Number of sets, repetitions, time under tension: what do you need to do to build muscle?

By Kalorya · July 28, 2026 · 8 min read

Barre de musculation chargée, posée au sol d'une salle d'entraînement

One room, two schools of thought. On the one hand: ‘Do 3 sets of 10’. On the other: ‘Forget about the reps; what matters is the time under tension — your muscle needs to work for 40 to 60 seconds per set’. Good news: both ideas have been put to the test head-to-head, stopwatch in hand. And the answer is neither one nor the other.

This ‘seconds’ issue has been tested directly

If time under tension were the correct unit of measurement, then two sessions with the same total number of seconds should result in the same muscle growth — regardless of how those seconds are divided up. This is exactly what a 2022 study set out to test: 38 untrained men, over 10 weeks, with sets calibrated to 36 seconds each, some consisting of 3-second repetitions and others of 6-second repetitions.

Result: similar gains in strength (1RM) and similar hypertrophy of the pectorals and triceps. With the same time under tension, the speed of execution made no difference. The stopwatch, however, did indeed ‘predict’ the result.

Except that the stopwatch keeps running even when you unplug it

The real test lies elsewhere: what happens when we vary the time under tension rather than standardising it? A meta-analysis brought together studies comparing different tempos, all carried out to the point of muscle failure. The verdict: muscle hypertrophy is similar for repetition durations ranging from 0.5 to 8 seconds.

Take another look at that figure. Between a half-second repetition and an eight-second repetition, there is a sixteen-fold difference in the time under tension. And the muscle cannot tell the difference. A set of 10 explosive repetitions (5 seconds under tension) and a set of 10 very slow repetitions (80 seconds under tension) produce the same growth. The authors do, however, set an upper limit: beyond 10 seconds per repetition, it would probably become less effective — but there are few controlled studies on this point.

What we varied	What we measured	Result
Duration of a repetition, with the time under tension standardised to 36 seconds per set (2022 study, 10 weeks)	Strength and muscle hypertrophy	There is no difference between 3 seconds and 6 seconds
Duration of a repetition, without normalising the time under tension (2015 meta-analysis)	Hypertrophy	No difference of between 0.5 s and 8 s per repetition
Load and repetition duration, sets performed to muscle failure (2019 study, trained men)	Fibre activation, anabolic signalling	The same regardless of the load and duration
Speed of repetition execution (2018 review, 6 studies, 119 beginners)	Muscle hypertrophy	Contradictory results; caution is advised

What it eliminates

Counting seconds does not predict muscle gain. If training programmes with the same time under tension yield the same results, it is simply because they were already equivalent in other respects — not because the second is the unit of measurement for muscle. You can put the stopwatch away.

An important point to note: not all seconds are equal

Time under tension is by no means insignificant — it is simply not sufficient. Two studies illustrate this clearly. In 2012, a study using light loads (30 per cent of 1RM) compared 6-second repetitions with 1-second repetitions for an equivalent training volume: the slow version produced a greater increase in muscle protein synthesis, notably a delayed but robust myofibrillar response 24 to 30 hours after the session.

But in 2015, another study compared three ways of maintaining the exact same time under tension using very different levels of force. It was the protocol that produced the most force (maximum eccentric) that triggered the strongest anabolic response — not the one that caused the most fatigue. In other words: a second spent generating a lot of force is not the same as a second spent dragging out the movement under a light load.

The right way to look at things

Time under tension is a consequence of your set, not its objective. Deliberately slowing down to ‘buy time’ forces you to reduce the load — you’re trading power for seconds, and the 2015 study suggests that this is a poor trade-off.

The unit that really works: the ‘difficult’ series

If it’s not the stopwatch, then what is? A 2021 systematic review asked the question head-on: is the total number of sets a valid way of quantifying training volume? After analysing 14 studies, the answer is yes — on one condition: that these sets are performed to failure or near failure. And this remains true across the entire range from 6 to more than 20 repetitions.

Better still: this particular unit has a measurable dose-response relationship. A 2017 meta-analysis (15 studies, 34 groups) calculated that each additional weekly set was associated with a 0.37 per cent increase in muscle mass. An extra second makes no difference; an extra set does.

A 2019 study involving 49 trained men tested 9, 18 or 27 weekly sets for the biceps over a period of 6 weeks. Muscle thickness increased in all groups — 9 sets per week are already sufficient to build muscle — although estimates of effect size suggest that 18 to 27 sets could lead to greater strength gains.

The balanced view: a 2017 analysis of upper-body exercises found no statistically significant benefit to doing three or more sets rather than fewer than three among beginners, and highlights the lack of well-designed studies among trained athletes. ‘More sets’ is therefore not a rule with no upper limit — it is a trend, with some real grey areas.

