*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Atlanta, GA | Posted: March 16, 2010
Don Fernandez
404-894-6016
Summary Sentence:
LRMC, the computer ranking system designed by three professors at the Georgia Institute of Technology, has forecast this year’s NCAA Final Four match-ups.
Full Summary:
LRMC (Logistic Regression Markov Chain), the computer ranking system designed by three professors at the Georgia Institute of Technology, has forecast this year’s NCAA Final Four match-ups will be Kansas vs. Syracuse and Duke vs. West Virginia.
LRMC BracketsThe calculations have been completed and the predictions are in: Kansas will face Duke in this year’s NCAA basketball finals with Kansas emerging as the victor.
LRMC (Logistic Regression Markov Chain), the computer ranking system designed by three professors at the Georgia Institute of Technology, has forecast this year’s NCAA Final Four match-ups will be Kansas vs. Syracuse and Duke vs. West Virginia with Kansas taking the title.
LRMC’s Final Four predictions differed from the NCAA’s seeding this year in selecting West Virginia to beat Kentucky to reach that coveted bracket. A few other surprises emerged as well.
“There are several upsets predicted in the earlier rounds,” said Joel Sokol, operations research professor in the Stewart School of Industrial and Systems Engineering at Georgia Tech. “Among others, San Diego State, Utah State, Murray State and Brigham Young could be surprises.”
Developed by Professors Sokol, Paul Kvam and George Nemhauser, LRMC utilizes data such as scores, teams competing, home court advantage and margin of victory in past performances to calculate likely victors along with potential underdogs that could emerge as statistical spoilers. The system has been impressively successful, particularly in 2008, when it predicted not only the Final Four and eventual victor, but also several upsets in earlier rounds.
This year, the professors have upgraded the system’s probability calculations, an improvement that could enhance the accuracy of the results. For this new research, Georgia Tech's LRMC team welcomed a new member: math professor Mark Brown of The City College of New York.
LRMC is not an infallible resource. There’s always the chance of an upset, injury or other factor that affects the outcome of competition.
“There’s a lot of randomness,” Kvam said. “Statistically, we’re going to have years where we’re off.”
Nonetheless, the system has proven more reliable with its predictions than the NCAA’s own Ratings Percentage Index (RPI). Historically, the upgraded LRMC method has picked the winner of more than 74 percent of NCAA tournament games correctly, while the RPI has been right less than 70 percent of the time.
This year’s LRMC bracket can be accessed at http://www2.isye.gatech.edu/~jsokol/profspicks/profspicks10.htm
