This study explored the accuracy of using the relationship between repetitions in reserve (RIR) and mean velocity (MV) to predict proximity to failure in bench press exercises across two modes (Smith machine and free-weight) and sexes (men and women). Twenty-eight recreationally trained individuals completed two sessions of each exercise mode, with sets performed to failure at 65% and 75% of one-repetition maximum. Individualized RIR-MV relationships showed a higher goodness-of-fit for the Smith machine compared to the free-weight bench press (R���=���0.79-0.87 vs. 0.41-0.78), without significant sex differences (���=���0.880). MV values were mostly similar between exercise modes across different RIRs (���>���0.077), but men demonstrated higher MV values than women for RIR 5-1 ( ��� 0.013). Individualized RIR-MV relationships were more accurate at estimating RIR 5 at 65%1RM in subsequent sessions than generalized RIR-MV relationships (absolute errors���=���0.64-1.36 vs. 1.25-2.00 repetitions; ���<���0.001), regardless of sex. However, the prediction accuracy was similar for both relationships at RIRs 2 and 0 (absolute errors���������1 repetition; ��� 0.164). These results suggest that RIR-MV relationships can effectively estimate proximity to failure across exercise modes and sexes, with individualized relationships offering greater accuracy.
