Game, Set, Tech
At its Game, Set, Tech media event ahead of the Roland Garros tournament in May, Sportradar brought together partners from ATP and TDI to examine how tennis is being reshaped by real-time data, AI-driven visualisations and player protection. For betting operators, broadcasters and sports bodies, the message was clear: tennis is becoming faster, richer and more personalised, but the technology powering that evolution also brings new responsibilities.
Tennis has always been a sport of momentum. A break point, a missed first serve, a sudden shift in rally length or shot selection can change the feel of a match long before the scoreboard fully reflects it. What Sportradar’s recent tennis and technology event made clear is that the industry is now building products capable of identifying, packaging and commercialising those shifts almost as they happen.
Across the afternoon, the conversation moved from the foundations of tennis data collection to live betting innovation, fan-facing visualisation, AI-generated content and the darker reality of online abuse directed at players. The common thread was data: how it is captured, how quickly it moves, how it is interpreted, and how it can be used responsibly.
Data Consistency
For David Lampitt, CEO of Tennis Data Innovations, the starting point was the transformation of tennis from a fragmented technology environment into a sport with a much more consistent data infrastructure. TDI, created by the ATP Tour and ATP Media, was established to manage and develop the ATP’s data and technology assets in much the same way that ATP Media manages its broadcast rights.
TDI’s role now covers a wide footprint: umpire technology, electronic line calling, player and ball tracking, non-televised streaming, performance data and the productisation of those assets for downstream partners such as Sportradar. The key challenge, Lampitt explained, was moving away from a world where ‘some data existed on some courts some of the time.’
Without consistency, sophisticated products are difficult to build. Lampitt compared the old model to having football data for only half a pitch at selected venues. Tennis needed a unified technology footprint before it could unlock the next generation of products. It’s this foundation that has enabled the ATP and TDI to build layered data products. Level one covers the traditional scoring and umpire data. Level two adds automated insights generated from tracking data, including shot type, winners and errors, rally length and other contextual markers. Level three goes deeper again, tracking ball and player position, bounce location, in/out calls and spatial information across a rally.
For Rainer Lichtmannegger, SVP Sports Content Products at Sportradar, it’s this depth that changes the betting conversation. Tennis is already one of the most naturally suited sports for in-play betting because it is structured point by point, with regular pauses, clear scoring segments and constant changes in probability. Rainer noted that in-play now accounts for an estimated 90 per cent of global tennis betting volume, although regional differences remain.
The richer the data, the more sophisticated the models can become, and faster and deeper inputs improve odds accuracy, support sharper risk management and allow sportsbooks to identify momentum shifts earlier than human traders alone. However, the more interesting commercial opportunity lies in market expansion.
Rather than betting only on match, set or game outcomes, operators can begin pricing markets around how a point or game is played: the last stroke of a rally, whether a point ends on serve or return, the number of shots in the next rally, whether a player records an ace in a service game, or whether a rally exceeds a certain length. These player-performance and micro-market products remain less mature in tennis than in other sports, but the underlying data now exists to support them.
The next step is latency. Lampitt described TDI’s work on what it calls ‘true real time,’ a system designed to compress the gap between what tracking technology already knows and when the umpire formally inputs the result. Electronic line calling already determines whether a ball is in or out before an audio call is played on court. However, the betting market can still be exposed to delay when the official score is not yet entered, particularly in situations such as clay-court mark inspections.
Reducing that gap is not simply about speed for its own sake, it addresses information asymmetry, improves trading efficiency and reduces the opportunity for courtsiding; where someone physically present at a venue gains a timing advantage over the official data feed. In Lampitt’s framing, sports betting increasingly resembles a financial market: the earlier reliable information reaches participants, the more efficiently risk can be managed.
Picture Perfect
Game, Set, Tech wasn’t only about betting. Sportradar’s visualisation presentation explored how the same data can be used to make tennis more accessible and engaging for fans. The company’s augmented streaming and visualisation products, developed with TDI, are designed to educate, inform and encourage action.
