RPE and RIR: How Accurate Is Perceived Effort, Really?

Every lifter has said it: "that set felt like an 8." The number usually comes from Reps in Reserve (RIR), the anchor scale that turned a vague sense of effort into a shorthand two decades of powerlifters and hypertrophy trainees now use daily. But a self-reported number only earns its place in a training log if it tracks something real. A decade of validation studies has now tested that assumption directly, set by set, against the one thing a barbell cannot lie about: how fast it moved.

Autoregulation exists because a fixed percentage of one-rep max cannot account for daily readiness. Sleep, stress, and cumulative fatigue all shift how heavy a "hard set" actually feels from one session to the next. RPE and RIR promised a fix, letting the lifter's own nervous system set the load rather than a spreadsheet. Whether that promise holds, and under which conditions it doesn't, changes how much weight a coach, an app, or a wearable should place on a lifter's self-report versus what the muscle and the bar are actually doing.

The Scale That Made Effort Measurable

The RIR-based RPE scale used across almost all of the research below was formalised by Zourdos and colleagues in 2016, adapting a 0 to 10 effort scale so that each whole-number rating corresponds to a specific number of reps left in the tank at failure: RPE 10 means no reps left, RPE 9 means one rep in reserve, and so on down to RPE 6 or below, where reserve becomes harder to pin down precisely.¹ In that original validation, barbell velocity fell in a strong, consistent relationship with stated RPE in both experienced lifters (r = -0.88) and novices (r = -0.77) across loads from 60% to 90% of 1RM and at a true one-rep max.¹ Helms and colleagues then built the practical framework for applying the scale to real training blocks, matching a stated RIR target to an expected percentage of 1RM so coaches could prescribe load without a barbell test every session.²

Perceived Effort Tracks Bar Speed Closely, Most of the Time

When 12 competitive powerlifters rated bench press RPE across a 3-week hypertrophy, power, and strength block, their self-ratings sat within an average of 0.33 points of the RPE researchers had actually prescribed, a level of precision that held up better on strength-focused days than on power days.³ A companion bench press study in 27 lifters split by training age found the same strong inverse relationship between rated effort and bar velocity in both experienced (4.7 years training) and novice (1.1 years) benchers (r = 0.85), though experienced lifters moved the bar slower and rated the same absolute load as harder near a true max.⁴ The relationship strengthens further with a second exercise pattern: in trained men and women performing front squats and hexagonal bar deadlifts to failure, RPE and velocity correlated at r = -0.98 to -1.00, essentially a straight line, with the gap between predicted and actual reps shrinking the closer a set got to RPE 9.⁵ A 2025 systematic review pooling 18 such studies confirmed the pattern holds broadly: RPE, movement velocity, and percentage of 1RM move together closely enough across the literature that the review authors describe rating scales as "a valid, economic, and practical tool" for everyday load management, not just a laboratory curiosity.⁶

The Closer to Failure, the More the Number Can Be Trusted

Two further findings complicate the tidy picture. First, accuracy is not constant across a set. When lifters called out their felt RIR mid-set and researchers cross-checked it against how many reps they actually completed, predictions were consistently sharper close to failure than earlier in a set, and sharper in later sets of a session than in the first.⁷ The size of the miss also shrank as fatigue built. At RPE 9, nominally one rep in reserve, the average gap between predicted and actual RIR was 2.05 reps, compared with 5.15 reps at RPE 5, and the correlation between the stated rating and total reps completed was moderate at best (r = 0.65 at RPE 5, r = 0.56 at RPE 7).⁷ A separate cross-sectional study replicated the same pattern and found that neither sex, training experience, nor prior rating experience meaningfully changed how accurate the mid-set predictions were.⁸ Second, and more surprising, accuracy does not reliably improve with practice alone. Tracking nine trained lifters across six weeks and 18 bench press sessions, one study found no significant improvement in RIR prediction error over time, and if anything, the tendency to underestimate how many reps were left grew slightly across sessions.⁹ Rating effort well, in other words, is a skill that plateaus quickly rather than one that keeps sharpening simply from repeated exposure, a genuine constraint on any coaching model that assumes self-report calibrates itself.

