Rankings are everywhere in sport — but how are they actually constructed, and what makes a ranking system fair?
In some sports, ranking competitors seems straightforward: runners can be ranked by time, swimmers by speed, and weightlifters by the amount they lift. However, the situation quickly becomes more complicated in many real-world settings. In the Paralympics, for example, performance must be considered alongside the athlete’s classification and specific disability. How should such adjustments be made fairly? What does “fairness” even mean in this context, and can it be measured statistically?
Other sports face different challenges. In chess, players are not ranked by a single measurable performance, but through a network of wins, losses, and draws against opponents of varying strength. Existing systems such as the Elo rating system attempt to estimate player ability from these results, but they are based on assumptions that may not always hold in practice. How accurately do current systems reflect true performance? Do some players systematically benefit or suffer from the way ratings are updated? Could alternative approaches perform better?
In this project, you will explore statistical questions surrounding ranking systems in sports and games using real or publicly available data. Depending on your interests, the project could involve statistical modelling, Bayesian methods, simulation, network-based approaches, or machine learning. Possible directions include comparing existing ranking systems, designing alternative methods, studying fairness and bias, or investigating how rankings change over time.
This project is suitable for students interested in statistics, sport analytics, games, decision-making, or fairness in quantitative systems. It offers plenty of scope for creativity and for tailoring the project toward your own interests.
Supervisors
Primary Supervisor: Elena Moltchanova
Key qualifications and skills
Solid grasp of probability theory, algebra and calculus and familiarity with software such as R or Python. Knowledge of basic optimisation theory and Monte-Carlo simulations is a bonus.
Does the project come with funding
No - Student must be self-funded
Final date for receiving applications
Ongoing
How to apply
Apply by email to primary supervisor
Keywords
statistics; sports analytics; equity and fairness in sports; ranking systems
