RPE / RIR Autoregulation

Autoregulation lets the trainee's performance on the day set the load, rather than a pre-written weight. The two linked tools for resistance training:

• RIR (Reps in Reserve) — how many more reps you could have done before failure. 0 RIR = failure, 2 RIR = two left in the tank, etc.
• RPE (Rating of Perceived Exertion, resistance version) — a 1–10 scale where RPE 10 = maximal (0 RIR), RPE 9 = 1 RIR, RPE 8 = 2 RIR, and so on (RPE = 10 − RIR). This is distinct from the Borg cardio RPE scales (see RPE and the Borg Scales (6–20 and CR10)).

How it programs:

• Target a stopping point, e.g., "work up to a set of 5 at RPE 8 (2 RIR), then do back-off sets." On a strong day the load that hits RPE 8 is higher; on a tired day it is lower — the system self-corrects.
• Cap effort to manage fatigue — most hypertrophy/strength work lives around RPE 7–9 (1–3 RIR); pushing to RPE 10 (failure) has no meaningful strength benefit and only a trivial hypertrophy edge, while costing roughly 2–3x the acute fatigue of stopping 1–3 reps short — a real, externally-verified but disproportionate trade (see verification_note and the 2026 update below).
• Autoregulate volume too — add or cut sets based on whether reps/RPE are holding, which connects directly to the MEV/MRV idea of training within a recoverable band.

Autoregulation embodies individuality and is the natural fix for the rigidity of percentage-based programming (a fixed % is too heavy on a bad day, too light on a good one). Its cost is that RIR is an estimate. A meta-analysis of 12 studies (414 people) found lifters under-predicted how many reps they had left by about 1 rep on average. Accuracy did not depend on training experience, but it was better on sets of 12 reps or fewer, closer to failure, and in later sets (Halperin et al. 2022). Many beginners therefore start with simpler double progression and add RIR targets once they've seen where their failure point actually is.

• Intermediate and advanced lifters whose strength varies day to day with sleep, stress, travel or diet.
• Anyone combining lifting with other demanding training (sport, running), where fixed loads are often wrong.
• Beginners, simply: use RIR as a stopping rule ("end the set with about 2 good reps left") rather than as a load-setting system.

• It's a skill-based estimate. Expect to be off by about a rep, and more on high-rep sets. Occasionally calibrate with a safe set to failure on a machine or isolation exercise.
• Not for maximal testing, and not a licence to grind to failure on heavy free-weight compounds without safeties.
• Medical exertion limits come first. If a clinician has set heart-rate or exertion limits (cardiac rehab, pregnancy, hypertension), follow those.

Back squat, plan: "1 top set of 5 @ RPE 8, then 3 × 5 at 90% of that load" (demo: Back Squat).

A fixed "3 × 5 @ 115 kg" would have been too light on the good day and close to failure on the bad one. RPE 8 kept the effort the same both days, which is the stimulus that matters.

Calibration drill: on your last set of leg press or a machine row, aim for "2 left", then keep going to failure. If you got 4–5 more, your "2" was really 4–5. Adjust your sense of effort next time.

Next lesson: Deload Weeks. When RPE creeps up at the same loads week after week, that's one of the clearest signs a deload is due.

• RIR = reps left before failure; 0 RIR = failure.
• RPE (resistance) is 1–10; RPE = 10 − RIR (RPE 8 ≈ 2 RIR).
• Target a stopping RPE/RIR so the load self-adjusts to daily readiness.
• Most work sits at RPE 7–9 (1–3 RIR); training to true failure (RPE 10) gives no meaningful strength benefit and only a trivial hypertrophy edge, at roughly 2–3x the acute fatigue cost of stopping short.
• RIR is an estimate: people under-predict remaining reps by ~1 on average (any training status); accuracy is best on ≤12-rep sets and near failure.