turnovers and missed shots being subtracted). Very carefully calibrated factors, or weights, are applied to the different elements. The calibration, as you would expect, is done to reflect the different value toward winning games that different actions on the court have. All of the good and bad combined together is divided by minutes, so we can tell the rate, which we need to determine the overall quality or value of the player.
REAL PLAYER RATINGS ARE ADJUSTED FOR DEFENDING NOT TRACKED BY SCOREKEEPERS STARTING IN 2009
Not counting purely subjective and abstract factors such as leadership, and not counting a few infrequently occurring actions on the court (not being counted or tracked by anyone yet) such as chasing down loose balls, the only thing a basketball player can regularly do on the court of any value that is not counted by NBA scorekeepers is preventing what would have been a score from being a score by defending against the shot or shots during a possession well enough to stop what would have been a score by the opposing team. In other words, what the player does to make the possessions of the opposing teams worthless other than what is already counted, which would be rebounds, steals, blocks, and personal fouls. These untracked or hidden actions would include effective man to man defending, effective rotation on defense off screens and picks, defensive recognition, and quickness of defensive reaction. These things would be counted by scorekeepers if it were possible. But, for example, there is no way to know exactly how many shots a good (or any kind of) defender has changed from being a score to a miss.
Quest for the Ring has developed a statistically valid way to accurately estimate the untracked or hidden aspects of defending. This is described in complete detail in the latter sections of this Guide.
SIMPLICITY, RELIABILITY, TRANSPARENCY, AND FOCUS ONLY ON "WINNING POWER"
Like everything statistical we do at Quest, we have kept this process as simple and reliable as possible, while at the same time spending as much time as necessary on design, quality control and performance evaluation. Unlike some other practitioners, we avoid what you might call layered complexity, which leads to formulas which can not be understood without studying them and which high traffic sites will not show on any of their web pages for fear that the public will rebel against the statistic. At Quest, we think that our rating systems can be understood and evaluated by most high school graduates, and we keep everything out in the open through User Guides such as this one.
Basketball statistical gurus frequently forget that no matter how intricate their formulas are, they are very heavily manipulating process items such as assists and rebounds while spending very little time on how these things fit together to produce wins and losses. We think that they are making the mistake, whether or not they are aware, of injecting value adjustments regarding how they think the game should be played and value adjustments about which playing styles are better than others.
Whereas, the primary objectives of the relative simplicity (small number of formulas, to be precise) of the Quest RPR is to avoid all how the game should be played and how players should play value judgments. We don't care about the styles, only about the results. The RPR is concerned first and foremost with the impact each player has on the potential for winning games.
Quest thinks it makes more sense to minimize the manipulation of process items, and to focus much more on coming up with the best possible estimation of how the process items impact points for and points against in games, which in turn of course determines wins. Whereas other "advanced statistics" might give you more depth and flavor regarding how a particular player plays, the Quest RPR is a way for the reader to, in a very quick and easy way, determine what the overall value of the player is with respect to producing wins or losses.
In other words, the foundation of RPR is and will always be measurement of a player's power to help win basketball games, whereas the foundation of other, more complicated statistics may include preferences about how the game should be played and about the style of players, with winning power measured less accurately as a result of those focuses.
THE MAIN REASON REAL PLAYER RATING IS SO VALUABLE
Because it is per time, RPR is the best possible measure of the net quality of a basketball player, or simply "how good" the player is (on average) for each minute of playing time.
REAL PLAYER RATING REPORTS CAN BE FOR THE WHOLE NBA, FOR A TEAM, OR FOR A GAME
With a Real Player Ratings Report for the entire NBA, you can see very rapidly who the best players in the NBA have been during the course of the season.
With a Real Player Ratings Report for a Team for the Regular Season, you can see very rapidly who the best players on the team have been during the course of the season. You can use this information to investigate the possiblity that the coach is not perfect. Well, we know that no coach is perfect. So really, with the benefit of 20/20 hindsight, we can investigate and determine what mistakes the coach has apparently made with regard to rotations and playing times. Furthermore, by using the Ratings, basketball knowledge, a little creativity, and logical deduction, we can also investigate and perhaps determine whether the coach has made incorrect decisions regarding which strategies and plays are best for his team's offense and defense.
Real Player Ratings for games are the most important component of Reports called Ultimate Game Breakdown: Players Reports.
CAUTIONS
To be completely honest and clear, although it is the best possible overall real life measure, RPR is still not a perfect or absolute, "final word" measure on any player. In general, you must remember that all performance measures including this one for the NBA are relative rather than absolute measures. The ratings are relative to the team context. Players do not exist in a vacuum, especially in basketball.
Several specific cautions will now be described.
Because basketball is a team game, and more so than most other sports, players who are on really good teams might have their own performances "crowded out" to some extent by even better players. So, paradoxically, ratings of good players on good teams will generally have slightly lower ratings than they would have if they were on a bad team. Conversely, great players on bad teams will have slightly higher ratings than they would have if they were on a good team.
Players need not only playing time but possession of the ball in order to produce many of the things that count in the rating. So if, for whatever reason, a player does not get the ball as often as he would on a different team, or with a different coach, or with whatever other circumstances you can dream of, then his RPR will be lower than what it could or would be.
If a good player is on a good team where there are players even better than he is, than his RPR will likely be lower than it would be if he were on a not as good team.
If a good player plays a certain position for which his team has an even better player, then it's probable that the better player will crowd out the lesser player to one extent or another, so that the lesser player's RPR will be lower than what it would be if he were the best player at the position on the team.
The ratings are only for the current season. It has recently been discovered that many player's ratings often change up or down by more than 10% from year to year, and by much greater amounts over many years.
Those who think defense in basketball is much more important than offense may consider the magnitude of the defensive adjustment to be inadequate. They will contend that defensive specialists who are poor offensive players should have a higher rating. While we realized that we needed to adjust the ratings for defending not tracked by NBA scorekeepers, we continue to believe that players who are great defensive specialists but poor or undeveloped offensive players should in most cases rank no higher than the major role player level.
Do not forget that RPR is a per time measure. RPP and not RPR measures total impact of a player. RPR measures how valuable a player has been toward winning basketball games, per unit of time.
The classification scheme, like the ratings, is relative. A role player on a bad team might be a solid starter on a very good team. A star on a bad team might be just a major role player on a really good team. And so on and so forth. A player is a star, a role player, or whatever only in the contexts of the particular season and the particular team involved. If he was on a different team, or if it was a different year, his classification might be different.
So in conclusion, don't think of RPR as the ultimate gospel or bible on how good players are. But do think of it as an extremely accurate and reliable summary of how good the players actually have been in real life in the specific time (season or playoffs) and place (team) involved.
A NOTE ABOUT REAL PLAYER RATINGS FOR INDIVIDUAL GAMES
However, not as many breakdowns of individual game ratings are going to closely track the overall average for the roster as you might think. This is because one of the interesting things about basketball that makes it different from most other sports is that "how good" a player is from game to game varies radically. The best players have terrible games where they do almost nothing sometimes, while players who normally do not do much can every once in a while have outstanding games, at least if you measure it per minute on the court anyway. If you just looked at actual production, and never at a reserve player's Real Player Rating, you would hardly notice any of his unusually outstanding games, since players who normally do not do much will normally not have much playing time.
INTERACTIONS BETWEEN PLAYING TIMES, PLAYER RATINGS, AND THE NEEDS OF TEAMS
There are certain things that only certain players can do very well, and if those things are crucial for the team, than those players will have to play more minutes than they might otherwise play. The extra minutes might tend to reduce the player's Real Player Rating, while his total production will rise with the additional minutes. So to fairly and completely evaluate any player, you must always look at both the Real Player Rating (RPR) and the Real Player Production (RPP).
Furthermore, it is strongly suspected that, in order to compete in the playoffs, a team must have as many players of as high a quality (RPR) as possible, while at the same time having at least one or two players whose actual production is among the highest in the NBA regardless of exactly how high the RPRs happen to be. (All high RPP players will be relatively high RPR players; some will be higher than others.) Specifically for example, LeBron James' actual massive amount of production is most likely just as important to the Cleveland Cavaliers as is his RPR or, in other words, as is his rate of production. Similarly, Kobe Bryant's quantity is probably at least as important to the Lakers as is his quality.
