Doug Moe is synonymous with the Denver Nuggets. No one had a longer tenure as coach of this team. And no coach led this franchise to more victories. So one suspects Moe knows the Nuggets.
Recently the “Big Stiff” (as Moe is known) was asked who is the greatest player in the history in the Nuggets. And here is Moe’s answer (HT: Phil Maymin):
“I’d have to say Melo,” Moe said. “His talent is going to win more games than anybody else. With him, I think the Nuggets are going to have more 50-win seasons than with any other player. And that’s not a knock against any of those players.”
Carmelo Anthony is currently leading the NBA in scoring. But when I look past scoring, I find myself disagreeing with Moe. According to Table One, Melo entered the current season ranked 12th on the Nuggets’ all-time Wins Produced rankings.
Table One: Ranking the Denver Nuggets from 1977-78 to 2008-09
As noted yesterday, Anthony is on pace to produce about nine wins this season. So when the 2009-10 season ends, Melo might catch David Thompson. But Anthony is going to have produce at his curren rate until 2018-19 (when he will be 35 years of age) to catch the number one ranked player in history. Yes, it’s possible that can happen. But given what happens to players once they pass the age of 30, it seems unlikely.
If Anthony does maintain his current pace for another decade — again, this is unlikely — he will eventually pass Fat Lever on the list. Lever– who played for Denver and Doug Moe – produced 114.2 wins in just six seasons in the Mile High City. In addition to leading Denver in Wins Produced, Lever’s 0.325 WP48 [Wins Produced per 48 minutes] also ranks number one in Denver history (Anthony has never posted a WP48 close to this mark) . Lever, though, never scored 20 points per game. So it’s understandable that people might think Carmelo Anthony is doing more today.
I should add… Lever is also not the only Moe player to surpass Anthony. T.R. Dunn, Alex English and Dan Issel produced more wins than Anthony in a Denver uniform. And Moe coached all these players. But apparently he thinks all fall short of Melo.
Again, Anthony scores. And as Moe notes, he plays for some of the most successful teams in Denver history. But I think Denver’s success is more about Carmelo’s teammates (i.e. Marcus Camby, Andre Miller, Chauncey Billups, Nene Hilario, and Chris Andersen) than it’s about Denver’s lone current All-Star. At least, these players — despite not being great scorers — appear to produce more wins than Anthony.
Now I am not saying Anthony isn’t a good player. But I do not think he is as productive as other top players today (i.e. LeBron James, Chris Paul, etc…). And he is not the most productive player in the history of the Nuggets.
Let me close by noting that this is the fourth time I have reported every player in the history of a team. Here are the previous three teams I have examined.
Ranking Every Player in the History of the Utah Jazz
Ranking Every Player for the Boston Celtics since 1977
Ranking Every Player for the LA Lakers since 1977
Perhaps this next summer – after the 2009-10 season ends – I will look at a few more teams.
– DJ
The WoW Journal Comments Policy
Our research on the NBA was summarized HERE.
The Technical Notes at wagesofwins.com provides substantially more information on the published research behind Wins Produced and Win Score
Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:
Simple Models of Player Performance
What Wins Produced Says and What It Does Not Say
Introducing PAWSmin — and a Defense of Box Score Statistics
Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.
Wages of Wins Journal 
Chicago Tim
January 22, 2010
Is it fair to compare players who play different positions? A long time ago, you noted that it is harder for guards to produce wins because there are more people capable of playing that position. A good center, on the other hand, can produce many wins because there are so few people capable of playing that position.
https://dberri.wordpress.com/2006/12/04/
Of course, all of that makes 6′ 3″ Fat Lever even more remarkable. I found this old SI article written while Fat Lever was playing, and theorizing about why he hadn’t received more recognition. Apparently he deliberately kept a low profile, and had a somewhat boringly efficient game. The anti-Allen Iverson, if you will.
http://sportsillustrated.cnn.com/vault/article/magazine/MAG1066964/index.htm
Don Taylor
January 23, 2010
Carmello Anthony, David Lee and Dwayne Wade are my favorite active players (I’m into footwork) even with that being the case, I immediately thought, “Lafayette Lever” as soon as I heard the question. Shame on Doug Moe. Fat played his hart out for him.
todd2
January 23, 2010
Knew Fat could play. Alex English would have been at the top of my list, or Dan Issel. Their WP48’s are surprisingly low. Bobby Jones’ numbers are surprisingly high—he had a rep for being a stopper. Thanks, Prof! Some great players and teams came and went in the Western Conference during Earvin Johnson’s time with the Lakers; nine times in the finals!
mrparker
January 23, 2010
Prof,
Can you comment on what made Fat Lever so productive. I’m looking at his bref page and I don’t see anything indicitive of a productive player besides an absurd steals percentage.
