Bayes-ically fair: A Bayesian Ranking of the Olympic Medal Table

2025/10/16 by MacDermott, Cormac, Scarrott, Carl J., Ferguson, John
#Applications (stat.AP) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.2510.14723

Abstract

Evaluating a country's sporting success provides insight into its decision-making and infrastructure for developing athletic talent. The Olympic Games serve as a global benchmark, yet conventional medal rankings can be unduly influenced by population size. We propose a Bayesian ranking scheme to rank the performance of National Olympic Committees by their "long-run" medals-to-population ratio. The algorithm aims to mitigate the influence of large populations and reduce the stochastic fluctuations for smaller nations by applying shrinkage. These long-run rankings provide a more stable and interpretable ordering of national sporting performance across games compared to existing methods.

Related

• Predication of Final Medal Counts in Olympic Games by Monte Carlo Simulations
• Forecasting the Olympic medal distribution during a pandemic: a\n socio-economic machine learning model
• Measuring the performance of nations at the Olympic Games using DEA models with different preferences
• STGCN-LSTM for Olympic Medal Prediction: Dynamic Power Modeling and Causal Policy Optimization
• Forecasting National Medal Totals at the Summer Olympic Games Reconsidered