RPE and RIR: measuring effort you can log

The two scales in one paragraph

RIR (reps in reserve) answers "how many more reps could I have done with honest technique?" A set stopped two reps before failure is 2 RIR. RPE (rating of perceived exertion) in modern strength training is the same idea flipped onto a 10-point scale: RPE 10 is 0 RIR, RPE 8 is 2 RIR, and so on. Zourdos and colleagues formalized this resistance-training-specific scale in 2016 and showed it tracks bar speed, which is the physical signature of proximity to failure. The two scales are one measurement wearing two notations.

Why proximity to failure is worth logging at all

Because it is the difference between a hard set and a warm-up that happened to take a while. The same 8 reps at 135 lb can be a 4-RIR placeholder or a 0-RIR grinder, and every downstream decision — whether theprogression gate was honestly passed, whether a set counts as a hard set, whether fatigue is accumulating — reads differently depending on which it was. A set log without effort context loses the most decision-relevant bit of the whole record.

The failure literature adds the practical target: taking most work close to failure matters for growth, but grinding every set to absolute failure adds fatigue faster than it adds adaptation. Grgic's meta-analysis found similar hypertrophy whether sets ended at failure or shy of it, and the ACSM consensus prescribes effort in exactly these terms. That is why "0–3 reps in reserve on most working sets" has become the practical consensus zone.

Can lifters actually estimate it?

Imperfectly, and it improves with exposure. The validation research shows experienced lifters correlate strongly with measured bar velocity when rating RIR, while novices run looser. Prediction accuracy also degrades the further a set stops from failure: calling 1 RIR is much more accurate than calling 5 RIR, because the last reps before failure feel unmistakably different.

Three practical consequences fall out of the evidence:

• Log RIR at low numbers. Ratings of 0–3 are meaningful; "maybe 6 left" is a mood, not a measurement.
• Calibrate occasionally. Taking a safe accessory movement to technical failure now and then re-anchors what 0 feels like.
• Expect novice inflation. New lifters usually have more reps in reserve than they think. Their "RPE 9" push-up set often has four honest reps left, which is one reason early progress feels magical when effort tightens up.

How Flexbound uses RIR

Every set in a Flexbound workout plan ships with an RIR prescription, not just a rep count — typically 1–3 RIR on main lifts and accessories, tighter or looser by goal and experience. The prescriptions come from versioned methodology rules citing the failure and consensus literature above, and they are deterministic: the app never lets a language model decide how hard you should train. When you log effort, it becomes part of your confirmed history, which the progression engine reads when deciding whether a rep-target pass was honest enough to earn a load increase.

Exercise swaps respect effort too: substituting a movement keeps the prescribed sets, reps, rest, and RIR unchanged, because the effort target is the part of the prescription that carries the stimulus.

Limits, stated plainly

Self-rated effort is a skill with error bars, not a lab instrument. It drifts with sleep, caffeine, and ego. It is least reliable in high-rep sets, where discomfort arrives before true muscular failure and most people under-count what is left. And no scale fixes dishonest logging: a diary of flat 2s recorded out of habit carries less information than an occasional honest "no idea, maybe 4". The scale earns its place by being logged plainly and calibrated occasionally, which is exactly the kind of record a written journal is good at keeping.