Exploring competition constraints at the CrossFit® Games: A systematic characterisation through an ecological dynamics lens

CrossFit® has evolved from a fitness methodology into a global competitive sport characterised by dynamic tasks, diverse movements, and variable constraints. Despite growing scientific interest, the lack of a structured task characterisation limits understanding of its performance demands and the design of representative training strategies. This study applied the Matrix of Analysis for Sports Tasks framework to qualitatively characterise CrossFit® through 218 tasks performed by individual competitors at the CrossFit® Games (2007-2025). Tasks were analysed across three phases: (i) taxonomy, classifying movements within weightlifting and throwing, gymnastics, and metabolic conditioning; (ii) environment description, contextualising competition formats and settings; and (iii) task description, examining single- and multi-disciplinary demands. Results showed that weightlifting and throwing exhibited the highest prevalence, gymnastics increased notably in recent years, and metabolic conditioning remained a consistent but less frequent component. A growing trend towards multidisciplinary tasks highlights the importance of adaptability, requiring integration of strength, skill, and endurance within single events. Although frequent changes in season format and task design pose challenges, the MAST framework provides a flexible, replicable tool to accommodate this variability and support evidence-based training and performance preparation.