A Bayesian analysis of home advantage in professional squash

Philip Greengard, Samer TakritiarXiv 2025 · DOI · arXiv

This paper provides a rigorous statistical look at one of the most debated topics in sports: the "home court advantage." By applying a Bayesian hierarchical model to years of professional squash data, the authors quantify exactly how much playing on home soil boosts a player's performance and win probability.

TL;DR

Playing at home in professional squash provides a statistically significant boost, adding roughly 0.4 games to the margin of victory for men and 0.3 for women. For two evenly matched players, this "home effect" swings the win probability from a 50/50 toss-up to approximately 58% in favor of the local hero.

What Problem Does This Paper Address?

While the concept of home advantage is well-accepted across sports—attributed to factors like crowd support, travel fatigue, and familiarity with the venue—quantifying it precisely is difficult. In squash, the Professional Squash Association (PSA) holds tournaments globally, but certain nations (like Egypt) dominate the rankings and host many major events.

The authors sought to determine if the home advantage is a measurable phenomenon in squash and how it compares to the inherent ability gap between players of different world rankings. They also investigated whether certain countries, specifically Egypt, England, and the U.S., exhibit stronger home effects than others.

Methods

The researchers employed a Bayesian hierarchical model to analyze match outcomes.

  • Outcome Variable: Instead of a simple win/loss binary, they used the margin of victory in games (e.g., a 3–0 win is a +3, a 2–3 loss is a -1). This captures more information about the "closeness" of a match.
  • Player Ability: They modeled player ability ($a_i$) based on world rankings. Crucially, they recognized that the gap in skill isn't linear; the difference between the World #1 and #2 is often much larger than the difference between #21 and #22. To account for this, they used a prior that included both a linear and a square root term for rank.
  • Home Advantage ($h$): This was modeled as an intercept that shifts the expected margin of victory when a player is in their home country and their opponent is not.
  • Inference: They used Markov Chain Monte Carlo (MCMC) sampling via Stan, a probabilistic programming language, to estimate the parameters.

Datasets

The study utilized data from squashinfo.com, covering PSA World Tour matches from December 2018 to March 2024.

  • Scope: Includes Bronze, Silver, Gold, and Platinum level tournaments, plus World Championships and Tour Finals.
  • Filtering: The analysis focused on matches between players ranked in the top 30 to ensure high-quality data and sufficient sample sizes.
  • Normalization: "Best of three" match outcomes were scaled by 1.5 to match the "best of five" format standard in major tournaments.

Key Insights and Ideas

  • The Non-Linearity of Excellence: The model confirms that the "ability curve" is steep at the top. The average difference in ability between successive rankings is about 0.15 units, but the gap between the very top players is significantly wider.
  • Home Advantage vs. Rank: The home advantage (0.3–0.4 games) is roughly equivalent to being ranked two to three spots higher than you actually are.
  • The "Mostafa Asal" Anomaly: The authors noted a "model deficiency" where the #4 ranked man appeared "better" than the #3. They attribute this to specific players (like Mostafa Asal) who may be under-ranked due to PSA suspensions despite having elite-level performance data in the matches they did play.
  • Egypt’s "Fortress" Effect: The home advantage is particularly pronounced in Egypt (estimated at 0.45 games for men). This may be due to the sheer volume of elite Egyptian players and the intense local support they receive during the many major tournaments held there.

Key Results

  • Global Impact: Home advantage adds 0.4 games to the expected margin for men and 0.3 games for women (standard error of 0.1).
  • Win Probability: For evenly matched players, the home player’s win probability increases to 58% (men) and 56% (women).
  • Country Specifics:
    • Egypt: Strongest effects (0.45 for men, 0.35 for women).
  • USA: Similar to Egypt for women (~0.3), but data was too sparse for a precise estimate for American men.
  • England: Showed smaller home advantages, though with higher uncertainty due to fewer matches involving home players in the dataset.

Limitations and Future Work

  • Crowd Support Nuance: The model defines "home" strictly by nationality. It doesn't account for "adopted" home crowds (e.g., a popular player receiving massive support in a foreign country).
  • Discrete vs. Continuous: The model treats the margin of victory as a continuous normal distribution, whereas squash games are discrete integers. While rounding helps, a generative discrete model might be more precise.
  • Causality: The study quantifies the existence of the advantage but does not distinguish between why it happens (e.g., is it referee bias, court conditions, or psychological comfort?).
  • Data Sparsity: Outside of Egypt, the U.S., and England, there wasn't enough data to estimate precise country-specific effects for other nations on the tour.

Summary updated

Abstract

We estimate the effect of playing in one's home country in professional squash using a Bayesian hierarchical model applied to men's and women's Professional Squash Association matches from 2018-2024. The model incorporates players' world rankings and whether they are competing in their home country. Using margin of victory in games as our outcome, we estimate that home advantage adds 0.4 games for men and 0.3 games for women to the expected margin, with standard errors of 0.1. For evenly matched players, this effect corresponds to an increase in win probability from 50% to roughly 58% for men and 56% for women. We estimate particularly strong home effects in Egypt, where many major tournaments are held, though data limitations prevent precise estimation of country-specific effects in many other nations.
BibTeX
@misc{greengard2025bayesian,
  author = {Philip Greengard and Samer Takriti},
  title = {{A Bayesian analysis of home advantage in professional squash}},
  year = {2025},
  eprint = {2506.09287},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2506.09287},
  howpublished = {arXiv},
}

Loading related content...

Figure