Dave Berri on Coaches

Moderator: Doctor MJ

sp6r=underrated
RealGM
Posts: 17,197
And1: 8,517
Joined: Jan 20, 2007
 

Dave Berri on Coaches 

Post#1 » by sp6r=underrated » Sat Apr 7, 2012 6:07 pm

Bryan Colangelo, in his 15 years as an NBA general manager, has employed eight coaches. And the way David Berri sees it, all that firing and hiring has mostly been in vain.

“Most NBA coaches have no impact on the performance of their players,” said Berri.

David Berri isn't an NBA GM. He’s a PhD in economics who recently co-authored a book that takes a dim view of the NBA’s coaching carousel. Stumbling on Wins, a wide-ranging work that attempts to stamp some empirical truth on the sporting world’s murky mumbo-jumbo, contains key results of a 2009 academic study that attempted to determine the impact of NBA coaches. And while the study concluded that some coaches make a difference — Phil Jackson, the 10-ringed Zen master currently helm of the L.A. Lakers, is heralded by the numbers as a transformative genius — the overwhelming gist is that an NBA team is only as good as its players.

“If the Raptors got a new coach, what would he say to (Andrea) Bargnani? ‘We'd like you to get 15 rebounds a game?’ ” Berri said in a recent interview. “Bargnani would probably say, ‘Well, I'd like to get 15 rebounds a game, too. But that's not going to happen unless you make the game 300 minutes long.’ That's the thing about (NBA players): You can't dramatically alter their performance.”

NBA teams, of course, often attempt to sell an opposing view. A new coach equals new hope, no matter that the players haven't much changed. But a read through Stumbling on Wins, which includes the key conclusions of a study of 62 NBA coaches from 1977-78 to 2007-08, suggests fans should be weary of such pitches.

While Raptors coach Jay Triano wasn’t included in the study, some of his predecessors were analyzed. The likes of Sam Mitchell and Lenny Wilkens, said Berri, “had no impact at all” on the performance of their players. The study suggests, on the other hand, that a team hiring Jackson could expect 17 additional victories in his first year on the scene.

Other coaches who ranked highly in the study: San Antonio’s Gregg Popovich (good for nearly 16 additional wins in his first year with your team), Golden State’s Don Nelson (plus-11 victories in his first year), Flip Saunders and Jim O’Brien. Only one of the eight coaches who have worked under Colangelo made the grade. The study suggests that the theoretical arrival of the late Cotton Fitzsimmons, who worked under Colangelo in Phoenix, could improve a team by about 16 wins in year one.

“Players tend to get better when they come to Phil Jackson,” write Berri and Schmidt. And players, Berri added in an interview, don't get worse after they leave Jackson, which suggests their improvement can't merely explained by their presence on the same floor as, say, Michael Jordan or Kobe Bryant, or their role in Jackson's triangle offence. The study found, for instance, that players saw drop-offs in production after leaving the unorthodox system run by Nelson.

What do NBA GMs think of the work of Berri and his colleagues? Berri said he has spoken to a few NBA executives, and he has done a “little bit” of paid number crunching for NBA teams. But he said working for a franchise isn't his goal, nor is gaining the approval of front-office executives.

“I think people in the NBA read what I write. They just won't publicly react to it,” Berri said.

Colangelo, for his part, said he has read only parts of the first book-length project in which Berri was involved, 2006’s The Wages of Wins. The GM has yet to crack Stumbling on Wins. But told of the crux of the NBA coaching study therein, Colangelo took umbrage.

“There's too many human elements in our game to rely solely on numbers to determine, ‘Oh, one guy's a good coach and one guy's not a good coach,’ ” said Colangelo. “There are all kinds of ways you can cut up data and statistics.”

If one detects skepticism emanating from Colangelo's corner office to Berri’s ivory tower, maybe it's mutual. Berri spent part of an afternoon this week mocking what he perceived to be the absurdity of many an NBA reality, including the coach-run sideline huddle.

