Reps in reserve, and whether you need to train to failure
Stop one to three reps short of failure on most sets. Research finds this produces nearly the same muscle growth as grinding to failure, without the fatigue cost — and for strength you can stop even further short. Failure has a narrow use, not a general one.
What reps in reserve means
Reps in reserve, or RIR, is how many more repetitions you could have completed before form broke down. RIR 0 is genuine failure. RIR 2 means you stopped with two good reps left in the tank.
It is a way of prescribing effort rather than just sets and reps. "Three sets of eight" tells you nothing about how hard those sets were. "Three sets of eight at RIR 2" does, and effort is one of the main variables that determines whether a set counts as a hard set at all.
| RIR | What it feels like | Use it for |
|---|---|---|
| 0 | Could not complete another rep. The bar stops moving. | Rarely. Final set of an isolation exercise at most. |
| 1–2 | Grinding, form still solid, one or two left. | Most hypertrophy work. |
| 3–4 | Clearly hard but controlled. | Heavy compound lifts, and strength work. |
| 5+ | Comfortable. Bar speed barely changes. | Warm-ups, technique practice, deload weeks. |
What the research actually shows
The gap between training to failure and stopping short is smaller than gym culture assumes.
Meta-analyses of failure versus non-failure training find no requirement to reach failure for either strength or muscle growth. More recent work looking specifically at proximity to failure finds hypertrophy does increase slightly as you get closer to it — but the effect is modest, in the region of a small effect size rather than a transformative one. A 2024 trial comparing failure against one to two reps in reserve over eight weeks in trained adults found similar quadriceps growth in both groups.
For strength the picture is clearer still: proximity to failure matters less, and stopping several reps short lets you accumulate quality practice at heavy loads without accumulating the fatigue that degrades the next session.
The trade-off in one sentence. Training to failure buys you a small amount of extra stimulus per set at a large cost in fatigue — and that fatigue reduces the quality of every set and session that follows.
Why failure costs more than it looks
- It wrecks the following sets. Take set one to failure and sets two and three will be markedly worse. You often end the exercise with less total quality volume than if you had stopped short on all three.
- It lengthens recovery. Failure training produces more muscle damage and central fatigue, which is a bill paid over the following days.
- Form degrades exactly when it matters. The last rep before failure on a squat or deadlift is the rep where position collapses.
- It is hard to be consistent with. Sessions built on maximum effort are the first ones you skip when tired.
You are probably worse at judging RIR than you think
Untrained lifters systematically underestimate how many reps they have left — often by three or more. If you think you have one rep left, you may well have four. This improves with experience, and there are a few ways to calibrate:
- Take one set to genuine failure occasionally — an isolation exercise, not a barbell squat — purely to learn what the last rep feels like.
- Watch bar speed. When reps visibly slow despite maximum intent, you are within one or two of failure. This is the most reliable cue available to you.
- Film a set. The gap between what a set feels like and what it looks like is instructive.
- Compare against your log. If you hit 10 reps at a weight where you predicted 8 was your limit, your estimation is running conservative.
When failure is worth it
There is a legitimate place for it, and it is narrower than most people train as though it is.
- Isolation work with a low fatigue cost: lateral raises, curls, leg curls, cable work. The last set to failure here is cheap.
- The final set of an exercise, where there is nothing left to degrade.
- Machines and cables, where failing is safe and you can stop mid-rep without risk.
- Low-load training — very high rep sets need to get close to failure to recruit anything worth recruiting.
Where it is a bad idea
- Heavy barbell squats and deadlifts, where the failing rep is the dangerous one.
- The first of several sets on the same exercise.
- Every set of every exercise, which is the most common error.
- Any session where you are already under-recovered.
The practical version
- Most sets: stop with one to three good reps left.
- Heavy compounds: stop three to four short. Strength comes from quality practice, not grinding.
- Last set of an isolation exercise: take it to failure if you want to.
- Deload weeks: RIR 4–5 throughout.
- Calibrate occasionally with one true failure set on something safe.
- If your logged weights are climbing, your effort is sufficient. That is the only test that matters.
Key takeaways
- One to three reps in reserve produces nearly all the growth without the fatigue cost.
- Meta-analyses find failure is not required for strength or hypertrophy.
- Failure degrades your following sets, so total quality volume often falls.
- Most people underestimate their remaining reps, sometimes by three or more.
- Save failure for the last set of low-fatigue isolation work.
- Never take heavy barbell squats or deadlifts to failure.
Related reading
General information, not medical or individualised advice. Speak to a doctor before starting a new programme, especially if you have a medical condition, are pregnant, or are returning from injury. If you train with a coach, their guidance takes precedence.
Track it, or none of this applies
MyFitsy pre-fills the weight you used last time, so progression is a decision instead of a guess.
