From left to right: Former NBA player and ESPN analyst Shane Battier, Boston Celtics Assistant GM Mike Zarren, WNBA player Sue Storm and former NBA head coach Mike D'Antoni speaking at MIT Sloan 2015.

Feb 28 2015

What Does Sports Analytics Even Mean? Experts Disagree

At the 2015 MIT Sloan Sports Analytics Conference, leaders in professional sports and data analytics attempt to define the term.

Everyone seems to agree that sports analytics has transformed and will continue to transform MLB, MLS and the NBA, NHL and NFL.

The problem? Few people working in either the professional sports industry or the analytics side agree on what “sports analytics” even means, let alone how it should be applied in at the player, team and league levels.

At the two-day 2015 MIT Sloan Sports Analytics Conference, held in Boston, defining and pinning down what analytics meant was a recurring theme.

It was a point of discussion during a panel on Friday titled “Is Analytics Taking the Joy Out of Sports?”

“Anytime we see data, it’s not called analytics,” said Ben Alamar, director of production analytics at ESPN. It’s just a bunch of numbers without context and analysis.

In that panel, Brian Burke, founder of the Advanced Football Analytics blog, defended a loose definition of the term and advocated for more use of data analytics across all sports leagues. Even those who call sports analytics nonsense often find themselves inadvertently relying on what they dismiss as worthless.

“Everybody uses stats. Even the guys who pooh-pooh it, in the next breath, they’ll throw out a number to prove why they don’t like statistics,” he said.

On Saturday, during a session titled “Analytics Illustrated,” professor and Grantland writer Kirk Goldsberry made a similar case for a broad and inclusive definition of analytics.

“Is the fact that Chris Bosh is left-handed analytics?” he asked. “I think so.”

Put more simply, in Goldsberry’s view, anything that can be used as evidence of a theory or hypothesis is fair game.

“My favorite definition of analytics is that they’re just reasoning artifacts,” he said.

Should Professional Sports Be Ruled by Numbers?

During the conference's last session, titled “The Future of the Game,” the panel batted around several competing theories about what analytics were and weren’t and whether they were meaningful or just noise.

Scott Pioli, assistant general manager for the Atlanta Falcons, explained that analytics to him went far beyond numbers.

“I don’t think analytics, to us, is a number. It’s a process, it’s a system,” he said.

Striking a much more defiant and aggressive position on the subject was Brian Burke, president of hockey operations for the Calgary Flames.

“This is still an eyeball business. Analytics is a tool in the toolbox, but it’s not the first tool and it’s not the most important tool,” Burke bellowed. He proved to be quite the tell-it-like-it-is personality throughout the panel, with his slicked-back white hair and untied necktie draped over his shoulders like a scarf. At times, Burke sounded more like he was waging a one-man war against the data drones.

“The notion that you can sit behind a computer and find athletes, it’s bullsh**t,” he declared.

The audience erupted into laughter at the candid remark. But the debate about what analytics is, who gathers and interprets that data, and how much that data should influence decision-making in sports remained unsettled by the end of the conference. And it’s likely to stay hotly divided in professional sports in the near future.

Ricky Ribeiro