Thanks
Robert
January 23, 2010
id assume it was the fact he was technically a PG but put up very gaudy rebounding totals.
mrparker
January 23, 2010
oh yeah,
I didn’t pay attention to those numbers. I was just looking at 51% true shooting percentage and got lost.
benamery21
January 24, 2010
Doug Moe’s response highlights a WOW premise: that even very experienced basketball tacticians have a hard time experientially evaluating the relative contributions of players to wins given the distractions of high scoring and entertaining balletics by some players. This goes a long way to explain why few coaches improve their teams winning percentage appreciably over the average coach–they are subject to the same errors of perception as most of us.
Case in point: Phoenix tonight (home win against GSW)
Alvin Gentry’s massive mistake (from a WOW perspective) with regard to his change in starting line-up (dropping Jason Richardson in favor of Barbosa and Channing Frye in favor of Robin Lopez) was partially ameliorated tonight by Barbosa’s return to the injured list (for about 4 weeks due to wrist surgery). Jason Richardson returned to his starting slot and Barbosa (whose productivity has been abysmal of late due to significantly below career level shooting stats) will no longer be hurting the Suns by playing at less than 100%.
Lopez, so far, has been productive in a starting role, but I contend that he reduces “all-star” Amare Stoudemire’s productivity by more than Frye (whose Win Score stats have been better than Lopez over the course of the full season). Lopez, Frye, and Louis Amundson (who has also lost minutes to Lopez’ new role) all outproduced Stoudemire tonight in fewer minutes.
Gentry also persists in giving (developmental) minutes to (highly unproductive) Earl Clark in close games. This has resulted in several unnecessary losses. Clark appeared tonight for 11 minutes, apparently in an attempt to keep the contest with Golden State sporting by spotting them Clark’s negative Win score.
benamery21
January 24, 2010
I would have said Deke Mutumbo if asked the same question. Ex-Sun Devil Lever definitely ranks as one of the most underappreciated players ever. I wonder if Moe’s response might have been more political than forthright, given that he’s still associated with the franchise and public opinion has Melo first on the current team?
benamery21
January 24, 2010
dberri: Where would Fat’s ’84-’85 Nuggets rank on your “all-time best starting squad” list (they didn’t make the top 25 you published in Nov ’09)? The ’84-’85 Lakers (who they lost to in the Western conference finals) ranked 4th on your all time list. I calc’d a 0.207 WP48 for Denver (Lever, Dunn, English, Natt, Cooper) which would put them about 15th (but is based on the PAWSmin conversion and doesn’t account for their torrid #1 in the NBA pace that season). They took a 1-1 split back home from LA to game 3 (even though Fat was injured and didn’t play in the series), but wouldn’t win again with 3 injured starters (Fat out whole series, English out from gm 3 on, and Natt playing with a sprained ankle).
Jimbo
January 24, 2010
Have you seen Ariza this year? His PER has fallen off dramatically and I’m sure his WP has also which really goes to show that when a player moves from a team where he is a secondary option and doesn’t have to hoist a ton of shots to a team where he does have to, his ‘overall’ stats will diminish. This observation goes to show that your model undervalues scorers b/c you think efficient scorers who don’t score a lot would score more at the same efficiency level if given more shots, which just isn’t true as Ariza’s experience demonstrates.
Something to think about…
dberri
January 24, 2010
Jimbo,
As I noted a few weeks ago, a number of other players on the Rockets have seen their shot attempts increase and their productivity was not changed. I guess that is something to think about also…
khandor
January 24, 2010
With all due respect …
What is quite hilarious is that some here actually think/believe that a statistical based formula for determining the relative value of a real life player’s actual level of “production” is a more accurate indicator of “the best” player in Denver Nuggets history, compared with the judgment of a relatively successful former NBA head coach/”tactician” by the name of Doug Moe.
IMO, Carmelo Anthony is so much better at what it is he just happens to do as a pro basketball player [i.e. be a scorer], compared to what it was that Fat Lever just happened to do as a pro basketball player [i.e. be an efficient multi-purpose guard], that it’s not even funny.
benamery21
January 24, 2010
khandor:
Almost as funny as the fact that you and I are both still here talking when you apparently think he’s an idiot, and he apparently thinks I’m one.
