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. In this installement (see the first, and a follow-up) of a TrueHoop series based on discussions with Winston, he begins by discussing how his metrics treat some of the NBA's elite names.
Tracy McGrady is a player who has never helped his team as much as people thought. Allen Iverson -- for one or two years he was really good.
The best player of the decade, though, I'd say, was Kevin Garnett. We have a rating over the last eight or nine years, and Garnett comes out number one. And I think everybody else [other stat experts] has that too, so that's nice.
Although I don't like Garnett. When I watch on TV, he's turning too edgy. Chippy attitude.
Another guy who is totally overrated is Amare Stoudemire. I mean, he's a stat stuffer. Troy Murphy gets great stats, but never does much for the team. (UPDATE: Winston amends this statement: "With Golden State Troy Murphy was a stat stuffer who did little to help the team but with Indiana the last two years he has improved to where he is an above average NBA player.") There's a bunch of guys like that.
Andre Iguodala, though. Whenever he's on the court for Philadelphia, they're great. Whenever he's off, they suck. God knows why he's a good player. I watch him play, and I don't know. (More on Iguodala.)
Jason Kidd is a little like that, but you can see why he makes guys better. But not Iguodala.
Sometimes I feel like I can see Kidd's greatness, but other times, at this stage of his career, I can't.
Kidd can't guard a fast guard. They go right by him like he's standing still. They always did. Against Chris Paul ... Jason Kidd might as well be standing still on defense.
But the interesting thing: Devin Harris can nail Tony Parker. But Steve Nash can beat Devin Harris. But Parker can beat Nash.
It's not transitive. We can show that. That's really interesting. That shouldn't be. But it is. There are probably a lot of other things like that.
If coaches see other examples of things like that, we can back them up with data. Del Harris really got to like us, I think, because a lot of times our numbers confirmed what he thought. It's hard to argue with the numbers when you've got a full amount of data on it.
Last year [Maverick assistant] Terry Stotts did a really great job asking us questions. Before the Spurs series, they asked us about Antoine Wright. He's not on the team anymore, thank god. OK, he had a bad rating in our system. But the fascinating thing was, when he played small forward, he was good. When he played shooting guard, he was terrible. So we can break that down. I can find every combination where he was small forward and he was good. Every combination where he was shooting guard he was terrible.
Against the Spurs, they used him as a small forward and he was great. Every time he played for Howard at small forward, they killed the Spurs.
Things like this ... I needed the coach to ask me the question because I would have never thought of it. You don't just throw the numbers at the coach, because, I mean, 500,000 numbers! But if the coach understands what he's doing, and says "I think Antoine Wright can play small forward can you tell me if that's true?" That's how you use the stuff.
What's the state of data for determining good and bad matchups?
It's hard. Because in the regular season, you don't have enough minutes that those two players have played against each other. In the playoffs, by the third game of the series, you pretty much have a good idea. But in the regular season, until they've played like the fourth game against each other ... then it might be worth something.
I don't know if last year's data on matchups ... I could use the last five years of the Spurs' data, which wouldn't be that bad. It's Manu Ginobili, Tony Parker and Tim Duncan. But is Ginobili healed? I don't know. I could tell last year he was never going to get better.
But during the playoffs, that matchup stuff ... you can say that when Paul Pierce, Rajon Rondo and Ray Allen are on the court; I can pull up every minute, and I can see what works. You can see, you'd want to do that. If you're going to Game 7, maybe you can find a combination of your players that can handle the killer combination for the Celtics. There's always something.
Against the Nuggets, J.R. Smith and Chris Andersen just really killed the Mavericks [in the 2009 playoffs]. But, see, Dahntay Jones -- the Pacers signed him, but I think he's terrible. Every time he was in with those four good players, he made them bad. I mean, the Pacers thought he was better than Jarrett Jack. I don't get it.
Why did Cleveland lose to Orlando? I did a post on that. Everybody thinks Mike Brown did a bad job coaching, right? OK, when Ben Wallace played, they lost by a point a minute.
They have a stat guy who does this kind of work.
They used to have Dan Rosenbaum.
They still have Dan Rosenbaum.
Then ask Dan Rosenbaum why he let this happen. I would never let his happen.
The 57 minutes when Ben Wallace played, and they didn't have Joe Smith in there, they lost by 58 points.
How could they let that happen? Ben Wallace didn't have to play. He totally sucked. The rest of the series, the Cavaliers won by 43 points.
Didn't you feel like Cleveland sunk into Lake Erie when the Cavaliers lost that series? It didn't have to happen. The 57 minutes they lost by 58 points ... you didn't have to play Ben Wallace. They had Joe Smith. When Joe Smith played instead of Ben Wallace they did OK.
How could they not know that?
Maybe they knew it and the coach didn't listen. I mean, I should be fair here. But I mean, we would have definitely told that to the coach. When I send the analysis to the coaching staff, I'll put little tidbits in like that.
But I'll tell you what, the Cavaliers are going to be good this year, and it's not because of Shaq. It's because of Parker. Anthony Parker is really good. What happens with Delonte West wit
h this thing ... I don't know.
Do they need Delonte West?
I can tell you that Anthony Parker was really good for Toronto. Not every year is he really good in our system, but he's usually really good.
Delonte West was an average player who plays a lot of minutes. If he can't play, and Anthony Parker plays ... well, they're OK.
Everybody used to think you had to play a five [center], a four [power forward], a three [small forward], a two [shooting guard] and a one [point guard]. But Mike D'Antoni changed all that. He made Boris Diaw play every position. And I think that's ingenious. I mean, people make these stupid molds.
Even the Mavericks, when they play Brandon Bass and Dirk Nowitzki, I would say "when you play Bass at the five" -- and they would say "no it's Dirk at the five and Bass at the four ..."
Who cares? I mean, the two guys are on the court. I don't care. They don't know they're supposed to be the five or the four. They just know they're supposed to play basketball. But people always think that way, and D'Antoni did a nice job of changing that, I thought.