Training frequency and rest days

How often should you train a muscle? And how many days off does training need? They sound like separate questions, one about programming and one about recovery, but they are the same question asked from opposite ends. A week has seven days, and every session you add for a muscle is a rest day you take away from it.

The evidence gives a clearer answer than the gym rules of thumb — once a week per body part, or always 48 hours between sessions. For muscle size, how you spread your sets across the week makes little difference. For strength, practising a lift more often helps, with diminishing returns. And how long a muscle needs to recover depends mostly on how hard you trained it, which is something you control.

What frequency means, and why it is hard to study

In the research, training frequency is the number of times per week a muscle or exercise is trained. ¹ That is not the same as how many days you go to the gym. Three full-body sessions train each muscle three times a week; a four-day upper/lower split trains each muscle twice; a five-day split with one day per body part trains most muscles once.

Frequency is hard to study because it rarely changes on its own. Train a muscle twice a week instead of once, doing the same session each time, and you have also doubled its weekly sets. More sets produce more growth — the subject of our guide to training volume — so a study like that cannot say whether the second session helped because of when it happened or because of the extra work.

Researchers deal with this in two ways. Some trials hold weekly volume constant and change only how it is distributed: the same twelve sets done in one session, or split across two or three. Others pool many studies and adjust for volume statistically. The 2025 meta-regression behind our volume guide did the second, including weekly volume in its frequency models, so its frequency estimates describe the effect of spreading training out over and above the effect of doing more of it. ¹ Its authors note that earlier meta-analyses of frequency had produced mixed results, which generally suggested no independent effect. ¹

Muscle size: spread the sets as you like

For muscle growth, the evidence is unusually consistent. A 2019 meta-analysis by Brad Schoenfeld, Jozo Grgic and James Krieger pooled 25 studies and found no significant difference between higher and lower frequencies when weekly volume was equated. ² Nor did it find one in any of the subgroups it checked: studies that measured muscle directly, studies of people who already lifted, and upper- and lower-body muscles analysed separately. When studies that did not equate volume were included, higher frequencies came out ahead, but the difference between training a muscle once a week and three or more times was modest. The authors concluded that, for a given volume, people can choose how often to train each muscle by personal preference. ²

The 2025 meta-regression, pooling 67 studies and 2,058 participants with volume adjusted for, landed in the same place: frequency’s effect on muscle growth was compatible with a negligible one. ¹

A trial in trained men shows what that looks like in practice. Twenty-seven resistance-trained men did the same weekly volume either three or six times a week for six weeks. ³ Both groups gained similar amounts of muscle at the back of the upper arm and the front of the thigh, with no significant differences between them; the elbow flexors grew significantly only in the three-day group. Strength gains were similar too. It is one small, six-week trial, but it fits the pooled evidence: putting the same sets into twice as many sessions bought nothing extra.

None of this makes frequency useless for building muscle; it changes what frequency is for. If you want more weekly sets than fit comfortably into one session, more sessions are how you fit them in. The benefit comes from the sets. Frequency is how you schedule them.

Strength: practice pays, up to a point

Strength is where frequency appears to matter, and where the two main analyses differ on why.

A 2018 meta-analysis led by Jozo Grgic reviewed 22 studies of frequency and strength, 21 of them in the pooled analysis. ⁴ Gains rose with frequency: standardised effect sizes of 0.74, 0.82, 0.93 and 1.08 for training once, twice, three times, and four or more times a week. (A standardised effect size expresses a gain relative to how much people vary, which makes studies comparable but does not translate into kilograms.) But when the analysis was limited to the seven studies that equated volume, there was no significant effect of frequency (P = 0.421), and the authors concluded that the advantage seemed to be driven mainly by volume. Whether frequency had any effect of its own, they wrote, remained unclear. ⁴

Two details in that paper point towards a mechanism. The effect of frequency was significant for multi-joint lifts but not for single-joint exercises, and the authors suggested motor learning as an explanation: more frequent sessions mean more practice of the movement that is later tested. ⁴ Population matters here as well. Only three of the studies used people with lifting experience, so the analysis mostly describes beginners.

The 2025 meta-regression, larger and newer, reached a firmer conclusion. After adjusting for volume, strength gains increased with frequency, with diminishing returns — and of the two outcomes, only strength showed a consistently identifiable effect of frequency. ¹ Its participants were 79 per cent male, with an average age of 25. On the current evidence, then, practising a lift more often helps strength beyond the sets it adds, though each additional session is worth less than the one before.

