Rate your sets, but stop pretending you can tell a 6 from a 7. You can judge effort well when you are close to failure and poorly when you are not, so use a scale with resolution only where your judgment actually has any.
What RPE means in a weight room
Outside the gym, RPE is a general "how hard was that" number. In lifting it means something specific: how many reps you had left. Zourdos and colleagues put the anchor in writing, and it's the version worth knowing. RPE 10 means zero reps left. RPE 9 means one. RPE 8 means two. Down it goes.
That reframe alone fixes most of the confusion. You are not rating pain, or how much you sweated, or whether you liked the set. You are answering one question: how many more could I have done?
The finding that should change how you rate
Hackett and colleagues took 81 lifters, had them call their remaining reps after a set, and then made them keep going to actual failure so the guess could be graded.
| How far from failure they really were | Average error |
|---|---|
| 0 to 5 reps | about 1 rep |
| 7 to 10 reps | more than 2 reps |
Two details matter more than the headline. First, accuracy collapses with distance from failure, which is intuitive once you see it: the last rep before failure announces itself, and the fifth-from-last does not. Second, training experience did not change the pattern. Ten years under a bar does not make you good at telling a 5 from a 6.
So a 10-point scale gives you fine gradations across a region where nobody, novice or veteran, can produce a trustworthy number. The bottom half of the scale is decoration.
Four words instead of ten numbers
Reps asks for one of four after a lift.
| Chip | Means | Sits at |
|---|---|---|
| Easy | Plenty left | 4 |
| Solid | A couple left | 7 |
| Grind | Maybe one left | 8 |
| Failed | Nothing left, or a rep missed | 10 |
Three of the four describe the near-failure end, where the research says your reading is worth something. One coarse bucket covers everything easier, because that whole region is a shrug and pretending otherwise would be fake precision.
The point is not that four is a magic number. The point is that a scale should not ask you for a resolution you cannot deliver.
What the rating is actually for
Effort is the brake on progression, not the accelerator. In Reps, an effort trend above 7 holds your weight where it is instead of adding to it, and a 9-or-above trend backs it off a step. Rate nothing and the app progresses you on reps alone, exactly like ordinary double progression.
Which means the honest way to use effort ratings is to log the sets that were hard. A "Grind" you record today is what stops the app from handing you five more pounds on a day you had nothing. Skipping the easy ones costs you very little.
Three ways lifters get this wrong
Rating the whole workout. Effort is per lift. A brutal squat session says nothing about how your curls felt.
Rating ambition instead of the set. If it moved fast and clean, it was easy, even if you wanted it to be heroic. The bar does not care about your intentions and neither should the number.
Rating after the fact. Log it while the set is still in your hands. By the third set of the next exercise you are remembering a story about the set, not the set.
If you take one thing
Ask "how many more could I have done" and answer honestly in coarse terms. Near failure, trust the answer. Far from it, do not overthink a number nobody in the literature can produce reliably either. The exact thresholds Reps applies to that answer are on how Reps decides your next set.
