Individualised Speed Thresholds: Absolute vs Relative Approaches to Quantifying High-Intensity Running
Prerequisites: This article assumes familiarity with high-intensity running metrics (HSR, sprint distance, acceleration, deceleration), external and internal training load concepts, and the basic principles of GPS and tracking systems. If any of these topics are new to you, start with:
Learning Objectives
- Explain the definition, historical context, and practical rationale behind arbitrary absolute speed thresholds (AAT).
- Describe the principles and methods for establishing individualised thresholds (IND) based on maximal sprint speed (MSS) and maximal acceleration/deceleration capacity.
- Compare over- and under-estimation issues arising from AAT versus IND application using empirical evidence.
- Analyse how contextual factors — position, tactical role, and match situation — affect the interpretation of speed thresholds.
- Design a practical strategy for combining AAT and IND approaches in applied settings.
Where Do Absolute Speed Thresholds Come From?
Arbitrary absolute thresholds (AAT) are fixed speed cut-offs applied uniformly to all players, regardless of individual physical capacity. A common classification system divides locomotor activity into six bands: standing (0–0.6 km/h), walking (0.7–7.1 km/h), jogging (7.2–14.3 km/h), running (14.4–19.7 km/h), high-speed running (19.8–25.1 km/h), and sprinting (≥25.1 km/h). This six-zone model was applied in a landmark analysis of 370 English Premier League players across 28 matches, establishing positional activity profiles that shaped subsequent research and practice (Bradley et al., 2009).
The appeal of AAT lies in simplicity and comparability. When every player is measured against the same cut-off, data can be aggregated across teams, leagues, and seasons. A five-season longitudinal study in the EPL used fixed thresholds of >5.5 m/s for high-intensity distance and >7 m/s for sprint distance, revealing a 12% increase in high-intensity distance and a 15% increase in sprint distance across the period (Allen et al., 2023). Without consistent absolute benchmarks, such trend analyses would be impossible.
The problem emerges when the same absolute speed is treated as if it carries the same physiological meaning for every player. In the EPL data, maximal running speeds varied by 6–8% between positions: wide midfielders reached 7.93 m/s while centre-backs peaked at 7.31 m/s (Bradley et al., 2009). A high-speed running threshold of 19.8 km/h therefore represents a substantially greater proportion of a centre-back’s capacity than a wide midfielder’s. AAT is also inconsistent across the literature. The same acceleration value of 3 m/s² has been classified as moderate, high, or maximal intensity depending on the study, creating confusion when practitioners attempt to compare findings (Pimenta et al., 2026).
AAT remains the dominant method in both research and practice for one reason: it enables between-team and longitudinal comparison with minimal logistical burden. That utility is real, but it comes at the cost of individual accuracy.
Why the Same Speed Means Different Loads
Individualised thresholds (IND) classify running intensity as a percentage of each player’s maximal capacity rather than a fixed cut-off. The principle is straightforward: a sprint at 25 km/h represents near-maximal effort for a player whose maximal sprint speed (MSS) is 28 km/h, but only moderate effort for a player whose MSS is 34 km/h. By anchoring thresholds to the individual, IND aims to ensure that “high intensity” reflects genuinely high physiological demand for each player.
For speed-based metrics, the most common approach uses percentage bands of MSS. A proposed classification sets >80% MSS as the threshold for high-intensity or sprint-level activity (Piñero et al., 2023). For acceleration and deceleration, maximal acceleration capacity (ACCmax) and maximal deceleration capacity (DECmax) serve as the reference points. A four-tier normalisation framework has been proposed: high intensity (>75% of individual maximum), moderate (50–75%), low (25–50%), and very low (<25%) (Pimenta et al., 2026).
Measuring individual maximal capacities requires specific protocols. ACCmax can be assessed through a 30-metre tracking sprint, while DECmax can be derived from deceleration-specific tests such as the 505 Test. An important asymmetry exists between the two: DECmax is typically larger than ACCmax and less dependent on initial speed, meaning that symmetrical thresholds (e.g., ±3 m/s²) are physiologically inappropriate (Pimenta et al., 2026). To account for fluctuations across a season, a rolling average of the top three values, updated longitudinally, has been recommended.
