Does Soreness Mean Muscle Growth? 110 Men Did 12, 24 or 60 All-Out Reps and Hurt About the Same.

You finish a session, and two days later the stairs are a problem. Somewhere in the back of your head a little scoreboard ticks over: that one worked.

Almost everybody who trains uses soreness as a gauge. It is free, it needs no equipment, and it arrives right on schedule. It is also the single most available piece of feedback your body gives you about a workout, which is exactly why it got the job.

So here is the question nobody asks about it: has anyone ever checked whether the gauge is connected to anything?

They have. Repeatedly. And it is not.

Five times the work, the same soreness

Start with the cleanest test of the gauge itself. Nosaka and colleagues took 110 male students and split them three ways [1]. One group did 12 all-out eccentric curls. One did 24. One did 60.

Eccentric means the lowering half of the rep — the part where the muscle is being pulled long while it is trying to hold on. It is far and away the best way to make yourself sore, which is why every study in this field uses it.

Then they measured what actually happened to the arm: how much force it could still produce, how far it would straighten, how far it would bend, how much it had swollen, and how much creatine kinase — a muscle protein that leaks into the blood when fibres are torn — had shown up in the bloodstream.

The objective damage behaved exactly as you would expect. Sixty reps did more harm than twelve, and took longer to come back from.

The soreness did not. Pressing on the muscle produced statistically the same pain in a man who had done twelve reps and a man who had done sixty. And soreness on palpation — pressing — correlated with none of the damage measures. Not one.

The best link they found anywhere was below 0.32. That number runs from 0 to 1, where 1.0 means two things move together perfectly and 0 means knowing one tells you nothing about the other. 0.32 is nearer the nothing end.

Their own conclusion, in the title of the paper: delayed-onset muscle soreness does not reflect the magnitude of the damage [1].

Fine — but does the damage build the muscle?

This is the fallback argument, and it deserves a straight answer, because for years it was a live hypothesis with real people behind it: tear the fibres, the body repairs them stronger, that is how you grow. Schoenfeld laid out the case for and against it in 2012 and the honest summary at the time was “unresolved” [7].

Then somebody went and looked inside.

Damas and colleagues put ten untrained young men through ten weeks of resistance training and took muscle biopsies — actual samples, cut out of the thigh — at week one, week three and week ten [2]. At each point they measured two things: how torn up the muscle was under an electron microscope, and how fast it was building new contractile protein.

Read the third row twice, because it is the whole argument.

At week one, damage was at its peak and protein synthesis was at its peak — and neither one predicted how much muscle those men ended up with. At weeks three and ten, when the damage had faded, protein synthesis and eventual growth moved together almost in lockstep [2].

The building response in week one was not building. It was repairing. It looks identical from the outside and it produces nothing you keep.

Damas summarised the whole line of work in a review six years later, and the sentence is about as blunt as physiology gets: muscle damage “is not the process that mediates or potentiates” the muscle growth you get from resistance training [3].

The cruellest part: in the early weeks you really do look bigger

Here is why this myth is so hard to shake. It has evidence. You feel it.

The same group ran the ultrasound version of the experiment [4]. Ten untrained men, ten weeks, thigh measured throughout. Cross-sectional area — the thickness of the muscle — was up about 2.7% by week three. Progress, surely.

Except the echo intensity was up 17.2%. Echo intensity is how bright the muscle looks on the scan, and it goes up when the tissue is waterlogged. Normalise the size gain against it and week three is the only timepoint where the muscle is disproportionately full of fluid [4].

And the strength? Maximum voluntary contraction did not move at all until week ten [4].

So in the first few weeks of training you are sore, the tape says you are bigger, and you cannot lift any more than you could before. Two of those three things are the same thing: swelling. The soreness and the size are both the flood, not the building.

Damas’s review puts numbers on the phases: for roughly the first four sessions the size change is mostly swelling; a modest amount of real growth arrives around ten sessions; and genuine hypertrophy — actual new muscle — is what you are looking at after about eighteen [3].

Then what is soreness measuring?

Novelty. That is the short answer, and it is a useful one.

A 2026 trial makes it very hard to argue with [5]. Twenty-three people did a punishing bout of eccentric work, waited three weeks, and did it again. The second bout was matched for total work — about 1,300 joules both times, and the difference between them was nowhere near statistical significance.

Identical work. And the second time, both the strength loss and the soreness recovered faster [5].

Nothing about the workout got easier. Something about the person got better at absorbing it — the researchers traced part of it to steadier firing in the motor units, the nerve-and-fibre bundles that actually pull [5].

Which gives you the thing that finally kills the gauge as a training tool. Soreness is highest when you are worst at an exercise, and it fades as you get good at it — over exactly the same weeks in which the muscle is growing fastest. The signal runs backwards to the thing it is supposed to be reporting.

If you have ever come back after a layoff, been destroyed by week one, and concluded that week one was your best week — that is the illusion, in one paragraph.