“I saw where Adrian Dantley (coaching the Denver Nuggets) kept saying, ‘We're giving up too many layups.’ And it's like the players were sitting there saying, ‘Yes, we know. We don't want them to get layups, either. Why don’t you tell us how to stop them?’ ” said Berri. “They keep putting the mic on (Boston coach Doc Rivers) and he keeps saying the same thing, ‘We've got to play like a team.’ I mean, I’m getting tired of that. I can't imagine what the players are thinking. . . . You've got to imagine (Celtics forward) Kevin Garnett, when Doc Rivers is talking, is, like, “Are you done yet? Can I go play? Yep. Play like a team. Got it.” That's coaching.”


http://www.thestar.com/sports/basketbal ... ort-claims
User avatar
Dr Positivity
RealGM
Posts: 59,691
And1: 15,484
Joined: Apr 29, 2009
       

Re: Dave Berri on Coaches 

Post#2 » by Dr Positivity » Sat Apr 7, 2012 11:19 pm

It's funny he uses Toronto as an example, because the Dwane Casey hiring clearly has led to extra wins, they're at 20 out of 56 vs 22 out of 82 last year, despite Bargnani missing half of this season
Doctor MJ
Senior Mod
Senior Mod
Posts: 50,597
And1: 19,350
Joined: Mar 10, 2005
Location: Cali
     

Re: Dave Berri on Coaches 

Post#3 » by Doctor MJ » Sun Apr 8, 2012 7:48 pm

Article is pretty old, no? I wonder if Berri would say anything different now, because it's been a telling couple of years.

First, we've had a good amount of unexpected success that seems to require us to credit coaching. Thibs first and foremost. Perhaps the answer is that if you hire a former NBA player with no coaching experience (as NBA teams have a weird tendency to do), he literally isn't going to do anything, but the wonks in the background actually can make a huge difference. Thibs has arguably been more valuable than any player over the past two years.

Second, we're seeing that without training camp, NBA players look terrible relative to normal NBA standards. While the skill to run a decent training camp is hardly unique, it's a little bit misleading to imply that coaching doesn't matter, because that makes people think players can just walk on the floor and do their thing. There is a difference between coaches having no impact, and coaches typically having similar, but positive, impact.
Getting ready for the RealGM 100 on the PC Board

Come join the WNBA Board if you're a fan!
User avatar
EvanZ
RealGM
Posts: 12,649
And1: 3,182
Joined: Apr 06, 2011

Re: Dave Berri on Coaches 

Post#4 » by EvanZ » Sun Apr 8, 2012 9:59 pm

Dr Mufasa wrote:It's funny he uses Toronto as an example, because the Dwane Casey hiring clearly has led to extra wins, they're at 20 out of 56 vs 22 out of 82 last year, despite Bargnani missing half of this season


Despite? ;)

Tibs deservedly gets a lot of credit. Don't forget about Doug Collins, though. That team hasn't added any significant free agents, yet it's much, much better since he arrived.
I was right about 3 point shooting. I expect to be right about Tacko Fall. Some coach will figure out how to use Tacko Fall. This movement towards undersized centers will sweep ng back. Back to the basket scorers will return to the NBA.
sp6r=underrated
RealGM
Posts: 17,197
And1: 8,517
Joined: Jan 20, 2007
 

Re: Dave Berri on Coaches 

Post#5 » by sp6r=underrated » Mon Apr 9, 2012 12:57 am

Doctor MJ wrote:
Second, we're seeing that without training camp, NBA players look terrible relative to normal NBA standards.


I am assuming by playing terrible you mean the lower effectiveness of offense this year in comparison to the last few years. If not what follows will be inaccurate.

Nothing to do with this lockout season, but the opinion you articulated here is why I think eras that have high defensive effectiveness and low pace end up being underrated. The unstated assumption underlying your opinion is that the most important sign teams and players are playing well is when they are scoring efficiently and playing poorly when they aren't. It is just as logical that the most important sign teams and players are playing well when they hold the opposition to inefficient views. While that later view is just as plausible as the former view, I think it is a distinctly minority opinion. In short I think most people agree with you.

As a result I think the historic perceptions of eras will be impacted by whether offense dominated. Teams that peaked in offensive era will be regarded as having great supporting cast and having to beat great opposition. Teams that peaked in defensive eras will be regarded as peaking when the league quality is at a low.
User avatar
Rerisen
Retired Mod
Retired Mod
Posts: 105,369
And1: 25,052
Joined: Nov 23, 2003

Re: Dave Berri on Coaches 

Post#6 » by Rerisen » Mon Apr 9, 2012 2:48 am

Doctor MJ wrote:First, we've had a good amount of unexpected success that seems to require us to credit coaching. Thibs first and foremost. Perhaps the answer is that if you hire a former NBA player with no coaching experience (as NBA teams have a weird tendency to do), he literally isn't going to do anything, but the wonks in the background actually can make a huge difference. Thibs has arguably been more valuable than any player over the past two years.


