It’s Time to Drop Total Running as a KPI in Football

In our latest instalment of the Behind the Science series, we invited Mauro Mandorino, Head of Sports Science for Italian Serie A side Parma Calcio 1913, to join Hudl’s Senior Solutions Consultant Will Sparkes for an insightful interview. 

In this video, they discuss his research on player performance evaluation, the imprecise premise many clubs have been operating on, and how he is putting his findings into practice at elite level.

Below, Mauro expands on these themes in his own words, in this blog adapted from his scientific journal article:

For many years, football performance has been judged through the lens of running metrics such as total distance and high-speed running. These numbers are easy to track, easy to compare, and look convincing on paper. 

However, our three-season analysis based on findings from a professional Italian football club demonstrates that this approach is fundamentally flawed and should be abandoned as a primary way of evaluating performance. Running, as affirmed by Martin Buchheit, is a consequence of football actions—not the cause of successful match outcomes. 

Our study introduces a more modern approach: following the Banister fitness-fatigue model, and using machine-learning models, the researchers estimated players’ fitness status (via heart-rate responses in training) and neuromuscular readiness (via player load responses). These indices were tracked weekly, offering a constant monitoring of the players’ true status than isolated fitness tests performed a few times per year. 

When these readiness indicators were compared with match outcomes and running activity across 73 matches, the results were striking. The highest probability of winning occurred when players were both fit and fresh, while simultaneously performing lower amounts of total distance than their usual baseline. 

In statistical terms, the combination of high fitness, high freshness, and low running volume produced the strongest association with winning (OR = 2.60). Conversely, teams that ran more—especially when players were not particularly fit—showed worse probabilities of winning.

Interestingly, running volume was actually highest during draws, not wins. This suggests that running often reflects the dynamics and tension of the game rather than the quality of performance. When both teams are trying to break a deadlock, they naturally accumulate more distance. 

The data also showed almost no meaningful relationship between running and either readiness metric: fitter or fresher players did not automatically run more during matches. Running simply does not capture the true nature of football conditioning or execution. 

The broader implication is clear: football performance is driven by the quality and frequency of football actions, not by the quantity of running performed. 

High readiness reduces internal load, sharpens decision-making, and supports better tactical execution. When players are not fresh or fit, limiting unnecessary displacement may even help maintain the structure and efficiency of the team. Running on its own, however, tells us nothing about why a team succeeds or struggles. 

This study reinforces the need for a shift away from traditional, overly simplistic KPIs. In the era of Sport Science 3.0, practitioners should focus on readiness, context, and football actions rather than relying on running volumes. 

The message is simple but important: stop treating running as a performance indicator—because it isn’t one.