Performance improvement, player monitoring, refereeing, commentary… How artificial intelligence is taking over the world of sports

Artificial intelligence (AI) has predicted that Ireland will emerge victorious in the Rugby World Cup, with France hot on their heels… And if its clairvoyant talents are anything to go by, AI has already proven itself in other aspects of the sport. From Ligue 1 clubs to Ligue 15 and tennis, franceinfo: sport decodes the advances in this technology.

In sports the use of statistics and data (“Data“in English) is not new. With its origins in the world of baseball, which developed in the 1990s, particularly in video games, data has occupied an essential place in sports since the beginning of the 2000s. Talking numbers have become part of sports terminology and no There is only a small step between data analysis and artificial intelligence, and you still have to know how to work with and understand it.

Data, artificial intelligence, men in the shadows

Fabien Galthie, France’s XV coach, did it A little over three years ago. After meeting the director of SAS Viya at a seminar, he invited this artificial intelligence company to create a platform adapted to the needs of the French Rugby Federation. “We wanted to answer simple questions, He explained in a press conference in November 2022. How do you train better to be better in matches? How will we defeat our next opponent? What did the teams that beat it do and what did the teams that did not do?“

Sahibi Shaib, a data scientist, is at the heart of this collaboration between high technology and rugby. For him, the difference between data and AI lies in the ability of AI to “It relies on a large volume of data in order for the machine to learn and be able to make decisions or predict events. Refers to predictive analytics, used to predict the future based on patterns established from the past. Thanks for that Interactive AI platform, Blues employees can ask questions with the goal of “o“Reject intuition”My friend Shayeb told France Info: Sport. “The coach can ask : ‘What is the best option for a camp outing? There, the AI ​​will conduct a full analysis to try to display the risks and opportunities of each option in the clearest way.”

In the service of performance

This predictive analysis is not divination. Rather, it is the logical continuation of what we call exploratory analysis, which consists of collecting and classifying data to extract patterns. Nicolas Bova, a video analyst for French rugby players, summarizes for 20 minutes Do it as you will “To extract one or two key pieces of information from this mass of data to share with the coaches, who will then develop a game plan and discuss it with the players.”

And he pays! A typical case is the Blues’ victory over the All Blacks in November 2021 (40-25). Using the platform, the Tricolores team simply asked the AI ​​what differences existed between the winning black teams and the losing teams. Bova explained in October 2022 in Crisis Podcast the team That they are finallyI was able to spread the information that was crucial in winning this match. “It was data on how long the opponent had possession of the ball.” Those who lost to the New Zealanders clearly exposed themselves by keeping the ball too long. Therefore the maximum time of ball possession and the maximum number of balls were set for the XV of France.

This effective tool is also applied in the football field. For two seasons, the League of Professional Football (LFP) implemented a system tracking Video on Ligue 1 and Ligue 2, written by artificial intelligence. Through these videos, together with GPS data of the players, all important information is shown to the staff: average speed, peak speed, distance traveled, decisive actions, passes… Artificial intelligence represents a real added value : “PTo process two million lines [les données d’un seul match]It will take hours and hours of work, appreciates Aurelien Duberne, director of video analysis at Toulouse FC. againstSave time, and that time can be reinvested elsewhere“. Then, it’s up to everyone to make the most of the LFP data to get the most out of it.

“It’s like a chef. You have the ingredients, but you don’t necessarily have to cook the same way.” […] “There is no miracle recipe.”

Aurelien Duberne, Director of Video Analysis at Toulouse FC

In franceinfo: Sports

One ever-troubling question remains: Will artificial intelligence, which can work day and night, replace men on the sidelines? Aurelien Dubern admits that with the speed at which he is progressing, he will take a special place, but there is no room to replace the human factor: “Clearly there will be a dedicated AI department in Teams, just as there is a video department, which didn’t exist ten years ago.” He mentions that as well “soccer [et le sport en général], He remains a human being, with all his feelings and what he generates. Will artificial intelligence be able to sense all the emotions in the locker room? It’s not tomorrow the day before.”

Filter, target, evaluate and compare information with a few clicks between individual player performance and strategy results… Ultimately, AI provides a quantitative, accurate and reliable report, which complements the employee experience. For Damien Comolli, president of the Toulouse club, these methods aim to “Removing as much as possible the doubts and irrationality that exist in football“Ve said the team.

The Toulouse club is considered a pioneer in its use of new technologies. As of 2020, its recruitment is based on an AI algorithm to identify talent. The tool developed allows to condense all the statistics related to the player and suggest the most suitable statistics for what the coach wants. “With our budget, which is not one of the largest in the first level, we have to try to find solutions to be more efficient“, Aurelien Duberne explained to France Football.

Referee, commentator… Artificial intelligence is coming to the stadiums

Artificial intelligence can also be useful for viewers. In view of the recent protests carried out by supporters of the French and then the English in a confrontationAccording to referee Ben O’Keefe’s insights during the Rugby World Cup, refereeing remains at the heart of discussions more than ever. According to reports, the artificial intelligence was able to count the mistakes made by O’Keefe Western France. According to her, he committed 10, including eight that would have put the French at a disadvantage against South Africa. Can artificial intelligence become a support for rulers in the future? until To make up for the shortage of official referees in lower divisions e.g ZDNet RequestA website that specializes in new technologies?

Pierre Miralles, head of artificial intelligence at Footovision, a start-up specializing in statistical analysis, points to the help in detecting match-fixing. “sWhen we look for players who perform well, we can also find those who perform poorly.”details for the team.

Thanks to the power of artificial intelligence, on football fields, 3D images can be projected onto screens to identify them Whether the player is offside or not. For Guy Akusebere, a former French XV player, the same model would be desirable in rugby, where the rules are sometimes unknown or poorly understood.

Towards the decline of humans?

The ChatGPT poll illustrates this; Generative AI, which can construct a real-time verbal or written response to a given situation, has made significant progress. In July, this revolution found a direct translation: Comments on summaries of some matches Wimbledon was implemented by artificial intelligence. According to the organizers of the Grand Slam Tennis Tournament, the long-term goal is to be able to commentate entire matches in tournaments that do not always benefit from commentary.

“Replace John McEnroe’s commentsThis is impossible!“, Nuance Bill Jenks, CTO of the prestigious London Championships.

“The human element must always be there, it is simply a matter of complementing it. The challenge is to find the right balance between tradition and innovation.”

Bill Jenks, Wimbledon’s chief technology officer

To Agence France-Presse

AI is neither omnipotent nor willing to play games in place of actors, and AI and its gradual democratization raise two questions: how far it will be able to go, and Will this lead to unification of playing styles?