Fans flocked to the Billie Jean King National Tennis Center in Flushing Meadows, New York, on Monday to see their favorite players battle it out over the course of two weeks for a chance to win the final Grand Slam tournament of the year: the U.S. Open.
The latest in A.I. technology is giving those same fans a unique opportunity to get an inside look at the most probable result for every match based on technology that analyzes more than 7 million data points.
Wimbledon runner-up Nick Kyrgios entered the U.S. Open ranked No. 25 on the ATP Tour. But according to IBM’s Power Index, the Australian tennis pro ranks second with improving marks following both his first and second-round matches.
"So, Nick Kyrgios is a great example and he's actually a 'one to watch.' Pre-tournament we identify 'ones to watch' as players who have a power index five slots higher than their tour ranking," Tyler Sidell, technical program manager for sports and entertainment partnerships at IBM, told FOX Business.
"The tour rankings are a rolling average of 52 weeks, so if you're really hot in January — you might not be represented in that tour ranking, but the power index takes in recent performance. Kyrgios was a finalist at Wimbledon and he's been on a roll lately. So, that takes into account as to why he has a higher power index, and he's very high up on the leaderboard right now."
IBM renewed its longstanding partnership with the United States Tennis Association (USTA) earlier this month in an effort to continue providing a "first-of-a-kind" fan experience through its digital offerings.
"The power index changes on a daily basis. We rerun the algorithms, we pick up any recent performance. So, the latest match — was it a blow out or was a close game? Take that into account, but we also pick up the new media commentary and the sentiment that's out there," Sidell continued.
Media commentary is just one component of the "Match Insights with Watson" feature, which is used to break down each match for the men’s and women’s singles tournament, but the A.I. analyzes more than a million sources — and for players like Kyrgios and Serena Williams, this certainly plays a significant role.
"The predictions take into account the recent performance, but it also takes into [account] media commentary," Sidell said. "So, we're using natural language processing, which is found in a product called Watson Discovery, to analyze what the media and pundits are saying about a particular player. So, is it positive? Is it negative? And how much punditry — how much is out there about that specific player?"
Sidell explained that there is a "weighted average" for each factor.
For Friday’s match against American tennis pro J.J. Wolf, Kyrgios has a 61% chance of winning based on "Watson’s Likelihood to Win" prediction — a feature unveiled last year. "Win Factors," a new feature unveiled this year, credits that to Kyrgios' recent success on the court and his overall "Power Index."
Fans also have a say this year with a new function that allows them to vote either for or against Watson’s prediction. For this match, the overall vote was in favor of Kyrgios.
"We're trying to explain to the fan and engage with them about Watson's projections and give them a little bit more context to debate there," Sidell told FOX Business.
With the announcement of their five-year renewal, USTA and IBM are hoping to continue on in their pursuit of giving fans a first-class experience using the latest in A.I. technology.
"The fan engagement is huge. That's a lot of what we do. We're using A.I. to spark some fan debate. The USTA is always trying to find those new fans. They're always trying to promote the new superstars. So, you know, you have the big three, you have [Rafael] Nadal and [Novak] Djokovic, and then you have [Roger] Federer, but they're not always going to be around. And we also see that with Serena as she's about to retire. So, it's about uncovering those new players and the up-and-comers," Sidell said.
"They always want to try to promote the new stars. They always want to try to promote their existing stars, but the support that A.I. is providing is that we're able to help fans uncover some insights about up-and-coming players that you might not know without the use of A.I."