3 Sources
[1]
IBM launched limb tracking at this year's U.S. Open - and it's scoring every tennis player's serve for fans to see | Fortune
When world number three and top-ranked American tennis player Jessica Pegula stepped up to the baseline to serve against Belarus' Aryna Sabalenka in the semifinals at the U.S. Open on Thursday night, it wasn't only cameras from fans, broadcasters, or journalists capturing her every move. That's
[2]
IBM Tracked 21 Joints to Score US Open Serves. The Data Could Reveal Performance Trends
A new AI-powered capability is enabling fans and athletes to tap into one of the complicated aspects of the US Open. This year, IBM, which is behind the tournament's digital app and website, launched a new feature called Serve Quality. As tennis fans can attest, the serve is considered to be one
[3]
IBM Is Using The U.S. Open To Prove Sports Are the Perfect Playground for AI
Opinions expressed by Entrepreneur contributors are their own. Many brands use the U.S. Open to entertain clients or put up flashy signage. IBM does both, but its involvement goes much deeper. Behind the scenes, the century-old tech titan is the invisible hand shaping the fan experience, powering
Share
Copy Link
IBM launched Serve Quality at the 2026 U.S. Open, using AI-powered limb tracking to score every singles player's serve. The feature analyzes 21 joints and body points 50 times per second using Hawk-Eye cameras, generating approximately 1.2 billion data points throughout the tournament and delivering real-time performance insights directly to fans through the official app.

IBM AI launched Serve Quality at the 2026 U.S. Open, marking the first time a Grand Slam tennis tournament has offered fans access to AI-powered limb tracking technology
1
. The feature analyzes 21 joints and body points on every singles player competing in the tournament, capturing movements 50 times per second through Hawk-Eye cameras positioned around Arthur Ashe Stadium2
. IBM's WatsonX processes skeletal data to generate a Serve Quality score out of 100, considering knee, wrist, and elbow movements along with ball toss, racquet positioning, and player-movement tracking data1
.The U.S. Open app delivers near-real-time performance insights by processing approximately 4.6 million data points per match
3
. IBM estimates that 1.2 billion joints will be analyzed by the end of the tournament, generating 7 million serve quality insights1
. Fans access these insights through the match recap section of the U.S. Open app, where an AI chatbot provides conversational responses about player performance1
. When Jessica Pegula faced Aryna Sabalenka in the semifinals, the AI-powered feature revealed Pegula posted a 72.32% serve quality score compared to Sabalenka's 71.95%, despite losing the match1
.Tyler Sidell, Technology Program Director of Sports & Entertainment Partnerships at IBM, explained that 12 Hawk-Eye cameras capture limb movements, feeding data into IBM's proprietary platform
1
. The company trained AI models on 2025 data and incorporated academic papers to weight the system appropriately, with 2026 marking the first production year for fans1
. IBM gained access to two years of raw optical limb-tracking data collected during the 2024 and 2025 seasons by Hawk-Eye, the sports technology company known for operating electronic line-calling systems2
. Brian Ryerson, Senior Director for Digital Strategy at the USTA, noted that skeletal data represents a rich and heavy data set that the team worked to present in a digestible format for fans1
.Serve Quality focuses on the serve because it represents the most important stroke in a tennis match, according to Sidell
1
. The AI-powered feature provides detailed analysis showing serve placement accuracy, with one example noting a player landed 75.95% of 79 total serves in the box with an average placement of 1.555 feet away from the optimal serve zone1
. Serve scores become available for every singles match only after completion, allowing fans to review performance trends and understand how technical execution impacts match outcomes1
.IBM powers the U.S. Open app and website, engaging 14 million fans worldwide through its digital infrastructure
3
. More than 14 million unique devices engaged with U.S. Open digital platforms in 2025, generating nearly 47 million visits, up 19% year over year3
. The 2026 tournament introduced multiple AI features including an AI chat feature, key moments highlighting during matches, and a redesigned Live Updates homepage offering personalized content1
3
. IBM protects against nearly 500 million suspicious requests targeting the USTA's infrastructure, using AI agents to identify and prioritize potential threats3
.Related Stories
The USTA's editorial team uses bespoke AI tools to generate match summaries, which humans then review and edit, increasing content production capacity by 300%
3
. This collaboration between IBM AI and the USTA demonstrates how a few-hundred-person team can scale operations during the three-week tournament window. The IBM 2026 Sports Survey found that 80% of fans see value in AI-powered sports experiences, particularly real-time statistics and personalized content3
. Among tennis fans specifically, 64% said they trust AI-generated sports content, while 73% use sports apps regularly and 91% use them during live events3
.Both IBM and USTA executives indicated that other shots, such as forehands and backhands, could eventually be tracked and shared with app users in coming years
1
. Ryerson defined success for Serve Quality through two factors: ensuring fans understand the technical data set and enhancing day-over-day storytelling with maximum accuracy1
. IBM's sports strategy extends beyond the U.S. Open to include partnerships with the UFC, the Masters, and Wimbledon, where it runs similar programs powering digital platforms and handling data for each event3
. The USTA launched the inaugural USTA Connect Innovation Challenge earlier this year, providing selected participants access to USTA and U.S. Open data sets to build next-generation digital solutions for tennis3
.Summarized by
Navi