RESUMO
Classification is a defining feature of Para-sport, and sports-specific classification systems determined through multidisciplinary scientific research are required, i.e., evidence-based and focused on the relationship between the impairment and the key performance determinants. Data envelopment analysis (DEA) was applied as a classification tool using a directional distance function (DDF) model. The aim of the study was to test the DEA as a possible classification tool in cerebral palsy football. We analyse the performance of 56 international para-footballers with hypertonia, ataxia or athetosis, who completed a 20-test battery with DEA models. Five of the tests are included in the model (change of direction: Illinois agility test; jumping: standing broad jump, four bounds for distance, and triple hop with the non-dominant leg; 10-m sprint/acceleration; and ball dribbling, both in a straight line and following a trajectory), showing that players with less impairment exhibit the highest efficiency. This outcome suggests that DEA models might be feasible for detecting and discriminating the performance and magnitude of impairment in cerebral palsy football, with an objective ranking of the athletes in relation to different physical performance tests. This study also provides reference scores for decision-making during classification and guidance for further research in team Paralympic sports.