Physical preparation in football: An umbrella review based on a relational evidence framework
Antonio Aguilar Gómez, Luis Manuel Martínez ArandaBackground
Physical preparation in football is embedded within microcycles through heterogeneous analytical and integrated tasks performed on and off the field. This fragmentation complicates synthesis and limits inferences about competitive performance. This umbrella review synthesised, classified, and evaluated systematic reviews and meta-analyses using a relational framework distinguishing four relationships: on-field training and game demands (C→A); training or competition and player response (C→B); complementary physical training and performance-support capacities (D→A); and complementary or preventive training and injury-related or functional indicators (D→B).
Methods
Following umbrella-review guidance, Scopus, Web of Science, PubMed, and SPORTDiscus were searched for systematic reviews, with or without meta-analysis, published between January 2016 and June 2026 and addressing physical preparation in football. Two reviewers independently performed searching, selection, and data extraction. Methodological quality was assessed using AMSTAR-2.
Results
Seventeen reviews were included: five addressed C→A, one C→B, three D→A, and eight D→B. Methodological quality was heterogeneous: two reviews were rated moderate-high, nine moderate, three low-moderate, two moderate-low, and one low. Evidence was strongest for physical support capacities, whereas training or competition and player response was least represented.
Conclusion
Physical preparation has stronger support for improving physical support capacities and preventive indicators than for demonstrating direct transfer to performance in play. The framework organises evidence by relationship type and delimits inferences, avoiding the assumption that improvements in capacities, player responses, or functional and injury-related indicators necessarily enhance competitive performance. Direct evidence of transfer to play remains limited; therefore, studies should evaluate how adaptations are expressed in representative tasks and competition.