Get the app (opens the App Store in a new tab)

Guides

RIR vs RPE: the same scale, counted two ways

RIR counts the reps you had left; lifting RPE rates effort from 1 to 10. RPE 10 = 0 RIR, RPE 8 = 2 RIR. Conversion table, %1RM chart and why we log RIR.

Key takeaways

  1. 01RIR = reps you could still have done. RPE (lifting) = 10 − RIR.
  2. 02RPE 8 = 2 RIR. RPE 9 = 1 RIR. RPE 10 = 0 RIR, nothing left.
  3. 03Lifters misjudge by about 1 rep on average, less when close to failure.
  4. 04For most working sets, stop at 1–3 RIR.
On this page

RIR (reps in reserve) is how many more clean reps you could have done when you stopped a set; RPE (rating of perceived exertion) rates the same set from 1 to 10. On the lifting RPE scale the two are the same thing counted differently: RPE = 10 − RIR, so RPE 8 is 2 reps left and RPE 10 is nothing left.

The difference is in the question you ask yourself between sets. RPE asks “how hard was that, out of 10?” RIR asks “how many more could I have done?” A count is easier to answer honestly when you’re out of breath. That’s why LiftLoggr logs RIR.

The conversion table

The lifting RPE scale was anchored to reps in reserve by Zourdos and colleagues in 2016: RPE 10 = 0 RIR, RPE 9 = 1 RIR, and so on (Zourdos et al. 2016).

RPE RIR What the set felt like Log in LiftLoggr as
10 0 Couldn’t do another rep 0
9.5 0–1 Maybe one more, maybe not 0
9 1 One more, definitely 1
8.5 1–2 One more for sure, maybe two 1
8 2 Two more 2
7.5 2–3 Two for sure, maybe three 2
7 3 Three more 3
6 or below 4+ Warm-up speed 4+

For half points, log the lower number. It keeps the record honest: if you weren’t sure the second rep was there, it probably wasn’t.

RIR vs RPE side by side

RIR RPE (lifting)
Question How many reps were left? How hard, 1–10?
Unit Reps Points
Typical working range 0–4 6–10
Half steps No (0–1, 1–2 as ranges) Yes (8.5, 9.5)
Easiest for Beginners, logging mid-session Coaches writing programs, singles and triples
Confused with — The Borg cardio scales

The cardio confusion is real. Searches for “RPE chart” return running and Borg-scale charts that rate whole-body effort. They don’t convert to reps in reserve. If a chart doesn’t mention reps, it isn’t the lifting scale.

Reps × RIR → % of 1RM

If you know the reps and the RIR, you know roughly which rep max you were working at: 8 reps with 2 left is about a 10-rep max. The table below uses the Brzycki formula, the same one LiftLoggr uses for estimated 1RM, on reps + RIR. Rows stop at 12 total reps, where the estimate starts to drift.

Reps done 0 RIR 1 RIR 2 RIR 3 RIR
1 100 % 97.2 % 94.4 % 91.7 %
3 94.4 % 91.7 % 88.9 % 86.1 %
5 88.9 % 86.1 % 83.3 % 80.6 %
8 80.6 % 77.8 % 75.0 % 72.2 %
10 75.0 % 72.2 % 69.4 % —

Read it both ways. A program says “5 reps at RPE 8” (2 RIR): load about 83 % of your 1RM. Or you did 100 kg × 5 with 2 left: your estimated 1RM is about 120 kg. The interactive version with your own numbers is the RIR chart; for the plain estimate, use the 1RM calculator.

How accurate is RIR?

Good enough, not exact. A 2022 meta-analysis of 12 studies and 414 participants found lifters underestimated their remaining reps by about 1 rep on average, with large differences between studies (Halperin et al. 2022). Three things made predictions better:

  • Closer to failure. Guessing 1 rep left is more accurate than guessing 5.
  • Heavier sets. Sets of 12 reps or fewer were judged better than long sets.
  • Later sets. You calibrate as the session goes on.

Training age didn’t help much. Experienced lifters were no better at guessing than newer ones.

What that means for logging:

  1. Cap the scale at 4+. Beyond 4 reps left, nobody’s guess is worth recording. LiftLoggr’s chips stop at 4+ for that reason.
  2. Calibrate now and then. On a safe exercise (a machine, a cable, a dumbbell curl), take the last set to failure and compare with what you would have logged.
  3. Log right after the set. Two minutes later, you’re guessing a guess.

A two-week calibration test

Pick two safe exercises, for example the leg press and the cable row. On the last set of each, stop when you think you have 2 reps left, log that number, then rest 20 seconds, pick the weight back up and do reps to failure with clean form. The reps you get are your real RIR (roughly; the short rest gives back a little).

