Premier League players sprint far more than a decade ago, and play more congested minutes

Footballers have spent several years arguing that there are too many matches and that the matches themselves have got harder. The argument has usually been made in interviews. It now has a dataset behind it.

A study published in the British Journal of Sports Medicine on 19 December analysed 3,750 English Premier League games across ten seasons, from 2015/2016 to 2024/2025 [s1]. It is descriptive rather than causal, and its authors are careful about that. What it establishes is that the physical demand has risen, and that the players most exposed to it are the ones at the top of the table.

What was measured

Four running metrics, defined by speed thresholds: team total distance; high-intensity distance (above 5.5 m/s); high-speed running distance (5.5–7 m/s); and sprint distance (above 7 m/s) [s1].

Tracking came from two semi-automated optical systems — TRACAB for the 2015/2016 to 2018/2019 seasons, and Second Spectrum from 2019/2020 onward [s1]. Match minutes came from Wyscout [s1].

The running-load analysis used 3,408 of the games; the player-minutes analysis covered 34 teams and 1,912 male players meeting an inclusion threshold of at least five games and more than 270 minutes [s1].

"Congested" minutes were defined precisely: three matches inside an eight-day window, with the third game counted as the first congested match, and any subsequent game within three days of it also counted as congested [s1].

What changed

All four running metrics rose significantly between 2015/2016 and 2024/2025, at p<0.001 [s2]. The effect sizes separate the story from the noise.

Total distance rose with an effect size of 0.56 — a small-to-moderate change [s2]. The high-intensity categories moved much more: high-intensity distance at 1.79, high-speed running at 1.67, sprint distance at 1.61 [s2].

Adjusted per minute of play, the pattern is sharper still. High-intensity distance per minute rose with an effect size of 1.55, high-speed running per minute 1.41, sprint distance per minute 1.46 — while total distance per minute fell slightly, at −0.27 [s2].

That combination is the finding. Players are not covering much more ground per minute. They are covering a great deal more of it fast. The game has become more intermittent and more explosive rather than simply longer.

Who absorbs it

On a rolling three-season basis, total match minutes rose by 1,600 and congested minutes by 1,415 across all players between the first and last seasons [s2].

For players at top-six clubs, the increases were far larger: 5,659 additional total minutes and 2,003 additional congested minutes [s2].

Success compounds exposure. European competition, deeper domestic cup runs and international duty land on the same squads, and the study quantifies how unevenly the calendar distributes its cost.

What the study does not show

This is where the paper deserves credit for restraint, and where coverage of it will probably go wrong.

The authors note that dense fixture periods are known to negatively affect physical performance and increase injury risk [s1]. They then state explicitly that this study does not examine the effect of increasing running load demands on injury, illness or mental health [s1]. There is no injury outcome in the dataset. Nothing here demonstrates that the 2024/2025 Premier League is more dangerous than the 2015/2016 one; it demonstrates that it is more demanding.

Their recommendation follows from that gap: injury and illness surveillance studies are needed to ground policy decisions in evidence, alongside continuous monitoring of how players respond to the increased demands [s1].

Three limitations are worth carrying into any reading of the numbers.

Running distance is a partial description of football's physical demand. The authors acknowledge that the sport's intermittent, multidirectional nature is also a critical factor [s1] — accelerations, decelerations, changes of direction and collisions are not captured by distance thresholds.

Optical tracking has known accuracy limits, particularly for acceleration and deceleration [s1].

And the tracking provider changed mid-series. The authors concede that data interchangeability issues between providers may account for some of the observed increases [s1]. That caveat applies most directly to the 2018/2019-to-2019/2020 boundary and is a real constraint on interpreting the decade-long trend as purely a change in how the game is played.

Why it matters

Fixture-calendar disputes have been conducted between leagues, federations and player unions largely on assertion. A ten-season description of what has actually changed — with the top-six exposure gap quantified [s2] — gives the argument a shared factual basis it has lacked.

It does not settle it. The health question is about injury, illness and career length, and that question requires surveillance data the study did not have [s1].

What to watch

Whether league-level injury surveillance is published in a form that can be linked to these load and congestion metrics. Until it is, the load side of the argument will keep getting better evidence while the harm side gets none.