en
Back to the list

Artificial Intelligence To Select Virtuoso Football Players

22 November 2019 05:42, UTC
Vsevolod Gnetii
The Italian professional football industry has launched an experiment to select promising young footballers at an early stage in order to make future professional football stars out of them. The flagship of the use of artificial intelligence for the above purpose was the Italian football club Serie A, Il Sassuolo, which signed a partnership with Italian startup Wallabies for the 2019–2020 season.

The company has been working in this field since 2016, and Il Sassuolo intends to use machine learning and artificial intelligence algorithms developed by Wallabies, with the aim of a scientific approach to finding promising football players and monitoring their progress on the football field. This technique is planned to be applied not only in Italy, but at a global level.

Luigi LIBROIA, managing director of Wallabies, said that the signed agreement creates such a scientific approach in the field of professional sports for the first time in world practice. Wallabies intends to use a system of algorithms to analyze a huge mass of data in a short period of time. This marks a fundamentally new approach: currently efforts are aimed at finding new players, while the Italian startup intends to focus on a comprehensive analysis of the player himself.

The machine learning used by Wallabies corresponds to the work of 400 people looking for a budding player. Luigi Libroia believes that a computer will not replace a human, but it can become a significant strategical help, opening up new opportunities for a person to overcome his inherent limitations. It is human nature to make decisions based on specific features, while a computer evaluates comprehensively the entire situation as a whole.

Wallabies takes a robotic approach to the selection of candidates for Series A, as its equipment allows to watch 8 thousand matches a year, marking and analyzing each specific player’s step. Wallabies analysis is based on 7 thousand variables for each player in each match and allows monitoring about 40 thousand players in 25 different sports leagues.