Stats Perform Partners with ENSAI for its Annual Data Challenge 

Data Science and high-level sports are at the heart of ENSAI’s 2021 Data Challenge. On March 3d and 4th, students worked in teams on issues submitted by the French Rugby Federation’s Research and Development Department. All data is provided by Stats Perform, a global SportsTech company. 

An interview with Andy Cooper, Marketing Manager at Stats Perform.

Andy, could you present Stats Perform?

Andy Cooper: Stats Perform is the global SportsTech leader in data and AI technology, providing the most trusted sports data and latest advancements in applying AI and machine learning to deliver better predictions and insights for fan engagement, betting and team performance. As a company, we have collected data across sports for 40 years and currently provide services to over 1,500 customers worldwide.

We possess a data archive which is faster, deeper and broader than any other sports database in the world.

This database is used to power our own AI Innovation Center, which is comprised of over 300 developers and 50 AI Scientists and Engineers.

They are actively building cutting-edge solutions which are providing fans with richer insights during live broadcasts, and teams with predictive performance insights to enhance their analysis and decision-making.

Stats Perform is the leader in its field: what particular technologies have made the difference?

There are a number of technologies we have pioneered, across different sports, which are making a big difference in terms of supporting analysts and coaches with applied use of data and AI-derived insights.

In rugby, we have a tool called RugbyHub which synchronizes match data and video. This synchronization helps coaches and analysts to analyze key phases of play which are specific to a game plan.

We also have the most advanced BI tool in professional rugby, DataEngine, where analysts can build customized reports to establish longer-term performance trends, across different positions on the field, which can inform decision-making across recruitment and performance analysis.

Speaking more broadly, we have recently developed and patented a computer vision tracking technology, AutoStats, which applies AI-enhanced body recognition to fluidly track a player’s location, movement and key in-game involvements, all directly from a broadcast video.

https://youtu.be/5nEzPzyfDdI

This has been pioneered in basketball to help NBA franchises evaluate elite draft prospects, using video footage from NCAA (university) matches, by comparing season-wide performance outputs which historically could only be generated through in-venue tracking hardware.

This is just one example of how AI technology is going to change the way analysis is conducted in sports during this decade and beyond.

How do Stats Perform’s technologies help the coaches and increase the athletes’ performance?

The key objective for all of our technologies is to ensure insights can be applied on the training ground.

A good example of this is Expected Goals in football (xG), which assigns a value to a shot based on the shot location and the probability that it will result in a goal, taking into account a number of different contextual factors.

From a coaching perspective, these outputs can be applied to teach attacking players about decision-making in front of the goal – identifying areas where, based on probability, it is better to pass instead of shooting, because of the difficulty to score from that location. Since xG was introduced, we are seeing fewer shots taking place from long range, which shows the impact that metric is having.

In rugby, we have also developed a success probability model to rate leading goal-kickers worldwide. The model takes into account a player’s historical kicking performance, the difficulty of a kick based on the field location and several venue-related adjustments to assign an ‘Expected Conversion Success Rate’ for goalkickers.

This has been used to power the Kick Predictor, as seen on the AWS match stats during the Six Nations championship, and can also be applied from a coaching perspective because if your opposition possesses a strong goal kicker, discipline in relation to avoiding conceding penalties is going to be crucial in the final outcome.

Could you name a team whose use of Stats Perform’s technologies led them to victory when perhaps it wasn’t expected in the first place?

Because we work with so many teams worldwide, it is very difficult to single out individual teams. For example, in football, every team that won the league title in the top five European leagues last season had services from Stats Perform, as did the winners of the UEFA Champions League and Europa League.

If we were to highlight successes, I would probably focus on two international football teams that used our Edge Analysis technology, which includes several AI prediction and simulation features, to help them achieve unprecedented success.

Croatia used the platform to attain insights into their opponents during their run to the 2018 World Cup final, whilst Qatar also used the platform to aid their match preparation at the 2019 Asian Cup, when they beat South Korea and Japan to win their first-ever confederation title.

How do fans and bettors benefit from Stats Perform’s technologies?

Every high-quality sportsbook and fantasy game chooses Stats Perform’s premium sports content to help them create more entertaining and trusted sports betting and fantasy experiences. Today’s sports bettor demands a frictionless sports betting experience.

Betting sites need to provide valuable resources for bettors and keep them entertained with live scores and rich, live statistics for thousands of events. Our data is relied upon by millions of bettors around the world and drives more confident decision-making for hundreds of thousands of sporting events per year.

As a part of the Data Challenge at ENSAI, Stats Perform will also hold a conference on Data Science and Rugby for its participants. Find out more about ENSAI’s corporate partners.