Early in the 2011 film “Moneyball,” Peter Brand, a baseball neophyte with an economics degree from Yale, shakily offers Oakland Athletics general manager Billy Beane his iconoclastic view of how the team should be run. “Your goal shouldn’t be to buy players; your goal should be to buy wins. And in order to buy wins, you need to buy runs. Baseball thinking is medieval. They are asking all the wrong questions. Using the stats the way we read them, we’ll find value in players that no one else can see.”The movie was based on Michael Lewis’s best-selling 2003 book of the same name, which told the true story of how Beane used smart and novel statistical analysis to gain unique competitive advantage with a payroll that was a fraction of his rivals’. Although Lewis’s book was groundbreaking, the phenomenon it chronicled seems quaint compared to the game-changing level of analysis being done today by nearly every team in every sport.
Not long ago, the proponents of sports analytics were considered less than credible by traditionalists who grew up making judgments on the basis of personal experience and instinct. But that has changed rapidly. At this year’s sixth annual MIT Sloan School of Management’s Sports Analytics Conference, 73 professional teams and 175 colleges were represented, all seeking insight from the best analytical minds about innovations, new research and best practices.
Sport has always used data to a rudimentary degree, mostly to evaluate and manage on-field performance, but even that has greatly expanded in scope and complexity in recent years. In professional football, for instance, where performance gauges like the vertical jump, the 40-yard dash and standard mental acuity tests have been relied upon for years, many teams are now going much further. The New England Patriots, for example, track their own proprietary set of data points on potential players with a “draft decision support system” that is updated daily with new reports from scouts.
Even in baseball, where an affinity for statistics has always been part of the game, the level and leverage of analysis has escalated dramatically. Old-school stats like ERA (earned run average) and RBI (runs batted in) that once set the standards of greatness have been moved aside for an array of esoteric calculations like OPS (on-base percentage plus slugging percentage), VORP (value over replacement player), BABIP (batting average on balls in play), DIPS (defense-independent pitching statistics) and LIPS (late-inning pressure situations).
According to Chris Marinak, senior director of labor economics for Major League Baseball, the sport is now working on devising better analytics for a different purpose: “We have an electronic medical history tracking system for over 200 major and minor league teams that can tell us what types of activities are conducive to what types of injuries. We can also tell where on the field certain injuries are most likely to occur. That opens the door to a lot of different analysis to improve care and reduce injuries.”
Data analysis is even beginning to affect how the games are played. In several sports, optical tracking systems are being used to map the positions of the players and the ball, as well as vectors and velocities on every play. “It adds up to more than a million data points per game,” said Kevin Goodfellow, an executive of Sportsdatahub.com. This allows teams to employ “spatial analytics.” In basketball, for example, they can study where all 10 players and the ball are on the floor, or in the air, at any given moment and determine which patterns are most likely to lead to scoring. They can also learn what combinations of players or player types (penetrating point guard, shot-blocking big man, three-point shooting specialist) are more effective when on the court together.
Off the field, analytics are having an equally big impact on the business of sports. The tracking of fan-related data and behavior patterns — much of which comes from their ubiquitous use of mobile devices, even while at events — will ultimately enable dynamic and variable pricing for tickets, pay-per-view events, fantasy sports games and even merchandise. “Helping venues sell more tickets starts with getting a holistic view of the fan, getting all your data in one place to make sense of it,” said John Forese, senior vice president at LiveAnalytics, a TicketMaster company.
Labor and contract talks, where once persuasion was the coin of the realm, are also increasingly driven by stark analytics of all kinds — payroll-to-performance ratios, gate and merchandise receipts, health and injury metrics. “Contract negotiations used to be a lawyer-driven process, but today the analytical people are far more important,” said Rob Manfred, executive vice president of Major League Baseball. According to Adam Silver, deputy commissioner of the National Basketball Association, predictive data modeling played a big role in resolving that sport’s lockout last December by allowing players and owners to see the implications of various salary cap systems, tax structures and revenue-sharing arrangements.
As companies spend more and more on sports-related sponsorships — a global total of about $50 billion annually — they too are looking for more hard data to objectively evaluate their investments. “Analytics are becoming more important on the front end of that,” said Phil de Picciotto, president of the sports agency Octagon. “We’ve got a research division that can separate into 20-some categories what makes people attracted to certain athletes or events.” That attraction often has less to do with on-field performance than with an athlete’s personal brand — unique style, solid values, likability, reach and influence. In today’s world, that brand is increasingly projected through and measured by social media.
“Social media is the next frontier for evaluating athletes and the exposure they bring,” said Lawrence Norman, vice president of global basketball for Adidas. “Look at (Orlando Magic center) Dwight Howard. He has 2 million friends on Facebook and 3 million followers on Twitter. Ten percent of those people actively engage about product. If you get even 10 percent of those people to buy shoes, that’s a great thing.”
As marketers have begun to recognize the power of social media to drive consumer behavior, social media analytics services like Klout, Kred and Reppify have sprung up to rate and continually update the reach and influence of social media users. Many companies, including Nike, have begun to pay close attention to these scores in their sponsorship and marketing efforts.
However, some still take a measured approach to this analytic age. “There’s no single analytic that would determine whether an athlete would be right for a campaign,” said Phil de Picciotto. “Analytics are important, but it all comes back to authenticity.” Sarah Robb O’Hagan, president of PepsiCo’s Gatorade, had a similar take: “Just adding up Facebook ‘likes’ doesn’t wind up being as important as the qualitative pieces. There’s still so much art that goes with the science.”