Thinking that perhaps Thibs impact has been even further highlighted by that the previous coach (Del Negro) was a negative influence, setting up a false baseline for the team which was less than it should have been. Plus/minus stabs might think VDN was a neutral... but by firsthand look... he was bad.

I think Vinny actually lost the Bulls team pretty early in 2010, where they had a whole string of games where they were repeatedly blown out, and playing with little effort. Self correction after that was more just the inherent competitive qualities of some of the players on the team (that Thibs would later put in perfect harmony) rather than any kind of regaining respect of Del Negro. Bulls were pretty much on auto-pilot without much direction after that. Toward the end of the year it was down to the GM having it out in physical melee with the head coach over a player's minutes. :nonono:

So going from bad to great, yeah a good coach can really stand out.
mysticbb
Banned User
Posts: 8,205
And1: 713
Joined: May 28, 2007
Contact:
   

Re: Dave Berri on Coaches 

Post#7 » by mysticbb » Mon Apr 9, 2012 8:58 am

Doctor MJ wrote:Article is pretty old, no? I wonder if Berri would say anything different now, because it's been a telling couple of years.


Yes, the article is older. And as typical for Berri there are many holes in that article. His method and data does not allow to conclude that coaching has no impact, the only conclusion possible is to say that in average coaches are average (who would have thought something different?). The only comparison he makes is the performance with coach A versus performance with coach B. Berri neither has the skills nor the knowledge to seperate here the influence by the coaches on the player performances.

And I'm really amused by him not answering the question about what GM's are thinking of his work. No GM takes his player performance and coaching performance "analyses" serious. It is a big hoax meant to sell books, nothing more. His articles are mainly published in journals with low impact factors, a few people are reading it, basically no people besides himself and his co-workers are citing his articles. Berri is a wannabe who seems to be blessed with diligence.

Also, his examples suggest that Berri never played any kind of sports on a competitive level.


@sp6r=underrated

Both, the recent season and 1999 showed a significant drop in quality. That has nothing to do with better defense, just with far worse and more inconsistent offensive performances. For 1999 you can see that one year before and one year after the ORtg/DRtg is higher. We can very well assume (unless some rule changes are happening) that we see an increase in ORtg/DRtg for next season again. In fact we saw a slightly constant increase througout the year, and we can also notice increased playing level.
Chicago76
Rookie
Posts: 1,134
And1: 228
Joined: Jan 08, 2006

Re: Dave Berri on Coaches 

Post#8 » by Chicago76 » Mon Apr 9, 2012 4:25 pm

EvanZ wrote:
Dr Mufasa wrote:It's funny he uses Toronto as an example, because the Dwane Casey hiring clearly has led to extra wins, they're at 20 out of 56 vs 22 out of 82 last year, despite Bargnani missing half of this season


Despite? ;)

Tibs deservedly gets a lot of credit. Don't forget about Doug Collins, though. That team hasn't added any significant free agents, yet it's much, much better since he arrived.


A big part of the problem with coaches is that there are too few of them to draw any sort of meaningful statistical inference of who is good vs. who is bad.

A lot of it boils down to a coach getting the right opportunity at the right time. The Bulls are a better team under Tibs w/ some excellent defense. They were doing pretty well under Skiles too. If there was no lag between the tenure of Skiles and Tibs, then that credit goes away (forget for a moment that players have changed over that period too). Does that make him any less of a coach?

Some coaches are really good at teaching a particular philosophy, and when they have the pieces to really make that work, they do well. When they don't, they tend to do about the same.

A great coach in one scenario can look pretty average in another and vice versa.
ElGee
Assistant Coach
Posts: 4,041
And1: 1,202
Joined: Mar 08, 2010
Contact:

Re: Dave Berri on Coaches 

Post#9 » by ElGee » Mon Apr 9, 2012 4:54 pm

sp6r=underrated wrote:
Doctor MJ wrote:
Second, we're seeing that without training camp, NBA players look terrible relative to normal NBA standards.


