Tech in Euro 2020 - Forging football champions with AI

Football frenzy is with us; people follow their favourite teams religiously, sipping all sorts of unhealthy beverages and munching truckloads of yummy snacks. TV stations bombard us with all kinds of analyses about what’s happening in the various international competitions and keep us informed about the odds of seeing our favourite team raise that much-desired cup. But did you ever ask yourself if Artificial Intelligence (AI) has any role in all this?

Of course, we know some of these technologies, such as Goal Line Technology (GLT) and Video Assistant Referee (VAR). These systems help referees to make the right decisions during matches. GLT checks whether the ball crossed the goal line or not, while VAR reviews the head referee’s decision. But under the hood, clubs are using other systems, which very few people realise. Let’s not forget that football is a multi-million euro sport. The human body is the principal machine used to play it. So it makes sense to use AI techniques to try to optimise that machine.

The first prominent use is to improve the single performance of the athlete. To do so, clubs typically rely on Gait analysis, amongst others. This analysis is the systematic study of human motion, using instruments to measure body movements, mechanics, and the activity of the muscles. Through this, they can figure out whether the athlete is running correctly, identify if he has any weaknesses and suggests ways to improve further. A typical Gait analysis system requires an expensive lab equipped with around 16 cameras carefully monitoring the movements of the individual. Apart from the fact that not everyone can have access to such a system, it also involves a lot of time and effort. But recent advances in AI technologies have made it possible to integrate such systems into a mobile device, thus allowing coaches to perform automated Gait analysis directly from the pitch and get the initial indications without any delays.

However, football is not an individual sport, and teamwork is critical. Because of this, the second obvious use is in the review of past matches. This analysis has two significant benefits: optimising the team and identifying flaws in the opponent. By following every team member during past games, the AI can extract essential statistics such as successful passes, interceptions, receptions, offences, kicking accuracy, goals, speed, fatigue, and so much more. The system maps group dynamics and highlights them. It then analyses this information and identifies the most crucial factors, thus helping the coach determine the best strategy for any player within the context of the team. The AI also processes the video feeds of the opposite team, but this time, its role is inverted. Rather than for optimisation purposes, it seeks to exploit weaknesses in the opposing team. This information will allow the coach to reorganise his team to use the opponents’ shortcomings in his favour.

On paper, such a system will sound fabulous. But what happens during the match? We all know that football is a live dynamic game. Even if the team tries to exploit the opponent’s weaknesses, the other coach might realise that and make adjustments to counteract such faults. Once again, AI comes into play.

The system is not only an analytical process that crunches data before the game. It is also an online coach monitoring the live game too. So changes happening in the opposing team are identified in real-time, and countermeasures proposed to the head coach. This analysis also applies to his team. Furthermore, we all realise that AI is much faster than any human. It stores vast amounts of information about every player, starting from his first match up to his most recent one. The system can realise if the player is doing well or underperforming. By using all of this information, the AI can play several virtual matches in its digital brain. It can take a peep into the future and try to identify the most successful strategy. Once it is confident enough with the plan, it can suggest it to the head coach for his consideration.

Of course, we have to keep in mind that this is not an exact science. There have been instances in the past where AI failed miserably in football. Just a year ago, a Scottish AI system designed to track the ball during a live football game constantly got confused by the bald head of the linesman repeatedly thinking it was the ball. After the match, they resolved the issue in no time, but the truth is that AI is not infallible.

In essence, there is no guarantee that AI predictions turn into reality. But given the enormous repositories of information that the AI has access to, the logic inbuilt in such systems and the considerable processing power it possesses, it is most probably the next best shot. What’s for sure is that the odds will favour those clubs that make early use of AI. And very soon, these enhancements will lead to a new level of football, one which was unheard of a few decades ago.