RPE and RIR Explained: How to Train by Effort, Not Just the Number on the Bar

Your spreadsheet says 225 today. It says 225 whether you slept eight hours or four, whether you’re fresh off a deload or three weeks deep into a hard block, whether your last meal was a steak or a vending machine granola bar.

The bar doesn’t care. The number is the number.

But your body cares — a lot. A fixed percentage tells you what to lift. Effort tells you what you can actually handle today. RPE and RIR are how you measure that effort, autoregulation is what you do with the measurement, and the right objective data is what keeps you honest about both. This is the layer of training that separates lifters who keep progressing from lifters who keep grinding.

What Is RPE in Weightlifting?

RPE stands for Rating of Perceived Exertion. In resistance training, the RIR-based version uses a 1-to-10 scale where 10 means you estimate that no additional repetition was possible with acceptable technique. Lower ratings correspond to more estimated reps in reserve. It is a useful language for effort, not an objective measurement of remaining repetitions.

The idea goes back to Swedish psychophysiologist Gunnar Borg, who built the original perceived-exertion scale in the 1970s and 80s. His version ran from 6 to 20 (mapped loosely to heart rate), and it was designed for endurance work, not barbells¹. It worked, but the math was clunky for a lifter counting reps.

The resistance-training version is more concrete than Borg’s endurance scale. Dr. Mike Zourdos and colleagues studied a 1-to-10 RPE scale anchored to estimated reps in reserve.² In this convention, RPE 10 corresponds to zero estimated reps left and RPE 8 to two. The word estimated matters: lifters can be wrong, especially farther from failure or in unfamiliar exercises.

Here’s the full conversion, the single most useful chart in this entire article:

RPE	Reps in Reserve (RIR)	What it feels like
10	0	Maximal — no more reps possible with good form
9.5	0–1	Maybe one more, maybe not
9	1	One clean rep left
8	2	Two reps left, bar speed slowing
7	3	Three reps left, still moving well
6	4	Four-plus reps left, comfortable
5 and below	5+	Far from failure; purpose depends on load and programming

Use the table as a shared vocabulary, then calibrate it against actual performance over time.

What Are Reps in Reserve (RIR)?

Reps in reserve is the number of additional reps you could have completed before hitting failure on a given set. Stop a set with two clean reps still in you, and you trained at RIR 2.

If RPE is the felt scale, RIR is the concrete input underneath it. RIR asks one specific question — how many more could you have done? — and RPE simply rates how that proximity to failure felt on a 1-to-10 scale. They’re two readouts of the same thing.

Many hypertrophy programs use sets relatively close to failure, often leaving some repetitions in reserve. Recent evidence suggests muscle growth generally increases as sets finish closer to failure, but the exact relationship remains uncertain and depends on load, exercise selection, volume, and training status.³ Strength can improve across a wider range of RIR targets.

The RPE/RIR Cheat Sheet

Use these cues mid-set to calibrate where you actually are:

RIR 0 (RPE 10): You estimate that no additional repetition was possible with acceptable technique. This is highly fatiguing and is not required for every set.
RIR 1 (RPE 9): You estimate that one more repetition was possible.
RIR 2 (RPE 8): You estimate that two more repetitions were possible.
RIR 3 (RPE 7): You estimate that three more repetitions were possible.
RIR 4+ (RPE 6 and below): The set is farther from failure. It may suit warm-ups, technique practice, power work, or planned lower-fatigue volume; it is not automatically “junk.”

RPE vs. RIR — What’s the Difference?

In the RIR-based resistance-training scale, RPE and RIR describe the same estimate from two directions: RIR estimates repetitions left, while RPE maps that estimate onto a 1-to-10 rating. RPE 8 conventionally maps to RIR 2.

The scale convention is easy to hold in your head:

Estimated RIR = 10 − RPE for the whole-number anchors. RPE 9 maps to 1 estimated rep in reserve; RPE 7 maps to 3.
Both are subjective estimates. The conversion does not make either value objectively correct.
Use whichever language clicks. Newer lifters often find RIR more concrete (“how many more?”). Experienced lifters tend to drift toward RPE because the felt scale becomes second nature.

Pick the term that is easiest to use consistently, and expect calibration to improve through repeated, well-controlled practice rather than arithmetic alone.

What Is Autoregulation, and What Does the Research Show?

Autoregulation means that a lifter or coach may adjust planned training from current performance rather than following a fixed load regardless of what happens in the session. RPE and RIR can inform that decision; they do not make it automatic or objectively correct.

Think of a fixed percentage as the starting route. Warm-up performance, technique, symptoms, and the program’s rules provide information a lifter or coach can use to stay on that route or choose a bounded change.

