You Can’t win running with the Ball
You Can't Win
Running With
The Ball
A new five-season EPL study just confirmed what the specificity crowd doesn't want to hear.
Tom Allen — Head of Sports Science and Performance at Arsenal — and his team recently published Possession in Motion, a five-season analysis of running in and out of possession, matched to outcomes across the English Premier League. They had access to optical tracking data from Second Spectrum across every EPL game over five seasons. That is not a small dataset.
The headline finding: running out of possession was the strongest predictor of success — points per game, goal probability, outcomes — across all the running variables tested. Total distance, high-speed running, sprint distance, high-intensity distance. All measured. The ratio of out-of-possession running to in-possession running was the dominant signal.
This matters. Not because it is surprising to those who have watched the game closely, but because it now has five seasons of EPL tracking data behind it.
The Specificity Vortex
There is an argument that gets repeated in performance departments, particularly in soccer: we warm up with the ball because that is what happens in a game. The reasoning sounds coherent. It is actually a trap.
Specificity is real. Skill acquisition demands it. But getting caught in the specificity vortex — where every physical preparation decision must mirror game conditions — leads practitioners to underweight the most fundamental physical requirement in field sport: the capacity to run.
Running without the ball. Running in defence. Running to recover a position before the skill moment even arrives.
Tom Allen's data makes this explicit. Running in possession is lower intensity. Players with the ball cover less ground and move less urgently. The athletes covering ground without the ball — pressing, tracking, recovering — are doing the harder physical work. Five seasons. Every EPL club. Same conclusion.
"Getting caught in the specificity vortex leads practitioners to underweight the most fundamental physical requirement in field sport: the capacity to run."
The Ticket To The Dance
Running is not the point of field sport. But it is the prerequisite.
Think of it like a ticket to the dance. You can be the best dancer in the room. It does not matter if you cannot get through the door. In any field sport — soccer, AFL, rugby, gridiron — your running capacity determines whether you can get to the positions where your skill becomes relevant. You cannot execute in a space you cannot reach.
The practical implication for performance staff is this: straight-line running capacity, assessed independently of ball-based training, is not a limitation of your preparation model. It is the foundation of it. The ball-work sits on top. The mechanics that determine how repeatably and sustainably an athlete can run are not developed or exposed through small-sided games and possession drills. They require dedicated assessment.
Turkey controlled 72% of possession, completed 704 passes to Australia's 271, and generated 30 shots to Australia's 9. By every conventional metric of game dominance, Turkey won. The scoreboard said otherwise. Irankunda and Metcalfe scored in transition — running fast, without the ball moments before — and the Socceroos won 2-0, with goalkeeper Patrick Beach making eight saves on his competitive debut.
What kept Australia in that game long enough to punish Turkey on the break was the fitness to run in defence, repeatedly, without the ball. That is the harder running. Tom Allen's paper quantifies exactly that across five EPL seasons. The World Cup opened with a live demonstration.
What SpeedSig Measures
SpeedSig exists specifically at this intersection. It does not measure game running. It does not attempt to capture the complexity of multidirectional, skill-integrated sport. That is not the point.
SpeedSig is built to answer one question: how does this athlete generate speed, and how mechanical is that process? From that baseline — five validated variables captured via IMU data during straight-line running sessions — we can identify the specific mechanical factors limiting an athlete's speed production, their injury risk profile, and the targeted interventions most likely to improve repeatable, sustainable running capacity.
What Tom Allen's paper confirms is that the quality of that underlying running capacity matters enormously — not just to sprint specialists, but to every athlete who needs to cover ground in a field sport. Which is all of them.
Warm-ups are an underutilised window. Athletes are available, building into the session. Straight-line work in a warm-up context is not wasted — it is where the mechanical baseline gets established and tracked over time.
The Uncomfortable Reframe
Most elite performance practitioners are not going to make their athletes significantly faster. The window for large speed gains closes early in athletic development. The realistic goal at the professional level is something more specific and more achievable: keep athletes running well, repeatably, sustainably, for longer.
That requires knowing how they run. Not just how far. Not just how fast. How.
Allen's paper gives us a five-season, whole-league argument for why that matters. The teams who run better without the ball win more games. The athletes who can keep running — who have not been ground down by accumulating mechanical dysfunction — are the ones on the field in the moments that matter.
The first principles question is not complicated. Can this athlete run? Can they run well enough, repeatably enough, to compete? If you cannot answer that with data, you are guessing.
See What Your Hardware Already Knows
SpeedSig extracts five validated biomechanical variables from Catapult and STATSports devices your organisation already owns. No new hardware. No lab. Straight-line sessions during warm-up.
Book a DemoReference: Allen, T., Taberner, M., Zhilkin, M., Harper, D., & Alexander, J. (2026). Possession in motion: a five-season analysis of running in-possession and out-of-possession with match outcomes in the English Premier League. Biology of Sport, 43(1), 899–909.
