One of the stars at the annual MIT Sloan Sports Analytics Conference was Kirk Goldsberry, who has applied his knowledge of geography to the ecosystem, or “court space,” of the basketball court. The event drew 2,200 people to Boston earlier this month.
One of the stars at the annual MIT Sloan Sports Analytics Conference, held earlier this month in Boston, was a scholar who has applied his knowledge of geography to the ecosystem, or “court space,” of the basketball court.
Kirk Goldsberry, an assistant professor in the department of geography at Michigan State University and a visiting scholar at Harvard University, made a splash by stepping out of his main field.
“Goldsberry does health care research,” notes Fast Company in “In Relentless Jocks-Nerds War, Hope For Peace Through Analytics.” “He creates maps that reveal a community’s lack of access to fresh produce, and he publishes his findings in academic publications such as the Journal of Hunger and Environmental Nutrition. But it’s Goldsberry’s latest research paper — “CourtVision: New Visual and Spatial Analytics for the NBA” — that has thrust him into the sports geek limelight. ‘I woke up yesterday never having been on national TV or in The New York Times and Sports Illustrated or interviewed by the Wall Street Journal,’ he says, incredulous that all those things have since happened.”
The annual conference has become a big event for what’s estimated to be a $400 billion business. It drew over 2,200 people, up from the 60 or so who attended the first conference six years ago. “ESPN, the lead sponsor, brings in more than that many staffers, nearly 100 this year, to use its promotional muscle and savvy to turn an event where much of the action involves presenting academic papers into a high-profile, entertaining affair,” writes Fast Company.
Videos from the conference are available for free at sloansportsconference.com. Among them: this conversation with Sports Data Hub founder Kevin Goodfellow on how Google, Facebook and Yahoo have turned sports analytics upside down:
Goldsberry’s basketball research uses “CourtVision, a new ensemble of analytical techniques designed to quantify, visualize, and communicate spatial aspects” of performance by members of the NBA. His case study: find out “Who is the NBA’s best shooter?”
The analysis works “by ignoring overall field-goal percentage — which turns out not to be very helpful — and instead finding the players who, like [Boston Celtics player] Ray Allen, make the highest percentage of shots from the most locations on the court,” writes the Boston Phoenix. Allen ranked second in Goldsberry’s study, with Steve Nash of the Phoenix Suns coming out on top. See the Times summary or Goldsberry’s paper [pdf] for technical details.
It’s been a hot six months for sports analytics. Actor Brad Pitt’s “Moneyball,” about the use of sabermetrics by the baseball team the Oakland As, has done well since its September release, with $75 million in U.S. ticket sales and $35 million in international sales.
True, two years ago, a character on the TV show “The Simpsons,” in an episode about sports analytics, said that “baseball is a game played by the dexterous but only understood by the Poindexterous,” but the mainstreaming is well underway. Goldsberry, for instance, became a sought-after commodity at the conference without even being a speaker. His paper was one of 20 selected to for display at the event, and it was named the runner up in the conference’s research paper showcase.