Building a Top-Performing Bracket for March Madness Using Advanced Analytics

A team at SAP is diving into the world of college basketball again this year to help bring sanity to March Madness through the latest innovations in the cloud analytics product SAP BusinessObjects Cloud.  

Using the 2017 NCAA Division I men’s basketball championships to highlight the value of analytics and hard data over intuition or sentiment, the Data Genius team at SAP is forecasting which of the 68 teams will advance in this year’s college basketball tournament.

To create its bracket this year, Nic Smith writes in a recent blog post, the Data Genius squad deployed a variety of techniques, including data preparation, modeling, data exploration, and automated smart data discovery to uncover hidden insights into which teams will perform best in the tournament, using an approach that is similar the way organizations deal with data and analysis in the business world.

Smith, who is marketing leader for cloud analytics at SAP, outlines the process and explains the steps involved in SAP's data analysis, including first pulling college stats from the web and using a Google Doc, scraping additional insights from the web and joining that data together to enhance the model, cleansing and exploring data, building a story, and then applying automated machine learning algorithms to the data to uncover relationships and insights that may not have considered previously.

Smith says that SAP’s Data Genius team will share its predictions, analysis, and results as the rounds of the tournament unfold and provide updated analysis and insights based on team performance.

The Data Genius squad is also challenging others to see if they have what it takes to beat the SAP Data Genius bracket using the SAP BusinessObjects Cloud. Details on how to do this are shared in his blog post on the SAP Community site.