Background: A recent advancement in velocity-based training involves estimating the maximum number of repetitions to failure (RTF) by analyzing the fastest velocity recorded within a set. A systematic review examining the fundamental characteristics of the RTF-velocity relationship is still lacking. Purpose: This study aimed to (1) determine the basic properties of the RTF-velocity relationships (goodness of fit, reliability, and accuracy) and (2) offer guidance on implementing various methodological factors that can impact the RTF accuracy prediction. Methods: Data were sourced from 3 databases: PubMed, SPORTDiscus, and Scopus. Studies were qualified for inclusion if they involved at least 2 sets performed to failure with different loads, utilized multijoint weight-lifting exercises, and monitored the RTF and fastest velocity for each set. Results: Six studies demonstrated (1) robust goodness of fit, (2) acceptable to high between-sessions reliability for the velocities associated to each RTF (1-15 RTF), and (3) acceptable RTF prediction accuracy during fatigue-free sessions (long interset rest), but, when fatigued (ie, short interset rest) the accuracy was compromised except for athletes with high training experience (eg, >2 y training-to-failure experience). Conclusions: The relationship properties remain unaffected regardless of the exercise (upper- vs lower-body), equipment (Smith- vs free-weight), velocity variable (mean and peak velocity), and resting time (from 5 to 10 min). However, the modeling procedure used (multiple- vs 2-point) did alter the accuracy. The individualized RTF-velocity relationships can be constructed through a linear regression model, but the failure experience seems to be a critical factor to increase its accuracy.
© Copyright 2025 International Journal of Sports Physiology and Performance. All rights reserved.