Chicago Tim
January 24, 2010
Khandor, prefacing your remark with “With all due respect” does not automatically make your remark respectful.
khandor
January 25, 2010
Chicago Tim,
Part One
1. Thanks for posting that link in your initial comment.
When I go back and read what was actually said there it becomes even more clear that David has some difficulty when it comes to distinguishing between:
i. The “Most Productive Player” … according to a specific mathematical-based metric; and,
ii. The “best player” [or, for that matter, the League MVP] within a specific environment and, therefore, a set of players.
2. It is fundamentally flawed to try to assert that:
“It is easier for a front-court player to produce wins than it is for a back-court player,”
based on the premise that,
“There is a short supply of tall people.”
———————————————————-
re: What does all this mean? One step in calculating Wins Produced is to compare a player’s performance relative to the average at his position. Because there is a “Short Supply of Tall People” (and I still like that phrase), frontcourt players consist of productive players – like Shaq, Robinson, and Rodman – and players that are quite unproductive (insert name of big stiff here). Hence a quality frontcourt player can produce at a level far above the average at his position. In contrast, guards tend to be shorter, so the supply of these players is much larger. Consequently Jordan was being compared to a population of players with relatively more talented athletes, and hence it was harder for Jordan to perform far above the average at his position.
———————————————————-
In fact, this premise is dependent on the poor judgment of many [some?] in the basketball community who insist on using “taller” people at the front-court positions, who happen not to be very good basketball players, in the first place, not the fact that there is a “Short Supply of Tall People”.
Although others might believe the answers to be straight-forward, when the following list of items is discussed, regarding the goings-on in the NBA:
– “MPP”
– “the best player”
– what a player’s specific “position” is
– what a player’s specific “height” is
– what a player’s specific “skill-set” is
– what constitutes a “short” person
– what constitutes a “tall” person
– what a player’s relative level of production is compared to the average player at a specific position
– how many back-court players there are in the NBA
– how many front-court players there are in the NBA
– how “qualitative” distinctions are made best amongst a specific set group of players
– the reason certain players are used most frequently at certain positions in the NBA
I do not share this same perspective.
Without clearly distinguishing between these different concepts and ideas, it is all too easy to think that one is proving something when, in fact, one is actually proving nothing of the sort.
Part Two
In what I’ve written here, thus far, if there is anything which you believe is disrespectful, then, please point it out. It is not my intention to be disrespectful, in any way.
Thanks, in advance.
feltbot
January 25, 2010
The fact that this stat has even a shred of currency is mind-boggling to me. A player like Carmelo Anthony is double and triple teamed regularly, something that lowers his own offensive efficiency, but raises that of his teammates and of his team immeasurably.
brgulker
January 26, 2010
Here’s the reason that people “believe” in Dr. Berri’s work.
1) We’ve read his book, not just his blog.
2) Teams that employ “productive” players are consistently among the top teams in the NBA.
3) I’ve yet to read a refutation of his metric that holds any substantive, empirical weight.
4) Dr. Berri readily admits the limitations of his metrics.
5) It bears repeating, teams that employ “productive” players are consistently among the top teams in the NBA.
And to me, that’s the clincher. At the end of the day, winning is what matters. Dr. Berri’s work describes player performance with respect to win consistently and accurately, more so than anyone else who does this type of work.
brgulker
January 26, 2010
Um, Dr. Berri has addressed this repeatedly, actually. Time and time again, he shows how players do not get any better when paired with a “superstar.”
benamery21
January 27, 2010
Yep. Spot on.
1) We’ve read his book, not just his blog.
2) Teams that employ “productive” players are consistently among the top teams in the NBA.
khandor
January 27, 2010
re: “2) Teams that employ “productive” players are consistently among the top teams in the NBA.”
Classic “chicken and egg”, right there.
re: “1) We’ve read his book, not just his blog.”
In a sense …
If the blog … which is based on the book … has specific difficulties and, therefore, doesn’t “work/function” properly, then, unfortunately, neither can the book, when placed under the appropriate type of scrutiny from a “basketball” perspective.
khandor
January 27, 2010
re: “Time and time again, he shows how players do not get any better when paired with a ‘superstar.'”
This supposed fact, according to the work of Dr. Berri, might not actually be fact at all, depending on how one actually decides to “measure” what constitutes [i.] “better” and [ii.] a “superstar”, in the first place.
Inter-dependency is a significant component of basketball and playing with the “right fit” teammate[s], and/or with the right fit “environment”, is a critical factor in determining someone’s actual level of success.
khandor
January 27, 2010
brgulker,
re: “3) I’ve yet to read a refutation of his metric that holds any substantive, empirical weight.”
