How Close to Failure Should You Train for Muscle Growth?

Train most working sets 0-3 reps from failure for muscle growth; keep heavy strength work 2-4 reps short, because load, not proximity, drives strength. The 2023-24 reviews agree on the range. The hard part is landing in it: most lifters misjudge their reps in reserve by a rep or more. Riven is the Apple Watch app that scores muscle failure.

I have watched lifters swear a set was "2 RIR" and then grind out 6 more reps when asked to keep going. That gap is the whole problem, and it is the reason this post spends as much time on measuring proximity to failure as on prescribing it.

Reps in reserve (RIR) in lifting is the number of additional reps you could have completed before reaching momentary muscle failure during a set (what reps in reserve really means, if you want the full explainer). 2 RIR means 2 reps left in the tank. Failure itself is the point where you physically cannot complete another rep with good form.

How close to failure should you train?

For hypertrophy, train most sets to 0-3 RIR, with the bulk of your serious working sets landing around 1-2 RIR. The dose-response research is fairly consistent: muscle growth is roughly maintained anywhere from 0 up to about 4-5 RIR when total volume is equated, then drops off once you're stopping sets more than 4-5 reps short of failure. A 2024 series of meta-regressions by Robinson and colleagues in Sports Medicine found that hypertrophy improves as sets get closer to failure — the slope pointed clearly in that direction, with a confidence interval excluding zero.

But "closer is better" is not "failure is mandatory." A 2022/2023 meta-analysis by Refalo et al. pooling 15 studies found only small, mostly non-significant advantages for training right to failure: momentary failure versus stopping short gave an effect size of just 0.12 (p = 0.343). Small. So you don't need to grind every set into the ground. You need to land in the productive zone — and reliably.

How close to failure for hypertrophy vs strength?

This is the most useful, least-discussed distinction in the whole conversation: proximity to failure matters for size, but not much for strength. For hypertrophy, train near failure (0-3 RIR). For strength, load matters more than proximity — you can train your heavy 3-5RM work at 2-4 RIR and gain just as well, because strength is driven by lifting heavy and practicing the movement, not by chasing the burn.

The same Robinson meta-regression that showed a hypertrophy benefit from getting close to failure found the opposite for strength — the slope's confidence interval contained zero. Their conclusion, word for word:

"Strength gains were similar across a wide range of RIR, while muscle hypertrophy improves as sets are terminated closer to failure." — Robinson et al., Sports Medicine, 2024 (PubMed 38970765)

That one sentence is the whole strength-vs-size split. Ruple et al. (2023) backs this up directly: trained adults running 0-1 RIR versus 4-6 RIR for 6 weeks made similar squat, bench, and deadlift 1RM gains and similar quad growth — though the near-failure group showed distinct motor-unit firing-rate adaptations. So "same hypertrophy" didn't mean "identical adaptations," a nuance most blogs flatten.

What this means in a real program: you can run heavy compound strength work several reps shy of failure to protect your technique and your nervous system, then push your accessory hypertrophy work much closer — 0-2 RIR on a cable curl or leg extension where the risk is low. Different goals, different RIR targets, same session.

RIR target by goal

The reviews at a glance, if you want to check the sources yourself:

• Refalo et al. (2023), Sports Medicine — 15-study meta-analysis: failure vs stopping short, effect size 0.19; momentary failure vs non-failure 0.12 (not significant).
• Robinson et al. (2024), Sports Medicine — meta-regressions: hypertrophy improves as sets end closer to failure; strength flat across a wide RIR range.
• Ruple et al. (2023), Physiological Reports — 6-week RCT: 0-1 RIR vs 4-6 RIR, similar 1RM gains and similar quad size.

I've put every review on training to failure, in one table — Grgic, Vieira, both Refalo papers, Ruple, Robinson — with the headline number from each abstract.

What are 'effective reps' and why proximity to failure matters

"Effective reps" (also called stimulating reps) is the idea that not every rep in a set contributes equally to growth — only the hard reps near the end, where you recruit your high-threshold motor units and the rep slows down, produce high muscle-fiber tension. That tension is the growth signal. Chris Beardsley's model puts full motor-unit recruitment at roughly 90% of max force — around a 5RM load — which is where the "last ~5 reps are the stimulating ones" heuristic comes from.