The case for autoregulation is well-developed. A 2020 Sports Medicine review by Greig and colleagues mapped the autoregulation literature and built the conceptual framework for why matching load to the individual — rather than to a number set weeks ago — makes sense⁴. The harder question is whether it actually beats fixed loading in practice.

In one 12-week trial, an RIR-autoregulated group improved strength more than a fixed-loading comparison group.⁵ That supports RIR as a practical option in that program; it does not prove every autoregulated design is superior to every percentage-based plan.

How Accurate Is RPE/RIR — and Who Should Be Careful?

RIR accuracy varies with exercise, load, proximity to failure, and familiarity with hard effort. Training experience may help in some contexts, but it does not guarantee an accurate estimate.

In Zourdos’s study, average lifting velocity and RPE had strong inverse correlations in both experienced (r = −0.88) and novice (r = −0.77) squatters.² That shows the measures moved together in this protocol; it does not show that experienced lifters estimated remaining repetitions almost perfectly.

Daniel Hackett and colleagues found errors were larger farther from failure and differed by exercise.⁶ In that study, resistance-training experience did not significantly affect accuracy. The practical lesson is to use RIR cautiously, particularly when a set stops far from failure.

This is the limitation of a subjective scale: the value is an estimate shaped by the task and the lifter’s reference points. Record it honestly, then compare it with subsequent performance rather than treating it as lab-grade data.

Eric Helms and colleagues described practical applications of the RIR-based RPE scale for resistance training.⁷ It can help lifters adjust load when a planned weight produces a different effort than expected, but the rating still needs context.

Wearable recovery data is not an objective validator of set-level RPE. It can add context about sleep, HRV, resting heart rate, and recent workouts, while the barbell set itself provides the relevant performance feedback.

How Wearable Context Fits Beside Set-Level Effort

A wearable cannot feel the bar in your hands or prescribe a validated RIR target. It can summarize recovery-related metrics that you consider alongside symptoms, motivation, warm-up performance, technique, and the plan.

Capacity and perceived effort can vary with time of day, sleep, stress, and accumulated fatigue.⁸ A planned percentage may therefore produce a different effort than expected, but neither a wearable score nor one subjective rating proves why.

Plews and colleagues reviewed longitudinal HRV monitoring in elite endurance athletes,⁹ while Shaffer and Ginsberg reviewed HRV metrics and norms.¹⁰ Neither source validates a universal rule that converts a morning HRV value into a resistance-training RIR prescription.

Apple Watch connects directly to HealthKit, and compatible Garmin, Oura, and WHOOP metrics can flow through HealthKit. Devices collect and summarize these signals with different schedules and methods; they do not all measure every metric continuously. SensAI can explain aggregated recovery context and use it during weekly program regeneration, but it does not convert those metrics into automatic set-level RIR targets.

If you want the deeper read on resolving the conflict between what your wearable says and what you feel, that’s covered here: Wearable data vs. perceived recovery: a deload-timing decision framework.

How to Use RPE and RIR for Hypertrophy and Strength

Use RPE and RIR to describe effort within a program designed for your goal, training age, exercise selection, and fatigue tolerance. There is no single optimal band for every set or lifter.

Hypertrophy: use proximity to failure deliberately

Hypertrophy tends to improve as sets finish closer to failure, but current evidence does not establish one exact optimum.³ With lighter loads, fatigue can increase motor-unit recruitment as a set progresses; with heavy loads, high-threshold motor units may be recruited earlier. Proximity to failure is one programming variable among load, volume, exercise selection, and recovery.

Training every set to momentary failure is not required for growth and can increase acute fatigue. The balance depends on the exercise and the rest of the program. Schoenfeld’s mechanisms paper discusses mechanical tension, metabolic stress, and muscle damage; it does not validate one universal RIR prescription.¹¹ For the full volume picture, see training volume for hypertrophy: sets per muscle per week.

Strength: use RPE to adjust planned loading

For strength work, RPE can help adjust a planned load when today’s performance differs from the assumption behind a percentage. The target should come from the program and exercise context. Technique breakdown does not guarantee injury, but it is a reason to stop or reduce load when control no longer meets the session’s standard.

Putting it together this week

Say your program calls for heavy squats, but the warm-up feels unusually slow and your technique deteriorates at loads that are normally comfortable. You might reduce the working weight while keeping the planned effort range, or choose a user-directed modification. The decision comes from the full context—not a fixed HRV percentage.

SensAI’s guided tracker records planned versus performed work, including completed sets, and supports user-requested exercise changes during a session. The current product contract does not include a target-RIR field for every set. For the broader strength application, see AI-automated progressive overload for strength training and how to build muscle: a complete guide.