Session Exercise Logged RIR Reps after the pause Gap
1 Leg press 2 4 +2
1 Cable row 2 3 +1
2 Leg press 2 3 +1
2 Cable row 2 2 0
3 Leg press 2 2 0

The example shows the usual pattern: most people underrate what they have left at first and get within a rep after a few tries. Repeat the test every couple of months, or whenever the logged RIR stops matching how sets feel. Don’t run it on squats, bench presses or deadlifts without a spotter or safety arms.

Which RIR to train at

Two findings frame the answer. In a 2024 series of meta-regressions, strength gains were similar across a wide range of RIR, while muscle growth improved as sets ended closer to failure (Robinson et al. 2024). And a 2022 scoping review found training very close to failure raises fatigue and discomfort more than it adds results (Refalo et al. 2022).

Set Target RIR Why
Heavy squat, bench, deadlift (1–6 reps) 2–3 Strength doesn’t need failure; form breaks first
Compound hypertrophy work (6–12 reps) 1–3 Close enough to grow, far enough to recover
Isolation and machines 0–2 Low risk; failure is easy to judge here
Deload week 4+ Recovery, not stimulus: see deload week

RPE and RIR loading work. In an 8-week study, lifters who picked loads to hit RPE targets gained as much as a group on fixed percentages, with a small strength edge in most individuals (Helms et al. 2018). The scale adjusts to good and bad days; percentages don’t.

Mistakes

  1. Logging the RPE you hoped for. Rate what happened.
  2. Treating 0 RIR as the goal. It’s a tool for the last set, not every set.
  3. Mixing scales. Pick RIR or RPE for your log and convert the program to it once.
  4. Logging RIR on warm-ups. Warm-ups are far from failure by design. LiftLoggr doesn’t ask.
  5. Ignoring a pattern. RIR 0 on every set of a lift for two weeks is a stall in progress; RIR 4+ on every set means the weight is too light.

How LiftLoggr uses RIR

RIR is optional and off by default. Turn it on in Settings → Logging → Log reps in reserve. After you check a working set, tap how many reps you had left: 0, 1, 2, 3 or 4+. The set shows a small badge such as “RIR 2”. Warm-up sets never ask.

Once you log it, the double progression rules use it:

  • Every working set at the top of the range with RIR 3 or more last time: the next session pre-fills two weight steps instead of one.
  • A first set at the top with RIR 3 or more: a chip offers the next step for the remaining sets in the same session.
  • A tough session isn’t automatically a stall. If your best set at that weight ended with RIR 2 or more, the session doesn’t count toward the three stalled sessions that trigger a deload suggestion.

LiftLoggr doesn’t log RPE. If your coach writes RPE, convert with the table above: RIR = 10 − RPE.

Questions

Is RPE 8 the same as 2 RIR?

Yes, on the lifting RPE scale built on reps in reserve. RPE 8 means you stopped with 2 reps left. The conversion is RPE = 10 − RIR.

What RIR should I train at?

1–3 RIR for most working sets. Strength gains look similar across a wide range of RIR, while muscle growth improves as sets end closer to failure. Save 0 RIR for the last set of safe exercises such as machines and isolation work.

Is RIR accurate?

Roughly. Across 12 studies, lifters underestimated how many reps they had left by about 1 rep on average, with wide variation. Predictions get better close to failure, in sets of 12 reps or fewer, and in later sets.

Is RPE in lifting the same as the Borg scale?

No. The Borg scales used in cardio rate whole-body effort (6–20, or 0–10). The lifting RPE scale rates one set by the reps left in it. A cardio RPE chart doesn't convert to RIR.

Does LiftLoggr log RPE?

No. LiftLoggr logs reps in reserve, 0 to 4+, after a working set. If your program is written in RPE, convert it with RIR = 10 − RPE.

Sources

  1. Zourdos et al. 2016, RPE scale measuring repetitions in reserve (J Strength Cond Res) pubmed.ncbi.nlm.nih.gov
  2. Helms et al. 2018, RPE vs %1RM loading in periodized programs (Front Physiol) pubmed.ncbi.nlm.nih.gov
  3. Halperin et al. 2022, Accuracy in predicting repetitions to task failure (Sports Med) pubmed.ncbi.nlm.nih.gov
  4. Robinson et al. 2024, Proximity to failure, strength and hypertrophy: meta-regressions (Sports Med) pubmed.ncbi.nlm.nih.gov
  5. Refalo et al. 2022, Proximity-to-failure in resistance training: scoping review (J Sports Sci) pubmed.ncbi.nlm.nih.gov