How Artificial Intelligence is changing football recruitment

One of the key lessons my team and I have learnt while working with professional football teams is around trust in AI and explainability of the model output.

Ryan Beal, founder of Sentient Sports, is poised on revolutionising the sports industry with its algorithim portal, which builds a dream football team. Ryan talks about how Artificial Intelligence is changing the health and fitness sector and how Sentient Sports is positioned for the future where AI plays a role in team sports.

“Professional sport has traditionally been a very subjective industry with decisions driven by opinion-based scouting.

However, in recent years, teams have adopted a range of technologies that collect large amounts of data from training and matches that include the movement of players during games, their health statistics, and their actual performance during such games.

This now means that more objectivity can be added to decision-making in sports, and we can add more assurance around multi-million dollar decisions.

Major teams around the world already use a variety of datasets to make decisions and improve their on-field performances but, on the wider scale, the new datasets and information can be hard to process.

This is where Artificial Intelligence (AI) comes into play.

Our innovation, Sentient Sports is a solution that uses research-developed algorithms to provide football clubs with a step change in their decision-making and scouting process for buying and selling players, promising to drive up footballing and financial results.

The way it came about was very organic. I have always been a big fan of all sports and especially football.

As the amount of data collected has grown, I was keen to explore how we could use AI to help improve decision making in sports and have been researching this as part of my Ph.D. at University of Southampton.

We came up with a new model to start evaluating how players work together and valuing them as a pair rather than individually.

This allows us to then predict how well a player will perform in a new team with new teammates. For this, we started to explore other AI methods for predicting how a player will fit a new team to help managers improve decision making for player recruitment.

The way it works is we extract the pairs of players in a game of football that often link up and contribute to the positive passage of plays (e.g. goals and shots).

From that, we can learn what makes a good pair by looking at a player’s on-field attributes and off-field factors such as player age, nationality and languages that are spoken.

One of the key lessons my team and I have learnt while working with professional football teams is around trust in AI and explainability of the model output.

Many AI models output black-box predictions which mean that it can be hard for humans to see how the AI has made its prediction.

This means that gaining the trust of users to make decision using models is a challenge. One key approach to do this is to sever more information from the model to the user – this is a big research area called Explainable AI (XAI).

In football, this is done by unpicking models to show users key features such as the languages the player speaks, the tactics he/she can play in, and which players have played more times together in the past.

This goes part of the way to explain model outputs but will be something that will be worked on significantly in the coming years.

Sentient Sports have received a huge amount of support from the startup accelerator, Future Worlds team, at the University of Southampton.

Through working with our partner, AI Abacus, a leading sports analytics company, we have launched a front-end portal for top football clubs across Europe to make use of these new AI techniques.

We’ve been researching the area for over three years and believe there is a real need for a service like this to help football decision makers.

Our AI algorithms help to provide assurance to player recruitment in football which has traditionally been a very subjective area to make decisions that cost teams hundreds of millions of pounds.”

Image Copyright – Vlad 1988 (Shutterstock.com) / Unsplash / Ryan Beal (founder of Sentient Sports) / edited by I Am New Generation Magazine

Updated – 22nd November 2021

More Stories

The most significant misconception about algorithms and AI is that it plays a central role in discriminating

In the age of AI – What new skills will organisations and their workforce need to keep up?

Using My Tech Startup To Stand Up Against Gender, Racial And Disability Bias In The Workplace