As Sportradar’s Senior Vice President (SVP) of Fan Engagement, Patrick Mostboeck noted, sports insiders often assume fans understand far more than they actually do. Visualisation can explain why a match is unfolding in a certain way, why one player is producing more aces, how rally length is changing, or where tactical pressure is being applied. For bettors, that context supports more informed decisions. For fans, it provides a deeper narrative layer.
The impact is already measurable. Mostboeck said its augmented tennis streaming product has been adopted by a significant share of sportsbook clients and has helped increase engagement and viewing time. A four per cent increase in average engagement time may sound modest, but at global scale it represents a meaningful extension of attention in an environment where sports consumption is becoming shorter, more fragmented and more mobile.
AI is accelerating this shift. Mostboeck previewed work-in-progress concepts showing how match data, automated narration, statistics and visual assets can be combined to create ready-made content clips. What once required hours of production could increasingly be generated near-instantly for publishers, social platforms, operators and rights holders.
The direction of travel is clear: more personalised, more automated and more interactive tennis content. The traditional world feed remains a high-quality product, but virtualised and data-driven content can allow fans to consume tennis in different ways depending on their interests, whether they are casual viewers, bettors, analysts or player followers.
Player Protection
The event’s final session introduced a different but equally important use of technology: protecting players from online abuse. Adam Pennock, Vice President of Risk and Investigations at Sportradar and Andrew Azzopardi, Director of Safeguarding at the ATP Tour, outlined the work being done through Sportradar’s integrity services alongside the ATP’s safeguarding team.
The numbers were stark. Since the launch of the service in July 2024, Sportradar has scanned close to five million comments across ATP player profiles, ATP channels and the wider tennis conversation. Around 300,000 were classified as abusive or threatening, representing roughly 6.5 per cent of all comments monitored. The issue is widespread: 93 per cent of monitored accounts had received at least one abusive comment. However, the abuse is not evenly distributed. A relatively small number of persistent users account for a disproportionate share, with some individuals operating dozens of accounts to harass players across multiple platforms and, in some cases, multiple sports.
Azzopardi framed the issue as a duty of care question. Tennis players travel constantly, often without stable support networks around them, and can be exposed to hundreds or thousands of abusive messages after a single match or incident. The instruction often given to athletes historically – ignore it, endure it, or leave social media – is no longer sufficient.
Pennock explained that Sportradar’s system uses AI to monitor comments in multiple languages and classify abuse, threats, spam and inappropriate activity. Where players opt in, the moderation layer can hide or delete abusive comments before they reach the athlete. The most serious cases are escalated for investigation, with Sportradar identifying individuals, assessing risk and working with tournament security, the ATP and law enforcement where necessary.
One case study illustrated the real-world value. A death threat against a player was intercepted by the system before the player saw it. Investigators identified the individual behind the message and discovered they were located in the same city as an upcoming tournament where the player was due to compete. Continued monitoring showed the person taking photographs from the venue. Security and police were alerted, the individual was detained and prevented from returning. This example underlined the session’s central point: online abuse is not always just online. Threats can move from digital hostility to real-world risk, and sports bodies need tools capable of identifying that escalation early.
Match Point
Taken together, the event presented tennis as a sport entering a new technological phase. Real-time tracking is changing the shape of betting markets. Visualisation is changing how fans understand the game. AI is changing how content is created and distributed. Integrity tools are changing how athletes are protected. And the commercial implications are significant.
Tennis offers the attributes operators increasingly want: global reach, year-round supply, point-by-point betting structure, rich data, strong player narratives and a growing toolkit for personalised engagement. However, the same technology that creates new revenue opportunities also raises questions about latency, fairness, athlete welfare, data rights and responsible use. The future of tennis betting and engagement will not be defined simply by having more data. It will be defined by how intelligently, quickly and ethically that data is used.
