Judging Your RIR

Where the guess goes wrong, and how to check it.

How good is your guess?

Weight and reps are facts. RIR is a judgement you make in the last second of a set, and Effective Work, your estimated max and your next suggested set are all built on it. So it is worth knowing how good people are at that judgement. The research is specific about where it goes wrong.

People stop sooner than they think

The largest review of the question pooled twelve studies in which lifters predicted how many reps they could do and then did as many as they could. On average they had about one more rep in them than they predicted. The error ran one way: people believed they were closer to failure than they were.

How far out the guess is made matters most. In one study, trained men squatted a moderate weight to failure and called out the moment they believed they had five, three and one reps left. At five they were out by about five reps. At three, by more than three. Even at one they were out by about two.

How many reps out each call was, on average
  • Called five reps left5.2 reps
  • Called three reps left3.7 reps
  • Called one rep left2.1 reps

Zourdos et al. (2021)

Shorter, heavier sets are easier to read. The same review found predictions were more accurate in sets of twelve reps or fewer than in longer ones, and on a fairly heavy bench press, trained men and women called their RIR to within about a rep.

Experience does not rescue you. One study of gym members saw a tendency for experienced lifters to do better when they predicted before the set began. The pooled evidence, and the studies that asked during the set, found no such advantage. This is a skill you calibrate on purpose, not one that arrives with years.

Why it matters here

An RIR logged too low flatters every number. A set you call RIR 1 that really had four reps left is scored as hard work it was not, and your next suggestion is built on strength you have not shown. An honest RIR is what makes those numbers mean something.

How to calibrate

Watch the bar, not the burnDiscomfort arrives early in a set and failure arrives late, so how much it hurts is a poor gauge. Speed is a better one. When researchers put sensors on the bar, the speed of a rep with a given number of reps left in reserve was nearly the same from lifter to lifter, whatever their experience. You do not need a sensor to use the idea: a last rep that moved as fast as your first means you had plenty left.
Test yourself on a safe liftPick a machine or cable exercise. At the point where you would normally stop, say your RIR out loud, then keep going until you cannot complete another clean rep. The gap between your call and what happened is your error. It is how the studies measure it, and a few honest tests tighten most people up.
Count clean reps onlyReps in reserve are reps with the same form. If the next one would need a shorter range or a swing, it does not count.
Read long sets with suspicionThe error is biggest on long, light sets, where most of the set is spent far from failure. If a set of fifteen or twenty felt like RIR 2, assume you had more.

What Stimuli does with it

You log RIR after every set, in half-steps, because near failure half a rep is a real difference. If you think in RPE, change the Effort scale in Settings, under Workout defaults: Stimuli treats the two as the same information.

A logged set with the RIR picker open, offering reps in reserve from zero to five in half-steps

Stimuli takes the number you give it at face value. It cannot see the bar, which is why the check has to be yours.

For why the reps near failure matter so much in the first place, see the “Effective Work” and “Training to Failure” guides.

Sources

  1. Halperin I et al. (2022). Accuracy in predicting repetitions to task failure in resistance exercise: a scoping review and exploratory meta-analysis. Sports Medicine.
  2. Zourdos MC et al. (2021). Proximity to failure and total repetitions performed in a set influences accuracy of intraset repetitions in reserve-based rating of perceived exertion. Journal of Strength and Conditioning Research.
  3. Refalo MC et al. (2024). Accuracy of intraset repetitions-in-reserve predictions during the bench press exercise in resistance-trained male and female subjects. Journal of Strength and Conditioning Research.
  4. Steele J et al. (2017). Ability to predict repetitions to momentary failure is not perfectly accurate, though improves with resistance training experience. PeerJ.
  5. Morán-Navarro R et al. (2019). Movement velocity as a measure of level of effort during resistance exercise. Journal of Strength and Conditioning Research.

Keep reading

Train with science

Stimuli gives every muscle an Effective Work score as you log sets, and shows you where it landed.

Download on the App Store Get it on Google Play