Computer Simulations Assessing the Potential Performance Benefit of a Final Increase in Training During Pre-Event Taper

Abstract
Thomas, L, Mujika, I, and Busso, T. Computer simulations assessing the potential performance benefit of a final increase in training during pre-event taper. J Strenth Cond Res 23(6): 1729-1736, 2009-A nonlinear model of training responses was utilized to test whether a 2-phase taper could be more effective than a traditional linear taper. Simulations were conducted using model parameters previously determined in 6 nonathletes trained on a cycle ergometer (non-ATH) and 7 elite swimmers trained in sport-specific conditions (ATH). Linear and 2-phase tapers were compared after a 28-day overload period at 120% of normal training. The 2-phase taper was assumed to be identical to the optimal linear taper, except for the final 3 days during which the training load was varied to maximize the final performance. The optimal linear taper was characterized by a mean training reduction by 32 ± 6% during 35 ± 6 days in non-ATH and by 49 ± 18% during 33 ± 16 days in ATH. The last 3 days of the 2-phase taper were characterized by a significant increase in training load by 23 ± 18% in non-ATH and 29 ± 42% in ATH (p < 0.005). The optimal taper characteristics were not statistically different between non-ATH and ATH. The maximal performance reached with the 2-phase taper was higher by 0.04 ± 0.02% in non-ATH and 0.01 ± 0.01% in ATH than with the optimal linear taper (p < 0.001). Positive and negative influences of training on performance were estimated as indicators of adaptation and fatigue, respectively. The negative influence was completely removed during both tapers, whereas the positive influence was slightly further enhanced during the 2-phase pattern. In conclusion, simulations showed that a 20 to 30% increase in training at the end of the taper, as compared to a prolonged reduction in training, allowed additional adaptations without compromising the removal of fatigue.

This publication has 18 references indexed in Scilit: