Communications of the ACM
As Tennis Stats Proliferate, Software Tries to Make Sense of It All
SlamTracker software predicts how tennis players could perform under a variety of circumstances.
Credit: U.S. Tennis Association
IBM researchers have developed SlamTracker, software that sifts through 20 years of data to predict how tennis players could perform under various circumstances.
Since 2011, SlamTracker has provided TV announcers with specific statistical talking points, known as Keys to the Match, to refer to before and during tennis matches. The keys are updated during matches to track any shift in momentum, and they correlate well with the final outcome, says IBM's John Kent. The findings are based on more than 8,000 matches IBM has cataloged over the past eight years at the four Grand Slam events. In total, the Keys to the Match are based on more than 41 million data points, including scores, match duration, winners, serve speed, number and types of shots, and serve percentages. "We've come up with a way to use the analytics to inform the fan," Kent says.
Fans at home can gain access to much of this information on their mobile devices and home computers. "There is a constant desire for the fan to have more and more information and make it bite-size," says the United States Tennis Association's Nicole Jeter West. This year, SlamTracker has been expanded to include a social sentiment function so fans can see what other fans are talking about.
From The New York Times
View Full Article
Abstracts Copyright © 2013 Information Inc., Bethesda, Maryland, USA
No entries found