Accuracy of Predicted Intraset Repetitions in Reserve (RIR) in Single- and Multi-Joint Resistance Exercises Among Trained and Untrained Men and Women

We assessed the accuracy of intraset repetitions in reserve (RIR) predictions on single-joint machine-based movements of trained and untrained men and women. Participants were 27 men (M age = 22, SE = 0.6 years; M weight = 90.8, SE = 4.0 kg; M height = 182.3, SE = 1.4 cm; M training experience = 66, SE = 9 months) and 31 women (M age = 20, SE = 0.4 years; M weight = 67.8, SE = 2.3 kg; M height = 167.6, SE = 1.1 cm; M training experience = 22, SE = 4 months). In one session, participants performed a five-repetition maximum (5RM) test on biceps curl, triceps pushdown, and seated row exercises; we then estimated one repetition maximum (1RM). Participants then performed four sets of each exercise, in a randomized order, to the point of momentary muscular failure at 72.5% of 1RM. During each set, participants indicated when they first perceived 5RIR and then predicted RIR on every repetition thereafter until failure. The difference between actual repetitions performed and predicted repetitions at each intraset prediction was determined to be the RIR difference (RIRDIFF). A 3-way repeated measures ANCOVA found that a 3-way interaction was not statistically significant (p = 0.435) and no covariates of sex (p = 0.917), training experience (p = 0.462) nor experience rating RIR significantly affected RIRDIFF (p = 0.462-0.917). There were significant main effects for the proximity to failure of the prediction and the set number (p < 0.01) but not for exercise (p = 0.688). Thus, intraset RIR predictions were more accurate when closer to failure and in later sets, but sex, training experience, and experience rating RIR did not significantly influence RIR prediction accuracy on machine-based single-joint exercises.