Training to failure isn't necessary for muscle growth — stopping 1–4 reps short works just as well

The bottom line up front

You don't need to push every set to the point where the weight stops moving.

Stopping 1–4 reps before failure builds the same amount of muscle as grinding to absolute zero. Four independent meta-analyses now agree on this. The grind-to-failure approach isn't wrong — it just isn't required.

So why does almost every gym-goer think it is? Because effort feels productive. And it is — up to a point. The research tells us exactly where that point is.

Stopping 1–4 reps short builds the same muscle as training to absolute failure.

— Grgic et al. (2022). Effects of resistance training performed to repetition failure or non-failure on muscular strength and hypertrophy. J Sport Health Sci.

What 'training to failure' actually means

Before the numbers, a quick definition. Momentary muscular failure means you physically cannot complete another rep with good form — the bar stops moving. This is different from stopping because it feels hard, or because you're breathing heavy.

Researchers also use 'set failure' more broadly: any stopping point that isn't momentary failure — like stopping because your technique breaks down, or hitting a target rep count that leaves a rep or two in the tank.

The distinction matters because most of the studies compare these two approaches directly. And the results are consistent across all of them.

The size numbers: no meaningful difference

A 2022 systematic review and meta-analysis pooling 15 studies found no statistically significant difference in muscle hypertrophy between training to failure and stopping short — effect size 0.22, which is small and non-significant (Grgic et al., 2022).

A separate 2023 meta-analysis in Sports Medicine confirmed this finding. Training to momentary muscular failure versus non-failure showed an effect size of just 0.12 for hypertrophy — not statistically significant (Refalo et al., 2023). Even when the researchers broadened the definition of 'set failure' to include any form of reaching a stopping point, the advantage for failure training was only effect size 0.19 — trivially small.

In plain terms: the extra reps you grind out at the very end of a set aren't doing extra muscle-building work. The muscle-growth signal — called mechanical tension, the pulling force on muscle fibres — is largely set by the hard reps you did before you hit failure, not the final failed rep itself.

The strength numbers: not reaching failure actually wins

Here's the more surprising finding. For strength gains, stopping short of failure may actually be the better strategy.

In Grgic et al. (2022), when training volume wasn't matched between the groups — the more realistic real-world scenario — non-failure training produced significantly greater strength gains (effect size -0.32, meaning non-failure won). The logic is simple: if you're grinding every set to failure, you fatigue faster, your technique deteriorates, and you can't sustain the same quality of effort across subsequent sets.

A 2024 meta-regression in Sports Medicine took a different approach — instead of comparing failure versus non-failure as two categories, it modelled the continuous relationship between how close to failure you train (measured in reps in reserve, or RIR — the number of reps you could still have done when you stopped a set) and both strength and hypertrophy outcomes. For strength, the relationship with RIR was negligible across all models. More proximity to failure didn't meaningfully improve strength (Robinson et al., 2024).

For hypertrophy, Robinson et al. (2024) found a modest signal: being closer to failure was associated with slightly more muscle growth — but the confidence intervals were wide, meaning the true effect could easily be near zero. The researchers called the fit of the models 'modest' — not a ringing endorsement for grinding every rep out.

When volume isn't matched, non-failure training produced significantly greater strength gains.

— Grgic et al. (2022). Effects of resistance training performed to repetition failure or non-failure. J Sport Health Sci.

The load question: does it matter what weight you're lifting?

One question the research has tried to answer: does load — how heavy the weight is — change things?

A 2021 network meta-analysis in Medicine & Science in Sports & Exercise looked at 28 studies with 747 adults, all training to volitional failure at low, moderate, or high loads. Muscle hypertrophy was similar across all load ranges when sets were taken to failure (Lopez et al., 2021). Whether you're doing 5 reps with a heavy weight or 20 reps with a lighter weight, reaching failure on each set produces roughly equivalent muscle growth.

Refalo et al. (2023) checked the same thing for proximity-to-failure comparisons — load didn't moderate the effect (p = 0.525). In other words, the 'train near failure' principle holds regardless of how heavy or light you're lifting.

This is actually reassuring. It means you don't need to go heavy AND grind to failure. Pick a load you can move well, take sets to within a rep or two of failure, and the muscle-growth signal is essentially the same as maxing out.

Hypertrophy was similar across low, moderate, and high loads — when effort was matched.

— Lopez et al. (2021). Resistance Training Load Effects on Muscle Hypertrophy and Strength Gain. Med Sci Sports Exerc.

There's one case where failure training has an edge

If you're more experienced in the gym, one subgroup finding is worth knowing.

Grgic et al. (2022) found that in resistance-trained individuals specifically, training to failure showed a significant effect for muscle hypertrophy compared to stopping short. The effect size wasn't reported in full in the abstract, but the significance threshold was crossed — which it wasn't for untrained participants.

The likely reason: experienced lifters have better technique, better tolerance for high-effort sets, and a higher baseline of fitness to absorb the extra fatigue. A beginner grinding to failure on every set is accumulating fatigue without the recovery buffer that experienced training builds over years.

If you're less than 1–2 years into consistent training, stopping 2–4 reps short is almost certainly the smarter play. If you're experienced and well-recovered, pushing closer to failure on your last 1–2 sets per exercise is a reasonable addition — not a replacement for good programming.

In trained individuals, failure training did show a significant hypertrophy edge — but not in beginners.

— Grgic et al. (2022). Effects of resistance training performed to repetition failure or non-failure. J Sport Health Sci.

What this means for how you actually train

Here's the practical takeaway from all four meta-analyses:

1–2 reps in reserve (RIR) — meaning you stop when you have 1–2 reps left in the tank — is the sweet spot for most lifters, most of the time. You're close enough to failure to generate a strong training signal, but not so close that you're torching your recovery and sacrificing form.

Some concrete guidelines:

- Working sets: Aim to stop when you could do 1–3 more reps. You don't need to hit absolute failure.
- Last set of an exercise: You can push closer to failure here — it won't carry fatigue into the next exercise.
- Compound lifts (squats, deadlifts, bench press, rows): Technical breakdown near failure is risky. Staying 2–4 reps short protects your form and your joints. Check out rpe vs rir for a practical way to track this in real time.
- Isolation exercises (curls, lateral raises, leg extensions): Lower injury risk near failure, so pushing closer is fine if you're recovered.
- Volume matters more than failure. The best predictor of hypertrophy is total sets per muscle group per week — see how many sets per muscle group per week — not whether you squeezed out that last grind rep.