Fantasy football is about to get a whole lot harder as an artificial football manager enters the competition.
Computer scientists have developed a simulated football manager that uses Artificial Intelligence software to pick a football team each week based on an extensive series of algorithms to analyse player’s performance and statistics
The software – developed by researchers at the University of Southampton – goes into much more depth than even the most avid fantasy football player would, and so far its performance has shown to be a runaway success. In tests, the artificial manager has ranked, on average, in the top 1% of players in the Official English Fantasy Football League run by the Barclays Premier League.
The team – Dr Sarvapali Rumchurn, Tim Matthews and George Chalkiadakis – tested their software by running a controlled experiment to see how it would have performed in last year’s competition. At one point it finished in the top 500 players.
“Our previous tests have shown that a machine working on its own will perform better than millions of humans,” said Dr Ramchurn, lecturer in Computer Science. “But a machine can’t take into consideration if a player is injured (and still plays), has low morale or has personal issues and may not perform at his best.”
To combat this problem, the team plan to enter their artificial football manager into the competition for real, but will introduce a human element.
“We will be using humans and the machine working together as a team so that the humans can add this subtle information into the system and, together with the software’s extensive analysis, it will hopefully improve the machine’s success rate, though in some cases, this could also potentially make it perform worse if humans put in inaccurate information,” said Ramchurn.
Researchers are also working on a web application which next season’s players to get advice from the artificial manager and play against it.