A couple years ago, Michael Lewis visited Google to talk about his new book The Blind Side. At 50:59, Lewis started talking about the Houston Rockets, and their then-new GM Darell Morey. In particular, Lewis was intrigued by the fact that the Rockets had orchestrated a big trade with the Memphis Grizzlies, and the central player in the trade was Shane Battier. Battier, interestingly, was a player the Grizzlies weren’t particularly interested in: he didn’t score a lot of points, didn’t have a ton of assists, didn’t block a lot of shots. But the Rockets had determined that, oddly enough, his team tended to play much, much better when Battier was on the court.
Lewis revisited that theme in a Sunday Times article this weekend called The No-Stats All Star, and the result is a fascinating essay on the importance of data in competetition. The Rockets are playing a different game than many other teams, in much the same way Billy Beane and the Oakland A’s were playing a different game in the 90s (the subject of Lewis’s fantastic Moneyball).
The Rockets have figured out how data can not only help Battier be a better player, but can actually convert their opponents’ biggest asset (in one example, the Lakers’ Kobe Bryant) into a liability. Executing still matters, of course – and a positive outcome doesn’t inevitably result. Lewis recounts how the Rockets assembled a huge mountain of data about Kobe Bryant. Does he go to the left or to the right. Does he score more off the dribble or from a pass. What’s his shooting percentage from 18 feet out. And so on.
What makes Battier so unique is that he wants that data, absorbs the data, and then puts it to good use:
People often say that Kobe Bryant has no weaknesses to his game, but that’s not really true. Before the game, Battier was given his special package of information. “He’s the only player we give it to,” Morey says. “We can give him this fire hose of data and let him sift. Most players are like golfers. You don’t want them swinging while they’re thinking.”
When Michael Lewis was at Google in ’07, I asked him whether coaches wanted their players aware of these sophisticated methods for evaluating their performance. The parallels to how we think about (and use) data to inform decisions at all levels of Google seemed pretty obvious to me. My question and Lewis’s answer start at 56m52s in:
Short answer: Beane wouldn’t want his players to be concerned with the data, wouldn’t trust that they could put it to good use. Just like Morey refers to most players being like golfers: “you don’t want them swinging while they’re thinking.”
Back to Battier. Check out what effect the Rockets data collection has when Battier is able to apply it in the game:
“If [Kobe] has 40 points on 40 shots, I can live with that,” Battier says. “My job is not to keep him from scoring points but to make him as inefficient as possible.” The court doesn’t have little squares all over it to tell him what percentage Bryant is likely to shoot from any given spot, but it might as well.The reason the Rockets insist that Battier guard Bryant is his gift for encouraging him into his zones of lowest efficiency. The effect of doing this is astonishing: Bryant doesn’t merely help his team less when Battier guards him than when someone else does. When Bryant is in the game and Battier is on him, the Lakers’ offense is worse than if the N.B.A.’s best player had taken the night off. [emphasis mine]
Not a bad way to think about how to compete: figure out what data matters, collect it, sift it, and apply it. Don’t be afraid to think while you swing. Indeed, if you can pull that off, you can often negate a competitor’s advantage, and even build an advantage of your own.