One of the most common questions performance practitioners ask is also one of the most difficult to answer: how does my team compare to others? Not just within the same league, but against higher-level competitions that are often used as performance references.
To help answer that question, the gpexe educational team analysed official match data from clubs competing in Bosnia and Herzegovina, Serbia and Slovenia during the 2024-25 season, creating league-level benchmarks and comparing them with reference values from the Italian Serie A.
The objective was simple: provide clubs with meaningful context for interpreting their own match demands. But as the project developed, it also reinforced another important lesson: reliable benchmarking depends on reliable data.
Building a benchmark across leagues
The project included 13 clubs across three Balkan competitions and analysed seven key performance indicators covering both locomotor and metabolic demands.
The metrics included:
- Total Distance
- High-Speed Running Distance (>20 km/h)
- Sprint Distance (>25 km/h)
- Speed Events
- Metabolic Power Events (MPE)*
- MPE Recovery Time
- MPE Recovery Power
*MPEs identify moments when energy demand exceeds aerobic capacity, requiring anaerobic contribution. Unlike speed or acceleration thresholds alone, they capture high-intensity efforts resulting from any combination of speed and acceleration, making them a more complete lens on match demands. Recovery Time and Recovery Power then describe how the athlete manages the phases between efforts: how long before the next MPE, and at what intensity. For a deeper dive, read our two-part article: The performance explained through MPEs.
To ensure consistency, only official matches were included, and the analysis focused on players who completed between 90 and 105 minutes of play.
“The goal wasn’t to create a ranking of teams,” explains the gpexe educational team. “The primary focus was on the Balkan clubs themselves, understanding where each team sits within its own domestic competition. Serie A values were included as an additional external reference point.”
It is worth noting that the Serie A dataset was collected using a slightly different method: rather than isolating only players who completed the full 90 minutes, all outfield players — both starters and substitutes — were included, and their combined data was divided by 10. This approach simulates a “composite” match average, as if each starting position were covered by a single player for the entire game. This differs from the criterion applied to the Balkan clubs, and may partially account for the higher values observed in Serie A. Comparisons should therefore be read as indicative rather than absolute.
Before benchmarking comes data quality
When data from multiple organisations are brought together, the first challenge is not statistical analysis, but consistency. Reviewing an entire season of match data across different clubs revealed several differences in how sessions had been managed and categorised. In some cases, substituted players remained active for the entire session. In others, half-time periods were included in the analysis, post-match top-up activities were added to official matches, or academy matches were classified using the same categories as first-team fixtures. Some clubs also included goalkeepers in team averages, influencing overall team values.
None of these situations was particularly surprising.
“In projects like this, you often see differences in how data collection workflows are managed,” confirms the educational team. “The project essentially confirmed patterns we’ve already observed in previous analyses.”
Rather than highlighting mistakes, the analysis reinforced the importance of standardisation when building league-wide benchmarks. After all, comparing teams only makes sense when everyone is measuring the same thing.
What did the data show?
Once the dataset had been cleaned and standardised, several clear trends emerged. Across all seven metrics analysed, Serie A produced the highest average values.
Average match demands across Serie A, Slovenia, Serbia and Bosnia & Herzegovina.
What the data show is not simply a gap in volume, but a difference in the structure of match intensity. Serie A players cover more ground and reach higher speeds, but the metabolic picture adds another layer: more high-intensity events per match, packed closer together, with recovery phases that are shorter and more active. In other words, the demands are not just greater, they are more continuous.
MPE average recovery time and average recovery power - Average match demands across Serie A, Slovenia, Serbia and Bosnia & Herzegovina.
This pattern holds consistently across all seven metrics, and that consistency is itself informative. It suggests that the gap between Serie A and the Balkan competitions analysed is not a matter of a single physical quality, but of an overall tempo of play that shapes demands across the board.
These findings suggest a greater density of high-intensity actions in the Italian top division, although tactical approaches, game models and contextual factors must always be considered when interpreting league-wide data.
“Benchmarking can tell you where you sit,” notes the gpexe educational team. “What it cannot do is tell you why. That’s where the context of the team, the coaching staff and the competition becomes essential.”
More than numbers
While the project generated valuable benchmark data, one of its most practical outcomes was the opportunity to support clubs in refining their data collection processes.
Benchmarking is often seen as an analytical exercise, but projects like this show that it is equally an operational one.
Session setup, player inclusion criteria and category management all have a direct impact on the quality of the insights that can be generated later.
For clubs that identified gaps in their own workflow during this process, the takeaways were concrete:
- use consistent session categories;
- correctly manage substitutions;
- exclude half-time periods from match analysis;
- avoid mixing academy and first-team fixtures under the same session category;
- regularly review data quality throughout the season.
Small details in the workflow can have a significant impact on long-term reporting and benchmarking accuracy.
Looking ahead
Training data could be even more interesting than match data, because it reflects how teams choose to prepare during the week — and that choice is entirely within the club’s control.
For the clubs involved, this project was a starting point: a clearer picture of where they stand, and a concrete prompt to ask better questions about their own processes. Benchmarking is often presented as a way to compare performance. In reality, its greatest value may be something else entirely: helping practitioners ask better questions about their own environment, their own processes and the decisions that shape performance every day.
The numbers matter. But the conversations they trigger may matter even more.
Author
gpexe educational team
References
- Match data source: gpexe GPS systems, official matches 2024–25 season (Bosnia & Herzegovina, Serbia and Slovenia)
- Gualtieri A., Franceschi A., La richiesta fisica della partita, in Sassi R. & Beato M. (eds.), Preparazione fisica nel calcio moderno. Aspetti teorici e pratici, Calzetti & Mariucci — Serie A data collected via gpexe
- The performance explained through MPEs, gpexe educational team, based on C. Osgnach’s research and gpexe data