Using Predictive Analytics to Bring Sanity to March Madness

Trying to put together the perfect bracket for the 2015 NCAA men's basketball championship? Data scientists are offering their expertise to help basketball fans solve this real-world challenge.

In a FiveThirtyEightSports blog post, Nate Silver, statistician and author of “The Signal and the Noise: Why So Many Predictions Fail - But Some Don't,” showcases the website’s NCAA tournament forecasting model. After stating that this year’s NCAA men’s basketball tournament has “a clear front-runner” with the  Kentucky Wildcats, Silver explains that the model takes into account not only computer generated data, but also “human rankings."

The model calculates the chance of each team reaching each round, taking into account a composite of power rankings, pre-season rankings, the team’s placement on the NCAA’s 68-team S-curve, player injuries and geography. 

According to Silver, FiveThirtyEightSports’ forecasts have predicted the winner of the tournament correctly in 2 of the past 3 years. 

And, SAP is also offering a March Madness challenge powered by SAP Lumira and SAP Predictive Analytics. SAP is inviting fans try out SAP Lumira and then share their insights and predictions via SAP Lumira Cloud. To join in, SAP provides instructions on how people can download SAP Lumira, download,SAP Lumira College Basketball sample data, and then publish their own data visualizations to the SAP Lumira Cloud.

Last year, the SAP Data Viz team correctly selecting 11 out of 16 Sweet Sixteen teams and 2 out of the Final Four (Florida and Wisconsin), and overall, 40 out of the 63 team picks were correct. See who the team picked this year at and whether their analysis is panning out.