The state of basketball analysis

Saturday marks the 2010 MIT Sloan Sports Analytics Conference, which is also commonly referred to as (thank you, Bill Simmons) Dorkapalooza.

This conference, now in its fourth year, is a funny thing. On the one hand, it's a curio. Most of the NBA is not here, and some meaningful parts of the NBA is convinced that nothing that happens here really matters.

On the other hand, it's undeniably the epicenter of some very smart NBA thinking, and it's no wonder this year's conference has moved to a bigger venue and is still very very very sold out. I have talked to more than a few agents and executives who never bothered to come in the past, but after hearing about it, were determined not to miss it this time around.

If you're here, it feels like this conference matters.

But does it?

To help us understand the import and role of advanced statistics in today's NBA, I turned to Dean Oliver, who is one of the field's founders. He wrote one of the field's core books, "Basketball on Paper," and is the Denver Nuggets' director of quantitative analysis.

I heard the other day about an NBA executive who couldn’t fathom why he’d attend the MIT Sloan Sports Analytics Conference. How could that possibly help his team win? What’s your reaction to that?

My reaction to that is what team is it? What’s their record?

In all seriousness, that’s a common sentiment. This is something that’s new over the last six or seven years. It hasn’t been a part of basketball. Anyone who has been in basketball for the long-term is used to making decisions without this kind of analysis. Historically, some people have made very good decisions.

But it’s like the stock market in the 50s and 60s. The stock market used to function without these kinds of numbers, and some people did very well. But once this kind of analysis was introduced, there was a new way to inform decisions, and it changed things, and some new people were able to get in and do well.

It’s about having good ways to make decisions, to make decisions with input from the numbers, which have an independent opinion. If you can ask the right questions, you can find it’s wonderful to have an independent opinion to complement what you’re doing.

Does it feel a little bit cultural? Sometimes I feel like basketball is the realm of the jocks, some of who object to nerds elbowing their way into the conversation.

You say cultural. I say territorial. But I think those kinds of labels are not helpful.

There are quants who used to be jocks. There are jocks who could be quants. It’s a matter of opening your mind. Think what Billy Beane did in baseball! He was a baseball guy who was open to a new way of looking at things. Or [Nuggets vice president of basketball operations] Mark Warkentien. He’s been in basketball for a long time, but he’s willing to listen, which allows you to incorporate more information into the process.

It’s an interesting time. Just a couple of weeks ago, I looked at teams that have stats people integrated into the decision process. (Boston, Cleveland, Dallas, Denver, Houston, Oklahoma City, Portland and I may have included Orlando -- I’m not certain what they do exactly.) It was seven or eight teams. They had won 60% of their games, and that’s counting Houston, which has only won half their games because they’re missing Yao Ming and Tracy McGrady wasn’t playing.

The teams that don’t have quants won 40-some percent. And it was pretty linear … the more or less they had someone integrated into their decision making, the more or less they were at the extremes of winning and losing.

Now, is that because teams with smart GMs have quant people, and teams with smart GMs do well anyway? Or is that because the quant people are making a difference?

The question of correlation vs. causation is hard to speak on -- I don’t know what the other teams do exactly. We all play our cards pretty close to our chests. But at the places I’ve been, I feel I’ve made a difference. I think it should be causal. It could be that these are just GMs who are open to different opinions, though.

So the influence of this analysis more on personnel or coaching?

Insight can be useful for both. My guess is that so far it has been more on the personnel side but I don’t have a good enough sense of what the other teams do to say for sure.

I remember a couple of years ago I saw you at the draft lottery, which is during the playoffs, and you pointed out that just about all of the teams still doing well were teams with analysts. It seems like it makes a difference. And yet there’s a lot of doubt. Do you get tired of people asking whether or not what you do matters?

I think that’s an important question. If people are asking that question, it shows they might be willing to change.

A lot of things people do can be quantified. People say all the time that you can’t quantify heart, for example. When people say "can’t" that means they have already ruled out the possibility.

But in fact, measuring things is just another way to try to make decisions right, and I think we can measure a lot more than some might think we can.

So how do you measure heart?

I haven’t had to yet. I have been told many times you can’t, usually in a dispute about a player. But it would be a worthwhile quest. I’ve thought about it. It would be an intense analysis. You’d have to think about how heart translates into performance, and then see if you can describe elements of that.

It seems to me that wins are easy to notice. So things that consistently lead to wins ought to be noticeable in some way.

Wins are easy to notice. How you get to wins is not easy to evaluate. Doing a quant job is hard. You have to accept that the win comes from eight or nine players doing a ton of different things all over the floor. The coach and his staff all do their jobs. The advance scout has given a strategy the team should use … parsing out all the responsibility for that win is very very hard.

Occasionally people will find something that looks crazy, in PER or adjusted plus/minus or whatever, and use it to condemn the entire enterprise. That seems unfair.

It’s also unfair that, from the other side, people use one mistake to condemn an executive. It happens both ways. One single event never means anybody’s horrible. I think we need to give everybody a wide berth to make a mistake or two, or have things that look funny sometimes.

Is it inevitable that a decade from now this kind of analysis will be much more commonplace?

Inevitable is a strong word. But we might be able to say it’s inevitable that eventually there will be more of this.

You mentioned seeing me at the draft lottery. Two seasons later, we’ve had now almost three seasons of seeing teams with quant people having success. That’s a trend that’s hard to ignore. You’d think that at some point people would just start following that trend blindly.