Wayne Winston is a professor at Indiana University and for the last nine years he has been Mark Cuban's stat guru for the Dallas Mavericks. Winston's recently published book "Mathletics," explains much of his work -- complete with formulas and spreadsheets. Earlier in the series (which is being discussed elsewhere, too) we have learned about:
Post 2: The player of the decade, Andre Iguodala's curious production and why the Cavs lost.
Post 3: The amazing Danilo Gallinari, buzzer-beating 3s, the perils of starting Greg Oden and great Laker lineups.
In the final installment, Winston focuses on coaches and referees:
On Mike D'Antoni and Going Small
Mike D'Antoni, I think, is the guy who changed the game. You spread the floor and go small, and these big guys become dinosaurs. That was a stroke of genius. Didn't win a championship, which is too bad, but that's the biggest change in the game I see.
Who'd have thought that Brandon Bass, Dirk Nowitzki, Josh Howard, Jason Terry and Jason Kidd is the Maverick's best lineup? It hardly makes sense. But for the last two years, that's been by far the best thing they've got.
On Tim Donaghy
They should have spotted him. I really feel there's a big debate. The NBA is more secretive about how they analyze referees than we are in chasing Osama bin Laden.
What I looked at is ... most people think that what he did was determine who won games. He didn't. He was into total points.
So I went back to every game where the line moved. If the line moves a lot, people bet money. And in those games, there were way more free throws called than you'd expect. And that's how you'd make scores higher. To make a game go over the over number, you would call free throws. Basically, he did it.
The NBA will never tell you who made which call. But if you make public, in those games, how many of the calls were made by Donaghy ... that's what they should be checking, and maybe they are.
But they found out about Donaghy from the FBI, right? They should have known. They have the right data. I know they hired Las Vegas people, they hired people to do this stuff.
On Referee Bias
Are refs biased against teams? Like Joey Crawford, I don't think he was biased against the Spurs [as has been suggested].
Honestly, every team has certain refs that they hate and are convinced are out to get them. The way you should analyze that is: Does your team play worse with those refs?
With Jeff Sagarin, my best friend and my partner in this stuff, we would analyze: When a ref refs your game, across the years, does your team play statistically worse than expected? That could mean either he hates you, or his reffing is so antithetical to the way you play that it affects your style. Either way is bad.
Now, most of the time I think you'll find a ref has no significant impact on a team's performance.
Dan Crawford was a negative effect on the Mavericks, if you analyze it this way, in the playoffs. Given the record, the Mavericks think Dan Crawford doesn't like them. Whether he does or not, I don't know. But I mean, I think the NBA should be doing that, and I don't know that they are.
They need to come clean on what they do. If there's another ref who's found to cheat ... I think the reffing thing is very important. The integrity of the game: Without that we don't have a game.
For People Who Hate Advanced Basketball Statistics
Every time I write about statistics, a certain portion of TrueHoop readers get upset. Whatever it is that they love about basketball, all these formulas and statistics seem to ruin that.
And I'm not entirely unsympathetic.
For instance, I have written that I'm against the use of performance enhancing drugs because you want to root for players who are doing things that you understand. Running and jumping and dunking the ball. A player does a good job, and we cheer that. But if it's really that the guy at the BALCO lab did a good job ... not as fun. If it becomes about drugs, then what you'd be seeing on the court is really in large part the work of someone in a lab somewhere ... and maybe we don't want to buy a ticket to watch that.
I could see that it would be the same with stats. If this is where the game is being played now, maybe that's not exciting to some people. What do you say to them?
I see what you're saying.
I just think we're trying to give the proper information they need to make good decisions. I don't think that's a bad thing. Wouldn't you want your coaches to have all the data needed to make informed decisions, like you'd want the doctor to order all the tests before deciding what operation to do on you.
I mean, should the doctor not do the tests? I don't know. This is getting to philosophy 101.
I think the skeptics would say: "I don't want the word 'data' in a discussion of coaching."
It's like the Mona Lisa. You can't quantify its appeal. But I think most fans want their team to win legally. And the stats are not the be-all and end-all. They are just one factor. The coach's intuition ... Rick Carlisle last year. Oh god last year was he good. I watch every game on League Pass. There were ten games last year that, going into the last five minutes they should have lost. And they won nine of them. He had all the information, but he just pulled the strings.
On Using Stats in Hiring a Coach
We did a study that showed he was the best coach in the NBA the last ten years. It's part of why Mark hired him.
How would you evaluate a coach?
We looked at every player that was traded to a coach's team. We looked at their rating the year before in our system and then how they did when that coach got them. And look at the difference. Carlisle crushed everybody. We sent that to Mark [Cuban] and a week later he hired Carlisle. (And I think he said that was part of it -- it wasn't the whole thing.)
But it wasn't close. He was at least two standard deviations better than average. He jumped out and there was nobody else close. He kicked everybody's butt on that.
Phil Jackson ... is he really good or did he just have Michael Jordan and Kobe Bryant? That debate will go on forever. It kills me that we don't have the data going back before 2000 to try to answer that.