How Fatigue Changes Sprint Mechanics in Football: What the Research Shows

10 minutes ago

9 min read

Fatigue doesn't just make a player slower — it changes how they sprint. That distinction matters more in football than in almost any other sprint-based context, because a footballer rarely produces just one maximal sprint in isolation. They produce repeated accelerations, high-speed runs and sprint actions across ninety minutes, on top of everything else the match demands, and the decisive sprint late in a game almost never happens when a player is fresh. A specific set of numbers describing this — how much power, velocity, and force actually drop across repeated sprints — has been circulating widely on social media recently. They turn out to be real, tracing back to a genuine peer-reviewed study, and they're really just the entry point into a broader, genuinely useful body of research on what fatigue does to sprinting.

The Study Behind the Viral Numbers

The figures come from Edouard and colleagues (2018), published in Frontiers in Physiology. Fourteen sprint-trained male athletes performed twelve 6-second maximal sprints on an instrumented treadmill, each separated by 44 seconds of recovery, while researchers tracked ground reaction forces, joint kinematics, and muscle activity throughout. The design wasn't primarily about describing fatigue for its own sake — the study's real focus was on the hamstrings, and specifically on what role the gluteal muscles play in protecting them as fatigue sets in. The mechanical fatigue data are effectively the setup for that bigger question, but they're worth understanding in their own right.

What Actually Drops — and What Doesn't

Across the twelve sprints, maximal power output fell by 17.5%, a large effect. Maximal velocity dropped by 6.6%. The horizontal component of ground force — the part of force production that actually propels a sprinter forward — fell by 8.6%. All three of these are meaningful, fatigue-driven declines.

What's genuinely interesting is what didn't change nearly as much: total (resultant) ground force fell by only 3.4%, and vertical force by just 3.1% — both small effects, not statistically meaningful on their own. That gap matters. It suggests fatigue during repeated maximal sprinting isn't simply "the muscles produce less force overall" — it's more specifically that less of the force being produced gets directed horizontally, in the direction that actually creates speed.

The Mechanical Adjustments

Alongside those force changes, the step pattern itself shifted. Ground contact time increased by 11.8%, swing time increased by 5.2%, and step frequency dropped by 6.0%. Knee speed just before ground contact — a marker of how aggressively the leg is being driven down and back at touchdown — fell by 14.1%, one of the largest changes in the entire dataset. Together, these describe a runner spending more time on the ground per step, taking fewer steps per second, and swinging the leg through with less speed into contact — a pattern consistent with the observed reduction in effective horizontal force production.

Football-Specific Fatigue Can Change the Entire Acceleration Strategy

Edouard's protocol is a useful, controlled window into repeated maximal-effort sprinting, but it isn't football. A separate line of research gets closer to match conditions specifically. Vial and colleagues (2023) tested 13 semi-professional male football players before and after a 45-minute soccer-specific fatigue protocol (a validated match simulation involving walking, jogging, sprinting, and change of direction), analysing the first three steps of a maximal acceleration from a standing start in each state.

The headline finding is almost the opposite of what might be expected: step-to-step acceleration did not decline. Despite the fatigue protocol significantly reducing single-leg jump height, the players accelerated just as quickly over those first three steps as they had before it. What changed was how they got there. After fatigue, players adopted a noticeably more upright posture, made ground contact further in front of their centre of mass, held longer ground contact times, and produced greater vertical impulse — while net horizontal impulse stayed essentially unchanged. Underneath that shift, hip flexion and extension velocity slowed, peak hip-extensor moments and power dropped at every step, and knee-extensor moments, power, and positive joint work rose to compensate — a measurable transfer of the workload from the hip to the knee. Ankle joint mechanics, by contrast, barely moved.

That last detail is a genuinely interesting wrinkle in the broader fatigue literature. In slower, continuous distance running, fatigue is typically associated with work shifting away from the ankle and toward the knee and hip — a distal-to-proximal shift. Here, in maximal football-specific acceleration, the shift ran the other way: away from the hip, toward the knee, with the ankle essentially untouched. The authors' interpretation is that fatigue doesn't have one universal signature — it shows up differently depending on the specific task and which muscle groups that task leans on hardest, and sprint acceleration leans heavily on the hip extensors (the glutes and hamstrings) in a way that steady-state jogging doesn't.

