And it keeps getting bigger and bigger (and better and better)…
Andres Alvarez has expanded his Wins Produced site (please see prominent link on the right). You can now see Wins Produced and WP48 numbers for every player from 2005-06 to 2009-10. And the numbers aren’t just for the regular season. Alvarez has also calculated Wins Produced for the playoffs. This is something that I have never done before, so I was very interested to see the most productive players in the playoffs for each of the past five seasons. These players are….
- 2006: Dirk Nowitzki , 7.25 Wins Produced
- 2007: LeBron James, 4.23 Wins Produced
- 2008: Kevin Garnett, 6.02 Wins Produced
- 2009: Dwight Howard, 7.36 Wins Produced
- 2010 (thus far): Rajon Rondo, 3.86 Wins Produced
Oddly enough, the player producing the most wins in the playoffs failed to win the NBA title three times from 2006 to 2009. If we focus on the team that won the title in each season, the top producer of wins was as follows:
- 2006: Dwyane Wade, 5.55 Wins Produced
- 2007: Tim Duncan, 3.71 Wins Produced
- 2009: Kobe Bryant, 5.64 Wins Produced
Ultimately these numbers will be expanded all the way back to the 1977-78 season. So soon, everyone is going to have an abundance of Wins Produced numbers to ponder.
Okay, obviously the big news today is the expanding data set from Andres. But there are some other Stumbling on Wins stories I want to note.
Two Toronto Star stories from Dave Feschuk
Dave Feschuk – of the Toronto Star — posted a story on Sunday that reviewed the story we told about aging in the NBA in Stumbling on Wins. Feschuk’s story began with our general finding, then focused on how Steve Nash managed to defy the pattern we identified.
The next day I spent more than an hour on the phone with Dave. From that conversation came the following column (appearing on Thursday):Changing NBA coaches mostly pointless, report claims.
From this column
- we see that Bryan Colangelo claims he has read Wages of Wins, but not Stumbling on Wins.
- we also see that Colangelo thinks you cannot just rely on numbers to make decisions. In fact, he seems to distrust numbers (at least, that is my impression).
- are some quotes from me on the subject of Andrea Bargnani, Adrian Dantley, and Doc Rivers. These are supposed to be funny (or reveal that I am a smart-ass).
The Sports Economist on Stumbling on Wins and the “Yay! Points” Thesis
Two economists from the Sports Economist – Brian Goff and Robert Macdonald – have commented on Stumbling on Wins. Goff’s column – LeBron “Stumbling on Losses” – details the tension between behavioral economics and mainstream economics (and says nice things about our book along the way). Macdonald’s column – (Ir)rational Reading – connects our work to The Myth of the Rational Market by Justin Fox. As Macdonald notes, both books challenge mainstream economics and leads him to wonder when sports economics will move away from relying so heavily on mainstream models.
Let me close by noting that WarriorsWorld.net recently posted Stumbling on Wins: The Right Stats. From this great review (any review that likes our book is a “great” review) we see a clever name for the tendency of NBA decision-makers (and fans) to over-value scoring. Hence forth, maybe we should call this the “Yay! Points!” thesis.
– DJ
The WoW Journal Comments Policy
Wages of Wins Journal 
Dre
May 19, 2010
I did enjoy the comments. One of them reminded me on an episode of where Homer becomes a manager:
Homer: (in his new role as supervisor) Um, are you guys working?
Employee: Yes, sir.
Homer: Can you . . . work any harder?
Employee: Sure thing, boss!
Next stop Toronto for a coaching gig!
ilikeflowers
May 19, 2010
KOBE!!!
2006 – 2010 playoffs:
Kobe: 0.175
Manu: 0.185
Wade: 0.290
Alvy
May 19, 2010
OmG, let’s talk about KoBe and comParE him!! DiD u GuYs sEe hiS seRies against the *JaZz*?? 30PtS on 50+% FG, hE so CoOl!!! OmG, diD u SeE his GaMe OnE against LoS SUNS??? He’S neXt MJ.
“~*Ilikeflowrs*~”, wHat does .175 meAN?
KOBE RULZZ!!!!
—Stay classy guys.
Now, another note. Thanks Andres Alvarez, I’ve always wanted to know about playoff WP numbers!
WXYZ
May 19, 2010
Cool. Now we can hear about how consistent these ratings are or are not with regular season.
WXYZ
May 19, 2010
I imagine that overall there is some dropoff, but how strong is the predictive power of regular season for individual players in the playoffs before or after the team adjustment?