IND is not without its own limitations. The percentage cut-offs themselves — 75%, 80% — are still arbitrary boundaries. MSS can vary depending on whether it is measured through a dedicated sprint test or extracted from match data, and the same player’s MSS may shift across the season. Furthermore, acceleration capacity is inversely related to initial speed: at 14.4 km/h, a player may reach 4–5 m/s², but at 23 km/h, the same player may barely exceed 2 m/s². Current IND frameworks for acceleration do not yet account for this speed-dependent relationship (Pimenta et al., 2026).
How Different Are the Numbers? AAT vs IND Evidence
The theoretical distinction between AAT and IND produces dramatic empirical differences when applied to the same match data.
A study of university-level and club-level footballers compared an absolute deceleration threshold of −4 m/s² (ARB) against an individualised threshold set at 75% of each player’s DECmax (IND) across 11 competitive matches (Moore et al., 2026). The results were striking.
| Metric | IND | ARB | Effect Size |
|---|---|---|---|
| Full-match deceleration distance (m) | 20.56 ± 19.03 | 149.84 ± 53.04 | g = 3.15 |
| Full-match deceleration count | 7.99 ± 7.60 | 28.14 ± 8.55 | g = 2.42 |
| MDP deceleration distance (m) | 1.64 ± 1.05 | 10.26 ± 10.25 | g = 1.15 |
| MDP deceleration count | 2.70 ± 2.03 | 19.78 ± 8.29 | g = 2.75 |
The absolute threshold classified more than seven times the deceleration distance as “high intensity” compared to the individualised approach. This occurred because the fixed −4 m/s² threshold fell below 50% of most players’ actual maximal deceleration capacity — meaning efforts that were objectively moderate were being labelled as high intensity (Moore et al., 2026).
The IND approach also revealed greater between-player variability (higher CV%), which reflects a feature rather than a flaw: players genuinely differ in how much high-intensity deceleration they perform relative to their capacity. AAT, by contrast, compresses these individual differences into a narrower range, masking the very information practitioners need to make personalised training decisions.
For speed-based metrics, an analysis of Spanish professional football used >80% MSS to define relative sprinting within most demanding passages of play (Piñero et al., 2023). This approach detected positional differences — wide midfielders recorded the longest relative sprint duration (2.4 s) and greatest distance (21.91 m) — and match-situation effects that absolute thresholds would aggregate away. Teams losing produced the greatest relative sprint distances, while no decline in relative sprint output was observed in the second half, challenging the common assumption that sprinting capacity always deteriorates over 90 minutes.
The pattern across studies is consistent: AAT systematically over-estimates the volume of high-intensity work, particularly for players with lower maximal capacities, while IND reveals the true spread of individual demands.
Context Changes Everything: Position, Tactics, and Match Situation
Even the most carefully chosen threshold — whether absolute or individualised — cannot fully describe a player’s load without contextual interpretation.
Positional granularity matters. An EPL analysis compared general positional classifications (e.g., “wide defender”) against specialised tactical roles (e.g., “full-back” vs “wing-back”) using an absolute HIR threshold of >19.8 km/h (Ju et al., 2023). Full-backs covered 34% less HIR distance than the wide defender average, while wing-backs covered 15% more. Using the general classification would over-estimate the demands placed on full-backs and under-estimate those placed on wing-backs. Specialised tactical roles also shifted the contextual meaning of specific actions: central defensive midfielders averaged only 2 close-down/press actions per match involving HIR, compared to 9 for attacking midfielders, yet both would be grouped under “central midfield” in a general classification.
Match situation shapes output. When analysed through individualised thresholds (>80% MSS), players produced their greatest relative sprint distances while losing, not while winning or drawing (Piñero et al., 2023). This suggests that tactical urgency — pressing to equalise, making recovery runs after losing possession — drives high-intensity output as much as physical capacity does.
Longitudinal trends differ by position. Over five EPL seasons, team-level sprint distance increased by 15%, but the magnitude and pattern of change varied substantially across positions. Centre-backs showed the largest increase in high-intensity distance, while wide midfielders showed no significant change across any metric (Allen et al., 2023). Team-level averages obscured these divergent positional trajectories entirely.
