We now hear about exciting new use cases for Generative AI every week. One that caught my eye recently involved the first top European football club to start using GenAI to inform its player recruitment decisions. The club in question is Sevilla FC in Spain which is using IBM's watsonx AI and data platform designed for building custom AI business applications.
Sevilla has won eight European titles, including seven UEFA Europa Leagues and a UEFA Super Cup. In addition to winning on the pitch, it's also an innovator in using data and tech to support player recruitment. Partnering with IBM to bring GenAI into the mix was just the logical next step.
In elite football, identifying the right players can involve multi-million dollar investments in transfer fees and player contracts. As any fan will tell you, significant uncertainty exists when it comes to bringing in the right football talent. So, clubs try to maximize every advantage they can to get an edge.
watsonx enables organizations to easily build custom AI applications, manage all data sources, and accelerate responsible AI workflows - all on one platform. Sevilla's new GenAI application, Scout Advisor, relies on watsonx's natural language processing and foundational models to analyze the huge volume of player information the club has collated in its existing databases. As well as quantitative data about players (height, weight, speed, number of goals scored and minutes played etc) it incorporates unstructured qualitative data including text from more than 200,000 reports about individual players that have been gathered by the scouting team.
With Scout Advisor, the team's scouting department can use simple natural language prompts to describe the key characteristics of the players they want to identify. The system will then generate curated lists of candidates summarizing the full set of scouting reports for each player, streamlining the process and ensuring all relevant data and insights are captured.
Sevilla FC Sporting Director, Victor Orta describes how by entering a query such as "extremo con desparpajo," which roughly translates to "winger with confidence," he can quickly receive results that capture the essence of his request.
"I don't need to review 45 reports for a player to know the opinion of my scouting department for the player. In perhaps two minutes, I can get all the information that is good for me to make the decision," Orta said. "This is a revolutionary tool for a director of football that gives time."
Traditionally, player recruitment has relied on a combination of subjective human observation and manual data analysis. This would involve scouring through multiple reports and data points and trying to connect the dots. As you can imagine, it's time-consuming and can only take account of a limited number of factors.
Scout Advisor is helping Sevilla using GenAI to maximize the potential of a detailed metrics-based scouting system with human-centric observations that cannot be as easily measured on the pitch. Given the importance of getting player recruitment right, other elite football teams around the world are sure to follow in its footsteps. It could help drive a more competitive transfer market, as clubs are able to identify and secure top talent more effectively.
Looking ahead, the application of GenAI in elite football is likely to become increasingly sophisticated. For example, there's the potential for real-time analysis of player performance during matches and the development of highly personalized training programs. You can imagine GenAI potentially being used to analyze opponent tactics, optimize training programs and even assist in predicting player injuries based on performance data.
More broadly I would not be surprised if the success of GenAI in football inspires other sports, such as basketball or hockey, to adopt similar technologies for player recruitment and development. By the same token, we're very likely to see GenAI being applied in a similar way in other industries, such as enhancing talent acquisition in the business world.
This blog was first published on the IBM Community.