That completes a picture from our volume guide, where strength stopped responding to extra sets after only a few a week. Strength still responds to more frequent practice. If a lift matters to you, spreading its weekly sets across two or more sessions is a better-supported way to improve it than adding sets to one.

How long a muscle needs between sessions

The most repeated rule about rest is that a muscle needs 48 hours before it is trained again. The evidence suggests the right wait depends on how hard the session was.

A 2017 study followed ten highly trained men through three sessions of bench press and squat at 75 per cent of their one-rep max, each four weeks apart. ⁵ The load allowed about ten reps per set. One session was three sets of five, one was six sets of five, and one was three sets of ten taken to failure — so the first two stopped halfway, and the six-by-five session matched the failure session’s total reps exactly. Recovery was tracked with jump height, bar speed and blood markers from the day before to 72 hours after.

After the sessions that stopped short of failure, jump height was back to baseline within six hours, and recovery between 24 and 48 hours was significantly faster than after the failure session. After three sets to failure, jump height and bar speed were still below baseline 24 and 48 hours later, and creatine kinase, a marker of muscle damage, stayed raised through 48 hours. The authors concluded that training to failure slows recovery by up to 24–48 hours. ⁵

The limits are real: ten men, one session of each kind, on a Smith machine, with recovery measured as performance and blood markers rather than as muscle growth. But the practical point stands. The 48-hour rule describes hard sessions, and how hard you train is your choice. Stopping most sets a rep or two short of failure — which our guide to progressive overload explains is enough to keep progressing — also shortens the wait before you can train the same muscle well again.

Recovery also changes as a programme goes on. In a study that followed ten young men through ten weeks of training, muscle damage was greatest after the first session, lower by the third week and lower again by the tenth. ⁶ Muscle protein synthesis was also highest after that first session, yet it was unrelated to later growth; only the responses at weeks three and ten tracked how much muscle each man gained. That suggests the after-effects of a new programme’s first sessions overstate how much rest the same training will need a few weeks later.

Rest days, overreaching and time off

If each muscle gets enough time between hard sessions, a whole day off matters less than it seems. In the trial above, men training six times a week did about as well over six weeks as men training three times on the same weekly volume; the one exception was the elbow flexors, which grew significantly only with three sessions. ³ Doubling the sessions, with the weekly total held constant, bought nothing extra over six weeks, which is as far as that evidence goes.

The fear behind most rest days is overtraining. A 2020 systematic review looked for it in 22 studies that deliberately tried to overreach or overtrain people through resistance exercise. ⁷ Eight produced a drop in performance and followed it up; four reported a drop without follow-up; ten found no drop at all. Inconsistent methods meant that functional overreaching, non-functional overreaching and overtraining syndrome could not be reliably told apart, and few studies had actually established any of them. The authors suggested overtraining may be related to frequent, high-intensity, monotonous resistance training, and found that no marker other than a sustained drop in performance had been established as a reliable sign. ⁷

That gives you a practical test: watch your performance, over more than one session. Our Progressive Overload Calculator applies the same logic. When your reps fall from one session to the next, it tells you to repeat the load rather than push, because one lower session is not a trend. If the drop continues, or a familiar load starts to feel unexpectedly hard — the warning sign described in our guide to RPE — that is the time to add rest, and sleep is the first place to look.

A longer break also costs less than most lifters fear. In a 2024 Finnish trial, 55 untrained adults, with an average age of 32 and nearly half of them women, trained twice a week. ⁸ One group trained for ten weeks, stopped for ten, then trained ten more; the other waited ten weeks, then trained for twenty without a break. During the break the first group lost some ground — about 5 per cent of leg-press strength and 7 to 10 per cent of muscle size — but regained it within five weeks of returning, and both groups finished with similar results. The participants were beginners, and the authors note the findings may not apply to trained lifters or to breaks caused by illness or injury. Still, if ten weeks off did beginners no lasting harm, a missed week is not worth losing sleep over.

The bottom line

Frequency and rest are one decision: how to spread a week’s training across its days.

For muscle size, it barely matters. Pick the schedule you will actually keep — full-body three times a week, an upper/lower split, or one day per body part — and get your weekly sets in. More sessions are mainly a way to fit in more sets.

For strength, practise the lifts you care about more than once a week. Frequency helps beyond the sets themselves, though each extra session adds less.

For recovery, the 48-hour rule is about hard sessions. Stopping most sets short of failure shortens the wait, and the first weeks of a new programme are the hardest to recover from. Judge whether you need more rest by your performance across several sessions, not by soreness or one bad day. And when life forces a break, the evidence says the ground you lose comes back quickly.