WXYZ
May 19, 2010
Coach-speak, at least during games, is often quite simplistic. But you’d have to know how detailed they were in pre-game and post-game film work about how to do things like stopping lay-ups to judge fairly. During a game maybe it might make sense to remind about technique or a specific thing to watch for, when you can. But maybe you already covered it and expect them to retain it and they don’t have a lot time to give detail then. Maybe an assistant can do that off-mike when there is more time.
WXYZ
May 19, 2010
And on-mike it would probably not be too smart to give away too much detail to the other side who will hear it.
Arturo
May 19, 2010
Andres,
Kudos. I love the site.
The NBA is the perfect example of an “irrational” market. Yay!Points! indeed. When was the last time the scoring champion played on championship team? Shaq in 2000, that’s a full ten years and counting.
WXYZ
May 20, 2010
What % of the top 10 or 20 on WP also scored 15+ per game?
WXYZ
May 20, 2010
Looking at the top15 by position data I guess it was just over half at PG, almost a 100% at SG, just over half at SF, almost 70% at PF and just over half at C.
WXYZ
May 20, 2010
What %s scored 10+ pts per game?
9o+% at PG, almost 80+% at SG, 80+% at SF, almost 80% at PF and two-thirds at C.
You don’t need many 20+ pts scorers but 15+ scorers are the majority of the top 15s at every position and are more strongly represented at 2 positions.
80% or more of the top 15s at every position other than Center scored 10+ pts per game.
There were only 127 players who scored 10+ in 60+ games last season and only 65 scored 15+. Other can do it, perhaps to a lesser degree, but on average teams only had 2 at 15+ and 2 more at 10+. I can see why getting scorers or good scorers is one concern.
Marparker
May 20, 2010
This data only goes back to 2006 but what jumps out at me is shooting guard play in the playoffs.
The sg play of the champs:
06- wade .278
07- ginobili .306
08- allen .132
posey- in backup minutes sg/sf .241
09- kobe .288
10-????
leaders so far
Allen- .112
Richardson .368
kobe .098
vince .079
Will Phoenix come back? Will someone else’s numbers sky rocket.
Side note- Joe Johnson Lebron’s people are on line 1.
Marparker
May 20, 2010
side note2- what if Mike Brown played Moon more minutes?
Marparker
May 20, 2010
I also went back and looked at the offensive ratings of the shooting guards of the championship teams during the regular season since 2006. Each had an ortg of 115 or above.
They were lower for years before but so was the league average offensive rating.
Allen 115
Richardson 116
Kobe 109
vince carter 112
sidenote- Late night bets on Boston to win just happened.7 to 4 bet.
Back to the getting Lebron a championship asap-
I’m going back to look at all the great players who never played with a highly efficient shooting guard or never won until they got to play with one and realizing how important that position seems to be.
And it makes me feel much worse for anyone who had to go through MJ who posted 120+ seasons routinely.
It also makes me scratch my head. I know this site is known for Kobe hating, but how bad must he have sucked in that playoff season(career 103ortg in playoffs, his lowest post 99-00 season) that a team with Rip Hamilton at sg won a championship.
For Lebron fans our mission seems to be to beam thoughts into Danny Ferry’s head. His mission: get Lebron an offensive 2-guard.
Leon
May 20, 2010
Sorry to bring this back up but I was wondering if you’ve had a chance to think about it Dberri? – ((https://dberri.wordpress.com/2010/04/08/another-qa-about-stumbling-on-wins/).
Erm, on another side note (word of the day?). It’s interesting that repeatedly (in your book and on here) that coaches don’t make a difference in the nba. Stefan Szymanski suggests that in the world of football/soccer, managers make a difference. We certainly see pretty constant production over years in nba players, and it wouldn’t surprise me if that was the case in soccer as well. So perhaps it is down to the format of the game and that in there are fewer players at any one time on a basketball court than a football pitch, and the scope for tactics and strategy reduced. It leads me to thinking that what effect a coach can have in the playoffs, after all, players play their game regardless of what is said, and they don’t seem to change so much, so it feels perhaps wasted on all the talk of match-ups, play by plays et al.
dberri
May 20, 2010
Leon,
What did you want me to think about?
Leon
May 20, 2010
Oops, sorry made a right pig’s ear of that ok, here’s the copy and paste :
“I have a question that is a bit off topic Dr. Berri, but I figure perhaps this is as good a time as any to ask. At what point does a player outweigh the advantages of scoring against the advantages of being more productive and thus helping your team become a more serious contender?”