It's also worth holding the two studies next to each other rather than treating them as describing the same thing. Edouard's athletes, repeating maximal sprints back-to-back with only 44 seconds of rest twelve times over, saw their actual output — force, power, velocity — decline. Vial's players, fatigued by a broader 45-minute match simulation but then asked for only a handful of acceleration steps, fully preserved their outcome by changing strategy. Read together, they raise an interesting possibility: compensatory strategies may preserve performance up to a point, but as fatigue becomes more severe, mechanical output can ultimately decline. That's a reasonable reading of two different experiments side by side, not something either one demonstrates on its own.

None of this is unique to a small, controlled study. Romero and colleagues (2022) tested 51 actual football players across twelve 30m sprints with 20-second recoveries and found the same broad pattern — a large drop in power output and less efficient horizontal force orientation as fatigue built. Their more important contribution, though, was showing that the average trend hides real individual variation: cluster analysis identified distinct groups of players whose trunk, hip, and knee angles changed in different ways under fatigue, with some patterns theoretically increasing hamstring strain and others theoretically reducing it. The population-level story is real, but it isn't the same story for every player.

Compensation, Not Necessarily Breakdown

It's tempting to read all of this as an athlete's technique falling apart under fatigue, but that framing undersells what's actually happening. In Vial's study specifically, players kept their acceleration intact despite real, measurable changes underneath — which looks a lot more like an adaptive reorganisation of effort than a simple deterioration. A more upright posture with a longer ground contact and reduced hip contribution isn't obviously the "wrong" way to accelerate when fatigued; it may be a reasonably effective way to keep producing the same outcome once the usual method starts to cost too much.

The more useful practitioner questions aren't "is this technique good or bad," but how much a given player's strategy actually changes under fatigue, at what point that change starts to compromise output rather than just look different, and whether the player can still produce enough horizontal force when the moment in the match actually calls for it.

The Hamstring Injury Angle

Both research groups land on a similar, carefully-hedged hypothesis, independently of one another. Edouard and colleagues proposed that, as hamstring capacity declines with fatigue, greater reliance on the hip extensor system — particularly the gluteus maximus — may help maintain horizontal force production while limiting additional demand on the hamstrings themselves. The theoretical vulnerability shows up specifically if that compensatory option starts to run out too: once gluteal or broader hip-extensor capacity is also compromised by fatigue, the hamstrings may be left more exposed to mechanical demand they'd otherwise have shared with the glutes. Vial and colleagues, watching a comparable hip-to-knee transfer show up in a completely different protocol and population, floated a related but not identical idea: that shifting work away from the hip extensors and onto the knee extensors as fatigue accumulates could itself function as a protective strategy for a muscle group — the hamstrings — that's already known to be vulnerable late in matches.

Neither team overstates this, and it's worth being precise that the two hypotheses aren't quite the same mechanism. Fatigue appears to trigger a compensatory redistribution of muscular work, but the direction and purpose of that redistribution look task- and athlete-dependent, and both proposals remain mechanistic hypotheses rather than a demonstrated injury-prevention effect — not a demonstrated cause of injury, and Romero et al.'s finding that individual players respond differently is a useful reminder not to treat either one as a fixed rule. What this does support fairly directly is why hip-extensor and hamstring-specific strength testing (covered in an earlier piece on physical testing) is worth taking seriously beyond just producing a number on a sheet: a player with genuinely strong, well-conditioned glutes may be better placed to keep absorbing that compensatory load as fatigue builds, rather than running out of that option and leaving the hamstrings more exposed.

What This Means for Speed Training

All of this points to a practical distinction that's easy to blur: training maximum sprint capacity and training the ability to keep reproducing speed under fatigue are two different jobs, even though both matter in football.

Developing the top end of a player's speed — maximum acceleration, high horizontal force production, high-quality mechanics — is best done relatively fresh, with full recovery between efforts. A large systematic review and meta-analysis of repeated-sprint training by Thurlow and colleagues (2023), pooling 176 studies, confirmed the intuitive part of this: how much fatigue a repeated-sprint session produces is directly shaped by programming choices like sprint distance, number of repetitions, and rest duration between efforts — shorter distances and longer recoveries keep the session closer to genuine speed work, while longer distances and shorter recoveries push it toward a conditioning stimulus instead. Treating every sprint session as interchangeable risks turning intended max-speed work into fatigue exposure by accident, or vice versa.

The other half of the job — repeated-sprint ability itself — is a trainable quality in its own right, not just a side effect of general fitness. A 2026 systematic review and multilevel meta-analysis by Chen and colleagues, pooling 17 randomised controlled trials and 386 soccer players, found field-based repeated-sprint training produced meaningful improvements in short-sprint ability, repeated-sprint ability itself, change-of-direction performance, and high-intensity running capacity — though, notably, it didn't meaningfully improve vertical jump performance or aerobic capacity, which is a useful reminder that repeated-sprint training is a specific tool for a specific set of qualities, not a general fitness intervention that happens to involve sprinting.