This is why proximity to failure matters at all (effective reps, explained goes deeper on the model and its critics). If you stop a 15-rep set at 7 RIR, you may have skipped most of your stimulating reps entirely. The early, easy reps were just fatigue accumulation getting you to the part that counts.

Now, be skeptical of the literal version. Greg Nuckols at Stronger By Science has argued the direct evidence for a fixed "exactly the last 5 reps" rule is weak — proximity and volume are entangled, and "5" is a recruitment-threshold estimate, not a law. I agree. Treat effective reps as a mental model for why ending sets too early bleeds stimulus, not as a rep-counting prescription. The takeaway survives either way: the back end of a hard set is where the growth lives.

Can you train too far from failure (junk volume)?

Yes. Sets stopped too far from failure become junk volume — they add fatigue and time without delivering the high-tension stimulating reps that drive growth. The practical cutoff from the dose-response work sits around 4-5 RIR: stay inside that and growth holds up (volume equated); drift past it — sets averaging more than ~5 reps short — and hypertrophy meaningfully drops off.

Here's the trap that catches even good lifters. Because reps-in-reserve is so often underestimated (more on that next), a lifter who thinks they're training at "3 RIR" may actually be at 6-7 RIR — squarely in junk-volume territory — while believing they're working hard. You can do 20 sets a week, the top of the 10-20 hard sets per muscle per week that the volume guidelines recommend, and under-stimulate every one of them if each set quietly stops 6 reps early. That's the worst-case scenario: maximum fatigue, minimum return. The fix isn't always more volume. Often it's pushing the intensity of effort on the sets you already do — and counting only the sets that actually got close as your real weekly volume.

The 1-rep error most lifters make

The productive zone is narrow, and the gauge most people use to find it is off. RIR self-estimation is systematically biased — and the bias runs in the dangerous direction. Steele et al. (2017), across 141 participants, found lifters underpredict their reps to failure: the least experienced were off by about 4-5 reps, and even lifters with over 3 years of training underpredicted by 1-2, with a standard error of measurement around 2.6-3.4 reps. So when someone says "2 RIR," they frequently had 4-6 reps left.

Now do the arithmetic against the table above. The hypertrophy target is 0-3 RIR and the sweet spot is 1-2. An experienced lifter who aims for 2 RIR and misses by a single rep is sitting at 3 — the outer edge of the zone. Miss by 2 and the set is at 4 RIR, past the point where the dose-response data says growth starts to fall away. A novice who misses by the average 4-5 reps is at 6-7 RIR, in junk volume, while believing the set was hard. That is what "the 1-rep error" means in practice: the error is small in reps and large in outcome, because the zone it has to land in is only 3 reps wide.

It is also invisible from the inside. Nothing about a set that stops at 3 RIR tells you it wasn't 2; the rep felt heavy, the muscle burned, and you racked it. That is not a rounding error. That is the difference between the productive zone and junk volume, hidden inside a set that felt like work. Why your reps-in-reserve estimate is wrong goes through the studies one by one if you want the full picture.

How to measure your proximity to failure

The honest answer: most people measure it badly, for the reason above. There are 3 ways to do better than gut feel (the long version is how to measure reps in reserve):

1. RPE/RIR logging with honest cross-checks. Periodically take a set to true failure and compare it to your estimate. It recalibrates your internal gauge — Steele's data shows experience helps, but only if you actually test yourself (how to calibrate your reps in reserve is the step-by-step).
2. Velocity loss. This is the validated objective proxy. As a muscle fatigues, reps physically slow down. Refalo's analysis used velocity-loss thresholds (high >25% vs moderate 20-25%) as a stand-in for proximity to failure, and growth trended upward with higher velocity loss. A linear position transducer measures this on the bar — accurately, and for about $300. The velocity-loss stop thresholds are worth knowing whichever device reads them.
3. Wrist-based scoring. Riven is the Apple Watch app that scores muscle failure. It reads your wrist motion and stays quiet during the set; about three seconds after you rack the weight you get the exercise, the rep count and a 0-100 score. Reps slow down as a muscle nears failure, and that is what the score is built on. No barbell clip, no camera, no extra hardware. It shows you which set was the real one.

