RPE and reps in reserve (RIR) let you pick the load based on how you are today, not on a percentage written down a month ago. RIR 2 means you stop when you could still do two more reps; RPE 8 says the same thing from the other direction. Here is the conversion table, how to use it to actually progress, and the three mistakes that make the system useless.
The scale, plainly
| RPE | RIR (reps left) | What it feels like |
|---|---|---|
| 10 | 0 | Failure: the last rep was maximal |
| 9.5 | 0-1 | Maybe one more, not sure |
| 9 | 1 | One clean rep left |
| 8 | 2 | Two reps of margin |
| 7 | 3 | Three left, bar speed still good |
| 6 | 4+ | Technique work or warm-up |
The reps-in-reserve scale has been validated by comparing lifters' estimates with what they could actually do: estimated and actual repetitions were strongly related in both the bench press (r = 0.95) and the squat (r = 0.93), with a clear inverse relationship between bar velocity and RPE (Zourdos et al., 2016). Read that carefully: the scale works on average, and it works best close to failure. At 4-5 reps in reserve, most people underestimate how much they have left.
Why a fixed percentage is not enough
Your max is not a stable number. It moves with sleep, stress, food and what you trained two days ago. Prescribing "4×5 at 80%" gives the same load to a good day and a terrible one.
With RIR the load adapts: "4×5 at 2 RIR" means 5 kg more on a good day, 5 kg less on a bad one, and in both cases the stimulus is the one you planned.
How to use it in practice
1. Learn to estimate. For two weeks, write down before each set how many reps you think you will have left. Occasionally, on a safe exercise (machine or isolation), go to failure and check the gap. Most people overestimate their margin at first.
2. Set the working range. Most sets: 1-3 RIR. Heavy compounds: 2-3 RIR. Isolation: 0-2 RIR.
3. Progress with the double criterion. Keep the same sets and load until the reported RIR rises by one point. Example: 4×5 at 100 kg reported at 1 RIR in week one; when the same 4×5 at 100 kg comes out at 2 RIR, add load.
4. Set a ceiling. If a set planned at 2 RIR comes out at 0, the session is asking for more than planned: drop a set, or take 5% off the remaining sets.
A four-week example (bench press)
| Week | Sets | Target RIR | Indicative load | Rule |
|---|---|---|---|---|
| 1 | 4×5 | 3 | 92.5 kg | Finish with clear margin |
| 2 | 4×5 | 2 | 95 kg | +2.5 kg if week 1 was a true 3 |
| 3 | 4×5 | 1 | 97.5 kg | If set three drops to 0, stop there |
| 4 | 3×5 | 3-4 | 85 kg | Lighter week |
The loads are indicative. The column that matters is RIR.
Three mistakes that break the system
1. Reporting the RIR you wish you had. If you log 2 RIR on a set that was actually failure, next week's plan is built on a lie. The data is for you, not for your coach's approval.
2. Using it only on heavy sets. Estimation is harder at high reps, but that is exactly where junk volume hides: a set of 15 stopped at 6 RIR trains almost nothing.
3. Changing exercises every week. Estimation improves with familiarity. Constant rotation means you never build the reference.
RPE for coaches
With clients, RIR has one huge practical advantage: it makes effort communicable without max testing, which most people should not be doing regularly.
Three details:
- Teach it with one calibration set: a single set to failure on a safe exercise, once, to give a concrete reference.
- Ask right after the set, not at the end of the session: the memory of effort fades fast.
- Compare reported with observed: if an athlete reports 2 RIR and then drops three reps on the next set, the estimate was optimistic.
In Athleex effort is logged alongside the reps of each set, so the history sits next to the load: when average RIR rises at the same load, progression is due; when it falls, fatigue is building and a deload week is the sensible next step.
For the first load on an exercise you have never done you need a procedure, not a guess: how much weight to use on a new exercise.
FAQ
Are RPE and RIR the same thing? Two ways of saying the same thing: RPE 10 minus RIR. RPE 8 = 2 reps in reserve. Use whichever feels natural, but do not mix them in the same program.
Does RIR work for beginners? With less precision: inexperienced lifters usually stop well short of true failure. In the first months, pair it with a fixed rep range and check occasionally with a controlled set to failure.
Do I still need to test my max? No, and that is one reason the system is practical. Test occasionally to verify, not to program.
How does it combine with percentages? Well: percentage as the starting point, RIR as the correction for the day. "4×5 at 80%, stop if you reach 0 RIR" is a complete prescription.
Will a bad estimate ruin my program? No, as long as you log what actually happened. The system self-corrects within two sessions precisely because the data comes from the field.
Sources
- 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;30(1):267-275. PubMed 26049792 — group-level correlations, not a guarantee of individual accuracy.
- 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;38(4):42-49. PMC4961270 — applied article, not an experiment.
- Rogerson D, et al. Deloading Practices in Strength and Physique Sports. Sports Medicine – Open, 2024;10:26. doi:10.1186/s40798-024-00691-y.
Educational content. It does not replace individual coaching or medical advice.