Wingate test, when time and overdue fatigue matter: Validity and sensitivity of two time-shortened versions

Abstract

This study aimed to analyze the validity and sensitivity of two time-shortened Wingate anaerobic tests (WAnTs), by means of three phases. In Phase A, 40 participants performed a traditional 30 s WAnT, whereas the first 15 s (WAnT15 ) and 20 s (WAnT20 ) were used to elaborate two predictive models. In Phase B, another 30 s WAnT was performed by 15 different volunteers to examine the error of these models (cross-validation). Finally, in Phase C, a 30 s WAnT was registered before and after a 10-week velocity-based training conducted by 22 different participants (training group, TRAIN = 11; control group that fully refrained from any type of training, CONTROL = 11). Power changes (in Watts, W) after this training intervention were used to interpret the sensitivity of the time-shortened WAnT. Adjusted coefficient of determination (R2 ) was reported for each regression model, whereas the cross-validation analysis included the smallest detectable change (SDC) and bias. Close relationships were found between the traditional 30 s WAnT and both the WAnT15 (R2 = 0.98) and WAnT20 (R2 = 0.99). Cross-validation analysis showed a lower error and bias for WAnT20 (SDC = 9.3 W, bias = −0.1 W) compared to WAnT15 (SDC = 22.2 W, bias = 1.8 W). Lastly, sensitivity to identify individual changes was higher for WAnT20 (TRAIN = 11/11 subjects, CONTROL = 9/11 subjects) than for WAnT15 (TRAIN = 4/11 subjects, CONTROL = 2/11 subjects). These findings suggest that the WAnT20 could become a valid and sensitive protocol to replace the traditional 30 s WAnT.