King's College London

At the first World Chess Championship held in 1886, only about a quarter of matches ended in a draw but, over the decades, that percentage has risen steadily, with the 2021 championship seeing almost two thirds of games end in a draw.

This proliferation of draws has proved a source of frustration for fans and players alike so, in a bid to break the cycle, researchers have shown how AI could be introduced to break the deadlock and offer a decisive view of matches.

Instead of resorting to lower quality tie-breaker games, which involve shorter time limits and more frantic moves, an AI engine would observe all moves made by players across a tied match and award the victory to one player based on the quality of moves made.

The suggestion was put forward in new research co-authored by Dr Mehmet Ismail, of King’s College London, and Professor Nejat Anbarci, of Durham University, in a paper forthcoming in the journal PLOS One.

The academics said: “In a nutshell, for each position in the game, our method measures the difference between the evaluations of a player’s actual move and the ‘best move’ as deemed by a powerful chess engine.

“In case of a tie in a chess tournament, the player with the higher ‘quality’ measure wins the tiebreak. We have applied this method to the tied World Chess Championship matches in history as well as the 2022 US Women’s Chess Championship, which went to a speed chess tiebreak between Jennifer Yu and Irina Krush. Although Irina Krush made an illegal move under time pressure and eventually lost the game and the championship, according to our tiebreak method she would have been the champion.”

The researchers also believe AI engines could be introduced as tiebreaker mechanisms in a host of sports and games in the future when they are accepted to be superior to humans, such as tennis and football.