A new procedure for race analysis in swimming based on individual distance measurements.

The aim of this study was to develop a new application based on the "individual distances" method to analyse swimming races, and to compare it with the traditional "fixed distances" method. One hundred and seventy-nine national level 100 m (four strokes) performances obtained from the 2008 "Open Comunidad de Madrid" (Spain) were analysed using a two-dimensional Direct Linear Transformation (2D-DLT) video analysis system. Average velocities in all race segments (P < 0.001) were faster using the "individual distances" method than when employing the "fixed distances" method. Specifically, start and turn times were shorter (P < 0.001) while free swimming times were longer (P < 0.001) when using the "individual distances" method. Correlations between methods were moderate to high, but several gender and stroke groups showed poor to no correlation, especially during the start and turn segments. Differences between methods were higher in some groups (female swimmers and freestyle stroke) where the start and turn distances were shorter. Measurements with the 2D-DLT technique provide distances and times employed during the race segments, which do not completely agree with times at fixed distances. Therefore, when evaluating swimming races, a combination of the individual and fixed distances methods should be used.