What’s the reason you are searching for “substantive, empirical weight”, in the first place? … other than the sort of indisputable evidence which is available through a set of specific data like this.
re: “At the end of the day, winning is what matters.”
Did “substantive empirical weight” make the correct assessment, in advance, that:
– the Miami Heat were going to beat the Dallas Mavericks for the 2005-2006 NBA Championship?
– the San Antonio Spurs were going to prevail as the 2006-2007 champions over the Cleveland Cavaliers?
– the Boston Celtics were going to capture the 2007-2008 title over the LA Lakers?
– the LA Lakers were going to be crowned NBA champions last season, at the expense of the Orlando Magic?
While a mathematical based formula can certainly do a terrific job of:
* Assigning specific values to different statistical categories associated with the game of basketball and, therefore, account retroactively for the successes and failures of specific teams/players in a given set of data; or,
* Predicting, in a general sense, how many “overall wins” a specific team is “likely” to achieve in an upcoming season, based on past production values for those same participants and mathematical probability;
IMO, it would be a mistake in basketball judgment to believe that such a metric/formula and, therefore, an associated “cookie-cutter” way of thinking about the game, itself, based on said formula, is capable of forming wholly accurate assessments of which players, teams and coaches, etc., are actually superior to their peers, when it comes to being able to win specific games and championships against specific opponents, in the real life NBA … which is REALLY what the bottom line is all about when it comes to properly understanding how this game actually works.
Basketball is not the same game as baseball … e.g. is there any doubt, whatsoever, as to who the New York Yankees started at 3rd base this past season, while they captured their 27th World Series title? As opposed to, let’s say, what position Tim Duncan actually plays for the San Antonio Spurs, both, this year and for the 2006-2007 NBA World Champions [not what position 82games.com has him listed at] … and, understanding it properly involves a great deal more than simply knowing the mathematics/economics of the game.
When Dr. Berri works within the “limits” of what his metrics can actually describe with accuracy, then, I’m a big proponent of his work.
brgulker
January 27, 2010
Either read the book or stop making these types of claims. If you do read the book, not just the blog, you’ll see how far from a “cookie cutter” this model actually is. You will also see how readily the authors admit the limitations of their own work.
But if you don’t read the book, your flawed assumptions will remain. No commenter here has the time to restate every single story and argument that appears in the book — that’s what the book is there for in the first place?
One last comment on your above paragraph:
No one, including Dr. Berri, claims any of that. You’re the only one who is.
His book makes an empirical argument. It is natural to seek out counterarguments that are also empirical.
No it’s not. When teams collect a bunch of productive players, they win games. You can’t get wins without players who produce them. Productive players cause wins.
I can’t think of any other ways to state it …
No person or model can predict who is going to win four 7-game series in a row. No one on this blog claims to be able to do that perfectly.
IIRC, Dr. Berri predicted that LA would be the best team during 08-09, and it turned out that the best team actually did win the championship.
But predictions are not what WoW is about. WoW is about evaluation. Evaluation is certainly used when people are making predictions, but that’s not the point.
Again, if you read the book, this becomes patently obvious.
khandor
January 27, 2010
brgulker,
1. re: “Productive players cause wins.”
Although I could well be wrong about this … I have yet to see any type of “substantive, empirical weight” which outlines a causal relationship that exists between “productive” players and “wins”.
2. I’m not saying that Dr. Berri is making an erroneous claim that his methodology is capable of “predicting” the outcome of specific future games with any degree of accuracy, or is capable of making wholly accurate assessments of specific players or teams.
What I’m saying is that:
– when it comes to “evaluating” properly which is actually “better”, using a mathematical-based formula to distinguish between the “quality” of a Rembrandt and a Monet, or a Picasso and a Mozart, or a Socrates and a Frank Lloyd Wright, or a Descartes and a MLK, or a Jordan vs Russell, or a Lakers vs Cavs, etc., is far from being the correct way to go
– when you start shifting ground from saying that,
“Player X was the Most Productive Player during the 2009-2010 NBA season, based on this specific metric,”
to something like,
“Player X should, therefore, have been awarded the MVP Award during the 2009-2010 NBA season, as the best player,”
you are taking a leap of faith and, in the process, making a fundamental mistake in basketball judgment.
Here’s a question for you and others to consider:
Is it possible to lose a game in which your team is outperformed in each of the following categories?
i.e. FG%, 3FG%, FT%, Rebound Differential, Assists, Turnovers, Steals, Blocked Shots, Personal Fouls, Charges Drawn, either FTAs or FGAs, and either FTMs or FGMs
If not, then, how come?
