top of page
  • Writer's pictureSarah Ruivivar

Google DeepMind's AI, TacticAI, Revolutionises Soccer Tactics

Updated: Mar 24

📸 ModelProp / Midjourney

In the world of soccer, the most thrilling coach you've never heard of isn't pacing the sidelines of a stadium but rather, it's nestled in the labs of Google DeepMind.


This AI powerhouse has been collaborating with Liverpool Football Club to bring artificial intelligence to the world's favourite sport, and their latest creation, TacticAI, is set to revolutionise the game.


In the past, DeepMind researchers developed models that could predict penalty kicks and player movements. But Liverpool wanted something more practical, something that could provide real-time, actionable insights. Enter TacticAI, an AI model designed to predict and optimise corner kicks, a critical yet often overlooked aspect of the game.


 

Want to learn more about AI's impact on the world in general and property in particular? Join us on our next Webinar! Click here to register

 

Corner kicks provide a unique opportunity in soccer. The action on the field freezes, and the attacking team gets a chance to swing the ball into the box. But with only one in 50 corners resulting in a goal, there's plenty of room for improvement. That's where TacticAI comes in.


Using player-tracking data from thousands of corners taken in the Premier League, TacticAI represents the arrangement of players as a graph, encoding players' positions, movements, heights, and weights as nodes, and relationships between players as lines. Through geometric deep learning, the AI model can predict which player is most likely to make the first contact on the ball and whether a shot will be taken as a result.


But TacticAI doesn't stop at predictions. It also generates recommendations for the best ways to adjust player position and movement to either maximise the chance of a shot being taken (for the attacking team) or minimise it (for the defending team). This kind of insight is invaluable for coaches and analysts, who spend hours analysing footage to identify weak points in their opponents' strategies or their own team's performance.


The future of TacticAI is even more exciting, with plans to build it into a natural language interface so that coaches can query it in text and get answers to their on-field problems. While it could be used during a game to help coaches refine their corner routines on the fly, it's most likely to be useful in the days leading up to a match, freeing up coaches' time for the creative part of coaching.


In the world of soccer, AI isn't just coming; it's already here. And with TacticAI, Google DeepMind is proving that AI can be a game-changer, both on and off the field.


 

Want to learn more about AI's impact on the world in general and property in particular? Join us on our next Webinar! Click here to register

 

Made with TRUST_AI - see the Charter: https://www.modelprop.co.uk/trust-ai

9 views0 comments

Comments


bottom of page