An artificial intelligence model can predict the outcome of corner kicks in football matches and help coaches design tactics that increase or decrease the probability of a player taking a shot at goal.
Petar Veličković at Google DeepMind and his colleagues developed the tool, called TacticAI, as part of a three-year research collaboration with Liverpool Football Club.
Corner kicks are awarded when the ball goes out of play over the goal line, and can be a good scoring opportunity for the attacking team. Because of this, football coaches develop detailed plans for various scenarios, which players learn ahead of games.
TacticAI was trained on data from 7176 corner kicks in England’s 2020 to 2021 Premier League season, including each player’s position over time and their height and weight. It learned to predict which player would be the first to touch the ball after the corner kick was taken. In tests, the receiver of the ball was among TacticAI’s top three candidates 78 per cent of the time.
Coaches can use the AI to generate tactics for attacking or defending corners that maximise or minimise the chance of a certain player receiving the ball, and of a team being able to take a shot at goal. It does this by mining real examples of corner kicks for similar patterns, then offering suggestions for how to change the tactics to achieve the desired outcome.
Abstract
Identifying key patterns of tactics implemented by rival teams, and developing effective responses, lies at the heart of modern football. However, doing so algorithmically remains an open research challenge. To address this unmet need, we propose TacticAI, an AI football tactics assistant developed and evaluated in close collaboration with domain experts from Liverpool FC. We focus on analysing corner kicks, as they offer coaches the most direct opportunities for interventions and improvements. TacticAI incorporates both a predictive and a generative component, allowing the coaches to effectively sample and explore alternative player setups for each corner kick routine and to select those with the highest predicted likelihood of success. We validate TacticAI on a number of relevant benchmark tasks: predicting receivers and shot attempts and recommending player position adjustments. The utility of TacticAI is validated by a qualitative study conducted with football domain experts at Liverpool FC. We show that TacticAI’s model suggestions are not only indistinguishable from real tactics, but also favoured over existing tactics 90% of the time, and that TacticAI offers an effective corner kick retrieval system. TacticAI achieves these results despite the limited availability of gold-standard data, achieving data efficiency through geometric deep learning.
Introduction
Association football, or simply football or soccer, is a widely popular and highly professionalised sport, in which two teams compete to score goals against each other. As each football team comprises up to 11 active players at all times and takes place on a very large pitch (also known as a soccer field), scoring goals tends to require a significant degree of strategic team-play. Under the rules codified in the Laws of the Game1, this competition has nurtured an evolution of nuanced strategies and tactics, culminating in modern professional football leagues. In today’s play, data-driven insights are a key driver in determining the optimal player setups for each game and developing counter-tactics to maximise the chances of success2.
When competing at the highest level the margins are incredibly tight, and it is increasingly important to be able to capitalise on any opportunity for creating an advantage on the pitch. To that end, top-tier clubs employ diverse teams of coaches, analysts and experts, tasked with studying and devising (counter-)tactics before each game. Several recent methods attempt to improve tactical coaching and player decision-making through artificial intelligence (AI) tools, using a wide variety of data types from videos to tracking sensors and applying diverse algorithms ranging from simple logistic regression to elaborate neural network architectures. Such methods have been employed to help predict shot events from videos, forecast off-screen movement from spatio-temporal data, determine whether a match is in-play or interrupted, or identify player actions.
The execution of agreed-upon plans by players on the pitch is highly dynamic and imperfect, depending on numerous factors including player fitness and fatigue, variations in player movement and positioning, weather, the state of the pitch, and the reaction of the opposing team. In contrast, set pieces provide an opportunity to exert more control on the outcome, as the brief interruption in play allows the players to reposition according to one of the practiced and pre-agreed patterns, and make a deliberate attempt towards the goal. Examples of such set pieces include free kicks, corner kicks, goal kicks, throw-ins, and penalties2.
Among set pieces, corner kicks are of particular importance, as an improvement in corner kick execution may substantially modify game outcomes, and they lend themselves to principled, tactical and detailed analysis. This is because corner kicks tend to occur frequently in football matches (with ~10 corners on average taking place in each match), they are taken from a fixed, rigid position, and they offer an immediate opportunity for scoring a goal—no other set piece simultaneously satisfies all of the above. In practice, corner kick routines are determined well ahead of each match, taking into account the strengths and weaknesses of the opposing team and their typical tactical deployment. It is for this reason that we focus on corner kick analysis in particular, and propose TacticAI, an AI football assistant for supporting the human expert with set piece analysis, and the development and improvement of corner kick routines.
TacticAI is rooted in learning efficient representations of corner kick tactics from raw, spatio-temporal player tracking data. It makes efficient use of this data by representing each corner kick situation as a graph—a natural representation for modelling relationships between players (Fig. 1A, Table 2), and these player relationships may be of higher importance than the absolute distances between them on the pitch. Such a graph input is a natural candidate for graph machine learning models, which we employ within TacticAI to obtain high-dimensional latent player representations. In the Supplementary Discussion section, we carefully contrast TacticAI against prior art in the area.
Uniquely, TacticAI takes advantage of geometric deep learning10 to explicitly produce player representations that respect several symmetries of the football pitch (Fig. 1B). As an illustrative example, we can usually safely assume that under a horizontal or vertical reflection of the pitch state, the game situation is equivalent. Geometric deep learning ensures that TacticAI’s player representations will be identically computed under such reflections, such that this symmetry does not have to be learnt from data. This proves to be a valuable addition, as high-quality tracking data is often limited—with only a few hundred matches played each year in every league. We provide an in-depth overview of how we employ geometric deep learning in TacticAI in the “Methods” section.
From these representations, TacticAI is then able to answer various predictive questions about the outcomes of a corner—for example, which player is most likely to make first contact with the ball, or whether a shot will take place. TacticAI can also be used as a retrieval system—for mining similar corner kick situations based on the similarity of player representations—and a generative recommendation system, suggesting adjustments to player positions and velocities to maximise or minimise the estimated shot probability. Through several experiments within a case study with domain expert coaches and analysts from Liverpool FC, the results of which we present in the next section, we obtain clear statistical evidence that TacticAI readily provides useful, realistic and accurate tactical suggestions.
In a blind test, football experts from Liverpool FC were unable to distinguish AI-generated tactics from human-designed tactics, and they favoured the AI-generated tactics 90 per cent of the time.
But despite its ability, Veličković says TacticAI is in no way intended to put human coaches out of work. “We are strongly in support of AI systems that amplify human capabilities and leave them more time for the creative part of their work, rather than a system that would replace them,” he says.
Veličković says the research also has wider applications beyond sport. “If we can model the game of football, we can model several aspects of human psychology better,” he says. “AIs, as they get more capable, they’re going to need to have a better understanding of the world, especially under uncertainty. Our system is capable of giving decisions and proposals under uncertainty. These are skills that we believe will be transferable to future AI systems, so it’s a good proving ground.”
Journal reference Nature https://www-nature-com.libproxy1.nus.edu.sg/articles/s41467-024-45965-x