TBall
May 20, 2010
Loved the comment on Doc in the Toronto Star article. I feel the same way. He is saying the same things these players have been hearing for 15-25 years. Work together. Play as a team. At some point, I sort of figure they have nothing else to say, but you can’t sit in the huddle silently, so you just start spewing cliches.
I also find Doc is one of those coaches that says things like “I told them to … and they [didn’t get it done]” when they lose and “I told them to … and we won because they [did it]” when they win. Disassociating with the team in losses and taking the captain’s chair in victory. Small things and, when the C’s win, the fans overlook it.
Coachingmatters
May 20, 2010
Substitutions, play calls and whether they are run or work are what matters most. There may not be many consistently above average coaches who produce a significant net impact against their counterpart on the other sideline but a poor coach can be outdistanced by an average coach and hurt a good amount.
It would be interesting to see a team with no coaches at all. That would test the literal ‘coaches have no impact at all on the performance of their players” theory
ilikeflowers
May 20, 2010
If I recall correctly what the professor studied was which coaches produce a lasting impact on player performance. Coaches do impact a team’s performance. For example, players improve their performance while playing for Don Nelson, but once they no longer play for him they tend to revert back to their previous production levels.
Coachingmatters,
that would be an awesome experiment although I imagine that one or more players would emerge to fulfill the coaching role vacuum. What I’d love to see would be how well various coaches’ lineups match what a strict wp48 lineup would be and whether or not these coaches tend to over or under-perform their team’s average wp48 or efficiency diff compared to other coaches.
ilikeflowers
May 20, 2010
From a GM’s perspective, I don’t care if a coach improves a player permanently, I only care if they improve him for my team. Thinking about it further, all things being equal, I would prefer the coach who makes players better for my team only.
brian
May 20, 2010
coachingmatters,
I don’t think the study says that a team with no coach is likely to be as good as the same team with a coach. What the study did do is compare coaches against each other. So what it claims is that *relative to other coaches*, many coaches have no statistically significant impact.
Coachingmatters
May 20, 2010
I accept that impact in the study meant “relative to other coaches”; but then I would speak of the findings with that attached, instead of without.
Coachingmatters
May 20, 2010
Without an official coach the PG or the star or the combo would indeed fill the void because there are managerial decisions to be made. This did happen a few times in the distant past.
Some probably would do at or near average, a few might do better. Maybe the ones who eventually become good coaches, but also some who could or should, if they wanted to.
Coachingmatters
May 20, 2010
Really distinct coach or player-coach, hard to imagine going without so I guess the experiment wouldn’t end up being a pure “without”, it would just be a shift of responsibility to somebody else.
I am not the one to judge how well Wilkens or Russell or whoever else is appropriate to name as having done it, did.
Coachingmatters
May 20, 2010
“I would prefer the coach who makes players better for my team only.”
So you might not explain everything about what you did to them and what was most important in crystal clear language.
That is getting pretty fancy though. I guess if it working you don’t have to crystal clear, but I’d tend to.
D-Fens
May 20, 2010
Hold on, so if players are better playing under the tutelage of Don Nelson, then Monta Ellis is an even bigger Black Hole than anyone gave him credit for?! I didn’t expect to ever deduce that Monta Ellis could in some way “overachieve”.
Still, $11m/year when you neglect defensive duties, refuse to distribute the ball, commit a ridiculous number of turnovers and solely concentrate on jacking up as many shots as possible for a serial loser is nice work if you can get it. What do you fine gentlemen think his WP48 is when Nellie departs then?
palamida
May 20, 2010
D-fens, be serious. The study and it’s conclusions are naturally limited to generalities.
Obviously not ALL players improve under Nelson.
All the study is telling us is that Nelson (and a few other coaches) have statically significant effect on player performance (in general, on average) whereas most coaches – do not.
As already stated above, the study only attests to the relative effect, meaning: How certain coaches differ from others. It’s entirely plausible that having no coach at all will have a drastic negative effect on player performance. This dataset is obviously self-selected since all the teams examined – did in fact employ a coach.
D-Fens
May 21, 2010
Palamida, regrettably as a Warriors fan I am being completely serious! I must say I was a little surprised to read that “Nellieball” statistically “improved” players, which as someone who wholeheartedly to many of the recurring conclusions on this blog regarding the personnel decisions in the NBA – scoring overvalued, efficency undervalued, defense undervalued – seemed almost counter intuitive to me.
palamida
May 21, 2010
De-fens, when I asked you’d get serious, I just meant that arguing: “How can this be true?, I mean – look at Monta Ellis” is obviously a logically flawed argument.
