How long to rest between sets

Most lifters rest by feel, which in practice means resting until the boredom outweighs the fatigue. That lands somewhere around 60 to 75 seconds on everything, and it is the wrong number for about half of what you do.

What the rest interval buys

One thing, mostly: the reps on your next set.

Cut rest short and the set that follows is worse. Fewer reps at the same load, or the same reps at a lower one, and either way the session’s hard-set count comes down. That is the entire mechanism by which short rests cost you hypertrophy — not some direct effect of the clock, but the volume you fail to complete because you started the next set half-recovered.

Which means the question is not “how long do I need to rest to grow”, it is “how long do I need to rest to do the work I planned”. Those have different answers per exercise, which is the part most apps get wrong.

Ninety seconds is a floor

The controlled work on this points the same direction each time it is redone. Below roughly 60 seconds, performance across sets falls off hard enough to cost you measurable growth. Between 60 and 90 it recovers. Past 90 the direct hypertrophy benefit stops showing up.

So 90 seconds is where the floor sits. It is not where the target sits, and the distinction matters: anything longer than 90 is bought for a different reason.

Longer rests are bought, not gifted

You take three minutes on a heavy squat because you want the third set to look like the first one. That is a performance-maintenance purchase, and on a compound at 5 to 8 reps it is worth the time — the alternative is a top set and then two increasingly cosmetic ones.

You do not take three minutes on a leg extension. Nothing systemic is limiting you, the local fatigue clears fast, and all a long rest buys there is a longer evening.

Rough shape of it:

Those are defaults, not prescriptions. What matters is that they differ, and that the difference tracks how much of you the movement taxes.

Rest belongs to the exercise

This is the practical version of everything above: a rest timer that is set once for the whole session is set wrong for most of it.

You decide the number when you programme the lift, calmly, knowing what the movement costs. You do not decide it mid-session, out of breath, with a sixty-second default counting down and a queue for the rack. Set it on the exercise and it is correct every time that exercise comes around, in every workout it appears in.

What a broken timer costs

A rest timer that counts its own ticks is wrong the moment your phone locks. The timer stops, you come back, it says 1:12 and you take that at face value. You have just taken a two and a half minute rest and logged it as ninety seconds.

Do that on the sets where the interval was doing real work and the third set underperforms for a reason that never appears in your log. Any timer worth having stores an instant and subtracts from wall time, so a locked screen, a backgrounded tab and a phone in a pocket all give the same answer.

When to break the rule

Two cases, both deliberate.

Antagonist supersets. Pairing a push with a pull lets the rest interval do double duty — you are resting one movement while working the other, so the wall-clock cost of the session halves without the second set being rushed. Rest between the pair, not inside it.

Deliberate density work. If the point of the block is work capacity rather than hypertrophy, short rests are the stimulus and reduced load is expected. That is a legitimate goal and it is not what most people are doing when they rest for 45 seconds — most people are doing it because they are in a hurry.

The short version

Never under 90 seconds. Longer where the set after it matters. Set it on the exercise, once. And check that your timer survives a locked phone, because a good number measured badly is the same as a bad number.

Where this comes from

The rest-interval literature is summarised in Grgic and colleagues’ 2018 meta-analysis of rest intervals and muscle hypertrophy (Sports Medicine), and in Schoenfeld and colleagues’ 2016 trial of 1 versus 3 minute rests in trained men (Journal of Strength and Conditioning Research), which found the longer interval produced more growth — by way of more completed volume, which is the mechanism described above.