How Big Data Has Overtaken March Madness

It’s no secret that big money...and Big Data...go into March Madness. The Chicago Tribune estimates that college basketball is the fastest-growing fantasy sport, and they aren’t wrong--NCAA fantasy could have grown up to 10 times what it was two years ago. But this isn’t the only “big” data point involved in NCAA basketball--the game has recently turned to analytics, and it’s leaning on the strength of the data more than ever.

Basketball fans (and coaches) have become so reliant on player stats that some say March Madness has become “America’s most popular exercise in statistical reasoning.” They might be right, since statistical models like those popularized by various ESPN pundits and Nate Silver have overtaken the sport.

And despite Charles Barkley’s assertions that only “idiots” believe analytics, I’m firmly on board with the newest wave of March Madness data analysis. And I’m not alone: top NBA teams are now scouting talent based on college stats, extending the “moneyball” concept beyond baseball and firmly into the NCAA’s court. Duke has installed SportsVU cameras to better track player movements and deliver NBA-quality stats. As ESPN has said, “Sooner or later, big data [would come] to college hoops.”

In the past, fans have relied on The Huffington Post’s Predict-o-Tron, Intel’s Kaggle, Kimono Labs’ March Madness API, and numberFire’s March Madness Helper in the past for basketball stats, but this year we need a more in-depth approach to the numbers.

Like HuffPo and Intel, we’ve created our own approach to basketball analytics. It’s called HoopWise, and it has crunched the NCAA basketball data to churn out a unique bracket. Although our bracket performed well originally, it’s fallen on tough times.