Relationships between ball speed and arm speed during the volleyball serve in youth elite male players,and why statistical significance might be misleading

The purpose of the study was three-fold: (i) to describe ball speed and arm speed during power jump serve in volleyball; (ii) compare the outcomes among playing positions, and (iii) analyze the relationships among outcomes. One middle blocker (MI), three opposites (OP), three setters (S), and six outside hitters or receiver-attackers (RA) youth elite volleyball players from the French national team participated in this study. This cross-sectional study analyzed the ball and arm speeds while serving over two consecutive weeks. A total of 150 valid power speed serves (ball speed = 103.0 ± 7.0 km h⁻¹; arm speed = 40.0 ± 4.2 km h⁻¹) were recorded during the six training sessions using a radar-gun and an inertial measurement unit. Ball speed varied from 90 to 110 km h⁻¹, while the coefficient of variation ranged from 1.5% to 7.5%. Arm speed varied between 35 and 45 km h⁻¹, while the coefficient of variation ranged between 2 and 14%. One-way ANOVA revealed significant differences among playing positions for the variables of ball speed (p = 0.018), arm speed (p = 0.019) and ratio of ball to arm speed (p = 0.002). Very large correlations between arm speed and the ratio of ball speed to arm speed were found (r = −0.77 [−0.83; −0.69]; p ≤ 0.001). The results suggest that players reduce their variability regarding the outcome. Finally, despite being related, the velocity of the arm and ball were not perfectly correlated. Thus, measuring both the arm and ball velocity may be necessary.