Reps per set: how many for size vs strength, per 4 studies

How many reps per set should you do?

Here's the thing almost every rep-range article gets backwards: reps per set isn't a number you choose. It's a number that falls out of two things you choose — the load, and where you stop.

A 2024 randomised trial makes that concrete. Twenty-two lifters did the same session on the barbell squat: 5 sets at 75% of 1RM (1-rep max — the most weight you can lift once with good form). The only difference was the stopping rule. One group stopped when bar speed dropped 10%. The other went to failure.

Same load. Same sets. 3.9 reps per set versus 13.8 (Romagnoli et al., 2024).

That's a 3.5× spread in reps per set from a single decision about when to rack the bar.

So if you came here for a number: 6–20 reps per set for muscle size, 1–8 for strength — and 6–8 where the two overlap. Those are bands, not targets — and where you land inside one is decided by your load and your stopping rule, not by the number on the page. That's why the useful version of the question is: what rep band fits my goal, and what stopping rule keeps me in it?

This article answers the first half, and closes with the one stopping-rule finding that changes how you read your own rep count. (For the full evidence on how close to failure you should train, see training to failure and rpe vs rir.)

Same load, same 5 sets — 3.9 reps per set versus 13.8, decided entirely by where they stopped.

— Romagnoli et al. (2024). Does Fatigue Affect the Perception of Velocity Accuracy During Resistance Training? J Strength Cond Res.

How many reps per set to build muscle (hypertrophy)

For hypertrophy — building muscle size — the research points to a broad range: roughly 6–20 reps per set, provided you're training close to failure.

Loads around 75% 1RM land most lifters in the 6–12 rep zone, but sets at lower loads (50–60% 1RM) taken close to failure show similar muscle growth to heavier, lower-rep work. The mechanism is the same: you're recruiting and fatiguing the muscle fibres that grow.

What this means practically:
- 6–12 reps is the classic hypertrophy range — heavy enough to create mechanical tension (the stretching and loading force on muscle fibres), high enough volume per set to accumulate fatigue.
- 10–20 reps works too, especially on isolation moves like the dumbbell bicep curl or lateral raises where joint stress is lower.
- Below 5 reps can build muscle, but it's better suited to strength — more on that below.

For a deeper look at how rep ranges interact with muscle growth, see how many reps for muscle growth.

The ceiling on hypertrophy rep ranges is higher than most people train. If you've been stuck at 3×10 for months, 3×15 at a lighter load — taken close to failure — will still drive growth.

How many reps and sets for muscle growth — does set count change the answer?

Set count and reps per set are separate levers, and the cleanest test of the set lever holds reps fixed. Lundby et al. (2024) did exactly that: 3×10 versus 5–10×10 on well-trained rowers over 8 weeks, ten reps per set in both groups. Strength went up in both. Lean body mass didn't move in either. There were no differences in training response between the groups on any measure — the higher-volume group tripled its set count and spent far longer in the gym for it.

So the answer to "does set count change the answer?" is, in this trial, no — at least not once you're already trained and already doing three hard sessions a week. That doesn't make sets irrelevant; it means more sets stopped buying anything here, which is a different claim from "volume is what matters."

The practical takeaway: pick a rep range you can sustain with good form and progressive overload — adding weight or reps over time (see progressive overload training) — rather than chasing a magic number. That progression is what actually drives long-term muscle growth.

For how to structure total weekly set counts by muscle group, see how many sets per muscle group per week.

How many reps and sets for strength

For strength — moving more weight — the standard prescription is 1–8 reps per set at 75–85%+ of your 1RM, with 3–6 as the practical zone. Heavy loads, low reps, long rests. That's the conventional reading, and for a trained lifter chasing a bigger squat or bench press it's a reasonable default.

But "heavier and fewer" isn't a universal law, and one paper in this package is a useful corrective. A 2025 Bayesian network meta-analysis pooled 13 trials on older adults with sarcopenia (age-related muscle loss) to find the dose that best improved handgrip strength. The winner wasn't the heavy-low-rep end of the range. It was 49% 1RM, 16 reps, 6 sets, three times a week (Hua-Rui et al., 2025).

That's a different population with a different goal — grip strength in clinically weak older adults, not a 1RM squat in a trained lifter — so don't read it as a prescription for your next strength block. Read it as evidence that the load-rep combination that builds strength depends on who's lifting. For an untrained or older lifter, moderate loads at higher reps did more than heavy singles.

Strength-focused rep targets for a healthy, training lifter:
- 1–5 reps: maximal strength, heavy compound lifts (barbell squat, deadlift, bench press)
- 3–6 reps: the practical strength zone — heavy enough to drive neural adaptations (your nervous system getting better at recruiting muscle), manageable enough to repeat
- 6–8 reps: the overlap zone — works for both strength and size

For how many sets to pair with this, see how many sets for strength.

In older adults with sarcopenia, the best dose for grip strength was 49% 1RM and 16 reps — not heavy singles.

— Hua-Rui et al. (2025). Optimal dose of resistance training to improve handgrip strength in older adults with sarcopenia. Front Physiol.

How many sets and reps for hypertrophy — a practical cheat sheet

You don't need a different number for every exercise. You need a range you can apply consistently and push hard within. Here's how to think about it by goal:

For muscle size (hypertrophy):
- 6–12 reps per set for compound movements (barbell squat, barbell row, bench press)
- 10–20 reps for isolation work (dumbbell bicep curl, lateral raises, leg extension)
- Stop 1–3 reps short of failure most sets — this keeps form intact and lets you repeat the effort next session

For strength:
- 1–8 reps per set (3–6 is the practical zone)
- At least 75% of your 1RM
- Full recovery between sets — 3 minutes minimum

For both (the overlap zone):
- 6–8 reps at a weight that feels genuinely hard by rep 7
- This is where most intermediate lifters live, and it works

The Janicijevic et al. (2026) RCT on basketball players is a useful reminder that how you structure the reps (supersets vs cluster sets — where you break rest into shorter intra-set pauses) matters less than the total reps and load. Both structures, at a matched 18 reps per exercise per session at 75% 1RM, produced the same neuromuscular improvements over 6 weeks.

Format is flexible. Load and proximity to failure are not.

What actually limits your reps per set — and what to do about it

Your rep count is an output — so the useful question is what it's an output of. Two inputs dominate: the load, and the stopping rule. Most lifters set the load deliberately and never make the stopping rule explicit at all.

There's a second finding in the Romagnoli trial that makes the stopping rule usable. The study's actual question was whether fatigue wrecks your ability to judge bar speed. It doesn't: lifters estimated barbell velocity just as accurately immediately after a hard training session as they did rested, even though the to-failure group's real bar speed had dropped at light and medium loads (Romagnoli et al., 2024).

Your internal read on how fast the bar is moving stays trustworthy when you're tired. That's what makes "stop when the bar slows down and your form shifts" a rule you can actually apply without a velocity tracker — and it's why your rep count will vary from the number printed in your programme.

The practical version: pick a rep band, not a rep number. If the plan says 10 and you get 8 at the same load, you haven't failed the session — you've measured it. Log it and let the load follow the trend over weeks (progressive overload training) instead of forcing a number.

Pick a rep band, not a rep number.