Reproducibility of pacing profiles in competitive swimmers.

This study aimed at determining the reproducibility of pacing profiles (PP) during simulated swimming trials as well as the comparison between simulated and real competitions (RC). Sixteen competitive front crawl swimmers (7 females, 9 males) performed 2 × 200 m, 2 × 400 m and 2 × 800 m tests, each test 7 days apart. All 100 m split (ST) and total times (TT) were recorded (additionally 50 m ST for the 200 m bouts). The PP of one RC within a maximum of 8 weeks before or after data acquisition was used for comparison. No difference was observed between test and retest for TT (p<0.16). Coefficients of variation (CV) for all ST during 800 m were between 0.9 and 1.8% (standard error of measurement (SEM)=0.6-2.1 s), except for the last 2 sections (CV=2.5% and 2.9%). During 400 m and 200 m, CV was below 1.7% for each section (SEM=0.4-1.7 s). Mean differences between test and retest ranged from 1.8 s (Cl: 0.1-3.4 s) in the 400 m bouts to 4.1 s (Cl: 1.3-9.5 s) for the 800 m races. Although section times were faster during all sections of RC compared to SC, PP was similar during both trials (p>0.22). However, swimmers were faster in each section during RC. In conclusion, PP seem stable, at least during the first three quarters of the race. Furthermore, simulated trials seem to be an acceptable model to analyse PP in competitive swimming.