Lifting Velocity as a Predictor of the Maximum Number of Repetitions That Can Be Performed to Failure During the Prone Bench Pull Exercise

Objective: To explore (1) the goodness of fit of generalized and individualized relationships between the maximum number of repetitions performed to failure (RTF) and the fastest mean velocity and peak velocity of the sets (RTF–velocity relationships), (2) the between-sessions reliability of mean velocity and peak velocity values associated with different RTFs, and (3) whether the errors in the prediction of the RTF under fatigued and nonfatigued conditions differ between generalized and individualized RTF–velocity relationships. Methods: Twenty-three sport-science students performed 4 testing sessions with the prone bench pull exercise in a Smith machine: a 1-repetition-maximum [1RM] session, 2 identical sessions consisting of singles sets of RTF against 4 randomized loads (60%–70%–80%–90%1RM), and 1 session consisting of 4 sets of RTF against the 75%1RM. Results: Individualized RTF–velocity relationships presented a higher goodness of fit (r2 = .96–.97 vs .67–.70) and accuracy (absolute errors = 2.1–2.9 repetitions vs 2.8–4.3 repetitions) in the prediction of the RTF than generalized RTF–velocity relationships. The reliability of the velocity values associated with different RTFs was generally high (average within-subject coefficient of variation = 4.01% for mean velocity and 3.98% for peak velocity). The error in the prediction of the RTF increased by ~1 repetition under fatigue (ie, set 1 vs sets 2–4). Conclusions: Individualized RTF–velocity relationships can be used with acceptable precision and reliability to prescribe the loads associated with a given RTF during the match a specific XRM during the prone bench pull exercise, but a lower accuracy is expected in a fatigued state.