Musings on Sloan: Inside Man

Posted by Rob Mahoney on March 10, 2010 under Commentary | 2 Comments to Read

Before this season began, I had the pleasure of chatting a bit with the Mavericks’ newly-christened stat head, Roland Beech. We haven’t had the opportunity to check in with him since, but Mark Cuban provided us with an update at the Sloan Sports Analytics Conference. Marc Stein asked Cuban about Beech specifically during the keynote basketball panel, entitled “Basketball Analytics,” and here was Cuban’s response:

If we were going to get to the point where we were able to collect more data and integrate the coaching staff more fully into using it, as opposed to just getting daily reports that we could sit down and review with the coaching staff, I wanted to have somebody there…on site that they could talk to at any time. Even during the game.

He talks to the players, don’t get me wrong — Dirk will give [Roland] as much of a hard time about his haircut as he gives anyone else. But that’s the type of thing where you don’t want too many voices. So we work back through the coaches.

Beech is officially a member of the Mavericks’ coaching staff, and he sits near the bench, often alongside Cuban, during the games. But even more important than the capacity in which Beech is working with the team is the note that Cubes touched on at the end, and one that was prevalent throughout the entire conference. Having the right statistics is one thing, and finding the right way to utilize them is something else entirely. You can’t just hammer raw data into a player’s head through their ear, it needs to be broken down. It needs to be analyzed.

Mike Zarren, Assistant GM and Team Counsel for the Celtics, hammered that point home perfectly:

I don’t know what it would mean to a player to tell him that “Some guy’s adjusted plus-minus is plus four.” I mean, how does that change what he does on a particular night? It doesn’t mean anything.

The most important thing to remember with advanced stats in basketball is that everything is contextual. That’s where the power lies. It’s the decoder that translates measures of all kinds into something usable, something real. It’s what separates those with an understanding of how to use the data — those who classify statistics as tools and never assume any metric to be omnipotent — from those who do not. The measures provided are simply descriptors of what we’ve seen on the floor. There are assigned values, calculations, and analysis that go along with those measures, but that’s simply another way to make sense of the information available.

But somewhere between that “adjusted plus-minus [of] plus four,” and the players on the court is a process. It involves the aforementioned decoding, it involves lineup analysis, and it involves evaluation of situational effectiveness. And it certainly involves Beech, the man working behind the scenes to ensure that the Mavs don’t miss a beat in the statistical revolution. Dallas is among the first franchises to not only embrace analytical research, but to have a full-time advisor on hand. Take pride, Mavs fans, because while franchises around the league are cavemen that run in fear of fire, your team, and more specifically your owner, realizes the value and application of it.

That level of analysis is the future of professional basketball. As teams accumulate more and more data, the advantages will become even clearer. There’s just so much that a coach can do when they know what Rick Carlisle knows. If basketball really is a game of match-ups, then the Mavs’ brain trust gives them a huge match-up advantage almost every night out; few teams are as  progressive as your Dallas Mavericks, and that’s something to hang your hat on.

  • Kirk Henderson

    That’s a good post. Thanks!

  • Blaine

    While I agree that its good the Mavs ar into this, lets not just overlook the downside. There was a post on TrueHoop this weekend that brought up how Avery over used the analysis, which resulted in us imploding in the first round in 2007.
    Common sense in that series would say, “Hey we have two big men, they have 0, sure they might like to run a lot, but we’ll kill them on the boards and get a lot of offensive rebounds to keep them from getting into transition” On top of that, Golden State was Damp’s previous team, and there was bad blood there. That doesn’t show up in the stats. I remember screaming at Avery to put in Damp bc he was gonna play a vendetta game. When he played in game 2, he was outstanding, but Avery was so into the analytics that the matchups that looked good on paper didn’t work on the court.
    Obviously, I’m still rather upset with the way Avery killed that 07 team. I still think there’s just too much more that goes into a game than just these analytical stats. Sure, they’re important, but when you think about only them, and forget intangibles, well, I guess you end up working for ESPN and never getting another head coaching job because you ask for too much money.