To permit the collection of quantitative data on start, turn and clean swimming performances in any swimming pool, the aims of the present study were to (1) validate a mobile in-field performance analysis system (PAS) against the Kistler starting block equipped with force plates and synchronized to a 2D camera system (KiSwim, Kistler, Winterthur, Switzerland), (2) assess the PAS's interrater reliability and (3) provide percentiles as reference values for elite junior and adult swimmers. Members of the Swiss junior and adult national swimming teams including medalists at Olympic Games, World and European Championships volunteered for the present study (n = 47; age: 17 ± 4 [range: 13-29] years; World Aquatics Points: 747 ± 100 [range: 527-994]). All start and turn trials were video-recorded and analyzed using two methods: PAS and KiSwim. The PAS involves one fixed view camera recording overwater start footage and a sport action camera that is moved underwater along the side of the pool perpendicular to the swimming lane on a 1.55 m long monostand. From a total of 25 parameters determined with the PAS, 16 are also measurable with the KiSwim, of which 7 parameters showed satisfactory validity (r = 0.95-1.00, p < 0.001, %-difference < 1%). Interrater reliability was determined for all 25 parameters of the PAS and reliability was accepted for 21 of those start, turn and swimming parameters (ICC = 0.78-1.00). The percentiles for all valid and reliable parameters provide reference values for assessment of start, turn and swimming performance for junior and adult national team swimmers. The in-field PAS provides a mobile method to assess start, turn and clean swimming performance with high validity and reliability. The analysis template and manual included in the present article aid the practical application of the PAS in research and development projects as well as academic works.

• Keywords
• competitive swimming, elite athlete, junior, youth,
• MeSH
• Video Recording methods   MeSH
• Adult MeSH
• Humans MeSH
• Adolescent MeSH
• Young Adult MeSH
• Swimming * physiology   MeSH
• Reproducibility of Results MeSH
• Athletic Performance * physiology   MeSH
• Check Tag
• Adult MeSH
• Humans MeSH
• Adolescent MeSH
• Young Adult MeSH
• Male MeSH
• Female MeSH
• Publication type
• Journal Article MeSH

INTRODUCTION: To investigate performance variation in all race sections, i.e., start, clean swimming, and turns, of elite short-course races for all swimming strokes and to determine the effect of performance variation on race results. METHODS: Comparing finalists and non-qualified swimmers, a total of 256 races of male swimmers (n = 128, age: 23.3 ± 3.1, FINA points: 876 ± 38) competing in the European short-course swimming championships were analyzed. The coefficient of variation (CV) and relative change in performance (Δ%) were used to compare intra-individual performance progression between rounds and inter-individual differences between performance levels using a linear mixed model. RESULTS: While most performance variables declined during the races (P < 0.005), performance was better maintained in 200 m compared to 100 m races, as well as in finalists compared to non-qualified swimmers. In 100 m races, Start Times improved between heats, semi-finals, and finals (P < 0.005) and contributed to the improved Split Times of Lap 1 in freestyle (P = 0.001, Δ = -1.09%), breaststroke (P < 0.001; Δ = -2.48%), and backstroke (P < 0.001; Δ = -1.72%). Swimmers increased stroke rate from heats/semi-finals to finals in freestyle (P = 0.015, Δ = 3.29%), breaststroke (P = 0.001, Δ = 6.91%), and backstroke (P = 0.005; Δ = 3.65%). Increases in stroke length and clean-swimming speed were only significant between rounds for breaststroke and backstroke (P < 0.005). In 200 m races, Total Time remained unchanged between rounds (P > 0.05), except for breaststroke (P = 0.008; CV = 0.7%; Δ = -0.59%). Start (P = 0.004; Δ = -1.72%) and Split Times (P = 0.009; Δ = -0.61%) only improved in butterfly. From the turn variables, OUT_5 m times improved towards the finals in breaststroke (P = 0.006; Δ = -1.51%) and butterfly (P = 0.016; Δ = -2.19%). No differences were observed for SR and SL, while clean-swimming speed improved between rounds in breaststroke only (P = 0.034; Δ = 0.96%). DISCUSSION: Performance of finalists progressed between rounds in 100 m but not 200 m races, most probably due to the absence of semi-finals. Progression in 100 m races was mainly attributed to improved Start and Split Times in Lap 1, while turn performances remained unchanged. Within round comparison showed higher performance maintenance in 200 m compared to 100 m events, which showed more pronounced positive pacing. Success of finalists was attributed to their overall higher performance level and superior progression between rounds.

• Keywords
• competition analysis, kinematic analysis, pacing, race parameters, swimming,
• Publication type
• Journal Article MeSH