Whereas, teams such as the Denver Nuggets, who have instructed a possible huge producer, Carmelo Anthony, to "not worry about scoring," may have made a fatal mistake relative to the playoffs, because teams with no extremely high rate producers may be generally doomed to lose quickly in the playoffs even if they have an unusually large number of high quality players as shown by RPR. This is because extremely high RPP players can by themselves "dominate a game" to some extent, meaning they can by themselves possibly win the game for their team, without worrying about complications that come in to play if you need to coordinate several high RPR but ultimately and theoretically limited RPP players.
Players who over the course of a season appear to rank higher in RPR (quality) but lower in RPP (quantity) may not be getting enough playing time. Players who over the course of a season appear to rank lower in RPR (quality) but higher in RPP (quantity) may be getting too much playing time. But as alluded to earlier, you must not automatically conclude this, because some skills are needed out on the court most of the time, but yet may be available only from a small number players on the roster. Such players may have to get more playing time due to that critical skill in short supply, even if their overall quality does not seem to justify all of that playing time.
A relatively common reason for unusual playing time will be players who are either truly outstanding defenders (who get extra playing time) or truly bad defenders (who get their playing time reduced).
Another common reason for extra playing time will be if a team has a point guard who has many more turnovers than the average point guard has. Because the point guard is so important, a good coach has to play his best guard who can make plays at the position for a full set of minutes every game, pretty much regardless of how many turnovers that player makes. If you take out your designated point guard due to "too many turnovers," it may end up sort of like cutting your foot off because you have a bad case of athletes foot!
MINIMUM PLAYING TIME RULES
Only players who played at least 10% of the minutes of whoever has played the most minutes on the team are included in these reports. Any player who has played for less than 10% of the minutes of the player who has played the most minutes is not included, since he didn't play for long enough to be fairly or reasonably compared with the other players. Furthermore, as described previously in the adjustment for defending section, only players who have played at least 300 minutes can have a defensive rating, or an overall RPR given to him. Both the 10% and the 300 minutes minimums must be met for a player to be rated.
REAL PLAYER PRODUCTION
Of course, looking at actual production (everything positive added together and everything negative subtracted out) is something that is extremely important too. The total production (everything good and everything bad combined together) is simply called Real Player Production or RPP.
There is no methodology for including defending (other than rebounding, steals, blocks, and personal fouls) in RPP at this time.
SOURCE OF TRACKED BASKETBALL COUNTS
The sources for the raw counts of scores, rebounds, steals, turnovers, and so forth is ESPN.com and NBA.com.
THE FORMULA
For 2008-09, the RPR formula has been very carefully and accurately tweaked again and is set to be as follows:
POSITIVE FACTORS
Points 1.00 (at par)
Number of 3-Pt FGs Made 1.00
Number of 2-Pt FGs Made 0.60
Number of FTs Made 0.00
Assists 1.75
Offensive Rebounds 1.15
Defensive Rebounds 1.25
Blocks 1.60
Steals 2.15
NEGATIVE FACTORS
3-Pt FGs Missed -1.00
2-Pt FGs Missed -0.85
FTs Missed -0.85
Turnovers -2.00
Personal Fouls -0.80
DEFENDING RATING
A quality of defending rating of between 0 and .230 is added to "Base or unadjusted RPR". In most cases, the defending rating is between 0.050 and .150. See the User Guide for the Defending Components" below for a very detailed explanation of how we determine how to defensively rate the players.
ACTUAL COMBINED AWARD OR PENALTY BY TYPE OF SHOT
3-Pointer Made 4.00
2-Pointer Made 2.60
Free Throw Made 1.00
3-Pointer Missed -1.00
2-Pointer Missed -0.85
Free Throw Missed -0.85
ZERO POINTS: PERCENTAGES BELOW WHICH THERE IS A NEGATIVE NET RESULT
3-Pointer 0 score % 0.200
2-Pointer 0 score % 0.246
1-Pointer 0 score % 0.459
This means that if a player has a lower percentage than any of the three above, then his RPR would be lower rather than higher as a result of his shooting that type of shot.
ASSISTS VERSUS TURNOVERS ZERO POINT
Assist/Turnover Ratio That Yields 0 Net Points: 1.143
This means that any player who has an assist/turnover ration of less than 1.143 is losing RPR rating when assists and turnovers are considered. He would have to either increase assisting or reduce turnovers to turn the combined effect from assists and turnovers positive.
QUALITY (RPR) AND QUANTITY (RPP} SUMMARIZED ONE LAST TIME
RPR reports show for each player the RPR (Real Player Rating) which tells you how good a player did (all the good things minus all the bad things) out on the court per unit of time. The RPP (Real Player Production) report tells you how much in total (the sum of the of the good things minus the sum of the bad things) a player did out on the court, without regard to playing time.
Many and maybe most sports watchers and an unknown but probably disturbingly large number of sports managers make the mistakes of exaggerating the importance of quantity and overlooking to some extent quality. These reports allow you to expand your horizons. These reports put quantity and quality side by side, which is extremely valuable, because both are roughly equally important in explaining accurately why and how the team is playing the way it is.
======== DEFENDING AND OFFENSIVE SUB RATINGS ======================
THE DEFENDING ADJUSTMENT TO REAL PLAYER RATINGS AND THE DEFENDING SUB RATING
THE DEFENDING COMPONENTS OR SUB RATINGS OF REAL PLAYER RATINGS--NEW AS OF JANUARY 2009
As of January 8, 2009, The Quest is proud to announce to you that the second major improvement to Real Player Ratings (RPR) in less than half a year is now fully up and running. The first major improvement were some needed changes in the factors used for RPR. The second major improvement (series of improvements, actually) is so far as I am aware the first ever effort to rate the defensive efforts of players that are hidden unless you watch all that player's games, because they are not scored or tracked by scorekeepers.
I have been talking about and working for and expecting the breakthrough in evaluation of defending for almost two years. Now that the breakthrough has come, I am now even more certain that RPR is the best overall rating system in existence, and that it is now roughly as good as it will ever or can ever be.
I recently developed a statistically valid way to rate the defending of players, that is, what they do to prevent scores other than rebounding, blocks, steals, and fouls, which were always included in RPR. This would include man to man defending, zone defending, pick and roll defending, defensive recognition, and defensive rotation.
Although the technique used had to be indirect and subject to a very small amount of statistical error, it validly awards the better defenders with bigger RPR bonuses. It has been validated by comparing results obtained with the defensive ratings shown on three different "advanced basketball statistics" web sites. Our results were shown to be extremely highly correlated with the results shown on the other sites. Where there are small differences, I believe mine are better, if only because mine uses simple, bedrock statistical theory rather than involved formulas.
HIDDEN DEFENDING
Before revealing what we do to reveal it, let's define "hidden defending." Exactly what is hidden defending? It's defending not tracked by the NBA. It's every action that helps to prevent the other team from scoring other than rebounding, stealing, and blocking, and fouling. So it would include man to man defending, zone defending, rotating in general, defensive recognition, and quick defensive response to various offensive tactics, such as pick and rolls. Obviously, if a defender is good at these things, the other team doesn't score as many points than if the defender is lousy at these things.
HOW TO REVEAL HIDDEN DEFENDING IN FOUR STEPS
STEP ONE: CALCUATION OF RAW HIDDEN DEFENDING RATINGS
Unlike most "advanced statistics" that are published on the internet or in print, we give you all the details about how we do ours, so that you can evaluate the evaluations, so to speak. The following is specifically what we are doing to be able to accurately and fairly compare players' defending:
Where do we start to discover what is hidden? We keep it as simple and yet as accurate as possible. We use the most official and therefore presumably the most reliable data as the building blocks for rating the defense of NBA players. We start with the player minutes and points scored by the other team while the player was on the court that are shown in the plus/minus statistical section at NBA.com.
There are no value judgments made regarding a player's defending style, or regarding a team's defending style for that matter. We don't care about style. Using points allowed per minute is looking at results, nothing more and nothing less.
