Using ELO ratings for match result prediction in association football

329 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Forecasting football match results in national league competitions using score-driven time series models (2019, methodology): “We refer to the original work of Irwin (1937) and Skellam (1946) for the derivation of the Skellam distribution based on the differences between two independent Poisson distributions.”
α-Rank: Multi-Agent Evaluation by Evolution (2019, methodology): “The Elo rating system has ubiquitously been used for ranking and predicting outcomes in board games [94], sports [95–99], and artificial intelligence [3, 100].”
Elo Ratings and the Sports Model: A Neglected Topic in Applied Probability? (2017, methodology): “For the cycle model, and the Ornstein–Uhlenbeck model with larger τ , we see numerically that the optimal value of δ for σ = 1 is around twice the optimal value for σ = 0.5.”
Eliminating Supportive Crowds Reduces Referee Bias (2021, background): “…when including 3Elo ratings, originally proposed by Elo (1978) for chess players and tournaments, are commonly used to estimate the strengths and form of football teams, both in practical applications (e.g., https://www.eloratings.net/) and in academic research (e.g., Hvattum and Arntzen, 2010).”