2019/02/06 by Phil Scarf, Mansour Shrahili, Scarf, Phil +8
Economics, Econometrics and Finance · Engineering · Medicine · #Applications (stat.AP) #FOS: Computer and information sciences #Sports Analytics and Performance #Sports Dynamics and Biomechanics #Sports Performance and Training
paper · pdf · doi:10.48550/arxiv.1902.02061
openalex publication_date 2019/02/06 · openalex created_date 2022/07/29 · openalex updated_date 2026/07/28
We suppose that performance is a random variable whose expectation is related\nto training inputs, and we study four performance measures in a statistical\nmodel that relates performance to training. Our aim is to carry out a robust\nstatistical analysis of the training-performance models that are used in\nproprietary software to plan training, and thereby put them on a firmer\nfooting. The performance measures we consider are calculated using power output\nand heart rate data collected in the field by road cyclists. We find that\nparameter estimates in the training-performance models that we study differ\nacross riders and across performance measures within riders. We conclude\ntherefore that models and their estimates must be specific, both to the\nindividual and to the quality (e.g. speed or endurance) that the individual\nseeks to train. While the parameter estimates we obtain may be useful for\ncomparing given training programmes, we show that the underlying models\nthemselves are not appropriate for the optimisation of a training schedule in\nadvance of competition.\n