Where the Scale Breaks Down

Population matters. In adults averaging 68 years old performing sets at RIR-2, RIR-4, and RIR-6 targets, actual velocity loss overshot the expected values at the higher-effort targets: sets stopped at a stated RIR-2 showed 16 percent velocity loss, well past what RIR-2 should represent, RIR-4 showed 10 percent loss, and only the more conservative RIR-6 target matched its expected near-zero loss.¹⁰ Lifters in that cohort underestimated their true proximity to failure by roughly 1.6 to 2.1 reps depending on the target, leading the authors to conclude that RIR "may lack precision" for prescribing exercise intensity in older adults specifically, even though the same scale performs well in younger trained cohorts.¹⁰ That is not a reason to discard the scale for older lifters, but it is a reason not to treat a stated RPE as interchangeable across every population a coach or app might serve.

RPE, RIR, and Velocity, Head to Head

The most direct test came in 2025, when a randomised crossover trial put four common autoregulation methods against each other in the same lifters: fixed percentage of 1RM, RPE, RIR, and velocity-based training.¹¹ Velocity-based training was the most accurate at consistently hitting its intended target, keeping every set within 5 percent of the intended bar speed, while both RIR and velocity-based training preserved more total training volume across the session than the rigid percentage-based approach, with no meaningful difference in the neuromuscular fatigue each method produced by the end.¹¹ A separate eight-week trial comparing RIR-based stopping points against training sets to literal failure found a further argument for RIR: lifters who stopped short of failure reported meaningfully better subjective feeling-scale scores and lower perceived discomfort than those who trained to failure, without sacrificing the training adaptations being chased.¹² Taken together, the evidence does not crown one method outright. It suggests each tool measures a slightly different thing, and that combining an objective signal with a subjective one closes gaps that either leaves on its own when used alone across a full periodised block.¹³ ¹⁴ ¹⁵

What This Means in Practice

None of this is an argument to abandon RPE or RIR. This body of work is exactly why the scale remains one of the most validated tools in strength training: it correlates strongly with real barbell velocity, especially in the last few reps of a hard set, which is precisely where the decision that matters most, stop or continue, actually gets made. What it does mean is that a stated number carries different confidence depending on when in the set it was given, how experienced or how old the lifter is, and whether that number has ever been checked against anything objective. This is precisely the gap a wearable that reads live muscle output can help close, not by replacing a lifter's own sense of effort, but by giving it something to be checked against: whether the set that felt like an 8 actually produced the recruitment and fatigue signature of one, in real time, rep by rep.

Key Takeaways

• The RIR-based RPE scale correlates strongly with barbell velocity in validation studies, particularly close to failure.

• Rating accuracy is highest in the final one to two reps of a set and tends to be noisier earlier on.

• Practice alone does not reliably sharpen RIR accuracy over weeks; underestimation can persist or even grow.

• Older adults show a wider, and practically meaningful, gap between stated and actual proximity to failure.

• Velocity-based and other objective signals do not replace perceived effort, they calibrate it.

References

1. Zourdos, M. C., Klemp, A., Dolan, C., Quiles, J. M., Schau, K. A., Jo, E., Helms, E., Esgro, B., Duncan, S., Garcia Merino, S., & Blanco, R. (2016). Novel resistance training-specific rating of perceived exertion scale measuring repetitions in reserve. Journal of Strength and Conditioning Research, 30(1), 267-275. https://doi.org/10.1519/JSC.0000000000001049

2. Helms, E. R., Cronin, J., Storey, A., & Zourdos, M. C. (2016). Application of the repetitions in reserve-based rating of perceived exertion scale for resistance training. Strength and Conditioning Journal, 38(4), 42-49. https://doi.org/10.1519/SSC.0000000000000218

3. Helms, E. R., Brown, S. R., Cross, M. R., Storey, A., Cronin, J., & Zourdos, M. C. (2017). Self-rated accuracy of rating of perceived exertion-based load prescription in powerlifters. Journal of Strength and Conditioning Research, 31(10), 2938-2943. https://doi.org/10.1519/JSC.0000000000002097

4. Ormsbee, M. J., Carzoli, J. P., Klemp, A., Allman, B. R., Zourdos, M. C., Kim, J. S., & Panton, L. B. (2019). Efficacy of the repetitions in reserve-based rating of perceived exertion for the bench press in experienced and novice benchers. Journal of Strength and Conditioning Research, 33(2), 337-345. https://doi.org/10.1519/JSC.0000000000001901

