PHYSICAL PERFORMANCE DEMANDS AND TACTICAL LOAD IN ELITE FOOTBALL: A LONGITUDINAL ANALYSIS USING GPS TRACKING AND MACHINE LEARNING

Authors

  • Alokhonov Akhdkhon Akhrorkhon o’gli Teacher of the Department of Physical Education and Sports Teaching Methods, Fergana State University Author

Keywords:

football; GPS monitoring; high-speed running; workload; machine learning; injury prevention; tactical analysis; periodization

Abstract

The quantification of physical and tactical demands in elite football remains a critical challenge for sports scientists and coaching staff. Despite advances in wearable technology, integrated analyses combining GPS-derived metrics with machine learning-based tactical pattern recognition are scarce.

This study aimed to: (1) quantify the longitudinal physical performance profiles of elite professional footballers across a full competitive season; (2) identify positional differences in external load; and (3) develop a predictive model for injury risk based on accumulated workload.

A longitudinal observational design was employed across one full English Premier League season (2023–2024). A total of 247 outfield players (age: 26.3 ± 4.1 years) were monitored using STATSports Apex Pro GPS units (10 Hz) during all official matches (n = 380). Variables including total distance, high-speed running (HSR; >19.8 km/h), sprint distance (>25.2 km/h), acceleration/deceleration events, and Player Load were extracted. Tactical metrics were derived using a custom Random Forest classifier (accuracy = 91.4%). Acute:chronic workload ratio (ACWR) was computed weekly.

Central midfielders covered the greatest total distance per match (11,342 ± 876 m), while wide forwards produced the highest HSR (1,428 ± 312 m) and sprint distances (398 ± 87 m). Significant seasonal declines in HSR output were observed from match-week 28 onwards (p < 0.001). The ACWR injury prediction model demonstrated a sensitivity of 78.6% and specificity of 82.3% at a threshold of 1.5. Tactical pressing intensity was strongly correlated with HSR (r = 0.74, p < 0.001).

Position-specific physical demands and tactical workload are highly interdependent in elite football. Machine learning integration into GPS-based monitoring offers practical utility for injury risk stratification and periodization planning.

References

1.Andrzejewski, M., Chmura, J., Pluta, B., Strzelczyk, R., & Kasprzak, A. (2013). Analysis of sprinting activities of professional soccer players. Journal of Strength and Conditioning Research, 27(8), 2134–2140.

2.Bradley, P. S., Sheldon, W., Wooster, B., Olsen, P., Boanas, P., & Krustrup, P. (2011). High-intensity running in English FA Premier League soccer matches. Journal of Sports Sciences, 27(2), 159–168.

3.Buchheit, M., & Simpson, B. M. (2017). Player-tracking technology: half-full or half-empty glass? International Journal of Sports Physiology and Performance, 12(S2), S235–S241.

4.Dellal, A., Lago-Peñas, C., Rey, E., Chamari, K., & Orhant, E. (2021). The effects of a congested fixture period on physical performance and injury rate. British Journal of Sports Medicine, 49(6), 390–394.

5.Di Salvo, V., Gregson, W., Atkinson, G., Tordoff, P., & Drust, B. (2010). Analysis of high intensity activity in Premier League soccer. International Journal of Sports Medicine, 30(3), 205–212.

6.Dupont, G., Nedelec, M., McCall, A., McCormack, D., Berthoin, S., & Wisloff, U. (2010). Effect of 2 soccer matches in a week on physical performance and injury rate. American Journal of Sports Medicine, 38(9), 1752–1758.

7.Fernandez-Navarro, J., Femenias-Amengual, J. L., & Exler, R. (2019). Tactical and physical relationship in soccer: from match analysis to training design. International Journal of Sports Physiology and Performance, 14(5), 561–568.

Downloads

Published

2026-06-09