Background: Although the biomechanical variables of individual starts have been widely studied, there is a notable lack of research on the specific techniques used in relay starts. Aim: This paper presents the preliminary results of a research project aimed at analyzing the movement patterns of two relay start variations using a markerless approach: S1 (no step with parallel feet) and S2 (no step with separated feet). Methods: Four recreational swimmers (two men, two women; mean height 167 ±7.75 cm, weight 65.5 ±12.37 kg, age 24 ±2.58 years) participating in a university sports league were recorded using a GoPro Hero4 (1080p,120Hz) positioned in the sagittal plane while performing both start types. Since the narrow configuration was used no distortion correction was applied. Video frames were processed using a markerless motion capture approach (Vailà software) to obtain joint angles for the shoulder, elbow, hip, knee, and ankle on both sides of the body. The joint angles at the moment of loss of foot contact (Event 1) and at water entry (Event 2) were compared between start types using the nonparametric Wilcoxon signed-rank test (* < 0.05). This study was approved under CAAE number 93853325.3.0000.5153. Results: At Event 1, for both S1 and S2, the mean joint angles (SD) were: shoulder 139.93° ±1.70°, elbow 173.29° ±2.06°, hip 158.25° ±1.99°, knee 176.05° ±2.37°, and ankle 123.19° ±4.11°. At Event 2, the mean angles were: shoulder 164.71° ±2.69°, elbow 175.17° ±2.48°, hip 166.21° ±2.99°, knee 176.83° ±2.11°, and ankle 124.81° ±0.69°. No significant differences were found between the two start variations at either event. Conclusions: Our preliminary results suggest that joint behavior followed the same pattern in both techniques across both analyzed events. Future work should correlate these
© Copyright 2026 XVth International Symposium on Biomechanics and Medicine in Swimming Proceedings. All rights reserved.