And what about the number of repetitions per set?

Here too, the effective range is much wider than is commonly believed. A 2017 meta-analysis of 21 studies compared light loads (≤ 60% of 1RM) with heavy loads (> 60%), with all sets taken to failure: there was no significant difference in hypertrophy. Gains in maximum strength, however, clearly favour heavy loads.

Study	What was compared	Muscle	Strength
2017 meta-analysis (21 studies)	≤ 60% vs > 60% of 1RM, until muscle failure	Equality	Heavy loads take precedence
2016 study (49 trained men, 12 weeks)	20–25 reps (30–50 per cent of 1RM) vs 8–12 reps (75–90 per cent)	Equality	A slight advantage for heavy loads
2015 study (18 fit men, 8 weeks)	25–35 reps vs 8–12 reps	Quadriceps: 9.3% vs 9.5%	Squat: 8.8% vs 19.6%
2020 Meta-analysis	Light loads vs heavy loads, muscle fibres	Equality (Type I and Type II)	—

The message is consistent: to build muscle, the effective range is from around 6 to over 20 repetitions. To get strong, you need to lift heavy weights. These are two separate goals, not just one.

The real hidden variable: how far you’re willing to go

That is what really makes the difference — and it is what time under tension captures by chance when it ‘works’. A 2022 study involving 25 untrained men combined the two factors: heavy or light loads, and sets performed to muscle failure or stopped short of it. Three out of four groups made equal progress; the fourth fell behind.

Muscle growth depends on the load and how close one is to muscle failure(Percentage increase in quadriceps cross-sectional area over 8 weeks)

2022 study, 25 untrained men, 8 weeks. Light loads = 30% of 1RM, heavy loads = 80% of 1RM.

Three bars on the graph are almost identical; the fourth drops sharply. With light loads, stopping well short of failure achieves virtually nothing: the effort is never hard enough to recruit the muscle. With heavy loads, on the other hand, going to failure brings no additional benefit — the load is already doing the job. (In terms of pure strength, heavy loads outperform light ones: +33.8 per cent and +33.4 per cent of 1RM compared with +17.7 per cent and +15.8 per cent.)

Two further studies confirm that, provided the set is sufficiently challenging, it is not essential to train strictly to failure: over 12 weeks in 32 untrained men, muscular failure and voluntary cessation resulted in similar gains in muscle cross-sectional area (6.1–7.5 per cent); and over 10 weeks in trained athletes, training to failure and not training to failure produced comparable gains (13.5 per cent versus 18.1 per cent in vastus lateralis cross-sectional area).

Why not end each set with a failure?

A 2022 meta-analysis of 20 studies assessed the cost: training to failure causes greater acute fatigue, more muscle damage and a higher perceived exertion, with a longer recovery time. If each set takes more out of you, you’ll do fewer of them over the course of the week — and it’s the total number of sets that helps you build muscle. Save training to failure for the last set of an exercise, not for every set.

How this works in practice

The question you’re asking yourself	What you need to sort out
How many seconds should a set last?	Don’t count the seconds: they depend on how hard you try
How fast should I do my reps?	At your natural pace. Anything between 0.5 and 8 seconds per repetition works
How many repetitions per set?	Between 6 and 20+ for muscle. Heavy (< 8) if you’re aiming for strength
How many sets should you do per week for each muscle group?	9 is enough to make progress. Increasing this to 18–27 may yield better results
How far should you push yourself during a set?	Close enough to muscle failure to make it challenging — essential if you’re training with light weights
Should you finish every set to muscle failure?	No. Save the muscle failure for the last set of each exercise, because of the recovery time it takes

In summary

Muscles don’t count the seconds: they count the tough sets. Time under tension is a gauge, not a lever — it goes up when you’re working hard, but artificially increasing it by slowing down won’t make you bulk up any more. Focus on three things: enough sets throughout the week, a load that puts you in the 6–20 rep range, and a real effort at the end of each set.

Since it’s the number of sets that counts, you still need to keep track of them. A programme that’s written down and followed session after session is better than any stopwatch.

View my programmes

One question remains: should these sets be spread over two, three or five sessions? We answer this, backed up by research, in ‘How many sessions a week to build muscle?’.

Every figure in this article is taken from the studies listed below, summarised on the basis of their abstracts alone. Please note: several of these studies focus on beginners or short-term periods (6 to 12 weeks), and there is still a lack of studies comparing training volumes among trained athletes. This is for information purposes only and is not a substitute for professional advice.