RPE Explained: Rate of Perceived Exertion for Lifters
"How heavy should it be?" has two possible answers. One is a number on a percentage chart that doesn't know how you slept. The other is RPE — a 1–10 scale that measures how hard a set actually was, today, in your body. Learn to rate your sets honestly and every program you ever run gets smarter.
What RPE is
RPE stands for Rate of Perceived Exertion: a 1–10 score you assign to a set the moment you rack the bar, answering one question — how many more reps did I have in the tank? In its original form (borrowed from endurance sports) it was a fuzzy feeling of effort. The lifting version, popularized by powerlifting coach Mike Tuchscherer, anchors every score to something countable: reps in reserve (RIR).
The anchoring is what makes it useful. "That felt like an 8" isn't a mood — it's a claim: I could have done exactly two more reps with good form. A set of 5 at RPE 8 and a set of 12 at RPE 8 involve very different weights, but they describe the same distance from failure, and that distance is what drives adaptation.
The RPE table
Memorize the top half; the bottom half exists mostly so warm-ups have a name.
| RPE | Meaning | Reps in reserve |
|---|---|---|
| 10 | Maximal effort — nothing left, the next rep fails | 0 RIR |
| 9.5 | Maybe one more rep, definitely no more weight | 0–1 RIR |
| 9 | One solid rep left | 1 RIR |
| 8 | Two reps left — hard but controlled | 2 RIR |
| 7 | Three reps left — the bar still moves fast | 3 RIR |
| 5–6 | Warm-up territory; speed and technique work | 4–5 RIR |
| 1–4 | Light warm-ups, bar work, mobility | Many |
Half points are legitimate above 8 — the difference between 9 and 9.5 is real and you'll feel it. Below 7, precision is pointless: nobody can tell 5 RIR from 6, and it doesn't matter.
Why RPE beats fixed percentages
Percentage-based programs prescribe weights off your one-rep max: "5 reps at 80%." The problem is that your true max isn't a constant — it's a moving target that swings meaningfully day to day with sleep, stress, food, and accumulated fatigue. On a great day, 80% for 5 is an easy RPE 7. After a bad week of sleep, the same bar weight might be a grinding 9.5.
A fixed percentage can't see any of that. RPE can — because you're the instrument. Prescribing "5 reps at RPE 8" instead of "5 reps at 80%" means the weight self-adjusts: heavier on strong days, lighter on weak ones, but the training stimulus — two reps from failure — stays exactly what the program intended. Coaches call this autoregulation, and it's the main reason RPE has taken over strength programming. It turns every session into the right session for the body that showed up.
Percentages aren't useless — they're a fine starting estimate, and pairing the two works well: pick the bar weight from a percentage, then adjust up or down until the set lands at the prescribed RPE.
How to calibrate your RPE
The honest objection to RPE is that it's subjective — and for new lifters, it is. Research and coaching experience agree on the direction of the error: beginners consistently underestimate effort. A set they rate RPE 9 often has four or five reps left in it, because they've never actually met failure and don't know what its neighborhood feels like — experienced lifters rate reps in reserve far more accurately than novices (Helms et al., 2016).
Calibration is straightforward:
- Take occasional AMRAP sets. Every few weeks, on a machine or a safe compound variation, take one set as far as it honestly goes. Before the set, predict your reps; afterward, compare. The gap between prediction and reality is your calibration error, and it shrinks fast.
- Check against an estimated 1RM. Plug the set into a 1RM calculator. If your "RPE 9" set of 8 estimates a max far above anything you've handled, you had more in the tank than you claimed.
- Watch bar speed. At a true RPE 7 the bar still moves crisply. Noticeable grinding starts around 9. If every rep of a "7" looks identical on video, it was probably a 5.
Most lifters are usefully accurate within a couple of months of paying attention — and accuracy above RPE 7, where it matters, comes first.
Where to program which RPE
Not every exercise deserves the same effort ceiling. The pattern most good programs follow:
- Big compounds (squat, deadlift, bench, press, rows): RPE 7–8. These lifts create the most systemic fatigue and carry the most technical risk. Two to three reps in reserve keeps form intact and lets you train them again in a few days.
- Isolation and machine work (curls, lateral raises, leg extensions): RPE 8–9. Low technical demand and low fatigue cost mean you can — and should — push closer to failure to make these count.
- RPE 10: rarely, and on purpose. Max-effort singles when testing strength, the final set of an AMRAP week, or a planned failure set on a machine. As a default setting, it buries you in fatigue for little extra gain.
Put together, most of your working sets should land between RPE 7 and 9. That's the zone where the stimulus is strong and the recovery bill is payable — the zone where progressive overload can keep compounding week after week.
Common mistakes
- Everything at RPE 10. Grinding every set to failure feels productive and tests well on social media, but the fatigue it generates suppresses your performance on every following set and session. Harder today, weaker all week.
- Sandbagging at 6 and calling it 8. The opposite failure, and the more common one. If the last rep of your "RPE 8" set looks exactly like the first, you were nowhere near two reps from failure — and sets that far from failure build very little.
- Using RPE to dodge hard work. "Felt like a 9 today" on a weight that was a 7 last week is occasionally true and usually an excuse. RPE is a tool for honesty, not a permission slip. When your ratings drift easy for weeks while the weights stand still, the scale isn't the problem.
RPE vs. training to failure
Training to failure — RPE 10 on every set — does work; failure is simply the far end of the effort scale. The question is cost. A meta-analysis comparing failure training with stopping short of failure found similar muscle growth and strength gains between the two — with strength actually favoring non-failure training in the volume-unmatched studies — at a fraction of the fatigue and joint stress (Grgic et al., 2022). Failure has a place: safe isolation exercises, final sets, occasional testing. As a philosophy, RPE 8–9 buys nearly all of the stimulus while leaving you able to train again on schedule — which, over a year, is where the results actually come from.
Log it or lose it
An RPE rating is worth little in the moment and a great deal in aggregate. "100 kg × 5" is an incomplete record; "100 kg × 5 @ RPE 7" tells future-you the set was easy and the weight should go up. The same weight at RPE 9.5 three weeks later flags fatigue before it becomes a stall. Logging RPE alongside weight and reps is what turns a training log from a diary into an instrument — it's the context that makes every other number in your progression interpretable.
Herculog puts an RPE field on every logged exercise, right next to weight and reps — one tap at the bar, no extra screens. And plans built in the plan builder carry their RPE targets into the app, so "3×8 @ RPE 8" shows up exactly where you need it: mid-set, deciding whether to add weight.
Program with RPE
Build a plan with RPE targets on every exercise in the free plan builder, then import it into Herculog and rate your sets as you train.
Herculog guides are general information for healthy adults, not medical advice. If you have a health condition or an injury — or pain that doesn't behave like normal soreness — talk to a medical professional before starting or changing a training program.