After simply dividing points allowed by minutes on the court, we adjust (we standardize, to be more precise) that rate for the pace of the team and for the quality of the team's defense. The two adjustments are needed so that the ratings of players who are on different teams can be fairly compared.
Players who are on teams with faster paces give up more points per minute through no fault of their own. Similarly, players who are on teams with less efficient defenses give up more points per minute, regardless of how well they defend, everything else held constant. You could not fairly compare players on two or more teams with different paces and different team defense qualities unless you standardized, or in other words controlled for those differences for all NBA players.
USE OF BASIC STATISTICAL SAMPLING THEORY
What we are doing is using an indirect and inexact yet accurate and statistically valid way to discover who the better defenders are. No two players are out on the court for all the exact same minutes. So although for every player, what the other players out on the court do defensively while they are out on the court is a very large factor determining what that player's points per minute allowed will be, when you look at many, many hundreds of minutes, what the individual player does, or does not do defensively, as the case may be, will eventually show up in that particular player's points allowed per minute statistic.
In other words, what any individual player does defensively has to sooner or later show itself in a differentiation from other players of his points allowed per minute. As the number of minutes rise above 500, and then 1,000 and then, for many players, above 2,000 and even 3,000 for a regular season, what a particular player does or does not do defensively becomes more and more exactly shown by the points allowed per minute number. This is very basic statistical sampling theory in operation. Statistical sampling theory is the easy to understand bedrock theory of statistics.
Due to the necessity of a large sample of minutes, we will not do defending estimates for any player who has played for fewer than 300 minutes. Quality of defending estimates will be slightly less accurate for players who have only played between 301 and about 600 minutes than they will be for players who have played for more than 600 minutes. We believe that the estimates are going to be extremely accurate for all players who have played 750 minutes or more. The idea is relatively simple: as the number of hundreds of minutes played goes up, the accuracy of this system improves, to the point where it gives you the same information you would have if you knew exactly how many possessions of the other team each player ruined with his defending.
For your information, all players allow between 1.85 and 2.18 points per minute; most allow between 1.94 and 2.11. The overall NBA average is about 2.03 points per minute allowed.
STEP TWO: CONVERSION OF RAW HIDDEN DEFENDING POINTS ALLOWED PER MINUTE TO FILTERED HIDDEN DEFENDING POINTS ALLOWED PER MINUTE
Since different players have different breakdowns between how much of their defending shows up in tracked statistics such as defensive rebounding and how much of it does not, in order to improve accuracy we need to have a method to filter, or in other words, separate, the two categories of defending. If we didn't do this, we would still have a useful statistic, but it would be biased in favor of players whose defending is counted in tracked statistics more so than other players. There would be in effect some double counting of defending for players who have most of their quality defending tracked by scored statistics.
The filter used is to multiply the raw hidden defending ratings by the percentage of the real player production that is offensive. In other words we take the inverse of the percentage of a player's real player production that is defensive and multiply the raw hidden defending ratings by that. The rationale to do this is that although the exact relation is unknowable, we know that for a given raw hidden defending performance level, there will be an inverse relation between scored defending and hidden defending. The more defensive rebounds, steals, and blocks a player is making for any raw level, the less he is relying on hidden defending to achieve the raw level. And vice versa. So multilying by the inverse of the percentage of all contributions that are defensive (in other words, multiplying by offensive contributions) filters out much of the bias that is in the raw hidden defending rating.
To be even more specific, we first extract out defensive rebounding, steals, blocks, and personal fouls, the sum total of which is called "Scored Defensive Contribution". All of the other components combined constitute "Scored Offensive Contribution". Now we can determine the percentages of the RPP that are offensive and defensive, and then we can use the offensive percentages to convert the raw hidden defending ratings to filtered hidden defending ratings.
STEP THREE: CONVERSION OF FILTERED ALLOWED POINTS PER MINUTE TO FILTERED HIDDEN DEFENDING RATING
We need to translate the adjusted or filtered points allowed per minute into numerical terms that are the most useful with respect to RPR. So with a very carefully designed translation scale, we amplify the very small differences in different player's points allowed per minute numbers into much larger different hidden defending ratings for each player. Then we simply add the hidden defending rating to the Base RPR to yield RPR.
STEP FOUR: USE OF HIDDEN DEFENDING RATING
We now have added in a reasonably good estimate of the value of actions of players that are not even kept track of by scorekeepers! The filtered hidden defending ratings are added to the Base, Unadjusted, or Scored RPR to give RPR. Technically, you could call the final result "Ajusted RPR," but we are trying to avoid that terminology because of how important we think it is to include the hidden defending in the performance measure.
SIZE OF THE DEFENDING ADJUSTMENTS
Base RPR's for most NBA players range between .400 and 1.000. The range of possible defending adjustments to the base RPRs is from 0 to about .230. In most cases, however, the adjustment will be between 0.020 and .170.
THE DEFENDING SUB RATING: PUTTING THE HIDDEN AND THE UNHIDDEN TOGETHER
Aside from the Hidden Defending Rating we can find out how well each player does in terms of unhidden or scored defending, can't we? Of course se can.
Aside from the hidden there is of course unhidden defending, which would be rebounding plus steals plus blocks minus personal fouls. If we extract the combination of those four out of the same counts that underlie the RPR as a whole, we get what we are going to call the Scored Defending Contribution. This could also be thought of as Tracked Defending Contribution if you prefer. Then if we divide this by minutes, we can have a Scored (or Tracked) Defending Rating.
Finally, if we combine Hidden Defending Rating (HDR) with Scored Defending Rating (SDR) we can have an Overall Defending Rating (ODR). I am for now going to simply multiply the HDR by two and add that to the SDR to yield the ODR. To combine them this way is more arbitrary than my usual standards allow; I am doing this because there is as of yet no non-arbitrary way of doing it. The formula of two times HDR plus SDR brings HDR almost up to par with SDR in terms of the actual numbers and the averages of those numbers involved.
In other words, I am saying for now that hidden defending is almost as important as scored defending. There appear to be many coaches and not a few hardcore basketball fans who believe that hidden defending is actually more important than scored defending, but I am very likely never going to agree with that. I think that although hidden defending is important, and plausibly almost as important as tracked defending, that it is like a quicksand in that there seems to be a tendency for a substantial minority of basketball people to get carried away with estimating the importance of it and then become more and more trapped by their error in terms of how they look at basketball or in terms of how they coach their team if they are coaching.
THE OFFENSIVE SUB RATING
The Offensive Sub Rating is all tracked actions other than the defensive ones (defensive rebounding, steals, blocks, and personal fouls) combined together using the RPR weights, divided by minutes. In other words, it is Total Offensive Production divided by minutes. For the list of all tracked actions and the weight factors assigned to each, see the secion titled "The Formula" above.
======== SUMMARY OF PRIMARY FORMULAS =================
Real Player Production or RPR = (All tracked or scored actions weighted according to best available analysis of importance / minutes) + Filtered Hidden Defending Adjustment
Real Player Production or RPP = Total Offensive Production + Total Defensive Production. (All tracked or scored actions weighted according to best available analysis of importance.)
Offensive Sub Rating = Total Scored or Tracked Offensive Production / Minutes
Defensive Sub Rating = (Total Scored or Tracked Defensive Production / Minutes) + 2 X Filtered Hidden Defending Adjustment
Filtered Hidden Defending Adjustment = Raw Hidden Defending Adjustment X Percentate of RPP that is Offensive
Raw Hidden Defending Adjustment = Assigned value based on chart. the objective of which, is to amplify seemingly minute differences in points allowed per minute.
REAL PLAYER RATINGS BY TEAM USER GUIDE
Updated Feb. 25, 2009
INTRODUCTION TO THE CONCEPT OF REAL PLAYER RATINGS
The Real Player Rating (RPR) is a very carefully constructed all inclusive performance measure. Everything of value that a basketball player can do is recorded by official NBA scorekeepers who sit right along the edge of the court, mid-court, and who are trained to observe and record everything that happens in a game.
Since these days all of these counts are immediately input into continually updated public data bases online, such as at ESPN, it is theoretically possible to combine everything together into an overall performance measure for each player. This is what the RPR does.