Whichever you pick, the point is the same: replace a guess that is wrong by a rep or more with a read that isn't. Velocity loss is the construct Refalo's review used as its stand-in for proximity to failure, so reading how much your reps slow down answers the question an RIR estimate is trying to answer — without the estimate.

The bottom line

Train most hypertrophy sets at 1-2 RIR (0-3 is the workable range). Keep heavy strength work 2-4 RIR — load does that job. Avoid the two failure modes: undershooting (junk volume from stopping too early, usually because you underestimated your reps left) and overshooting (grinding every compound to failure and torching your recovery and weekly volume). The single biggest upgrade for most lifters isn't a new program — it's getting an honest read on how close to failure they actually train. Pick one of the 3 methods above and run it for a week before you change anything else.

FAQ

How close to failure should I train for muscle growth?

Take most working sets to 0-3 RIR, with serious sets landing around 1-2 RIR. Growth holds up across roughly 0-5 RIR when volume is equated, but drops off once you're stopping more than 4-5 reps short of failure.

Is training to failure necessary to build muscle?

No. Controlled comparisons show 1-2 RIR produces hypertrophy equivalent to failure when volume is matched, while failure adds fatigue cost. Save true failure for low-risk isolation work and final sets.

How close to failure should I train for strength vs hypertrophy?

For hypertrophy, get close — 0-3 RIR. For strength, proximity matters far less; you can train heavy 3-5RM work at 2-4 RIR because load and practice drive strength gains, not how near failure you get.

Can you train too far from failure?

Yes. Sets stopped more than ~4-5 reps short become junk volume — fatigue and time without the high-tension stimulating reps that drive growth. This often happens unknowingly because lifters underestimate their reps in reserve.

Why are my reps-in-reserve estimates probably wrong?

Research shows lifters chronically underestimate reps to failure — novices by 4-5 reps, even experienced lifters by 1-2. A self-reported "2 RIR" is often really 4-6, which silently under-stimulates growth.

Can an Apple Watch tell how close to failure I am?

Reps slow down as a muscle nears failure, and that slowdown can be read from the wrist. Riven is the Apple Watch app that scores muscle failure; it scores each set the moment you rack it.

Is there a review on training close to failure (1-3 RIR) and hypertrophy?

Yes. Refalo et al. (2023, Sports Medicine) pooled 15 studies and found only a trivial edge for going all the way to failure; Robinson et al. (2024, Sports Medicine) ran meta-regressions showing hypertrophy improves as sets end closer to failure while strength does not; and the Ruple (2023) and Refalo (2024) trials found 0-2 RIR matched failure for growth. Practically, 1-3 RIR captures the benefit.

Sources

• Robinson, Pelland, Remmert, Refalo, Jukic, Steele & Zourdos (2024), Exploring the Dose-Response Relationship Between Estimated Resistance Training Proximity to Failure, Strength Gain, and Muscle Hypertrophy: A Series of Meta-Regressions, Sports Medicine 54(9):2209-2231 — https://pubmed.ncbi.nlm.nih.gov/38970765/
• Refalo et al. (2023), Influence of Resistance Training Proximity-to-Failure on Skeletal Muscle Hypertrophy: A Systematic Review with Meta-analysis, Sports Medicine — https://pmc.ncbi.nlm.nih.gov/articles/PMC9935748/
• Ruple et al. (2023), The effects of resistance training to near failure on strength, hypertrophy, and motor unit adaptations in previously trained adults — https://pmc.ncbi.nlm.nih.gov/articles/PMC10161210/
• Steele et al. (2017), Ability to predict repetitions to momentary failure is not perfectly accurate, though improves with resistance training experience — https://pmc.ncbi.nlm.nih.gov/articles/PMC5712461/
• Chris Beardsley, How many stimulating reps are there in each set to failure? — https://sandcresearch.medium.com/how-many-stimulating-reps-are-there-in-each-set-to-failure-9d179f594dd
• Greg Nuckols / Stronger By Science, The Evidence is Lacking for Effective Reps — https://www.strongerbyscience.com/effective-reps/