90% mental, 10% physical? Psychological and physiological predictors of ultrarunning performance

Within the ultrarunning community, performance is often considered ‘90% mental, 10% physical’. This demonstrates the influence of both psychological and physiological factors, particularly emphasising the perceived dominance of psychology. However, extant research typically considers psychological or physiological variables in isolation. Additionally, studies frequently recruit participants from a single ultramarathon, thereby neglecting heterogeneity between events. The current study addresses these limitations by employing an interdisciplinary approach and recruiting participants from a range of races to develop a robust, predictive model of ultrarunning performance. Ultramarathon participants (n = 69 across 24 events) underwent graded exercise testing, psychometric surveys, and physiological/cognitive assessments (Cold Pressor, Stop-Signal, and Serial Sevens tasks). Performance was quantified as percentile finishing time relative to gender-matched event winners. Selection of the initial 15 candidate variables was informed by evidence from relevant literature. A single global model was developed using backwards stepwise regression (Aikake Information Criterion selection), including four covariates: age, gender, apparatus, and distance. The overall model (p < .001, adj. r2 = .426) included VO2max (ß = .609), emotional intelligence (ß = .281), inhibitory control (ß = .259), and sleep behaviours (ß = .226) as key significant predictors. These

Results: demonstrate that the predictability of ultrarunning performance is improved by moving beyond traditional metrics of fitness and considering both psychological and physiological variables. Overall, this model provides an integrated approach for maximising endurance sports training and performance, with broader application to other fields involving physical and psychological exertion (e.g., military, pain management).