The implication is clear. Threshold selection is a necessary first step, but it is insufficient on its own. Practitioners must layer positional, tactical, and situational context over any threshold-based metric to arrive at a meaningful interpretation. A high-intensity running distance of 900 metres means something fundamentally different for a centre-back in a low-block defensive system than for a wing-back in a high-pressing formation.
Technology Limits: When Thresholds Meet Measurement Error
The accuracy of any threshold — absolute or individualised — depends on the measurement system that produces the underlying data.
GPS inter-unit variability can reach up to 50%, even among devices from the same manufacturer (Buchheit & Simpson, 2017). This means that two players wearing different GPS units may receive meaningfully different speed and acceleration values for identical movements. Acceleration and deceleration data are particularly vulnerable: the values change substantially depending on the time window used for calculation (0.2–0.8 s) and the filtering technique applied. No consensus exists on which settings are optimal.
A critical paradox exists in current practice. The variables that practitioners rate as most important for load monitoring — high-speed running distance, acceleration, and deceleration — are the ones with the lowest validity and reliability (Buchheit & Simpson, 2017). Tactical factors such as formation, coaching instructions, and scoreline often exert a larger influence on a player’s activity pattern than their physical fitness, further complicating the interpretation of locomotor data as a proxy for training status.
Field-based assessments of maximal deceleration capacity also show limited reliability. The acceleration-deceleration assessment (ADA) protocol produced ICC values of 0.48 at 20 metres and 0.63 at 30 metres, with coefficients of variation of 11.7% and 20.8% respectively (Moore et al., 2026). In-match measurements of peak deceleration exceeded ADA-derived values, suggesting that dedicated tests may underestimate true maximal capacity.
One advantage of the individualised approach in this context is that it uses higher absolute values as thresholds. Because GPS measurement error is relatively constant in absolute terms, a higher threshold means that a smaller proportion of the recorded efforts will be misclassified due to device noise. An IND threshold of, say, −7 m/s² (75% of a DECmax of −9.3 m/s²) sits further from the noise floor than an AAT of −4 m/s², reducing the risk of false positives (Pimenta et al., 2026).
Regardless of the threshold approach chosen, foundational measurement practices remain essential: players should wear the same GPS unit across sessions, between-system calibration equations should be applied when training and match data come from different technologies, and practitioners should maintain awareness that system changes invalidate direct data comparison (Riboli et al., 2023).
Putting It Together: A Practical Integration Strategy
AAT and IND are not competing alternatives — they serve different purposes within a comprehensive monitoring system.
AAT is suited for comparison and benchmarking. When the goal is to compare a player’s output against league averages, track season-to-season trends, or evaluate positional demands across teams, absolute thresholds provide the common currency that makes such comparisons possible. Profiling and benchmarking should be conducted on an individual basis where possible, using standardised scores (z-scores, percentiles) to contextualise a player’s data against relevant norms (McGuigan, 2022).
IND is suited for individual load management and training prescription. When the goal is to determine whether a specific player has been exposed to genuinely high-intensity efforts — and to prescribe training that replicates match demands at the individual level — normalised thresholds provide a more physiologically meaningful picture. The same external load produces different internal responses depending on a player’s training status, nutrition, and psychological state, reinforcing the need for individualised interpretation (Impellizzeri et al., 2019).
A practical integration strategy involves several components.
Establish individual maximal capacity profiles. Measure MSS, ACCmax, and DECmax during pre-season and update these values longitudinally using a rolling average of the top three recorded values. Use both dedicated testing protocols and in-match peak values, recognising that match-derived values may exceed test-derived values.
Build a drill database with dual metrics. For each training drill, record both absolute and individualised output. Over time, this creates an evidence base for prescribing drills that target specific intensity zones relative to each player’s capacity while also allowing comparison of drill demands against match benchmarks (Riboli et al., 2023).
Layer contextual information. Report threshold-based data alongside positional role, tactical context, match situation, and playing time. A player’s individualised sprint distance in a match where the team was chasing the game should not be directly compared to a match where the team held a comfortable lead.
Combine external and internal load measures. External load metrics — whether absolute or individualised — describe the stimulus imposed on the player. Internal load measures (heart rate, RPE, subjective wellbeing) describe the response. When a standardised external load produces an increasing internal response over time, this signals declining fitness or accumulating fatigue. Neither load type alone tells the full story (Impellizzeri et al., 2019).
