This study aimed to establish the optimal cut-off point(s) between classes in a new, evidence-based classification system for VI judo. We collected results from international VI judo competitions held between 2012 and 2018. Data on visual acuity (VA) and visual field (VF) measured during classification were obtained. Performance was determined by calculating a win ratio for each athlete. VA was significantly associated with judo performance (r = -.33, p <.001), VF was not (r =.30, p =.15). Decision tree analysis suggested to split the data into two groups with a VA cut-off of 2.5 logMAR units. Stability assessment using bootstrap sampling suggested a split into two groups, but showed considerable variability in the cut-off point between 2.0 and 3.5 logMAR. We conclude that to minimise the impact of impairment on the outcome of competition, VI judo should be split into two sport classes to separate partially sighted from functionally blind athletes. To establish an exact cut-off point and to decide if other measures of visual function need to be included, we argue for continued research efforts together with careful evaluation of research results from a multidisciplinary perspective.