Real Player Rating or RPR is everything tracked by scorekeepers that a player does, good and bad, added and subtracted (with negative things such as turnovers and missed shots being subtracted). Very carefully calibrated factors, or weights, are applied to the different elements. The calibration, as you would expect, is done to reflect the different value toward winning games that different actions on the court have. All of the good and bad combined together is divided by minutes, so we can tell the rate, which we need to determine the overall quality or value of the player.
REAL PLAYER RATINGS ARE ADJUSTED FOR DEFENDING NOT TRACKED BY SCOREKEEPERS STARTING IN 2009
Not counting purely subjective and abstract factors such as leadership, and not counting a few infrequently occurring actions on the court (not being counted or tracked by anyone yet) such as chasing down loose balls, the only thing a basketball player can regularly do on the court of any value that is not counted by NBA scorekeepers is preventing what would have been a score from being a score by defending against the shot or shots during a possession well enough to stop what would have been a score by the opposing team. In other words, what the player does to make the possessions of the opposing teams worthless other than what is already counted, which would be rebounds, steals, blocks, and personal fouls. These untracked or hidden actions would include effective man to man defending, effective rotation on defense off screens and picks, defensive recognition, and quickness of defensive reaction. These things would be counted by scorekeepers if it were possible. But, for example, there is no way to know exactly how many shots a good (or any kind of) defender has changed from being a score to a miss.
Quest for the Ring has developed a statistically valid way to accurately estimate the untracked or hidden aspects of defending. This is described in complete detail in the latter sections of this Guide.
SIMPLICITY, RELIABILITY, TRANSPARENCY, AND FOCUS ONLY ON "WINNING POWER"
Like everything statistical we do at Quest, we have kept this process as simple and reliable as possible, while at the same time spending as much time as necessary on design, quality control and performance evaluation. Unlike some other practitioners, we avoid what you might call layered complexity, which leads to formulas which can not be understood without studying them and which high traffic sites will not show on any of their web pages for fear that the public will rebel against the statistic. At Quest, we think that our rating systems can be understood and evaluated by most high school graduates, and we keep everything out in the open through User Guides such as this one.
Basketball statistical gurus frequently forget that no matter how intricate their formulas are, they are very heavily manipulating process items such as assists and rebounds while spending very little time on how these things fit together to produce wins and losses. We think that they are making the mistake, whether or not they are aware, of injecting value adjustments regarding how they think the game should be played and value adjustments about which playing styles are better than others.
Whereas, the primary objectives of the relative simplicity (small number of formulas, to be precise) of the Quest RPR is to avoid all how the game should be played and how players should play value judgments. We don't care about the styles, only about the results. The RPR is concerned first and foremost with the impact each player has on the potential for winning games.
Quest thinks it makes more sense to minimize the manipulation of process items, and to focus much more on coming up with the best possible estimation of how the process items impact points for and points against in games, which in turn of course determines wins. Whereas other "advanced statistics" might give you more depth and flavor regarding how a particular player plays, the Quest RPR is a way for the reader to, in a very quick and easy way, determine what the overall value of the player is with respect to producing wins or losses.
In other words, the foundation of RPR is and will always be measurement of a player's power to help win basketball games, whereas the foundation of other, more complicated statistics may include preferences about how the game should be played and about the style of players, with winning power measured less accurately as a result of those focuses.
THE MAIN REASON REAL PLAYER RATING IS SO VALUABLE
Because it is per time, RPR is the best possible measure of the net quality of a basketball player, or simply "how good" the player is (on average) for each minute of playing time.
REAL PLAYER RATING REPORTS CAN BE FOR THE WHOLE NBA, FOR A TEAM, OR FOR A GAME
With a Real Player Ratings Report for the entire NBA, you can see very rapidly who the best players in the NBA have been during the course of the season.
With a Real Player Ratings Report for a Team for the Regular Season, you can see very rapidly who the best players on the team have been during the course of the season. You can use this information to investigate the possiblity that the coach is not perfect. Well, we know that no coach is perfect. So really, with the benefit of 20/20 hindsight, we can investigate and determine what mistakes the coach has apparently made with regard to rotations and playing times. Furthermore, by using the Ratings, basketball knowledge, a little creativity, and logical deduction, we can also investigate and perhaps determine whether the coach has made incorrect decisions regarding which strategies and plays are best for his team's offense and defense.
Real Player Ratings for games are the most important component of Reports called Ultimate Game Breakdown: Players Reports.
CAUTIONS
To be completely honest and clear, although it is the best possible overall real life measure, RPR is still not a perfect or absolute, "final word" measure on any player. In general, you must remember that all performance measures including this one for the NBA are relative rather than absolute measures. The ratings are relative to the team context. Players do not exist in a vacuum, especially in basketball.
Several specific cautions will now be described.
Because basketball is a team game, and more so than most other sports, players who are on really good teams might have their own performances "crowded out" to some extent by even better players. So, paradoxically, ratings of good players on good teams will generally have slightly lower ratings than they would have if they were on a bad team. Conversely, great players on bad teams will have slightly higher ratings than they would have if they were on a good team.
Players need not only playing time but possession of the ball in order to produce many of the things that count in the rating. So if, for whatever reason, a player does not get the ball as often as he would on a different team, or with a different coach, or with whatever other circumstances you can dream of, then his RPR will be lower than what it could or would be.
If a good player is on a good team where there are players even better than he is, than his RPR will likely be lower than it would be if he were on a not as good team.
If a good player plays a certain position for which his team has an even better player, then it's probable that the better player will crowd out the lesser player to one extent or another, so that the lesser player's RPR will be lower than what it would be if he were the best player at the position on the team.
The ratings are only for the current season. It has recently been discovered that many player's ratings often change up or down by more than 10% from year to year, and by much greater amounts over many years.
Those who think defense in basketball is much more important than offense may consider the magnitude of the defensive adjustment to be inadequate. They will contend that defensive specialists who are poor offensive players should have a higher rating. While we realized that we needed to adjust the ratings for defending not tracked by NBA scorekeepers, we continue to believe that players who are great defensive specialists but poor or undeveloped offensive players should in most cases rank no higher than the major role player level.
Do not forget that RPR is a per time measure. RPP and not RPR measures total impact of a player. RPR measures how valuable a player has been toward winning basketball games, per unit of time.
The classification scheme, like the ratings, is relative. A role player on a bad team might be a solid starter on a very good team. A star on a bad team might be just a major role player on a really good team. And so on and so forth. A player is a star, a role player, or whatever only in the contexts of the particular season and the particular team involved. If he was on a different team, or if it was a different year, his classification might be different.
So in conclusion, don't think of RPR as the ultimate gospel or bible on how good players are. But do think of it as an extremely accurate and reliable summary of how good the players actually have been in real life in the specific time (season or playoffs) and place (team) involved.
A NOTE ABOUT REAL PLAYER RATINGS FOR INDIVIDUAL GAMES
However, not as many breakdowns of individual game ratings are going to closely track the overall average for the roster as you might think. This is because one of the interesting things about basketball that makes it different from most other sports is that "how good" a player is from game to game varies radically. The best players have terrible games where they do almost nothing sometimes, while players who normally do not do much can every once in a while have outstanding games, at least if you measure it per minute on the court anyway. If you just looked at actual production, and never at a reserve player's Real Player Rating, you would hardly notice any of his unusually outstanding games, since players who normally do not do much will normally not have much playing time.
INTERACTIONS BETWEEN PLAYING TIMES, PLAYER RATINGS, AND THE NEEDS OF TEAMS
There are certain things that only certain players can do very well, and if those things are crucial for the team, than those players will have to play more minutes than they might otherwise play. The extra minutes might tend to reduce the player's Real Player Rating, while his total production will rise with the additional minutes. So to fairly and completely evaluate any player, you must always look at both the Real Player Rating (RPR) and the Real Player Production (RPP).
Furthermore, it is strongly suspected that, in order to compete in the playoffs, a team must have as many players of as high a quality (RPR) as possible, while at the same time having at least one or two players whose actual production is among the highest in the NBA regardless of exactly how high the RPRs happen to be. (All high RPP players will be relatively high RPR players; some will be higher than others.) Specifically for example, LeBron James' actual massive amount of production is most likely just as important to the Cleveland Cavaliers as is his RPR or, in other words, as is his rate of production. Similarly, Kobe Bryant's quantity is probably at least as important to the Lakers as is his quality.
