The purpose of this investigation was to determine whether the concepts of critical swimming speed, critical stroke rate and anaerobic swimming capacity could be used by coaches as a reliable index in order to monitor endurance performances in competitive swimmers. The results of this study conducted with well-trained swimmers showed that the 30-min test velocity (V30) is not different from the critical swimming speed determined from 200- and 400-m tests but is overestimated by 3.2 %. Furthermore, a regression analysis of the number of stroke cycles on time calculated for each swimmer showed a linear relationship (r(2) greater than 0.99 and p less than 0.01). The 30-min stroke rate test (SR30) was not different from the critical stroke rate determined from 200- and 400-m tests after a correction of minus 3.9 %. These data suggest that the slope of this regression line represents the critical stroke rate defined as the maximal stroke rate value, which can theoretically be maintained continuously without exhaustion. Coaches could easily use critical swimming speed combined with critical stroke rate in order not only to set aerobic training loads but also to control the swimming technique during training. Besides, anaerobic swimming capacity (ASC) values defined as the y-intercept of the regression line between distance and time were not correlated (p > 0.05) with the determined distance over which a significant drop in the maximal speed could be noticed on a 25-m test. Thus, ASC does not provide a reliable estimation of the anaerobic capacity.

The purpose of this study was to analyse the effects of an exhaustive exercise on arm coordination and intracyclic velocity variations (IVV) to better understand the ways in which they are modified under fatigue conditions. Seventeen competitive swimmers performed a 200-m all-out test and a set of two 25-m (before and after the 200-m) at maximal intensity to measure stroking parameters, IVV, and the relative duration of the different parts of the stroke cycle and identify the model of arm coordination by using the index of coordination (IdC). Results showed an increase in the relative duration of the propulsive time, which induced a change in arm coordination as fatigue developed in relation to a decrease in stroke length (SL) and stroke rate (SR) (p<0.05). The evolution of IdC corresponds to a reduction of the non-propulsive lag time between the two arms' propulsive actions. Despite these modifications, IVV were not significantly modified (p<0.05). The present results highlighted that IdC and relative durations of each part of the cycle (particularly the pull phase) could be assessed to complete the "SL x SR" model and to partly understand the technique modifications under fatigue condition.