Put together, football speed development probably shouldn't be framed as fresh sprinting or fatigued sprinting, as if one has to be chosen over the other. Fresh, high-quality sprint work builds the ceiling; repeated-sprint exposure under real fatigue builds the ability to keep accessing a meaningful share of that ceiling deep into a match — and given everything above, a fair share of what actually needs training is the second one, since the fastest player when fresh isn't necessarily the player who can still produce effective sprint mechanics in the 85th minute.

Key Takeaways

The viral "sprint fatigue curve" figures are real and trace to Edouard et al. (2018): across twelve 6-second treadmill sprints, power fell 17.5%, velocity 6.6%, and horizontal force 8.6% — while total and vertical force barely changed, pointing to a change in force direction, not just magnitude.
In a football-specific 45-minute fatigue protocol, Vial et al. (2023) found players fully preserved their acceleration by adopting a more upright posture and shifting work from the hip to the knee — a reminder that visible mechanical change doesn't automatically mean reduced output.
The direction of fatigue-driven joint-work redistribution isn't universal: it shifted from hip to knee during football-specific acceleration, but tends to shift from ankle to knee/hip during slower continuous running — the task itself shapes the compensation pattern.
Fatigue appears to trigger compensatory redistribution of muscular work, but the direction and purpose of that redistribution are task- and athlete-dependent. Both Edouard et al. and Vial et al. proposed potential hamstring-protective mechanisms, though these remain mechanistic hypotheses rather than demonstrated injury-prevention effects, and individual players respond differently (Romero et al., 2022).
Training maximum speed and training repeated-sprint ability are different jobs with different programming demands; evidence supports both as genuinely trainable, distinct qualities rather than one being a byproduct of the other.

Conclusion

The numbers behind this particular piece of viral content held up under scrutiny, which isn't always the case — they come from a real, well-designed, peer-reviewed study, even though the posts sharing them don't say so. The fuller picture behind those numbers is more useful than the numbers alone: fatigue reorganises how a player produces speed, not just how much of it they have left, and that reorganisation looks different depending on the task, isn't identical from player to player, and may in some cases be protective rather than purely negative. For anyone training speed in football, the practical upshot is the same one the research keeps pointing back to — the decisive sprint rarely happens when a player is completely fresh, so the ability to reproduce effective mechanics under fatigue deserves training attention in its own right, not just as an afterthought to maximum speed work.

References

Chen, A., Zhuang, M., Huang, Z., Zhao, L., & Wang, L. (2026). The effects of field-based repeated-sprint training on physical performance in soccer players: A systematic review and multilevel meta-analysis. Frontiers in Physiology, 17, 1745959. https://doi.org/10.3389/fphys.2026.1745959
Edouard, P., Mendiguchia, J., Lahti, J., Arnal, P. J., Gimenez, P., Jiménez-Reyes, P., Brughelli, M., Samozino, P., & Morin, J-B. (2018). Sprint acceleration mechanics in fatigue conditions: Compensatory role of gluteal muscles in horizontal force production and potential protection of hamstring muscles. Frontiers in Physiology, 9, 1706. https://doi.org/10.3389/fphys.2018.01706
Romero, V., Lahti, J., Castaño Zambudio, A., Mendiguchia, J., Jiménez Reyes, P., & Morin, J-B. (2022). Effects of fatigue induced by repeated sprints on sprint biomechanics in football players: Should we look at the group or the individual? International Journal of Environmental Research and Public Health, 19(22), 14643. https://doi.org/10.3390/ijerph192214643
Thurlow, F., Weakley, J., Townshend, A. D., Timmins, R. G., Morrison, M., & McLaren, S. J. (2023). The acute demands of repeated-sprint training on physiological, neuromuscular, perceptual and performance outcomes in team sport athletes: A systematic review and meta-analysis. Sports Medicine, 53(8), 1609–1640. https://doi.org/10.1007/s40279-023-01853-w
Vial, S., Cochrane Wilkie, J., Turner, M., Scanlan, M., & Blazevich, A. J. (2023). Does fatigue influence joint-specific work and ground force production during the first steps of maximal acceleration? Scandinavian Journal of Medicine and Science in Sports, 33(6), 894–906. https://doi.org/10.1111/sms.14318