In theory, the Carolina Panthers and Denver Broncos will meet in the Super Bowl this Sunday because they are the two best teams in the NFL this year. But why were they the best? A football romantic might point to Cam Newton's maturation or Peyton Manning's veteran savvy, but those things are hard to quantify.
What's easier to put into numbers is the role of health: The Broncos and Panthers were both among the six healthiest teams in the NFL this season, in terms of the number of games lost to injury by their players.
That's no coincidence, according to Stephen Smith, CEO of the Dublin-based sports-tech startup Kitman Labs. After analyzing four years of NFL injury data, Kitman has calculated the precise cost of winning a professional football game: $2.3 million. That is, for every $2.3 million in salary cap dedicated to players who spend Sundays on the disabled list rather than on the field, the average team will lose one more game.
The relationship is strong. Andy Shelton, an applied sports scientist at Kitman, says 89 percent of the variance in NFL teams' win-loss records can be accounted for by the amount of player payroll lost to injury. "It's all about your key player availability," Shelton says. "Teams that have the best players tend to win more. It's not rocket science."
Yet preventing sports injuries is, if not astrophysics, a decidedly devilishly complex science of its own. For all the advances in training, nutrition, and equipment over the past few decades, injury rates in the major sports have only gone up. The magnitude of the problem and the obvious financial incentive to solve it have given rise to a slew of startups using proprietary technology to try to keep elite athletes healthy. There's Catapult Sports, which puts sensor-packed devices into shirts to gauge the loads being placed on players' bodies; Fusionetics, which uses motion capture to diagnose flaws in their movements; Motus Global, which analyzes the patterns of baseball pitchers and hitters to gauge their vulnerability to repetitive-stress injuries; and the list goes on.
Kitman collects largely the same sorts of data as its competitors: Athletes perform daily movement tests in front of 3-D motion-capture device (basically a Microsoft Kinect) and answer questions about their sleep, diet, soreness, and mood in a mobile app.
What sets Kitman apart is what it does with this data. Rather than apply a one-size-fits-all model of injury-risk assessment, it focuses on teasing out the patterns in an individual athlete's vulnerability over time. A risk factor that's highly predictive for one player might be irrelevant to his teammate, and vice versa.
"We cannot apply the same analytics across every person," Smith says. "What we have found statistically is there is very little evidence to support that what happens with one person is replicated then across other people. But we have found very strong significance in understanding similar relationships and patterns in the same person. That makes perfect sense because human beings are so complex, so unique."
One person who buys into this approach is Billy Beane, the executive vice president of operations for the Oakland Athletics. Beane's then-novel embrace of advanced analytics propelled the A's to the top of baseball and provided the subject of the book Moneyball. One of the biggest challenges in running a team, he says, is that every new player who joins the team is a black box when it comes to his health. "We would then become more reactive to injuries that would happen, as opposed to proactive," he says. Kitman's longitudinal approach could solve that conundrum; when a player changes teams, he or she can bring his data along to the new one. In addition to signing the A's on as a Kitman client, Beane has joined the company as an adviser. "I'm a bit myopic in my thinking, but ultimately I believe more data and more analysis is going to be the answer," he says.
Smith came up with the idea that would become Kitman Labs seven years ago while working as a strength and conditioning coach at Leinster, an Irish Rugby team. Like other pro sports organizations, Leinster had started to collect a large volume of data on its athletes without seeming to have any clear idea how to use it. Meanwhile, it had the same injury problems as all of its competitors. Smith was upset when players he'd grown close to were forced by injuries to retire early. "I was like, why are we collecting all this information unless we're going to actually do something with it?" Smith says. "I felt like the athletes deserved more from us, that we owed them a duty of care. It wasn't for want to trying. We just didn't have the tools to actually do it."
So he set out to earn a master's degree, choosing "combined risk factors as predictors of injury" as his research area. The success he had in applying his findings--a healthy Leinster won the European championship in 2009 and 2011--provided validation and the springboard for a startup.
Kitman counts among its clients the healthiest teams in soccer's English Premier League and British rugby as well as the Miami Dolphins, the second-healthiest team in the NFL this season. But Smith is hesitant to take credit for those successes or to make any public claims about the efficacy of Kitman's platform. That credit, he says, should go to the trainers, coaches and other support personnel who use Kitman's insights to make decisions about their athletes' care and feeding. "All we do is mine the data and surface the information back to the practitioners," he says.
That attitude shows in the company's name. In rugby and English football, a kitman is the equivalent of an equipment manager, but with more far-reaching duties. He's the member of the staff who makes sure every aspect of the locker room is just as it should be without ever drawing the slightest bit of focus. "We wanted Kitman to be exactly the same," says Smith, "operating in the background, not standing up taking all the accolades."