The effect of performance fatigue on sprint running technique

Author Identifier

Shayne Vial

Date of Award


Document Type

Thesis - ECU Access Only


Edith Cowan University

Degree Name

Doctor of Philosophy


School of Medical and Health Sciences

First Supervisor

Jodie Cochrane Wilkie

Second Supervisor

Anthony Blazevich


Sprint running is a complex, multi-joint movement that has been well studied in human research, largely through the analysis of 3-D kinematics and force-time signals obtained during non-fatigued running. In team sport, however, the unplanned nature of match play determines if and when sprint efforts occur, meaning that some efforts may be performed with minimal recovery and therefore likely in the presence of fatigue. While fatigue-induced changes in running technique have been shown to incite potentially injurious kinematic patterns, it is unclear whether these modifications occur because of altered lower limb joint torques, power generated and absorbed, or muscle lengths and contraction speeds adopted. Therefore, to explore the effect of running-induced fatigue on sprint running technique, 3-D kinematics and ground reaction forces were collected in thirteen intermediate-level soccer players during the first three steps of acceleration and sprinting at maximum speed before and after completing 45-minutes of a simulated soccer match.

Study 1: During non-fatigued accelerative sprinting, horizontal impulse progressively decreased with each step (step 1 [S1] to step 2 [S2] to step 3 [S3]), while similar positive work was performed at the hip and ankle joints in S1 and S2 but greater relative ankle joint contribution was observed in S3. Thus, the proximal hip extensors and distal ankle plantarflexors contributed equally to forward acceleration. After fatiguing exercise, acceleration was maintained even though running technique was substantially altered. The sagittal trunk angle increased by ~15% (i.e., more erect posture) with a clear reduction in hip flexion angular velocity (22%) and hip extensor moment (8%). An increase in knee joint moments (7%) and powers (4%) compensated for reduced hip joint function. This acute adaptation alleviated hip flexor-extensor force requirement and preserved propulsive ground force application. These findings indicate that the muscles spanning the hip joint may be more susceptible to fatigue than those around the ankle and knee joints during accelerative sprinting, although technique alterations were sufficient to retain acceleration once fatigued.

Study 2: In non-fatigued maximum speed sprinting, the dominant limb (DL) produced greater propulsive impulse as a result of greater work being done at the ankle joint, whilst the non-dominant limb (NDL) produced more vertical impulse. After fatiguing exercise, the mean maximum running speed decreased by 4%, with a notable shift toward horizontal force production in NDL, mainly resulting from an increase in plantarflexion (i.e. distal joint) moments and powers, while few kinematic and kinetic changes were detected in DL. Due largely to these changes in NDL, interlimb asymmetries were substantially reduced with fatigue. Thus, these results are consistent with a theory in which speed is prioritised during non-fatigued maximum sprinting, which may benefit from a less symmetrical sprint gait (i.e. each limb is utilised ‘maximally’ regardless of force imbalances), but injury risk reduction may be prioritised over speed during fatigued sprinting, with a more uniform sprint gait being adopted to distribute muscle forces more equally between limbs. Importantly, interlimb asymmetry decreased following fatigue as a result of changes in NDL, contrary to the hypothesis that asymmetry should increase with fatigue.

Study 3: Musculoskeletal modelling was used to estimate lengths of the three bi-articular hamstring muscle-tendon units (MTU), which were found to operate at significantly longer lengths during maximum speed than accelerative sprinting, irrespective of fatigue. If peak hamstring MTU lengths are indeed a risk factor for injury, then this is speculatively unlikely to occur within the first three steps of acceleration. After fatiguing exercise, peak anterior pelvic tilt increased, and peak hip flexion decreased while peak knee extension angle remained unchanged during maximum speed sprinting. The adaptations, collectively, resulted in a lack of change in estimated hamstring MTU lengths in DL. Conversely, fatigue led to an increase in the peak length of biceps femoris long head (BFlh) MTU in NDL. Using simple linear regression, increased peak knee extension angle (before foot-strike) was associated with a longer BFlh MTU (in NDL), which may have some utility as a forewarning for an increasing BFlh MTU length in practical settings. An important finding from this study was that overall BFlh MTU excursion remained constant despite peak length increasing, demonstrating a longer mean operating length when fatigued. Given that the length at which a muscle is actively lengthened influences the magnitude of muscle damage incurred, the longer operating length observed within BFlh MTU (NDL) with fatigue may induce greater magnitudes of muscle damage, and thereby increase injury risk.

The combined findings of this thesis demonstrate that the acute technique adaptions in response to fatiguing, running-based exercise varied considerably between each of the sprint phases. These findings may help inform performance and injury preventive practices, in particular, interventions targeting the gluteal and hamstring muscle groups to promote fatigue resistance.

Access Note

This thesis is embargoed until 17th February 2026.

Access to this thesis is restricted. Please see the Access Note below for access details.