Chéry, C and Ruf, L. Reliability of the load-velocity relationship and validity of the PUSH to measure velocity in the deadlift. J Strength Cond Res 33(9): 2370–2380, 2019—This study investigated the reliability between load and mean velocity, mean propulsive velocity, peak velocity, mean power, and peak power in the deadlift. Also, we looked at the validity of an inertial sensor (PUSH) and a linear-position transducer (Tendo) to measure velocity variables. Ten strength-trained men (23.4 ± 1.3 years) were involved in three 1 repetition maximum (1RM) testing sessions, separated by at least 72 hours. The protocol used 6 different lifting intensities, comprising 20, 40, 60, 80, 90, and 100% of 1RM. Reliability of the load-velocity and load-power relationship was assessed by comparing velocity and power measurements from session 2 and 3 for each relative %1RM. The validity of tested devices was analyzed regarding to each relative intensity by comparing results from each instrument to a reference instrument (GymAware). The findings revealed that intermediate intensities (ranging from 60 to 90% of 1RM) seem to be reliable. Furthermore, extreme points of the load-velocity curve (20, 40, and 100% of 1RM) were found to be less reliable and should therefore be used with caution when implemented as part of autoregulating strategies. Tendo produced measurements that were highly correlated with GymAware and thus, constitutes a valid and cheaper alternative. By contrast, measurements from the PUSH presented a low level of precision and accuracy. Therefore, PUSH cannot be considered as a valid tool to measure velocity variables in the deadlift.