Whereas, teams such as the Denver Nuggets, who have instructed a possible huge producer, Carmelo Anthony, to "not worry about scoring," may have made a fatal mistake relative to the playoffs, because teams with no extremely high rate producers may be generally doomed to lose quickly in the playoffs even if they have an unusually large number of high quality players as shown by RPR. This is because extremely high RPP players can by themselves "dominate a game" to some extent, meaning they can by themselves possibly win the game for their team, without worrying about complications that come in to play if you need to coordinate several high RPR but ultimately and theoretically limited RPP players.
Players who over the course of a season appear to rank higher in RPR (quality) but lower in RPP (quantity) may not be getting enough playing time. Players who over the course of a season appear to rank lower in RPR (quality) but higher in RPP (quantity) may be getting too much playing time. But as alluded to earlier, you must not automatically conclude this, because some skills are needed out on the court most of the time, but yet may be available only from a small number players on the roster. Such players may have to get more playing time due to that critical skill in short supply, even if their overall quality does not seem to justify all of that playing time.
A relatively common reason for unusual playing time will be players who are either truly outstanding defenders (who get extra playing time) or truly bad defenders (who get their playing time reduced).
Another common reason for extra playing time will be if a team has a point guard who has many more turnovers than the average point guard has. Because the point guard is so important, a good coach has to play his best guard who can make plays at the position for a full set of minutes every game, pretty much regardless of how many turnovers that player makes. If you take out your designated point guard due to "too many turnovers," it may end up sort of like cutting your foot off because you have a bad case of athletes foot!
MINIMUM PLAYING TIME RULES
Only players who played at least 10% of the minutes of whoever has played the most minutes on the team are included in these reports. Any player who has played for less than 10% of the minutes of the player who has played the most minutes is not included, since he didn't play for long enough to be fairly or reasonably compared with the other players. Furthermore, as described previously in the adjustment for defending section, only players who have played at least 300 minutes can have a defensive rating, or an overall RPR given to him. Both the 10% and the 300 minutes minimums must be met for a player to be rated.
REAL PLAYER PRODUCTION
Of course, looking at actual production (everything positive added together and everything negative subtracted out) is something that is extremely important too. The total production (everything good and everything bad combined together) is simply called Real Player Production or RPP.
There is no methodology for including defending (other than rebounding, steals, blocks, and personal fouls) in RPP at this time.
SOURCE OF TRACKED BASKETBALL COUNTS
The sources for the raw counts of scores, rebounds, steals, turnovers, and so forth is ESPN.com and NBA.com.
THE FORMULA
For 2008-09, the RPR formula has been very carefully and accurately tweaked again and is set to be as follows:
POSITIVE FACTORS
Points 1.00 (at par)
Number of 3-Pt FGs Made 1.00
Number of 2-Pt FGs Made 0.60
Number of FTs Made 0.00
Assists 1.75
Offensive Rebounds 1.15
Defensive Rebounds 1.25
Blocks 1.60
Steals 2.15
NEGATIVE FACTORS
3-Pt FGs Missed -1.00
2-Pt FGs Missed -0.85
FTs Missed -0.85
Turnovers -2.00
Personal Fouls -0.80
DEFENDING RATING
A quality of defending rating of between 0 and .230 is added to "Base or unadjusted RPR". In most cases, the defending rating is between 0.050 and .150. See the User Guide for the Defending Components" below for a very detailed explanation of how we determine how to defensively rate the players.
ACTUAL COMBINED AWARD OR PENALTY BY TYPE OF SHOT
3-Pointer Made 4.00
2-Pointer Made 2.60
Free Throw Made 1.00
3-Pointer Missed -1.00
2-Pointer Missed -0.85
Free Throw Missed -0.85
ZERO POINTS: PERCENTAGES BELOW WHICH THERE IS A NEGATIVE NET RESULT
3-Pointer 0 score % 0.200
2-Pointer 0 score % 0.246
1-Pointer 0 score % 0.459
This means that if a player has a lower percentage than any of the three above, then his RPR would be lower rather than higher as a result of his shooting that type of shot.
ASSISTS VERSUS TURNOVERS ZERO POINT
Assist/Turnover Ratio That Yields 0 Net Points: 1.143
This means that any player who has an assist/turnover ration of less than 1.143 is losing RPR rating when assists and turnovers are considered. He would have to either increase assisting or reduce turnovers to turn the combined effect from assists and turnovers positive.
QUALITY (RPR) AND QUANTITY (RPP} SUMMARIZED ONE LAST TIME
RPR reports show for each player the RPR (Real Player Rating) which tells you how good a player did (all the good things minus all the bad things) out on the court per unit of time. The RPP (Real Player Production) report tells you how much in total (the sum of the of the good things minus the sum of the bad things) a player did out on the court, without regard to playing time.
Many and maybe most sports watchers and an unknown but probably disturbingly large number of sports managers make the mistakes of exaggerating the importance of quantity and overlooking to some extent quality. These reports allow you to expand your horizons. These reports put quantity and quality side by side, which is extremely valuable, because both are roughly equally important in explaining accurately why and how the team is playing the way it is.
======== DEFENDING AND OFFENSIVE SUB RATINGS ======================
THE DEFENDING ADJUSTMENT TO REAL PLAYER RATINGS AND THE DEFENDING SUB RATING
THE DEFENDING COMPONENTS OR SUB RATINGS OF REAL PLAYER RATINGS--NEW AS OF JANUARY 2009
As of January 8, 2009, The Quest is proud to announce to you that the second major improvement to Real Player Ratings (RPR) in less than half a year is now fully up and running. The first major improvement were some needed changes in the factors used for RPR. The second major improvement (series of improvements, actually) is so far as I am aware the first ever effort to rate the defensive efforts of players that are hidden unless you watch all that player's games, because they are not scored or tracked by scorekeepers.
I have been talking about and working for and expecting the breakthrough in evaluation of defending for almost two years. Now that the breakthrough has come, I am now even more certain that RPR is the best overall rating system in existence, and that it is now roughly as good as it will ever or can ever be.
I recently developed a statistically valid way to rate the defending of players, that is, what they do to prevent scores other than rebounding, blocks, steals, and fouls, which were always included in RPR. This would include man to man defending, zone defending, pick and roll defending, defensive recognition, and defensive rotation.
Although the technique used had to be indirect and subject to a very small amount of statistical error, it validly awards the better defenders with bigger RPR bonuses. It has been validated by comparing results obtained with the defensive ratings shown on three different "advanced basketball statistics" web sites. Our results were shown to be extremely highly correlated with the results shown on the other sites. Where there are small differences, I believe mine are better, if only because mine uses simple, bedrock statistical theory rather than involved formulas.
HIDDEN DEFENDING
Before revealing what we do to reveal it, let's define "hidden defending." Exactly what is hidden defending? It's defending not tracked by the NBA. It's every action that helps to prevent the other team from scoring other than rebounding, stealing, and blocking, and fouling. So it would include man to man defending, zone defending, rotating in general, defensive recognition, and quick defensive response to various offensive tactics, such as pick and rolls. Obviously, if a defender is good at these things, the other team doesn't score as many points than if the defender is lousy at these things.
HOW TO REVEAL HIDDEN DEFENDING IN FOUR STEPS
STEP ONE: CALCUATION OF RAW HIDDEN DEFENDING RATINGS
Unlike most "advanced statistics" that are published on the internet or in print, we give you all the details about how we do ours, so that you can evaluate the evaluations, so to speak. The following is specifically what we are doing to be able to accurately and fairly compare players' defending:
Where do we start to discover what is hidden? We keep it as simple and yet as accurate as possible. We use the most official and therefore presumably the most reliable data as the building blocks for rating the defense of NBA players. We start with the player minutes and points scored by the other team while the player was on the court that are shown in the plus/minus statistical section at NBA.com.
There are no value judgments made regarding a player's defending style, or regarding a team's defending style for that matter. We don't care about style. Using points allowed per minute is looking at results, nothing more and nothing less.
