Taming March Madness with BI

With the NCAA Men's and Women's Basketball Tournaments tipping off, talk of March Madness, brackets and bubble-busters is reaching a fever pitch.

Two business professors and self-professed sports fanatics are trying to take some of the madness out of the month -- and in the process provide insight into how technology solves critical business problems.

Jay Coleman, an operations management professor at the University of North Florida in Jacksonville, and Allen Lynch, an economics professor at Mercer University in Macon, GA, have once again applied analytic software from SAS to predict "at-large" teams - those teams that did not get automatic bids to the basketball tournaments.

Over the past 11 years, their NCAA Dance Card, powered by SAS® software, has boasted an impressive 94 percent accuracy rate. On Selection Sunday this year, the professors missed three teams, which put them at 91 percent, slightly below their average.

Predicting game winners and losers
Now, using another model powered by SAS, the Score Card, Coleman and Lynch are predicting tournament game winners. Over the last five years, the Score Card has correctly predicted the winner three-quarters of the time, including many upsets.

These basketball-themed models are an entertaining way to point out a broader business trend: the momentum for business intelligence (BI) software.

In addition to basketball picks, BI and predictive analytic software help businesses succeed every day by helping predict such things as:

• Which customers will respond to a bank's direct mail on home-equity loans.
• Which floor plan will help a retailer sell more designer shoes.
• Which combination of medical treatments will keep diabetic patients healthy and out of intensive care.

The "intelligence" that many other software vendors provide -- basic reporting on what happened last year, last quarter or even last week -- falls woefully short of the reliable, predictive intelligence that companies need to remain competitive.

"SAS can analyze a lot of information, pull out what's important, ignore what's not important and predict what's coming in the future, as opposed to simply describing what happened in the past," says Coleman, who noted that the Dance Card represents an overall trend in business intelligence toward more prediction and less description.

"Predictive analytics is a key part of business intelligence that helps companies move from information to a decision, from the descriptive to the predictive," says Coleman. "With forecasting, data mining, optimization and other advanced predictive software from SAS, businesses can go beyond simply reporting on what has already happened and instead understand what will happen and where they need to go next."

College professors Jay Coleman (left) and Allen Lynch