Do not rely on a single threshold or ratio. Benchmarks and cut-offs are guides, not gospel. Any threshold should be interpreted within the broader context of the player’s profile, positional demands, and the monitoring system’s measurement characteristics (McGuigan, 2022).
Key Takeaways
- AAT (e.g., ≥19.8, ≥25.2 km/h) enables between-team and longitudinal comparison, yet the same absolute speed represents different relative intensities across positions and fitness levels — a 19.8 km/h threshold may fall below 50% of one player’s maximal capacity while exceeding 80% of another’s.
- IND is set as a percentage of individual maximal capacity (MSS, ACCmax, DECmax), with high-intensity commonly defined at 75–80% of maximum; measuring and longitudinally updating these individual capacities is a prerequisite for accurate individualised monitoring.
- AAT can systematically over-estimate high-intensity efforts — empirical evidence shows absolute-threshold deceleration distance exceeded individualised values by more than seven-fold in the same matches, with very large effect sizes across all metrics.
- Speed thresholds alone cannot determine a player’s actual load; integration with contextual factors such as specialised tactical role, match situation, and formation is essential for avoiding systematic over- or under-estimation of positional demands.
- In practice, AAT (for benchmarking and between-team comparison) and IND (for individual load management and training prescription) should be used in parallel, with individual maximal capacities updated throughout the season and all threshold-based data interpreted alongside internal load measures and tactical context.
References
- Allen, T., Taberner, M., Zhilkin, M., & Rhodes, D. (2023). Running more than before? The evolution of running load demands in the English Premier League. International Journal of Sports Science & Coaching, 19(2), 779-787. https://doi.org/10.1177/17479541231164507
- Bradley, P. S., Sheldon, W., Wooster, B., Olsen, P., Boanas, P., & Krustrup, P. (2009). High-intensity running in English FA Premier League soccer matches. Journal of Sports Sciences, 27(2), 159-168. https://doi.org/10.1080/02640410802512775
- Buchheit, M. & Simpson, B. M. (2017). Player-Tracking Technology: Half-Full or Half-Empty Glass?. International Journal of Sports Physiology and Performance, 12(s2), S2-35-S2-41. https://doi.org/10.1123/ijspp.2016-0499
- Impellizzeri, F. M., Marcora, S. M., & Coutts, A. J. (2019). Internal and External Training Load: 15 Years On. International Journal of Sports Physiology and Performance, 14(2), 270-273. https://doi.org/10.1123/ijspp.2018-0935
- Ju, W., Doran, D., Hawkins, R., Evans, M., Laws, A., & Bradley, P. (2023). Contextualised high-intensity running profiles of elite football players with reference to general and specialised tactical roles. Biology of Sport, 40(1), 291-301. https://doi.org/10.5114/biolsport.2023.116003
- McGuigan, M. (2022). Profiling and Benchmarking. In D. N. French & L. Torres Ronda (Eds.), NSCA’s Essentials of Sport Science. Human Kinetics.
- Moore, L., Drury, B., & Hearn, A. (2026). Hitting the Brakes in Soccer: Individualised Thresholds for Assessing High-Intensity Decelerations during Matches. International Journal of Strength and Conditioning, 6(1). https://doi.org/10.47206/ijsc.v6i1.565
- Piñero, J. Á., Chena, M., Zapardiel, J. C., Roso-moliner, A., Mainer-pardos, E., Lampre, M., & Lozano, D. (2023). Relative Individual Sprint in Most Demanding Passages of Play in Spanish Professional Soccer Matches. Sports, 11(4), 72. https://doi.org/10.3390/sports11040072
- Pimenta, R., Antunes, H., Silva, H., Ribeiro, J., & Nakamura, F. Y. (2026). The Need for GPS Data to be Normalized for Performance and Fatigue Monitoring in Soccer: Considerations for Accelerations and Decelerations. Strength & Conditioning Journal. https://doi.org/10.1519/ssc.0000000000000958
- Riboli, A., MacMillan, L., Calder, A., & Mason, L. (2023). Player monitoring and practical application. In A. Calder & A. Centofanti (Eds.), Peak performance for soccer: The elite coaching and training manual. Routledge.