Prediction of performance in a 100-km run from a simple equation

This study aimed to identify predictive variables of performance for a 100-km race (Perf(100-km)) and develop an equation for predicting this performance using individual data, recent marathon performance (Perf(marathon)), and environmental conditions at the start of the 100-km race. All runners who had performed official Perf(marathon) and Perf(100-km) in France, both in 2019, were recruited. For each runner, gender, weight, height, body mass index (BMI), age, the personal marathon record (PR(marathon)), date of the Perf(marathon) and Perf(100-km), and environmental conditions during the 100-km race (i.e., minimal and maximal air temperatures, wind speed, total amount of precipitation, relative humidity and barometric pressure) were collected. Correlations between the data were examined, and prediction equations were then developed using stepwise multiple linear regression analyses. Significant bivariate correlations were found between Perf(marathon) (p<0.001, r = 0.838), wind speed (p<0.001, r = -0.545), barometric pressure (p<0.001, r = 0.535), age (p = 0.034, r = 0.246), BMI (p = 0.034, r = 0.245), PR(marathon) (p = 0.065, r = 0.204) and Perf(100-km) in 56 athletes The, 2 prediction equations with larger sample (n = 591) were developed to predict Perf(100-km), one including Perf(marathon), wind speed and PR(marathon) (model 1, r² = 0.549; standard errors of the estimate, SEE = 13.2%), and the other including only Perf(marathon) and PR(marathon) (model 2, r² = 0.494; SEE = 14.0%). Perf(100-km) can be predicted with an acceptable level of accuracy from only recent Perf(marathon) and PR(marathon), in amateur athletes who want to perform a 100 km for the first time.