Abstract
The objective of this study was to determine the methods of quantification for training and performance, which would be the most appropriate for modeling the responses to long-term training in cadet and junior judo athletes. For this, 10 young male judo athletes (15.9 6 1.3 years, 64.9 6 10.3 kg, and 170.8 6 5.4 cm) competing at a regional/state level volunteered to take part in this study. Data were collected during a 2-year training period (i.e.,702 days) from January 2011 to December 2012. Their mean training volume was 6.52 6 0.43 hours per week during the preparatory periods and 4.75 6 0.49 hours per week during the competitive periods. They followed a training program prescribed by the same coach. The training load (TL) was quantified through the session rating of perceived exertion (RPE) and expressed in arbitrary unit (a.u.). Performance was quantified from 5 parameters and divided into 2 categories: performance in competition and performance in training. The evaluation of performance in competition was based on the number of points per level. Performance in training was assessed through 4 different tests. A physical test battery consisting of a standing long jump, 2 judo-specific tests that were the maximal number of dynamic chinup holding the judogi, and the Special Judo Fitness Test was used. System modeling for describing training adaptations consisted of mathematically relating the TL of the training sessions (system input) to the change in performance (system output). The quality of the fit between TL and performance was similar, whether the TL was computed directly from RPE (R2 = 0.55 6 0.18) or from the session RPE (R2 = 0.56 6 0.18) and was significant in 8 athletes over 10, excluding the standing jump from the computation of the TL, leading to a simplest method. Thus, this study represents a first attempt to model TL effects on judo-specific performance and has shown that the best relationships between amounts of training and changes in performance were obtained when training amounts were quantified simply from RPE.