In this study, we aimed to examine the maximal aerobic speed (sVO2max) prediction capability of different modalities in amateur soccer players.
Eighteen male soccer players with at least 3 years of training experience were recruited. Players randomly attended the following 4 testing sessions: Yoyo intermittent recovery test-1 (IRT1), multistage 20 m shuttle test (20-MST), 5 minutes Time Trial (5 min TT) and Université de Montréal Track test (UMTT). Differences and relationships between values of UMTT sVO2max and final speed (FS) recorded with the Yoyo IRT1 and 20-MST, mean speed of 5 min TT and sVO2max equations suggested in literature were tested.
sVO2max obtained with UMTT was significantly different to that obtained with Yoyo IRT1, 20-MST and 5 min TT (−1.7 ± 0.8, 2.1 ± 0.7, −0.6 ± 1.1 km.h−1 respectively, p < 0.05). The highest correlations were found between UMTT sVO2max and 20-MST FS (r = 0.70) and Yoyo FS (r = 0.69). The Bland–Altman plots showed high bias when comparing sVO2max UMTT with Yoyo IRT1 and 20-MST FS, indicating that these two methods systematically generate different results. The 5 min TT had the smallest bias and the largest limits of agreement with UMTT sVO2max, whereas the FS Yoyo IRT1 and 20-MST methods had relatively narrow limits of agreement. Additionally, the regression equation model used for predicting sVO2max was more robust using Yoyo IRT1 results (R2 = 0.70 vs. R2 = 0.63 for 20-MST). These findings suggest that after sVO2max prediction using the provided equations, coaches may establish precise aerobic training loads in such population.