After simply dividing points allowed by minutes on the court, we adjust (we standardize, to be more precise) that rate for the pace of the team and for the quality of the team's defense. The two adjustments are needed so that the ratings of players who are on different teams can be fairly compared.
Players who are on teams with faster paces give up more points per minute through no fault of their own. Similarly, players who are on teams with less efficient defenses give up more points per minute, regardless of how well they defend, everything else held constant. You could not fairly compare players on two or more teams with different paces and different team defense qualities unless you standardized, or in other words controlled for those differences for all NBA players.
USE OF BASIC STATISTICAL SAMPLING THEORY
What we are doing is using an indirect and inexact yet accurate and statistically valid way to discover who the better defenders are. No two players are out on the court for all the exact same minutes. So although for every player, what the other players out on the court do defensively while they are out on the court is a very large factor determining what that player's points per minute allowed will be, when you look at many, many hundreds of minutes, what the individual player does, or does not do defensively, as the case may be, will eventually show up in that particular player's points allowed per minute statistic.
In other words, what any individual player does defensively has to sooner or later show itself in a differentiation from other players of his points allowed per minute. As the number of minutes rise above 500, and then 1,000 and then, for many players, above 2,000 and even 3,000 for a regular season, what a particular player does or does not do defensively becomes more and more exactly shown by the points allowed per minute number. This is very basic statistical sampling theory in operation. Statistical sampling theory is the easy to understand bedrock theory of statistics.
Due to the necessity of a large sample of minutes, we will not do defending estimates for any player who has played for fewer than 300 minutes. Quality of defending estimates will be slightly less accurate for players who have only played between 301 and about 600 minutes than they will be for players who have played for more than 600 minutes. We believe that the estimates are going to be extremely accurate for all players who have played 750 minutes or more. The idea is relatively simple: as the number of hundreds of minutes played goes up, the accuracy of this system improves, to the point where it gives you the same information you would have if you knew exactly how many possessions of the other team each player ruined with his defending.
For your information, all players allow between 1.85 and 2.18 points per minute; most allow between 1.94 and 2.11. The overall NBA average is about 2.03 points per minute allowed.
STEP TWO: CONVERSION OF RAW HIDDEN DEFENDING POINTS ALLOWED PER MINUTE TO FILTERED HIDDEN DEFENDING POINTS ALLOWED PER MINUTE
Since different players have different breakdowns between how much of their defending shows up in tracked statistics such as defensive rebounding and how much of it does not, in order to improve accuracy we need to have a method to filter, or in other words, separate, the two categories of defending. If we didn't do this, we would still have a useful statistic, but it would be biased in favor of players whose defending is counted in tracked statistics more so than other players. There would be in effect some double counting of defending for players who have most of their quality defending tracked by scored statistics.
The filter used is to multiply the raw hidden defending ratings by the percentage of the real player production that is offensive. In other words we take the inverse of the percentage of a player's real player production that is defensive and multiply the raw hidden defending ratings by that. The rationale to do this is that although the exact relation is unknowable, we know that for a given raw hidden defending performance level, there will be an inverse relation between scored defending and hidden defending. The more defensive rebounds, steals, and blocks a player is making for any raw level, the less he is relying on hidden defending to achieve the raw level. And vice versa. So multilying by the inverse of the percentage of all contributions that are defensive (in other words, multiplying by offensive contributions) filters out much of the bias that is in the raw hidden defending rating.
To be even more specific, we first extract out defensive rebounding, steals, blocks, and personal fouls, the sum total of which is called "Scored Defensive Contribution". All of the other components combined constitute "Scored Offensive Contribution". Now we can determine the percentages of the RPP that are offensive and defensive, and then we can use the offensive percentages to convert the raw hidden defending ratings to filtered hidden defending ratings.
STEP THREE: CONVERSION OF FILTERED ALLOWED POINTS PER MINUTE TO FILTERED HIDDEN DEFENDING RATING
We need to translate the adjusted or filtered points allowed per minute into numerical terms that are the most useful with respect to RPR. So with a very carefully designed translation scale, we amplify the very small differences in different player's points allowed per minute numbers into much larger different hidden defending ratings for each player. Then we simply add the hidden defending rating to the Base RPR to yield RPR.
STEP FOUR: USE OF HIDDEN DEFENDING RATING
We now have added in a reasonably good estimate of the value of actions of players that are not even kept track of by scorekeepers! The filtered hidden defending ratings are added to the Base, Unadjusted, or Scored RPR to give RPR. Technically, you could call the final result "Ajusted RPR," but we are trying to avoid that terminology because of how important we think it is to include the hidden defending in the performance measure.
SIZE OF THE DEFENDING ADJUSTMENTS
Base RPR's for most NBA players range between .400 and 1.000. The range of possible defending adjustments to the base RPRs is from 0 to about .230. In most cases, however, the adjustment will be between 0.020 and .170.
THE DEFENDING SUB RATING: PUTTING THE HIDDEN AND THE UNHIDDEN TOGETHER
Aside from the Hidden Defending Rating we can find out how well each player does in terms of unhidden or scored defending, can't we? Of course se can.
Aside from the hidden there is of course unhidden defending, which would be rebounding plus steals plus blocks minus personal fouls. If we extract the combination of those four out of the same counts that underlie the RPR as a whole, we get what we are going to call the Scored Defending Contribution. This could also be thought of as Tracked Defending Contribution if you prefer. Then if we divide this by minutes, we can have a Scored (or Tracked) Defending Rating.
Finally, if we combine Hidden Defending Rating (HDR) with Scored Defending Rating (SDR) we can have an Overall Defending Rating (ODR). I am for now going to simply multiply the HDR by two and add that to the SDR to yield the ODR. To combine them this way is more arbitrary than my usual standards allow; I am doing this because there is as of yet no non-arbitrary way of doing it. The formula of two times HDR plus SDR brings HDR almost up to par with SDR in terms of the actual numbers and the averages of those numbers involved.
In other words, I am saying for now that hidden defending is almost as important as scored defending. There appear to be many coaches and not a few hardcore basketball fans who believe that hidden defending is actually more important than scored defending, but I am very likely never going to agree with that. I think that although hidden defending is important, and plausibly almost as important as tracked defending, that it is like a quicksand in that there seems to be a tendency for a substantial minority of basketball people to get carried away with estimating the importance of it and then become more and more trapped by their error in terms of how they look at basketball or in terms of how they coach their team if they are coaching.
THE OFFENSIVE SUB RATING
The Offensive Sub Rating is all tracked actions other than the defensive ones (defensive rebounding, steals, blocks, and personal fouls) combined together using the RPR weights, divided by minutes. In other words, it is Total Offensive Production divided by minutes. For the list of all tracked actions and the weight factors assigned to each, see the secion titled "The Formula" above.
======== SUMMARY OF PRIMARY FORMULAS =================
Real Player Production or RPR = (All tracked or scored actions weighted according to best available analysis of importance / minutes) + Filtered Hidden Defending Adjustment
Real Player Production or RPP = Total Offensive Production + Total Defensive Production. (All tracked or scored actions weighted according to best available analysis of importance.)
Offensive Sub Rating = Total Scored or Tracked Offensive Production / Minutes
Defensive Sub Rating = (Total Scored or Tracked Defensive Production / Minutes) + 2 X Filtered Hidden Defending Adjustment
Filtered Hidden Defending Adjustment = Raw Hidden Defending Adjustment X Percentate of RPP that is Offensive
Raw Hidden Defending Adjustment = Assigned value based on chart. the objective of which, is to amplify seemingly minute differences in points allowed per minute.
Monday, February 9, 2009
User Guide for the Defending Components or Sub Ratings of Real Player Ratings
THE DEFENDING COMPONENTS OR SUB RATINGS OF REAL PLAYER RATINGS--NEW AS OF JANUARY 2009
As of January 8, 2009, The Quest is proud to announce to you that the second major improvement to Real Player Ratings (RPR) in less than half a year is now fully up and running. The first major improvement were some needed changes in the factors used for RPR. The second major improvement (series of improvements, actually) is so far as I am aware the first ever effort to rate the defensive efforts of players that are hidden unless you watch all that player's games, because they are not scored or tracked by scorekeepers.
