Moneyball 2.0

Or should that be 3.0?:

The technology was originally developed to track missiles. Now, SportVU systems hang from the catwalks of 10 NBA arenas, tiny webcams that silently track each player as they shoot, pass, and run across the court, recording each and every move 25 times a second. SportVU can tell you not just Kevin Durant’s shooting average, but his shooting average after dribbling one vs. two times, or his shooting average with a defender three feet away vs. five feet away. SportVU can actually consider both factors at once, plus take into account who passed him the ball, how many minutes he’d been on the court, and how many miles he’d run that game already.

It’s big data in a relatively small pool, and it has the potential to impact everything about basketball, from how it’s coached, to how it’s recruited–even to how we calculate a player’s worth. Sportvision, another sports data collection system based on the same underlying big data idea, has already massively impacted baseball since it came into play in 2006. Now SportVU is generating more basketball data than anyone ever has. And its potential has only begun to be tapped–health care researcher Kirk Goldsberry, who recently wowed the stats geeks at MIT Sloan Sports Analytics Conference with his spatial analysis to determine the best shooters, has begun mining SportVU’s data for new insights. But only 10 teams in the NBA are currently using SportVU. Four of them made the playoffs. One even made it to the finals: The Oklahoma City Thunder.

Here is more, and for the pointer I thank Jim Olds.


Only 10 teams are using it, four made the playoffs, one made the finals.

This doesn't necessarily sound like a successful adoption of information. The choice to use it may have been highly correlated with existing strength of the team, and the desire was to improve chances of moving farther up the ladder.

Which teams, and what were their pre and post adoption statistics. I'd like to see a control group and treatment group here.

If it works, and all teams adopt, the better teams should have smaller marginal benefits than the lesser skilled teams. Higher scoring or more competitive games?

Would this promote league expansion since more players just below the margin would become NBA material? Or it's truly Moneyball, supranormal profits should induce entry.

Moreover, there are 30 NBA teams and 16 make the playoffs. 16/30 = 53% > 40%.

Also, the willingness of teams to pay for this system probably correlates with their willingness to pay the luxury tax and buy a better team by exceeding the salary cap. The whole Moneyball concept doesn't make all that much sense in the NBA, really -- you can build a good baseball team by identifying undervalued individuals. Basketball success is much more dependent on team chemistry, so you wind up with lots of trades that benefit all parties (this happens in baseball as well, but more in terms of trading instant success for prospects that will pay off in a few years). You have teams like San Antonio who spend a ton of money to find diamonds in the rough in foreign associations, but that's more about scouting than finding some set of statistics that other folks overlook.

This innovation looks likely to lead to a costly and socially-wasteful arms race among all pro teams -- that is, the relative advantage of investing in this technology will be cancelled out, since every team will feel compelled to install the SportVU system (and hire statisticians to review the massive amounts of data generated) to prevent other teams from using this technology to their advantage

As Yogi Berra said, "You can observe a lot just by watching."

In general, Bill James exaggerated the inefficiencies in the baseball system by focusing on examples, such as voting for league MVP by sportswriters, where evaluators didn't have much first hand information with games and thus had to go by hackneyed statistics like RBI totals. In contrast, voting by fans for home team MVPs were more sophisticated because people merely listening to 100+ games a year on the radio could tell pretty well who was most valuable.

When he finally got a job with a baseball team, James quickly won a World Series, but that was heavily due to his sluggers using steroids (a topic he religiously avoided writing about even though it was the most important story in baseball).

So how did the sportswriters fail to see the connection between the better players and the more relevant statistics? How did the public at large, as well as agents, players, managers and owners get duped into believing batting averages and RBI's were more relevant than OBP and slugging? It seems that after a hundred years of millions of people watching baseball much of the conventional wisdom was wrong, even to the point that the sportswriters themselves believed in the hackneyed statistics, and managers used hackneyed plays like the hit and run and having the 2nd batter in the lineup sacrifice bunt to move the runner on first to second in the first inning.