Common RPE/RIR Mistakes (and How to Calibrate)

RIR errors can run in either direction. Accuracy generally worsens farther from failure, and exercise selection, load, and familiarity can all change the estimate.⁶

Watch for these traps:

Sandbagging the rating. You felt the set was brutal but logged it as moderate because the number felt less intimidating. The log lies, your fatigue accumulates, and you wonder why you’re always tired.
Treating an early in-set estimate as exact. Accuracy is usually worse farther from failure. Record the end-of-set estimate consistently and compare it with later performance.
Ignoring performance feedback. If warm-up speed, technique, pain, or symptoms differ materially from normal, reassess the planned load rather than forcing a number.
Assuming failure testing is mandatory. It is not. If a coach uses occasional calibration sets, exercise selection, training experience, spotting, equipment, and medical context should determine whether that is appropriate.

Feedback helps. You can tell SensAI that a set felt harder than expected and request a shorter session, less volume, or an exercise change in plain language. Its memory can retain injuries, preferences, and constraints; the documented product does not automatically store an RPE calibration model or rewrite future prescriptions from one rating.

The Bottom Line

The number on the bar and perceived effort describe different parts of the same set. Neither one is sufficient alone.

RPE and RIR give lifters a consistent language for estimated effort. Autoregulation uses that estimate alongside actual performance. Recovery data can add context, but it does not make a subjective rating objectively correct.

SensAI combines the performance you log with aggregated recovery context when it regenerates your program each week. Current-session changes remain under your control through conversation or quick actions. Pair that with a thoughtful deload week and an AI-automated progressive overload approach, and the plan can respond to evidence without pretending to know more than the data shows.

References

Borg GA. “Psychophysical bases of perceived exertion.” Med Sci Sports Exerc, 1982. https://pubmed.ncbi.nlm.nih.gov/7154893/ ↩
Zourdos MC, Klemp A, Dolan C, et al. “Novel Resistance Training-Specific Rating of Perceived Exertion Scale Measuring Repetitions in Reserve.” Journal of Strength and Conditioning Research, 2016. https://pubmed.ncbi.nlm.nih.gov/26049792/ ↩ ↩²
Robinson ZP, Pelland JC, Remmert JF, Refalo MC, Jukic I, Steele J, Zourdos MC. “Exploring the Dose-Response Relationship Between Estimated Resistance Training Proximity to Failure, Strength Gain, and Muscle Hypertrophy: A Series of Meta-Regressions.” Sports Medicine, 2024;54(9):2209-2231. https://pubmed.ncbi.nlm.nih.gov/38970765/ ↩ ↩²
Greig L, Stephens Hemingway BH, Aspe RR, et al. “Autoregulation in Resistance Training: Addressing the Inconsistencies.” Sports Medicine, 2020. https://pubmed.ncbi.nlm.nih.gov/32813181/ ↩
Graham T, Cleather DJ. “Autoregulation by ‘Repetitions in Reserve’ Leads to Greater Improvements in Strength Over a 12-Week Training Program Than Fixed Loading.” Journal of Strength and Conditioning Research, 2021. https://pubmed.ncbi.nlm.nih.gov/31009432/ ↩
Hackett DA, Cobley SP, Davies TB, Michael SW, Halaki M. “Accuracy in Estimating Repetitions to Failure During Resistance Exercise.” Journal of Strength and Conditioning Research, 2017. https://pubmed.ncbi.nlm.nih.gov/27787474/ ↩ ↩²
Helms ER, Cronin J, Storey A, Zourdos MC. “Application of the Repetitions in Reserve-Based Rating of Perceived Exertion Scale for Resistance Training.” Strength and Conditioning Journal, 2016. https://pubmed.ncbi.nlm.nih.gov/27531969/ ↩
Chtourou H, Souissi N. “The Effect of Training at a Specific Time of Day: A Review.” Journal of Strength and Conditioning Research, 2012. https://pubmed.ncbi.nlm.nih.gov/22531613/ ↩
Plews DJ, Laursen PB, Stanley J, Kilding AE, Buchheit M. “Training Adaptation and Heart Rate Variability in Elite Endurance Athletes: Opening the Door to Effective Monitoring.” Sports Medicine, 2013. https://pubmed.ncbi.nlm.nih.gov/23852425/ ↩
Shaffer F, Ginsberg JP. “An Overview of Heart Rate Variability Metrics and Norms.” Frontiers in Public Health, 2017. https://pubmed.ncbi.nlm.nih.gov/29034226/ ↩
Schoenfeld BJ. “The Mechanisms of Muscle Hypertrophy and Their Application to Resistance Training.” Journal of Strength and Conditioning Research, 2010. https://pubmed.ncbi.nlm.nih.gov/20847704/ ↩