More to the point, I suspect WoW “over values” Nellie, because it fails to capture the counterpart element. In most cases, while it causes WoW to marginally under value or over value a certain player – In “Nellieball” this becomes a serious issue:
Yes, Nellie players use the mismatches offensively to their advantage, but that contribution is negated but what they allow at the other end.
Amare is a perfect example of this (even though he has never played under Nellie).
While he himself has produced some decent numbers throughout his career, he has consistently been “feasted” on by opposing big men, especially when he was deployed as Center.
This isn’t some random stat that exposes what the naked eye can never see :p
Amare’s defensive woes are quite plain.
Counterpart data can probably assist us in performing the “team adjustment” better – meaning allocating defensive wins and losses responsibilities in a more exact manner. The main problem with this, (as Dberri has pointed out in the past) is the reliability of the data to begin with.
With that caveat, it’s worth noting that most players “grade” very similarly when you incorporate counterpart data, but some – do not. I think Golden state is a real outlier in that regard as anyone with a set of eyes and\or common sense, can attest to.
Jimbo
May 21, 2010
Ronnddooooooooooooooooooo!!!!
Andrew
May 24, 2010
I’d guess that being the most productive player in the playoffs is probably tied to playing a lot of playoff games. And a lot of times, the worse team in the Finals has played more games, so I think that might partially explain it.
kevin
May 26, 2010
Apropos to nothing, it appears Charlotte is offering the head coaching position to Tom Thibodeau.
Which is highly ironic to me, as I played with Thibo in college. Me being 3 years older, I always thought of him as a dumbshit freshman who didn’t hustle back on defense and had a tendency to bleed out a little too quickly towards the opposing basket after a shot went up or we were in transition. Now he’s reagarded as one of the best defensive minds in the basketball world.
Go figure.
dberri
May 26, 2010
Kevin,
I think is see a potential guest post topic. The scouting report on Thibodeau from someone who knows!!
kevin
May 26, 2010
Well, Dave, to be fair to Thibo, he was only 19 or 20 when I knew him. I think after all these years I have come to terms with my experience with him. He was a 3/4 for us. He was really too small defensively to carry the 4 position but was very clever in getting his shot up over the tall trees in traffic. He could overpower the other 3’s in our league underneath. He was also very tough on the boards, especially the offensive boards (we always th9ought he was better on the offensive boards because that’s where he could smell easy points).
He was a crappy ballhandler and wouldn’t know a pass if it kicked him in the shins. And the aformentioned defensive shortcomings were a constant source of friction between us elder statesmen and the brash but clueless young freshman. I was a 1/2 and sometimes I would find myself freezing him out on offense, just to send a message that he better get his ass back on D. It pissed me off he was always slapping his hands together for the ball but leaving me constantly facing 2 on 1 and 3 on 2’s, with their 3 on one wing. I silently seethed but our center, who was the team captain (first name Jon. Incredible athlete) and not a person you wanted to get on the bad side of, would openly rip on him verbally. It never happened but I was afraid Jon would lose it one day and beat the crap out of Thibo. And Jon was just the man to do it too.
Maybe all the verbal poundings he got then sank in at some point. But I can tell you, I sure wasn’t there to witness it.
Marparker
May 27, 2010
His wiki entry:
One of the finest defensive coaches in the league, he helped the Houston Rockets rank among the Top 5 in the league in scoring defense and field goal percentage defense from 2004 to 2007,[1] and has helped his team finish in the league’s Top 10 in team defense 15 times.[2] He coached in 87 playoff games and was part of the 1999 NBA Finals as an assistant coach with the New York Knicks prior to joining the Celtics, with whom he won the 2008 NBA Championship.
*******************
Or to paraphrase. Tom Thibodeau has coached Marcus Camby, Yao Ming, and Kevin Garnett.
*****************
I don’t think those guys will be walking through that door in Charlotte.
kevin
May 27, 2010
The negativity of your post sucks, Marparker. It stinks and it sucks and it stinks…:)
kevin
May 27, 2010
Speaking of Pitino, is he one of the great egoists of our age? I mean, how else to explain the title of his biog “Success is a Choice”. Like, all those boosters who bought with cars and clothes and hookers the players he won a championship with didn’t have anything to do with it. It was all him.
So, I’m assuming that since success was a choice for him at Kentucky, can we also assume that failure was a choice for him with the Celtics?
Now THAT stinks and sucks and stinks.
marparker
May 28, 2010
took me a minute to get the joke