Born, DennisDennisBornRüeger, EvaEvaRüegerBeaven, C MartynC MartynBeavenRomann, MichaelMichaelRomann2024-11-192024-11-192022-062045-232210.24451/arbor.19392https://doi.org/10.24451/arbor.1939210.1038/s41598-022-13837-3https://arbor.bfh.ch/handle/arbor/33890To provide percentile curves for short-course swimming events, including 5 swimming strokes, 6 race distances, and both sexes, as well as to compare differences in race times between cross-sectional analysis and longitudinal tracking, a total of 31,645,621 race times of male and female swimmers were analyzed. Two percentile datasets were established from individual swimmers’ annual best times and a two-way analysis of variance (ANOVA) was used to determine differences between cross-sectional analysis and longitudinal tracking. A software-based percentile calculator was provided to extract the exact percentile for a given race time. Longitudinal tracking reduced the number of annual best times that were included in the percentiles by 98.35% to 262,071 and showed faster mean race times (P < 0.05) compared to the cross-sectional analysis. This difference was found in the lower percentiles (1st to 20th) across all age categories (P < 0.05); however, in the upper percentiles (80th to 99th), longitudinal tracking showed faster race times during early and late junior age only (P < 0.05), after which race times approximated cross-sectional tracking. The percentile calculator provides quick and easy data access to facilitate practical application of percentiles in training or competition. Longitudinal tracking that accounts for drop-out may predict performance progression towards elite age, particularly for high-performance swimmers.enSport Trainingswissenschaftarticle