I have been talking about and working for and expecting the breakthrough in evaluation of defending for almost two years. Now that the breakthrough has come, I am now even more certain that RPR is the best overall rating system in existence, and that it is now roughly as good as it will ever or can ever be.
I recently developed a statistically valid way to rate the defending of players, that is, what they do to prevent scores other than rebounding, blocks, steals, and fouls, which were always included in RPR. This would include man to man defending, zone defending, pick and roll defending, defensive recognition, and defensive rotation.
Although the technique used had to be indirect and inexact, it validly awards the better defenders with bigger RPR bonuses. It has been validated by comparing results obtained with the defensive ratings shown on three different "advanced basketball statistics" web sites. Our results were shown to be extremely highly correlated with the results shown on the other sites. Where there are small differences, I believe mine are better, if only because mine uses simple, bedrock statistical theory rather than involved formulas.
HIDDEN DEFENDING
Before revealing what we do to reveal it, let's define "hidden defending." Exactly what is hidden defending? It's every action that helps to prevent the other team from scoring other than rebounding, stealing, and blocking. So it would include man to man defending, zone defending, rotating in general, defensive recognition, and quick defensive response to various offensive tactics, such as pick and rolls. Obviously, if a defender is good at these things, the other team doesn't score as many points than if the defender is lousy at these things.
HOW TO REVEAL HIDDEN DEFENDING IN FIVE STEPS
STEP ONE: CALCUATION OF RAW HIDDEN DEFENDING RATINGS
Unlike most "advanced statistics" that are published on the internet or in print, we give you all the details about how we do ours, so that you can evaluate the evaluations, so to speak. The following is specifically what we are doing to be able to accurately and fairly compare players' defending:
Where do we start to discover what is hidden? We keep it as simple and yet as accurate as possible. We use the most official and therefore presumably the most reliable data as the building blocks for rating the defense of NBA players. We start with the player minutes and points scored by the other team while the player was on the court that are shown in the plus/minus statistical section at NBA.com.
After simply dividing points allowed by minutes on the court, we adjust that rate for the pace of the team and for the quality of the team's defense. The two adjustments are needed so that the ratings of players who are on different teams can be fairly compared.
Players who are on teams with faster paces give up more points per minute through no fault of their own. Similarly, players who are on teams with less efficient defenses give up more points per minute, everything else held constant. You could not fairly compare players on two or more teams with different paces and different team defense qualities unless you standardized, or in other words controlled for those differences for all NBA players.
USE OF BASIC STATISTICAL SAMPLING THEORY
What we are doing is using an indirect and inexact yet accurate and statistically valid way to discover who the better defenders are. No two players are out on the court for all the exact same minutes. So although for every player, what the other players out on the court do defensively while they are out on the court is a very large factor determining what that player's points per minute allowed will be, when you look at many, many hundreds of minutes, what the individual player does, or does not do defensively, as the case may be, will eventually show up in that particular player's points allowed per minute statistic.
In other words, what any individual player does defensively has to sooner or later show itself in the points allowed per minute. As the number of minutes rise above 500, and then 1,000 and then, for many players, above 2,000 and even 3,000 for a regular season, what a particular player does or does not do defensively becomes more and more exactly shown by the points allowed per minute number. This is very basic statistical sampling theory in operation. Statistical sampling theory is the easy to understand bedrock theory of statistics.
Due to the necessity of a large sample of minutes, we will not do defending estimates for any player who has played for fewer than 300 minutes. Quality of defending estimates will be slightly less accurate for players who have only played between 301 and about 600 minutes than they will be for players who have played for more than 600 minutes. We believe that the estimates are going to be extremely accurate for all players who have played 750 minutes or more. The idea is relatively simple: as the number of hundreds of minutes played goes up, the accuracy of this system improves, to the point where it gives you the same information you would have if you knew exactly how many possessions of the other team each player ruined with his defending.
For your information, all players allow between 1.87 and 2.16 points per minute; most allow between 1.94 and 2.11. The overall NBA average is about 2.03 points per minute allowed.
STEP TWO: CONVERSION OF RAW HIDDEN DEFENDING POINTS ALLOWED PER MINUTE TO FILTERED HIDDEN DEFENDING POINTS ALLOWED PER MINUTE
Since different players have different breakdowns between how much of their defending shows up in tracked statistics such as defensive rebounding and how much of it does not, in order to improve accuracy we need to have a method to filter, or in other words, separate, the two categories of defending. If we didn't do this, we would still have a useful statistic, but it would be biased in favor of players whose defending is counted in tracked statistics more so than other players. There would be in effect some double counting of defending for players who have most of their quality defending tracked by scored statistics.
The filter used is to multiply the raw hidden defending ratings by the percentage of the real player production that is offensive. In other words we take the inverse of the percentage of a player's real player production that is defensive and multiply the raw hidden defending ratings by that. The rationale to do this is that although the exact relation is unknowable, we know that for a given raw hidden defending performance level, there will be an inverse relation between scored defending and hidden defending. The more defensive rebounds, steals, and blocks a player is making for any raw level, the less he is relying on hidden defending to achieve the raw level. And vice versa. So multilying by the inverse of the percentage of all contributions that are defensive (in other words, multiplying by offensive contributions) filters out much of the bias that is in the raw hidden defending rating.
To be even more specific, we first extract out defensive rebounding, steals, blocks, and personal fouls, the sum total of which is called "Scored Defensive Contribution". All of the other components combined constitute "Scored Offensive Contribution". Now we can determine the percentages of the RPP that are offensive and defensive, and then we can use the offensive percentages to convert the raw hidden defending ratings to filtered hidden defending ratings.
STEP THREE: CONVERSION OF FILTERED ALLOWED POINTS PER MINUTE TO FILTERED HIDDEN DEFENDING RATING
We need to translate the adjusted or filtered points allowed per minute into numerical terms that are the most useful with respect to RPR. So with a very carefully designed translation scale, we amplify the very small differences in different player's points allowed per minute numbers into much larger different hidden defending ratings for each player. Then we simply add the hidden defending rating to the Base RPR to yield RPR.
STEP FOUR: USE OF HIDDEN DEFENDING RATING
We now have added in a reasonably good estimate of the value of actions of players that are not even kept track of by scorekeepers! The filtered hidden defending ratings are added to the "Base or Scored RPR" to give RPP. The range of possible defending adjustments to the base RPR is from 0 to about .230. In most cases, however, the adjustment will be between 0.030 and .150.
STEP FIVE: OVERALL EVALUATION OF DEFENDING
Aside from the Hidden Defending Rating we can find out how well each player does in terms of unhidden or scored defending, can't we? Of course se can.
Aside from the hidden there is of course unhidden defending, which would be rebounding plus steals plus blocks minus personal fouls. If we extract the combination of those four out of the same counts that underlie the RPR as a whole, we get what we are going to call the Scored Defending Contribution. This could also be thought of as Tracked Defending Contribution if you prefer. Then if we divide this by minutes, we can have a Scored (or Tracked) Defending Rating.
Finally, if we combine Hidden Defending Rating (HDR) with Scored Defending Rating (SDR) we can have an Overall Defending Rating (ODR). I am for now going to simply multiply the HDR by two and add that to the SDR to yield the ODR. To combine them this way is more arbitrary than my usual standards allow; I am doing this because there is as of yet no non-arbitrary way of doing it. The formula of two times HDR plus SDR brings HDR almost up to par with SDR in terms of the actual numbers and the averages of those numbers involved.
In other words, I am saying for now that hidden defending is almost as important as scored defending. There appear to be many coaches and not a few hardcore basketball fans who believe that hidden defending is actually more important than scored defending, but I am very likely never going to agree with that. I think that although hidden defending is important, and plausibly almost as important as tracked defending, that it is like a quicksand in that there seems to be a tendency for a substantial minority of basketball people to get carried away with estimating the importance of it and then become more and more trapped by their error in terms of how they look at basketball or in terms of how they coach their team if they are coaching.
As of January 8, 2009, The Quest is proud to announce to you that the second major improvement to Real Player Ratings (RPR) in less than half a year is now fully up and running. The first major improvement were some needed changes in the factors used for RPR. The second major improvement (series of improvements, actually) is so far as I am aware the first ever effort to rate the defensive efforts of players that are hidden unless you watch all that player's games, because they are not scored or tracked by scorekeepers.
