Playing defense with team data
March, 1, 2014
By Daniel Nowell
Special to ESPN.com
Special to ESPN.com
Jim Davis/The Boston Globe/Getty ImagesTracking the NBA's referees could be one of the next breakthroughs in NBA analytics.Aside from the media, there are two classes of people at the Sloan Sports Analytics Conference: people who have jobs and currency, and the people who would like jobs and currency. The difference between these classes, behaviorally, is disclosure: The jobs and currency class comes to Sloan and keep their secrets, while the seeker class comes to try and sell theirs.
By this point, it is a trope of the conference that nobody with any real skin in the analytics game can talk about what they’re doing. For as much as the executives, scouts and analysts here have convened to share knowledge and to celebrate that sharing, the obvious competition between them prevents any real honesty. Broadly speaking, everybody here is engaged in “analytics,” but when it comes time to discuss specifics, no teams will step forward and detail their nutrition programs, the ways they use optical-tracking data, their psychological assessments, their training regiments and so on. Most smiles here are pleasant poker faces.
Conversely, the people after the jobs and money come to Sloan to tell as many people as possible as much as they can. John Ball is one such one person. As ESPN’s Kevin Pelton described yesterday, Ball is at Sloan with a business -— backed by a research paper presented at the conference —- that sells referee analytics to NBA teams. The idea, more or less, is to let teams see the tendencies that various referees have so that the teams can capitalize on, or at least prepare for, those tendencies.
As Ball tells it, he’s not the first to have this idea: “Mark Cuban has gone on the record about how he’s tracked the referees, and NFL teams and scouts have made adjustments and game plans based on referee crews. It’s prevalent in many sports, and some forward-thinking teams have already done it.” To Ball’s mind, he’s tracking an observable phenomenon, and some teams have already gotten hip to the benefits. Unlike the teams, though, he stands to benefit from talking about his work as much as possible, and so his information raises the questions of how, and whether, information can really move between organizations anymore.
A certain strain of idealism floats in the air at the Sloan Conference, an idealism that is probably the biggest reason the conference has picked up an increasing number of critics. It’s kin to the idealism of TED talks, South by Southwest and other increasingly derided conclaves of pop intellectualism. There is an idea here that the gathering of knowledge -— surely a virtue -— can only make sports better, and why choose not to make something better?
What critics of Sloan perceive, correctly, is that the ideology masks an essential reality: men (it has become mandatory to note that the conference is almost entirely male) have gathered here to apply the rigor and technology of derivatives trading to winning sports games. My suspicion is that the critics of Sloan are so vitriolic because sports fans get enough of this margineering in the real world -— these guys are doing just fine with their tech ventures and capital firms. Why are they polluting our games?
This tension may not be resolvable, because the ideal of opening sports up to new ways of thinking is directly at odds with the competitive marketplace. In the age of Big Data -— a phrase directly incorporated into the titles of two panels, both specially marked with sponsorship from tech companies —- information has become as valuable as oil. Sharing extraction techniques is, in the judgment of the market, foolish.
There is skepticism among NBA executives whether Ball’s information about referees could be an advantage to teams. The league, after all, charts every call made by every referee in every game. And if they have charted all of the referee behaviors there are to chart, how could Ball have proprietary information? Of course, the NBA won’t release its data, and Ball, while he can release more than the league, can’t chart and tell, either. For fans, there’s no way to tell who’s bluffing, or even whether somebody is intentionally bluffing.
So Sloan is the hybrid trade show-conference that serves as the ostensible hub of sports information sharing, and nobody here knows whether anybody else knows anything. If there is a way to resolve this paradox, it seems as though the various leagues in attendance are the likeliest source. Sports have always functioned as a social laboratory, and the rulebook serves as the controls. On the field, on the court, we are trying to achieve the optimal conditions for objective evaluation -— if you subscribe to this theory, then the best-designed sport is the sport in which the better team wins most often. But when you can’t even tell what assumptions teams make about officials, how does the framework of competition hold up?
The question Sloan asks, then, is what responsibility leagues have to regulate information? Should the NBA allow information to move at the speed of the market, to be collected and even hoarded by the most forward-thinking (or cynical, depending on your vantage) teams? Or should it put its thumb on the scale and standardize information-gathering practices? To put it another way: If one team has gathered information about the very fabric of the sport that another team doesn't have, are they even playing the same game?
As always, off-court life is bleeding into the sport, and the league will have a chance to shape the response. Sports have been a theater and a treatment for our dramas of race, sexuality and every other social issue. The extent to which we are defined by our data, and who has the rights to that data, and who governs the trading of that data are central issues that sports now have a chance to help us make sense of. We, like general managers and everybody else at Sloan, are locked in a cycle of paranoia and fascination with our data -— look at all the things we can do now, but stop at nothing to avoid leaks!
Talking about the conference with Paul Flannery, Celtics Assistant GM Mike Zarren said, “The big-data revolution was happening whether this conference happened or not. The world was already changing.” He’s right. But the precepts of that revolution —- that information is currency, and that it should be guarded and allowed to behave like currency —- can be tested here. We can literally play with how we will treat data, and find out how many points certain information is worth. But before we can do that, we have to learn who knows what, and perhaps the most enduring lesson of Sloan this year is how far away we are.