How duped were fans? Bill James crunched a zillion numbers to figure out who the greatest baseball player of all time is and came up with the shocking revelation that he was actually ... Babe Ruth (i.e., the most popular player ever).

The reality is that the contribution of the Bill James / Moneyball number-crunching is real but fairly marginal. Contrast that with how corrupt James' behavior was during the long steroid era:

Compare Bill James' avoidance of the topic of steroids -- a subject begging to be analyzed statistically -- all the way until the late 2000s with Thomas Boswell pointing out in the Washington Post way back in October 1988 that Jose Canseco is "the most conspicuous example of a player who has made himself great with steroids."

It reminds me of fundamental, versus technical, versus value analysis. I'm not sure why Bill James would be concerned with steroids aside from attempting to predict when the steroid era would end and what statistics might be affected by it. For example, pitching seems to be getting the better of "repeal" of the steroid era, but the change is surprisingly marginal. Presumably 'roids help hitting and hitters more than pitchers, but James' point seems to partly be that if two tennis players (to make the analogy more clear) take steroids, their serves will get faster, but the ultimate game scores will be much the same because they are competing against eachother.

In other words, if everyone in baseball took steroids, then the competition itself would still be, in a sense, fair. But it's not just about being fair. It is that because both sides are getting better, it's hard to tell that they both have an advantage over the rest of us, as opposed to an unfair advantage against the other competitor. Also, the point of steroids was to fool people. So he must be held to a higher standard than everyone else?

" I’m not sure why Bill James would be concerned with steroids "

Uh, because Bill James say his job is to analyze with numbers what people are talking about? What were people talking about more than McGwire and Sosa "restoring the innocence to the game" or Bonds hitting 73 homers?

And it's hardly holding Bill James to a higher standard than other sportswriters when his performance was vastly worse than, say, Thomas Boswell's, who publicly called out Jose Canseco in 1988 (at risk of a libel suit, by the way). In 1993, a baseball player's agent told me Canseco was "the Typhoid Mary of steroids." But, not a peep out of the sainted Bill James on the biggest statistical story in baseball until about 2009.

Why not? The most likely reason is sheer corruption: talking about steroid use would have kept James from getting a baseball job, and once he got the job with the Red Sox, he helped assemble a World Series champ built around juicers.

Yeah, but Babe Ruth was the man who changed the game. Before Ruth, there were no homerun hitters. A typical score in the pre-Ruth era was 1-0. So much of the strategy in those days was about scoring just one run. Hence all the sacrifice bunts, which managed to stay in the game post-Ruth until about the 90's. Skippers lived in the past for about 70 years, somehow. I'd say fans were very, very, extremely duped by the inertia of conventional wisdom from the pre-Ruth era. If you don't believe me, see Larry Dierker on the subject.

Interesting how you consider Bill James to be "corrupt". At any rate, attacking James' ''corruption" regarding steroid use is neither here nor there regarding his statistical analysis. The facts on the ground are that baseball is managed very differently in the post-Sabermetrics age than the pre-Sabermetrics age and Bill James was the biggest single influence in that change. When, for instance, was the last time you saw someone sacrifice bunt in the first inning? That change in tactics isn't about steroids. It's about managers finally learning the math.

"Yeah, but Babe Ruth was the man who changed the game."

Right, so the biggest victory in the triumph of Moneyball was won by a semi-literate slugger named Babe Ruth around 1920, and by the fans who flocked to see him. What Bill James and Billy Beane did were mopping up exercises compared to what Babe Ruth had accomplished.

Don't let the lack of data get in your way of attacking something, nor your ability to cherry-pick examples from books hundreds of pages long.

Big Blue is scheduled to coach my favorite basketball team next year.

During intermission, it will play a game of chess.

With any luck, soon we won't even have to watch the NBA games anymore.

I wonder what would happen if ubiquitous surveillance of professors started. Keystrokes, video, observation of grading.

Surely, nothing would go wrong.

See, defense is a public good (ducks).

Comments for this post are closed