I have been talking about and working for and expecting the breakthrough in evaluation of defending for almost two years. Now that the breakthrough has come, I am now even more certain that RPR is the best overall rating system in existence, and that it is now roughly as good as it will ever or can ever be.
I recently developed a statistically valid way to rate the defending of players, that is, what they do to prevent scores other than rebounding, blocks, steals, and fouls, which were always included in RPR. This would include man to man defending, zone defending, pick and roll defending, defensive recognition, and defensive rotation.
Although the technique used had to be indirect and inexact, it validly awards the better defenders with bigger RPR bonuses. It has been validated by comparing results obtained with the defensive ratings shown on three different "advanced basketball statistics" web sites. Our results were shown to be extremely highly correlated with the results shown on the other sites. Where there are small differences, I believe mine are better, if only because mine uses simple, bedrock statistical theory rather than involved formulas.
HIDDEN DEFENDING
Before revealing what we do to reveal it, let's define "hidden defending." Exactly what is hidden defending? It's every action that helps to prevent the other team from scoring other than rebounding, stealing, and blocking. So it would include man to man defending, zone defending, rotating in general, defensive recognition, and quick defensive response to various offensive tactics, such as pick and rolls. Obviously, if a defender is good at these things, the other team doesn't score as many points than if the defender is lousy at these things.
HOW TO REVEAL HIDDEN DEFENDING IN FIVE STEPS
STEP ONE: CALCUATION OF RAW HIDDEN DEFENDING RATINGS
Unlike most "advanced statistics" that are published on the internet or in print, we give you all the details about how we do ours, so that you can evaluate the evaluations, so to speak. The following is specifically what we are doing to be able to accurately and fairly compare players' defending:
Where do we start to discover what is hidden? We keep it as simple and yet as accurate as possible. We use the most official and therefore presumably the most reliable data as the building blocks for rating the defense of NBA players. We start with the player minutes and points scored by the other team while the player was on the court that are shown in the plus/minus statistical section at NBA.com.
After simply dividing points allowed by minutes on the court, we adjust that rate for the pace of the team and for the quality of the team's defense. The two adjustments are needed so that the ratings of players who are on different teams can be fairly compared.
Players who are on teams with faster paces give up more points per minute through no fault of their own. Similarly, players who are on teams with less efficient defenses give up more points per minute, everything else held constant. You could not fairly compare players on two or more teams with different paces and different team defense qualities unless you standardized, or in other words controlled for those differences for all NBA players.
USE OF BASIC STATISTICAL SAMPLING THEORY
What we are doing is using an indirect and inexact yet accurate and statistically valid way to discover who the better defenders are. No two players are out on the court for all the exact same minutes. So although for every player, what the other players out on the court do defensively while they are out on the court is a very large factor determining what that player's points per minute allowed will be, when you look at many, many hundreds of minutes, what the individual player does, or does not do defensively, as the case may be, will eventually show up in that particular player's points allowed per minute statistic.
In other words, what any individual player does defensively has to sooner or later show itself in the points allowed per minute. As the number of minutes rise above 500, and then 1,000 and then, for many players, above 2,000 and even 3,000 for a regular season, what a particular player does or does not do defensively becomes more and more exactly shown by the points allowed per minute number. This is very basic statistical sampling theory in operation. Statistical sampling theory is the easy to understand bedrock theory of statistics.
Due to the necessity of a large sample of minutes, we will not do defending estimates for any player who has played for fewer than 300 minutes. Quality of defending estimates will be slightly less accurate for players who have only played between 301 and about 600 minutes than they will be for players who have played for more than 600 minutes. We believe that the estimates are going to be extremely accurate for all players who have played 750 minutes or more. The idea is relatively simple: as the number of hundreds of minutes played goes up, the accuracy of this system improves, to the point where it gives you the same information you would have if you knew exactly how many possessions of the other team each player ruined with his defending.
For your information, all players allow between 1.87 and 2.16 points per minute; most allow between 1.94 and 2.11. The overall NBA average is about 2.03 points per minute allowed.
STEP TWO: CONVERSION OF RAW HIDDEN DEFENDING POINTS ALLOWED PER MINUTE TO FILTERED HIDDEN DEFENDING POINTS ALLOWED PER MINUTE
Since different players have different breakdowns between how much of their defending shows up in tracked statistics such as defensive rebounding and how much of it does not, in order to improve accuracy we need to have a method to filter, or in other words, separate, the two categories of defending. If we didn't do this, we would still have a useful statistic, but it would be biased in favor of players whose defending is counted in tracked statistics more so than other players. There would be in effect some double counting of defending for players who have most of their quality defending tracked by scored statistics.
The filter used is to multiply the raw hidden defending ratings by the percentage of the real player production that is offensive. In other words we take the inverse of the percentage of a player's real player production that is defensive and multiply the raw hidden defending ratings by that. The rationale to do this is that although the exact relation is unknowable, we know that for a given raw hidden defending performance level, there will be an inverse relation between scored defending and hidden defending. The more defensive rebounds, steals, and blocks a player is making for any raw level, the less he is relying on hidden defending to achieve the raw level. And vice versa. So multilying by the inverse of the percentage of all contributions that are defensive (in other words, multiplying by offensive contributions) filters out much of the bias that is in the raw hidden defending rating.
To be even more specific, we first extract out defensive rebounding, steals, blocks, and personal fouls, the sum total of which is called "Scored Defensive Contribution". All of the other components combined constitute "Scored Offensive Contribution". Now we can determine the percentages of the RPP that are offensive and defensive, and then we can use the offensive percentages to convert the raw hidden defending ratings to filtered hidden defending ratings.
STEP THREE: CONVERSION OF FILTERED ALLOWED POINTS PER MINUTE TO FILTERED HIDDEN DEFENDING RATING
We need to translate the adjusted or filtered points allowed per minute into numerical terms that are the most useful with respect to RPR. So with a very carefully designed translation scale, we amplify the very small differences in different player's points allowed per minute numbers into much larger different hidden defending ratings for each player. Then we simply add the hidden defending rating to the Base RPR to yield RPR.
STEP FOUR: USE OF HIDDEN DEFENDING RATING
We now have added in a reasonably good estimate of the value of actions of players that are not even kept track of by scorekeepers! The filtered hidden defending ratings are added to the "Base or Scored RPR" to give RPP. The range of possible defending adjustments to the base RPR is from 0 to about .230. In most cases, however, the adjustment will be between 0.030 and .150.
STEP FIVE: OVERALL EVALUATION OF DEFENDING
Aside from the Hidden Defending Rating we can find out how well each player does in terms of unhidden or scored defending, can't we? Of course se can.
Aside from the hidden there is of course unhidden defending, which would be rebounding plus steals plus blocks minus personal fouls. If we extract the combination of those four out of the same counts that underlie the RPR as a whole, we get what we are going to call the Scored Defending Contribution. This could also be thought of as Tracked Defending Contribution if you prefer. Then if we divide this by minutes, we can have a Scored (or Tracked) Defending Rating.
Finally, if we combine Hidden Defending Rating (HDR) with Scored Defending Rating (SDR) we can have an Overall Defending Rating (ODR). I am for now going to simply multiply the HDR by two and add that to the SDR to yield the ODR. To combine them this way is more arbitrary than my usual standards allow; I am doing this because there is as of yet no non-arbitrary way of doing it. The formula of two times HDR plus SDR brings HDR almost up to par with SDR in terms of the actual numbers and the averages of those numbers involved.
In other words, I am saying for now that hidden defending is almost as important as scored defending. There appear to be many coaches and not a few hardcore basketball fans who believe that hidden defending is actually more important than scored defending, but I am very likely never going to agree with that. I think that although hidden defending is important, and plausibly almost as important as tracked defending, that it is like a quicksand in that there seems to be a tendency for a substantial minority of basketball people to get carried away with estimating the importance of it and then become more and more trapped by their error in terms of how they look at basketball or in terms of how they coach their team if they are coaching.
Subscribe to:
Posts (Atom)