And there is a decent chance the pain is not even in the muscle

This last one is a hypothesis rather than a settled finding, and we are labelling it that way rather than letting it carry weight it has not earned.

Wilke and Behringer make the case that delayed soreness may originate not in the muscle fibres but in the fascia — the sheet of connective tissue wrapped around and through the muscle [6]. Their argument is that fascia is far more richly supplied with pain nerve endings than muscle is, and that it hurts a great deal more when you provoke it.

The experiment that makes the point: irritate the fascia and the muscle in the same leg with a chemical injection. In an untrained, un-sore leg, the fascia is about 42% more painful than the muscle. Now do it in a leg that already has delayed-onset soreness — and the muscle hurts exactly as much as it did in the fresh leg. It is the fascia that has become 39% more sensitive [6].

Read that plainly. In a sore leg, the muscle was not tender. The wrapping was.

The authors go as far as proposing the phenomenon be renamed — delayed onset soft tissue stiffness rather than muscle soreness [6]. Whether that survives is an open question. But if the pain is substantially a connective-tissue signal, then asking it about muscle growth is asking the wrong tissue.

Where the belief came from, and it is older than the gym

The first paper ever written on this is from 1902. Theodore Hough, in the American Journal of Physiology, noticed that an untrained muscle working against a strong spring produces a soreness that — in his words — cannot be regarded as a phenomenon of pure fatigue [8].

Sit with that for a second. In the very first study of muscle soreness, the finding was that soreness tracks how unaccustomed the work is, not how hard it is. That has been the answer for over a century. It just never made it out of the journal.

What made it out instead was lactic acid. Soreness got explained as acid pooling in the muscle after hard work, which recast the pain as the exhaust fumes of effort — a receipt. That explanation has been largely rejected, and the killer detail is this: concentric exercise, the lifting half, produces considerably more lactate than eccentric work at matched power output, and it almost never makes you sore [6].

From there it needed a slogan, and the aerobics boom supplied one. Jane Fonda’s Workout tapes arrive in 1982 with “feel the burn” and “no pain, no gain”, and a whole generation learns to read discomfort as the unit of account.

And where does it end up? Ask Google the exact question at the top of this page and its AI Overview answers it partly by citing a supplement company and a mobility app, sitting alongside the Cleveland Clinic and Henry Ford Health. Which is the tell. A question about physiology gets answered, at the top of the internet, partly by people who sell you the thing that makes the soreness go away.

None of which is your fault. It is the most intuitive feedback signal in fitness, it was explained to you wrongly for a century, and it got printed on the wall of every gym you have ever walked into.

So how do you actually tell if you are growing?

The honest answer is that you use the boring instrument, and you use it over a longer window than you want to.

Is the work going up? Load times reps times sets. If you are doing more of it this month than last month with the same form, you are training in a way that grows muscle, whether or not it hurt afterwards. This is the only readout that updates weekly and means anything.

Is the tape moving over eight to twelve weeks? Not eight days. The Damas data says the early size change is largely fluid [4], so measuring at three weeks is measuring water. Give it a training block.

Are you expecting a sane amount? This is where most people go wrong before they ever get to soreness — and it is worth reading what happened when 585 people ran the identical programme for twelve weeks. The spread in what they got is the single most clarifying thing in this whole area.

And the unglamorous inputs still decide it: enough total hard work, enough recovery, and enough protein — though probably less than you have been told.

What this does not mean

It does not mean train easy. This is the misreading to avoid, so let us be blunt: the training still has to be hard. The gauge is broken, not the engine. Progressive overload built every result in the Damas data [2]. Those men trained for ten weeks and their fibres got bigger. Nothing here is an argument for doing less.

It does not mean soreness is a bad sign either. It is not damage you need to fear, and it is not damage you need to chase. It is mostly information about how new the movement was to you.

It does not mean you should go out of your way to prevent it. Some of the things sold for that are actively counterproductive — routine post-lifting cold water reduces soreness and blunts the growth, which is the one place where chasing the soreness number genuinely costs you. Others simply do less than advertised: massage, compression boots, tart cherry juice.

And it does not mean the absence of soreness is a warning. If you have been training the same movements for months and they stopped hurting, that is the repeated bout effect doing precisely what it is supposed to [5]. You did not get lazy. You got adapted.

What this is rated, and what the rating covers

Unsupported. The claim is that soreness indicates muscle growth, and it has been tested from four directions and failed all four.

It is rated Unsupported because soreness does not track the damage it is supposed to be reporting [1], because damage does not drive the growth it is supposed to be signalling [2] [3], because soreness falls across exactly the weeks when growth accelerates [5], and because the early size change that seems to confirm it is fluid [4].

What is not rated here: whether hard training builds muscle, which it plainly does; whether eccentric work is a good tool, which it is; and whether the fascia hypothesis is correct, which is genuinely unsettled and is flagged as such above [6].

How we read a study, and what each confidence tier means, is set out here. Related: whether cardio burns the muscle you just built.