Acute and Residual Soccer Match-Related Fatigue: A Systematic Review and Meta-analysis

Silva JR, Rumpf MC, Hertzog M, Castagna C, Farooq A, Girard O, Hader K

Background: Understanding soccer players’ match-related fatigue and recovery profiles likely helps with developing conditioning programs that increase team performance and reduce injuries and illnesses. In order to improve match recovery (the return-to-play process and ergogenic interventions) it is also pivotal to determine if match simulation protocols and actual match-play lead to similar responses.


To thoroughly describe the development of fatigue during actual soccer match play and its recovery time course in terms of physiological, neuromuscular, technical, biochemical and perceptual responses.
To determine similarities of recovery responses between actual competition (11 vs. 11) and match simulations.
Methods: A first screening phase consisted of a systematic search on PubMed (MEDLINE) and SportDiscus databases until March 2016. Inclusion criteria were: longitudinal study with soccer players; match or validated protocol; duration ≥ 45 min; and published in English.

Results: A total of 77 eligible studies (n = 1105) were used to compute 1196 effect sizes (ES). Half-time assessments revealed small to large alterations in immunological parameters (e.g. leukocytes, ES = 1.9), a moderate decrement in insulin concentration (ES = – 0.9) and a small to moderate impairment in lower-limb muscle function (ES = – 0.5 to – 0.7) and physical performance measures (e.g. linear sprint, ES = – 0.3 to – 1.0). All the systematically analyzed fatigue-related markers were substantially altered at post-match. Hamstrings force production capacity (ES = – 0.7), physical performance (2–4%, ES = 0.3-0.5), creatine kinase (CK, ES = 0.4), well-being (ES = 0.2-0.4) and delayed onset muscle soreness (DOMS, ES = 0.6–1.3) remained substantially impaired at G ≥ 72 h. Compared to simulation protocols, 11 vs. 11 match format (CK, ES = 1.8) induced a greater magnitude of change in muscle damage (i.e. CK, ES = 1.8 vs. 0.7), inflammatory (IL-6, ES = 2.6 vs. 1.1) and immunological markers and DOMS (ES = 1.5 vs. 0.7) than simulation protocols at post-assessments. Neuromuscular performances at post-match did not differ between protocols.

Conclusion: While some parameters are fully recovered (e.g. hormonal and technical), our systematic review shows that a period of 72 h post-match play is not long enough to completely restore homeostatic balance (e.g. muscle damage, physical and well-being status). The extent of the recovery period post-soccer game cannot consist of a ‘one size fits all approach’. Additionally, the ‘real match’ (11 vs. 11 format) likely induces greater magnitudes of perceptual (DOMS) and biochemical alterations (e.g. muscle damage), while neuromuscular alterations were essentially similar. Overall, coaches must adjust the structure and content of the training sessions during the 72-h post-match intervention to effectively manage the training load within this time-frame.

View this research