Data analytics from artificial intelligence may have assisted a professional tennis player’s next shot at Wimbledon.
Like many sports, tennis is adapting to AI, using data analytics into sports science methodologies to better model how the human body performs, strengthens and recovers.
Analytics can improve an athlete’s split-second decision making when they compete, measuring and improving performance, analyzing opponents, and prioritizing off-court recovery, said Peter Spivack, who co-created Carnegie Mellon University’s Introduction to Tennis Analytics course with Ainika Hou.
AI and data analytics will be in the background of the prestigious Wimbledon tennis tournament, which began Monday and goes until July 12 in the London suburb.
“There’s a lot of trends and patterns that occur within a match, as well as over a season or over years, and it’s very interesting to see how that develops and how players are able to take what they’ve done already and adapt that to further their strategies in the future,” Hou said.
Student of the game
Many elite tennis players have added data analysts to their teams, alongside personal coaches, trainers and physiologists.
Analysis happens before, after or between matches. Spatial-temporal data through a match recording can identify the movements of the player and the ball over time. Other models can predict where the ball is going to land.
“We’re able to use artificial intelligence models to be able to identify the players, identify the movement,” Spivack said.
Data can be used to help a player choose a serve or stroke type given a match score, the weather and sun position or their opponent’s skills and fatigue; a player’s court position for returns or recovery given the last shot type, ball position and opponent position.
“It’s really the job of the tennis analyst to look at that data and say, ‘What should I do about that?’ ” Spivack said. “Should I devote more time to practicing endurance? Should I devote more time to go for riskier shots earlier in the rally because I want to give up those long points?”
Data is usually kept internal because of a player’s competitiveness, but analytics firms report success: Golden Set Analytics says its clients have won 12 men’s and women’s Grand Slam victories in the last five years.
Tournaments themselves collect their own data: AI software firm TennisViz uses AI and computer vision to interpret a video recording of a match into temporal-spatial, or time-space, data and measurements such as a player’s movement, time frame and specific shot type.
Wimbledon also taps into AI to boost the viewer’s experience. The “Live Likelihood to Win” responds to data from live matches to update a player’s probability of winning, taking into account things like scorelines and momentum shifts. The “Match Chat” is an AI assistant that can answer questions fans have about matches and athletes.
“This won’t be as in-depth as the models you’d expect the players to be using to try to improve their game, but this will be more of the basics of, ‘This person is hitting it wide a lot. Why is that happening?’” Spivack said. “ ‘This person looks really tired in the third set and in the fourth set, he looks more refreshed. What changed?’ ”
Grass court strategies
The nature of Wimbledon makes applying inferences from data less certain but still important, Hou said. All Wimbledon matches are played its famous grass courts.
Generally, on a grass court, the ball moves faster and bounces lower compared to clay or hard courts, Hou said. It can be harder to keep longer rallies going because of the lower bounces.
“Players have to get to the ball faster and get their racquet below the ball and over the net sooner than those other courts,” she said. “We can see how the strategies in grass court matches differ from some of those other surfaces.”
But factors like weather and physical demands can make things less predictable. “A match on the first day of Wimbledon might be different grass than the final day,” Hou said. “How that affects the players, how that affects the different statistics that we like to study in tennis could also be interesting.”
Kellen Stepler is a TribLive reporter covering education in Allegheny County. He joined the Trib in April 2023. He can be reached at kstepler@triblive.com.
Daily and weekly updates delivered straight to your inbox.
Get news alerts first, right in your browser.
Daily and weekly updates delivered straight to your inbox.
Support our journalism and get an ad-free experience on all your devices.
1-800-909-8742
Fax (724) 779-8743
210 Wood Street
Tarentum, PA 15084
Leave a Reply