Artificial intelligence is becoming part of the live experience at the US Open, with fans using the tournament app to follow real-time predictions, match analysis and a new measure of serving technique.
During Coco Gauff’s first-round match against Turkey’s Zeynep Sönmez on Tuesday, the app tracked the players’ chances of winning, highlighted significant moments and provided a score for “serve quality” as the contest unfolded.
The technology is being developed by IBM in partnership with the United States Tennis Association. Cameras positioned around Arthur Ashe Stadium follow the ball, racquet and players’ movements, recording details including knee and elbow flexion, wrist speed and the transfer of energy through the body.
Those measurements are processed using IBM’s AI systems to assess the efficiency, accuracy and consistency of a serve. The resulting score is presented to fans after matches, with the analysis breaking down the mechanics behind a player’s delivery.
In Gauff’s case, the system said her successful serves featured “controlled racket preparation and deep knee bend during her setup”. The new feature is available across all 254 singles matches at this year’s tournament.
IBM said the system tracks 21 points across a player’s body and racquet 50 times a second. By the end of the fortnight, it expects the technology to have generated about 1.2 billion data points.
Tyler Sidell, IBM’s technical programme director for sports and entertainment partnerships, said the intention was to give spectators another way into the action without attempting to replace the unpredictability of the sport.
“There’s so much unpredictability in sports — you could take a look at all the data, but anything can happen out there on the court,” he said. “That’s the fun about it. We want to provide an insight, but still, let’s watch the matches play out.”
The technology is also being used by players. Jessica Pegula, who reached the fourth round, said she studies data on opponents’ serving patterns before matches, using it to prepare for the decisions she may face on court.
“Tennis is a lot of problem solving on the court, so it’s a lot of patterns, and I think serve is a really big one,” she said. “It’s the one controllable shot we have in tennis.”
Pegula acknowledged that opponents can change their tactics once a match begins, making instinct and adaptability just as important as preparation.
“You can get a lot of analytics, but sometimes things change during the match, sometimes someone changes their strategy, or they maybe go against the grain of what you thought they were going to do, and you still have to really trust your instinct,” she said.
IBM’s “Likelihood to Win” feature has been part of the US Open’s digital coverage for around six years, but this is only the second tournament in which supporters have been able to watch the probability change live from point to point.
The prediction draws on current match statistics, previous performances, expert opinion and momentum. Gauff was given a 72 per cent chance of beating Sönmez before the match, with the figure shifting as the contest developed.
A related feature, “Key Moments”, provides short explanations of why the balance may have changed. During Gauff’s match, one update said she had three chances to secure victory while Sönmez needed a “heroic response”.
The volatility of the system was illustrated in the men’s draw when top seed Alexander Zverev faced Italy’s Lorenzo Sonego. Zverev began with an 87 per cent chance of winning, but Sonego’s probability rose to 93 per cent late in the fourth set before the German recovered to win the five-set contest.
The tournament app also includes Match Chat, an AI assistant that answers questions about players, fixtures and the grounds. Asked where to find the US Open’s signature Honey Deuce cocktail, it supplied the locations of nearby bars and offered to identify the closest one to a fan’s entrance gate.
IBM said its US Open digital services are used by more than 14 million tennis fans around the world. This year’s expanded features also include a personalised live-updates page designed to bring together news, scores and analysis centred on each supporter’s favourite players.
