In elite rowing a 2.000 m maximum test performed on a rowing ergometer is generally used for monitoring individual physical abilities and team selection. Due to the high physical and psychological load and lactate accumulation of such tests a regular examination isn't recommendable. Different performance prediction models based on various anthropometric and metabolic parameters have been proposed (Maestu et al. 1999, Ingham et al. 2002, Riechman et al. 2002). Aim of the study was the development and evaluation of a 2.000 m performance prediction model for elite rowers based on anaerobic threshold and maximum sprint power. Methods: 18 male and 7 female elite rowers performed each a 2*8 min submaximal incremental test for the determination of power at 4 mmol/l blood lactate level (P4mmol/l) and a 20 s sprint for the determination of maximum sprint power (P20s) and maximal lactate formation rate (VLAmax). The 2.000 m race simulation time (T2.000) was measured separately in the course of a team selection competition. Using linear (multiple) regression analysis various combinations of parameters (also for gender specific subgroups) were tested for the prediction of T2.000. Results: The prediction model based on the parameter combinations P4mmol/l and P20s and the combination P4mmol/l, P20s and VLAmax revealed the most precise predictions of T2.000 (with R2 = 0.97 and R2 = 0.98, respectively). The standard deviation SD of residuals for both prediction models was 4.6 s and 3.8 s, respectively, for the entire group. Gender specific subgroups revealed a standard deviation SD of 4.7 s and 3.2 s, respectively, for the male and 2.7 s in both models for the female subjects. The distribution of residuals for both prediction models considering the entire group displayed a Gaussian distribution with a standard deviation sd of 3.5 s for the parameter combination P4mmol/l and P20s and sd of 3.8 s for the combination P4mmol/l, P20s and VLAmax. Discussion: The combination of a submaximal incremental test and a maximum sprint test allows for a precise prediction of the individuals 2.000m race time. We recommend the test procedure of our prediction model for regular examination of the athlete's ability. To further improve the prediction quality or enable a gender-specific model additional investigations with a larger and more diverse sample are needed.
© Copyright 2012 17th Annual Congress of the European College of Sport Science (ECSS), Bruges, 4. -7. July 2012. Veröffentlicht von Vrije Universiteit Brussel. Alle Rechte vorbehalten.