This is a preview and has not been published.

Court Surfaces: A Compositional Approach to Tennis Analytics

Authors

  • Pepus Daunis-i-Estadella Univ. of Girona
  • Ernest Baiget National Institute of Physical Education of Catalonia (INEFC). University of Barcelona https://orcid.org/0000-0002-2059-2846
  • Martí Casals National Institute of Physical Education of Catalonia (INEFC). University of Barcelona and Sport and Physical Activity Studies Centre (CEEAF), Faculty of Medicine. University of Vic-Central University of Catalonia https://orcid.org/0000-0002-1775-8331

DOI:

https://doi.org/10.17713/ajs.v54i5.2102

Abstract

Compositional data analysis (CoDA) has been extensively applied in fields such as microbiomics, geosciences, and health sciences, yet its potential in sports analytics remains largely untapped. This study applies CoDA to analyze the proportional distribution of matches played on different tennis court surfaces —clay, hard, and grass— and its influence on player rankings and performance. Data from 1,171 Association of Tennis Professionals (ATP) players ranked in the top 100 were examined, revealing significant differences in surface composition across ranking categories (p < 0.001). High-ranked players dedicated 57.8% of their total minutes to hard courts, compared to 32.6% on clay and 9.6% on grass, while lower-ranked players showed a lower proportion of tournaments on hard and a higher proportion on grass. Temporal trends indicated a shift from clay dominance in earlier decades to a predominance of hard courts, with players from the 2010s dedicating 59.8% of their tournament play to hard surfaces. These findings underscore the potential of CoDA to uncover nuanced relationships in sports data, providing actionable insights for optimizing strategies in tennis performance.

Downloads

How to Cite

Daunis-i-Estadella, P., Baiget, E., & Casals, M. Court Surfaces: A Compositional Approach to Tennis Analytics. Austrian Journal of Statistics, 54(5), 160–177. https://doi.org/10.17713/ajs.v54i5.2102

Issue

Section

Special Issue on Compositional Data Analysis and CoDaWork 2024