Women’s Rugby World Cup premiered GenAI for broadcast match analysis
England’s triumph at the Women’s Rugby World Cup also saw the debut of generative AI for real-time broadcast rugby match analysis
The Women’s Rugby World Cup, which culminated with a 33-13 victory for England’s Red Roses over Canada, saw the debut of generative artificial intelligence (GenAI) for real-time match analysis, through a partnership with supplier Capgemini.
England’s triumph and female empowerment through sport were the headline stories coming out of the finals of the Women’s Rugby World Cup – where New Zealand also beat France 42-26 to third place – but for sports data analytics enthusiasts, there was much interest in the use of real-time AI-generated statistics, relayed on television screens globally.
The IT services and consulting firm developed a system called TryZone IQ for the tournament – described as generative AI technology that supports commentary teams by providing real-time match analysis and contextual insights during live broadcasts. Capgemini claimed this was the first-ever use of GenAI at a Rugby World Cup.
In collaboration with World Rugby and sports data provider Opta, the supplier said it studied fan behaviour and preferences to shape features that convert raw data into concise narratives.
Steven Webb, UK chief technology and innovation officer at Capgemini, said: “As women’s rugby reaches new heights and the World Cup draws global attention, we felt there was a unique opportunity to apply innovation to make the game more engaging for a broader fanbase.
“With TryZone IQ, we’re infusing the transformative power of generative AI into the heart of the game to meet the expectations of fans today who are looking for real-time stories, context and clarity as they follow the sport. It is a great example of how innovation can help fans feel closer to the action, while enhancing their overall experience and enjoyment along the way.”
The application, which was used by broadcast media during the tournament, brings together historic and live match data, which is updated minute-by-minute in what Capgemini describes as a generative AI engine – but there is also a “human in the loop” element to what is then broadcast in an editorial workflow.
Rugby analysts select and refine the data, ensuring it is not out of line with common-sense analysis of the game. These insights can be used to support live commentary or integrated across digital platforms and social channels, adding context to what viewers and consumers of social media are seeing.
World Rugby, Opta and Capgemini also claimed they made a conceptual breakthrough for rugby analysis with two new on-screen visualisations. One was “expected tries”, using the same approach seen in football with the xG measure for expected goals, which shows how many try-scoring opportunities have been created compared to actual tries scored. The other was “attacking threat”, which translated live metrics, such as possession, territory and momentum, into a percentage power-bar on screen that revealed who was dominating play, and so most likely to score.
Capgemini also claimed it has played a pivotal role in advancing inclusivity in the sport as a global partner of the Women in Rugby initiative and enabling the Capgemini Women in Rugby Leadership Programme to support a new generation of female leaders in rugby.
Rugby is no stranger to analytics. Welsh professional rugby player Dan Biggar gave his thoughts on its role on the playing side to Computer Weekly earlier this year, before his retirement. It can only develop more as the years roll by.