I am assuming by playing terrible you mean the lower effectiveness of offense this year in comparison to the last few years. If not what follows will be inaccurate.

Nothing to do with this lockout season, but the opinion you articulated here is why I think eras that have high defensive effectiveness and low pace end up being underrated. The unstated assumption underlying your opinion is that the most important sign teams and players are playing well is when they are scoring efficiently and playing poorly when they aren't. It is just as logical that the most important sign teams and players are playing well when they hold the opposition to inefficient views. While that later view is just as plausible as the former view, I think it is a distinctly minority opinion. In short I think most people agree with you.

As a result I think the historic perceptions of eras will be impacted by whether offense dominated. Teams that peaked in offensive era will be regarded as having great supporting cast and having to beat great opposition. Teams that peaked in defensive eras will be regarded as peaking when the league quality is at a low.


From a neutral starting point, it might be fair to say "why be partial to (bad) offense and not (good) defense?" However...

Look at the NBA tendencies across a season. The pattern is for teams to increase their offense/shooting numbers as the year progresses, which implies one of two things:

(1) that there is some kind of cohesion and chemistry at work on offense (something related to coaching)
(2) that defenses are getting tired

I have a real hard time with the second explanation, because it would require offenses and defenses to be at 100% operating efficiency from day 1 (or equal operating efficient, only to crest up to 100% together), then for offense to remain there while defense dropped. This seems highly improbable given that in all kinds of basketball competitions for years, team's thrown together without cohesion and preparation struggle on offense out of the gate (empirically, certainly many HS and college coaches will allege this).

But we can also look at the lockout seasons as two large points to contradict #2. If fatigue or conditioning were responsible for the yearly trends we see, then the 1999 out of shape players would have struggled mightily on defense producing an increase in offense. We saw exactly the opposite. Then we saw offense increase as the crunched season progress way more than it normally does during a season. The exact same pattern held true in the 2011 lockout season.

All of this makes perfect sense, of course, because even today, defensive cohesion is theoretically simpler than offensive cohesion -- there is no ball, and really there are no plays (and no fine motor skill in shooting).

All of this is not to denigrate a period like 2004 (although rules were relevant in that strange season), but to note that the lack of training camps are a serious issue for quality of play, particularly on offense, and not as a result of great D, but because of choppy O.
Check out and discuss my book, now on Kindle! http://www.backpicks.com/thinking-basketball/
eitanr
General Manager
Posts: 8,318
And1: 291
Joined: Nov 26, 2003

Re: Dave Berri on Coaches 

Post#10 » by eitanr » Mon Apr 9, 2012 11:04 pm

I do agree though with the concept though that coaches do make very little difference. If we got rid of all the coaches in the league would the standings change much? If Miami lost Coach Spo for the year and essentially had no coach at all, would they not be favored to win the title?

I happent to believe there are a few coaches like Pops, Phil, and I'd agree on Tibs that make a difference, but it's hard to distinguish the other 27 from one another, except maybe Del Negro.
Read the best NBA Articles on the Web right here, delivering innovative insights and a unique perspectives on all the happenings of the league.

http://fullcourtanalytics.blogspot.com/
giberish
RealGM
Posts: 15,824
And1: 5,808
Joined: Mar 30, 2006
Location: Whereever you go - there you are

Re: Dave Berri on Coaches 

Post#11 » by giberish » Tue Apr 10, 2012 12:17 am

There isn't a big difference between average coaches - just like there isn't a big difference between average rotation wing players.

There are a few very good coaches, a few very bad coaches (not just VDN), a bunch of averageish coaches, and some averageish coaches who have 'lost' their teams to the extent that they are worse then even the bad coaches (at which point you do have to make a change - if only to bring in a different average coach).

That Berri doesn't understand the difference between the basic stuff mic'd for the media and actual coaching just makes him look silly.
User avatar
tclg
Lead Assistant
Posts: 5,194
And1: 0
Joined: Jan 15, 2007
Location: Chicago

Re: Dave Berri on Coaches 

Post#12 » by tclg » Tue Apr 10, 2012 4:08 pm

wow jackson with 17 wins is pretty amazing

Return to Statistical Analysis