Tier-specific evolution of match performance characteristics in the English Premier League: it’s getting tougher at the top

Abstract

This study investigated the evolution of physical and technical performances in the English Premier League (EPL), with special reference to league ranking. Match performance observations (n = 14,700) were collected using a multiple-camera computerised tracking system across seven consecutive EPL seasons (2006–07 to 2012–13). Final league rankings were classified into Tiers: (A) 1st–4th ranking (n = 2519), (B) 5th–8th ranking (n = 2965), (C) 9th–14th ranking (n = 4448) and (D) 15th–20th ranking (n = 4768). Teams in Tier B demonstrated moderate increases in high-intensity running distance while in ball possession from the 2006–07 to 2012–13 season (P < 0.001; effect size [ES]: 0.68), with Tiers A, C and D producing less pronounced increases across the same period (P < 0.005; ES: 0.26, 0.41 and 0.33, respectively). Large increases in sprint distance were observed from the 2006–07 to 2012–13 season for Tier B (P < 0.001; ES: 1.21), while only moderate increases were evident for Tiers A, C and D (P < 0.001; ES: 0.75, 0.97 and 0.84, respectively). Tier B demonstrated large increases in the number of passes performed and received in 2012–13 compared to 2006–07 (P < 0.001; ES: 1.32–1.53) with small-to-moderate increases in Tier A (P < 0.001; ES: 0.30–0.38), Tier C (P < 0.001; ES: 0.46–0.54) and Tier D (P < 0.001; ES: 0.69–0.87). The demarcation line between 4th (bottom of Tier A) and 5th ranking (top of Tier B) in the 2006–07 season was 8 points, but this decreased to just a single point in the 2012–13 season. The data demonstrate that physical and technical performances have evolved more in Tier B than any other Tier in the EPL and could indicate a narrowing of the performance gap between the top two Tiers.

Keywords:

• Longitudinal
• football
• standard
• passing
• sprinting

Acknowledgement

Thanks to Prozone Sports for providing access to the data used in this study.

Disclosure statement

No potential conflict of interest was reported by the authors.