Training Simulation with Nothing but Training Data - Simulating Performance based on Training Data Without the Help of Performance Diagnostics in a Laboratory

Traditional and contemporary approaches to mathematical fitness-fatigue models in exercise science: A practical guide with resources. Part I.

“…It is only in the last decade or so that data collection on training input and performance has substantially increased making these types of investigations more feasible across many sporting domains. Notably, some recent studies have begun to take advantage of the opportunity to utilise field data with fitness-fatigue models [27,29,55,66,74]. However, the majority of the research has focused on quantifying relationships between the underlying physiological processes of training, through iterative refinement of model structure to map with observed physiological phenomena [45].…”

“…Where (τ g ≥ τ h ≥ 0) are constants that describe the rate of the fitness and fatigue decay, respectively. Initial conditions of the system are generally set at g(0) = h(0) = 0 if the athlete has not exercised for a considerable amount of time, else these initial values may need to be estimated a priori [26,27] or within the optimisation problem [11], or be selected based upon a previous modelling period. Larger values of ω(t) represent greater training doses and therefore a greater stimulus for adaptation.…”