5. Odgers, J. B., Zourdos, M. C., Helms, E. R., Candow, D. G., Dahlstrom, B., Bruno, P., & Sousa, C. A. (2021). Rating of perceived exertion and velocity relationships among trained males and females in the front squat and hexagonal bar deadlift. Journal of Strength and Conditioning Research, 35(Suppl 1), S23-S30. https://doi.org/10.1519/JSC.0000000000003905

6. Petro, J. L., Ferrari, G., Cardozo, L. A., Vargas-Molina, S., Carbone, L., Kreider, R. B., & Bonilla, D. A. (2025). Validity of rating of perceived exertion scales in relation to movement velocity and exercise intensity during resistance exercise: A systematic review. Sports Health, 17(3), 621-628. https://doi.org/10.1177/19417381241260412

7. Zourdos, M. C., Goldsmith, J. A., Helms, E. R., Trepeck, C., Halle, J. L., Mendez, K. M., Cooke, D. M., Haischer, M. H., Sousa, C. A., Klemp, A., & Byrnes, R. K. (2021). Proximity to failure and total repetitions performed in a set influences accuracy of intraset repetitions in reserve-based rating of perceived exertion. Journal of Strength and Conditioning Research, 35(Suppl 1), S158-S165. https://doi.org/10.1519/JSC.0000000000002995

8. Remmert, J. F., Laurson, K. R., & Zourdos, M. C. (2023). Accuracy of predicted intraset repetitions in reserve (RIR) in single- and multi-joint resistance exercises among trained and untrained men and women. Perceptual and Motor Skills, 130(3), 1239-1254. https://doi.org/10.1177/00315125231169868

9. Remmert, J. F., Robinson, Z. P., Pelland, J. C., John, T. A., Dinh, S., Hinson, S. R., Elkins, E., Canteri, L. C., Meehan, C. M., Helms, E. R., Hall, M. E., Laurson, K. R., & Zourdos, M. C. (2023). Changes in intraset repetitions in reserve prediction accuracy during six weeks of bench press training in trained men. Perceptual and Motor Skills, 130(5), 2139-2160. https://doi.org/10.1177/00315125231189098

10. Gómez-Redondo, P., Alcazar, J., Valenzuela, P. L., Ara, I., Alegre, L. M., & Mañas, A. (2025). Validity of repetitions in reserve for prescribing resistance exercise in older adults. Experimental Gerontology, 210, 112884. https://doi.org/10.1016/j.exger.2025.112884

11. Cowley, N., Nicholson, V., Timmins, R., Munteanu, G., Wood, T., García-Ramos, A., Owen, C., & Weakley, J. (2025). The effects of percentage-based, rating of perceived exertion, repetitions in reserve, and velocity-based training on performance and fatigue responses. Journal of Strength and Conditioning Research, 39(4), e516-e529. https://doi.org/10.1519/JSC.0000000000005026

12. Refalo, M. C., Helms, E. R., Hamilton, D. L., & Fyfe, J. J. (2025). The effect of proximity-to-failure on perceptual responses to resistance training. European Journal of Sport Science, 25(3), e12266. https://doi.org/10.1002/ejsc.12266

13. Helms, E. R., Byrnes, R. K., Cooke, D. M., Haischer, M. H., Carzoli, J. P., Johnson, T. K., Cross, M. R., Cronin, J. B., Storey, A. G., & Zourdos, M. C. (2018). RPE vs. percentage 1RM loading in periodized programs matched for sets and repetitions. Frontiers in Physiology, 9, 247. https://doi.org/10.3389/fphys.2018.00247

14. Helms, E. R., Cross, M. R., Brown, S. R., Storey, A., Cronin, J., & Zourdos, M. C. (2018). Rating of perceived exertion as a method of volume autoregulation within a periodized program. Journal of Strength and Conditioning Research, 32(6), 1627-1636. https://doi.org/10.1519/JSC.0000000000002032

15. Helms, E. R., Kwan, K., Sousa, C. A., Cronin, J. B., Storey, A. G., & Zourdos, M. C. (2020). Methods for regulating and monitoring resistance training. Journal of Human Kinetics, 74, 23-42. https://doi.org/10.2478/hukin-2020-0011