If so, then, how come?
Inter-dependency in the game of basketball is one of the fundamental things which define its character and provide its overall artistic beauty.
brgulker
January 27, 2010
Productive players do the following: score more points than their opponents (by scoring efficiently), get more rebounds than their opponents, get more steals than their opponents, get more assists than their opponents, commit fewer fouls than their opponents, turn the ball over less than their opponents, et al.
Productive players — players who do those things — create wins for their team.
Regardless of one’s position on any particular statistical model, I don’t see how this can be disputed. IMO, you might as well dispute that the team who scores the most points wins.
You’re fundamentally missing the point. You’re using a micro-level case study to try to argue against macro-level analysis. It simply doesn’t work.
The object of WoW is not a random, specific game but rather large sets of data across large amounts of time. In any given game, just about anything can happen. But over the course of 82 games, 164 games, and so on randomness is mitigated.
khandor
January 27, 2010
burgulker,
1. When Allen Iverson’s Philadelphia 76ers went to the NBA Finals, was he a particularly efficienct basketball player?
2. re: “Productive players do the following: score more points than their opponents (by scoring efficiently), get more rebounds than their opponents, get more steals than their opponents, get more assists than their opponents, commit fewer fouls than their opponents, turn the ball over less than their opponents, et al.
Productive players — players who do those things — create wins for their team.
Regardless of one’s position on any particular statistical model, I don’t see how this can be disputed. IMO, you might as well dispute that the team who scores the most points wins.”
i. What about the players who do some of those things but not others? Are they, therefore, properly classified as being “unproductive”?
ii. Undoubtably, “productive players” who do some or all of those things can “be a part” of a team that wins a great many games in an NBA season.
Unfortunately, it is a different thing entirely to then try to assert that it’s the productive players, as designated by a specific statistical-based metric, who are the cause of their team winning X number of games in a NBA season.
3. I think that you and, perhaps, many others are essentially missing “the point”.
Basketball, properly understood, is an idiosyncratic micro-level game, best understood and evaluated in anecdotal terms, not a generic macro-level one, best understood and evaluated strictly by an examination of “substantive, empirically weighted” evidence … even though this specific type of “data analysis” can be useful to promote a further and more comprehensive understanding of how the game, itself, actually works.
IMO, doing this specific type of academic research, and then implementing its associated strategies for success will consistently fail to produce a real life NBA title-winning team.
Real life … and, the game of basketball … unfortunately, does not happen to work this way.
benamery21
January 27, 2010
khandor:
Can you predict accurately which side of a die will show up on a given throw? No. But you know what the probability of a given side showing up is. If you bet me you could throw ‘7’ s with a pair of dice, we agree on what the odds are, but what actually happens on a given night is subject to chance. dberri’s work doesn’t create that level of precision with respect to the odds but it significantly improves the estimation of what those odds are. Randomness is a factor just as when throwing dice: Did you predict the Clippers beating the Celtics even once this season? But, the odds are that over a large sample size, the model will be fairly accurate in predicting a team’s season wins even when composed of a changed cast of players (if adjusted for minutes played). Is it perfect? No, of course not. There are a number of admitted limitations, and I feel, a few unadmitted ones. But it’s better than anything else I know about.
Some of the limitations could be addressed by improved box score data collection (tracking more and different things). Some are inherent assumptions that there is no easy way around. The limitations are where an analyst/coach using the tool has room to use informed intuition.
benamery21
January 27, 2010
khandor: Anybody who believes ANYTHING is ‘best evaluated in anecedotal terms’ isn’t going to find a lot of common ground with folks here.
khandor
January 28, 2010
benamery21,
re: khandor: Anybody who believes ANYTHING is ‘best evaluated in anecedotal terms’ isn’t going to find a lot of common ground with folks here.
No disrespect intended, but …
IMO, finding “common ground” isn’t necessarily what this web site is about.
The question, in this specific instance, is:
If the goal is, in fact, to be able to create a real-life championship-winning NBA team …
Is the best way to evaluate individual players, teams and coaches, etc., accurately based on
i. A “substantive, empirical weighted macro-level ‘quantitative’ analysis of the readily available statistical data,”
or,
ii. An “idiosyncratic, micro-level ‘qualitative’ analysis, predicated on anecdotal case studies and the fundamental interdependency inherent in the game.”
No one who reads what I have to say about the game of basketball should ever think that I am someone who fails to recognize the value of Dr. Berri’s work, and its actual limitations.