Modelling training response in elite female gymnasts and optimal strategies of overload training and taper.

The aim of the study is the modelling of training responses with a variable dose-response model in a sport discipline that requires highly complex coordination. We propose a method to optimise the training programme plan using the potential maximal performance gain associated with overload and tapering periods. Data from five female elite gymnasts were collected over a 3-month training period. The relationship between training amounts and performance was then assessed with a non-linear model. The optimal magnitude of training load reduction and its duration were investigated with and without an overload period using simulation procedures based on individual responses to training. The correlation between actual and modelled performances was significant (R² = 0.81 ± 0.02, P < 0.01). The standard error was 2.7%. Simulations revealed that taper preceded by an overload period allows a higher performance to be achieved compared to an absence of overload period (106.3 ± 0.3% vs. 105.1 ± 0.3%). With respect to the pre-taper load, the model predicts that optimal load reductions during taper were 48.4 ± 0.7% and 42.5 ± 1.0% for overloading and non-overloading strategies, respectively. Moreover, optimal durations of the taper period were 34 ± 0.5 days and 22 ± 0.5 days for overloading and non-overloading strategies, respectively. In conclusion, the study showed that the variable dose-response model describes precisely the training response in gymnasts.