Velocity Variables: Determining Predictive Metrics During The Bench Press To Failure At Different Relative Intensities

PURPOSE: To investigate differences in repetitions to failure (RTF) between % one repetition maximum (1RM), and to determine which velocity metric, fastest repetition velocity (FRV) or average concentric velocity (ACV), best predicts RTF during the bench press (BP) at 70, 80, and 90% 1RM by comparing group versus individual data. METHODS: Eleven (n = 11) resistance-trained subjects (7 males & 4 females) reported to the lab on two separate days. 1RM testing was completed for BP on the first visit. During the second visit, subjects completed three sets to failure in descending order at 90, 80, and 70% of their 1RM. Using a linear position transducer, ACV was calculated from the first 2-3 repetitions, and FRV was the single rep with the fastest concentric velocity recorded for each set. A one-way repeated measures ANOVA was performed to examine possible differences in RTF among 70%, 80%, and 90% 1RM. Post hoc analyses included paired samples t-tests with Bonferroni corrections. Separate linear regression analyses were conducted for ACV and FRV as the predictor variables and RTF as the outcome variable. The coefficient of determination (R2) and Standard Error of the Estimate (SEE) were calculated from linear regression models using the group data and each subject individually. A dependent t-test was performed on the SEE from the individual linear regression analyses to determine differences in the prediction error between ACV and FRV. RESULTS: For RTF, there was a significant interaction (p < 0.001; η2p = 0.944). Post hoc tests indicated RTF differed significantly between all 3 (p < 0.001) %1RM comparisons. Linear regression from the group data indicated a moderate level of predictability for ACV (R2 = 0.53; SEE = 2.43 reps) and FRV (R2 = 0.52; SEE = 2.46 reps. Linear regression using the individuals’ data improved the prediction model for ACV (R2 median = 0.98; SEE = 0.69 reps) and FRV (R2 median = 0.97; SEE = 1.18 reps) with no statistically significant differences in SEE between ACV and FRV (p = 0.11). CONCLUSION: ACV and FRV show moderate predictability of RTF using the group data. Individual data provided the highest predictive models of RTF. These findings support the use of individual data to predict RTF over the group data, suggesting ACV and FRV can predict RTF during the BP within <1.5 reps.