Ben Alamar once worked for the Thunder as a analytics consultant. Analytics is the discovery and communication of meaningful patterns of data. Simply put in sports, analytics is the deep study of statistics.
Now Alamar is a professor of management at Menlo College in California, and he has written a book that soon will be available: Sports Analytics: A Guide for Coaches, Managers and Other Decision Makers.
The book should be a fabulous peek behind the Thunder veil. Sam Presti’s secretive organization is wondrously successful but maddeningly frustrating for followers of the team who want to learn more about how and why decisions are made. Presti seldom speaks in detail, and his lieutenants never speak at all.
But last week, Alamar spoke at the Sloan Sports Analytics Conference in Boston, and he sat down for an interview with Grantland’s Zach Lowe, which you can view here. It’s a fascinating look at some inside Thunder decisions. Here’s a sampling:
* In the 2008 draft, the Thunder debated between Russell Westbrook and Brook Lopez with the fourth pick. The Thunder chose Westbrook. Rightly, most would say.
“It was an analytics-supported decision,” Alamar said. “They really liked Russell, for a lot of basketball reasons. Personality, performance, everything. The fundamental question was, could he play point guard? He had not played a lot of point guard in college.”
Enter analytics. “We had to do some data collection,” Alamar said. “Look at the film. There was a time he was the primary point guard (at UCLA), because Darren Collison was injured. So we were able to use that data and other instances where he was making decisions with the basketball. And compare those decisions and results, when passes resulted in shots, how that impacted the probability the shot was going to be made.
“We compared Russell’s ability to do that to Collison’s, to the point guards in the draft, like Derrick Rose, and to NBA level point guards. We could look in the NBA and see how Steve Nash did.”
Analytics go deep. “We correct for the general ability of the team and for distance of the shot and all that kind of stuff,” Alamar said. “What our analysis showed, while Russell wasn’t Derrick Rose, he was better than Darren Collison in that metric, and he was better than O.J. Mayo in that metric, and he was on par with a lot of high-level NBA point guards.
“While that was not definitive, it was a piece of evidence to suggest that look, given the time and experience of doing this, he does have this ability. He can do this. One more piece of the puzzle to say yes, this is the guy we want.”
* Analytics is not bedrock science. It’s ever-evolving. Some of the sharpest stat geeks remain some of the most vibrant learners.
Alamar said his analytics showed Lopez to be a “fine” pick as well. But Alamar argued for Westbrook.
“With all the analytic work I had done on Russell Westbrook, I argued probably more vehemently than the number allowed me to in that situation, that Russell Westbrook was the guy,” Alamar said. “Which eventually comes back to haunt you, because Brook Lopez was a good player. We made the right choice, but in terms of my discussion of the two, I overstated my case. It was a learning process for me in that draft, as an analyst, in how far you can go, and the truth is, you should never go further than the data allows.”
* Part of the Westbrook/Lopez debate concerned position. NBA tradition says, it’s hard to find a center.
Analytics sometime search for which position is more central to success – point guard or center. But analytics go deeper than that.
“Little more specific, given where we thought we were going to go,” Alamar said. “We had Kevin Durant on the team. There was a plan where he was headed. It was pretty clear what was going to happen.
Given that we had that piece, what pieces do we want to add to that?
Given the analysis I had done, which player would have a bigger impact on you being an elite team, it was the point guard position.”
Good big men are the league’s most scarce resource, NBA people always say, but “but the truth is, that’s not necessarily the case,” Alamar said. “But even if they were the scarcest resource, that doesn’t mean they have the biggest impact.”
* Analytics have their limits and sometimes mean squat. The drafting of Serge Ibaka, for instance, from that same 2008 draft.
“At that point, the international data was so shaky, we weren’t utilizing any of the international data,” Alamar said. “Now, the data has become a little better. We have a little better track record of players, too, coming from the international game to the NBA and how they perform at different areas. But the international data is tough because the levels of competition vary so much, and how they move from team to team, league to league.”
Alamar says he doesn’t even know what went into the Thunder selecting Ibaka 24th overall in 2008.
“To be honest, what went into that pick, was not part of me,” Alamar said. “It wasn’t part of analytics.”