Muscle soreness after exercise: why it happens and whether to train

You did walking lunges for the first time in six months. It felt fine at the time. Now it is Sunday morning, the stairs down to the kitchen are a project, and you are standing there wondering whether this means you did something right or something stupid.

It is a fairer question than it sounds, because the answer has been wrong for two generations. Almost everyone has been told that lactic acid is sitting in the muscle. It is not, and this is not a trivial correction. If you believe soreness is a waste product to be massaged or stretched out, you will do things that do not work while missing the one thing that does.

It is not lactic acid. That was tested directly

There is one experiment that settles this so neatly it is worth knowing. Schwane and colleagues had the same people run for 45 minutes on a treadmill on two separate occasions. Once on the level. Once on a 10 percent downhill grade. They measured blood lactic acid before and during each run. Soreness and lactic acid were then assessed at intervals for 72 hours afterwards. The results were published in The Physician and Sportsmedicine in 1983 (PMID 27409551).

The outcome was about as clean as sports science gets. Level running raised lactic acid significantly, yet the participants reported no meaningful post-exercise soreness. Downhill running never raised lactic acid at all, yet the participants developed significant delayed onset soreness. The two variables went in opposite directions in the same experiment, in the same people.

So why did the myth stick? Because lactic acid genuinely is involved in a different pain. There are several kinds of exercise-related pain with different time courses and different causes: the pain felt during or right after a session, delayed onset soreness, and cramp. The burning in your quads on the last hill is one thing. The stiffness on Sunday is another. Lactic acid is at least a candidate for the first, never for the second (Miles and Clarkson, Journal of Sports Medicine and Physical Fitness 1994, PMID 7830383).

What happens inside the muscle

Here it is only honest to say that nobody has the full answer. Sweden's national health service, 1177, states plainly that it is not entirely clear why the muscles swell. The major review of the field lists six different proposed explanations and concludes that a combination of two or more is likely needed to account for what happens (Cheung, Hume and Maxwell, Sports Medicine 2003, PMID 12617692).

What is well described is the pattern. Eccentric work induces micro-injury more often and more severely than other types of muscle action. The damage is followed by an inflammatory response, and it is that delay which explains why you feel great on Saturday evening and can barely sit down on Sunday. The soreness does not come alone either: it is accompanied by prolonged loss of force, reduced range of motion at the joint, and raised levels of the muscle enzyme creatine kinase in the blood.

The timeline

Both trained and untrained people get sore. What decides it is not your fitness but how unaccustomed that specific load was for that specific muscle.

Soreness is no proof of a good session

This is the most important point on the page, because it drives behaviour more than anything else here. A great many people use soreness as a gauge: if it hurt afterwards the session was worth something, and if it did not, the session was wasted. That logic leads straight into chasing unfamiliar exercises instead of getting good at a few.

There is a study that measures exactly this. Damas and colleagues followed ten men through ten weeks of resistance training, taking muscle biopsies in week one, week three and week ten. They measured both muscle damage and protein synthesis day by day (The Journal of Physiology 2016, PMID 27219125).

Damage was highest in week one. Protein synthesis was also highest in week one. And neither was related to how much muscle the participants had actually built by the end. Only at three and ten weeks, once damage had subsided and become minimal, did protein synthesis begin to correlate with real growth. The authors conclude that muscle hypertrophy results from accumulated intermittent increases in protein synthesis mainly after a progressive attenuation of muscle damage.

Put practically: the week when everything hurts most is the week that builds least. Your body is still learning the movement. The building comes afterwards, once you can repeat the same thing without being wrecked by it.

The study has ten young participants with a mean age of 27, all men, and rests on biopsies and protein synthesis measurement rather than on direct comparisons between training programmes. It shows a direction, not a verdict. But the direction points against the received wisdom, which is reason enough to stop using soreness as a grade.

Can you train while sore?

Usually yes. The advice is the same from both directions here, which is unusually reassuring for this topic. 1177 says it is good to take it a little easier or to train the muscles that do not hurt, and that light, gentle training can relieve the soreness. The research review says the same thing in other words: anyone who has to train daily should reduce the intensity and duration of exercise for one to two days after a session that caused intense soreness, or alternatively train less affected body parts.

There is also a curious finding in that same review: exercise is the most effective means of relieving pain during soreness. The catch is that the analgesic effect is temporary. Movement does not clear the soreness, it buys you a break from it. That is still enough to make the warm-up feel like a miracle and the cool-down feel like a relapse.

The reason not to go hard with the same muscles, however, is concrete and worth taking seriously. While you are sore, force is reduced and range of motion is restricted, and muscle recruitment patterns shift. Your body solves the task using different muscles than usual, which places unaccustomed stress on tendons and ligaments. That compensation raises the risk of a genuine injury if you return too early and too hard.

What helps, and what only feels like it helps

Here it is possible to be unusually specific, because the area is well studied. The most comprehensive review pooled 99 studies on recovery techniques and ranked them (Dupuy and colleagues, Frontiers in Physiology 2018, PMID 29755363).

The foam roller has had a review of its own: a meta-analysis of 21 studies found the effects on performance and recovery to be rather minor and partly negligible, though rolling after a session did reduce perceived muscle pain. The authors' summary is that the evidence justifies the foam roller as a warm-up activity rather than as a recovery tool (Wiewelhove and colleagues, Frontiers in Physiology 2019, PMID 31024339).

Stretching is the most misunderstood

Cochrane reviewed twelve randomised studies on exactly this question (Herbert, de Noronha and Kamper 2011, PMID 21735398). Stretching before exercise reduced next-day soreness by an average of half a point on a 100-point scale. Stretching afterwards, by about one point. In the single large trial, with 2,377 participants, stretching both before and after reduced peak soreness over a week by four points out of a hundred. That was statistically significant, though the authors immediately point out that the effect is still very small. Their conclusion is that stretching, whether before, after, or both, does not produce clinically important reductions in delayed onset muscle soreness in healthy adults.

1177 lands in the same place and adds a sensible nuance: many people still find it pleasant to warm up before training and to stretch afterwards. Do it if you enjoy it. Just do not expect it to buy you anything on Sunday.

The caveat about cold water

This is a point where the research disagrees with itself, which is worth stating plainly rather than picking a side. The older 2003 review found no effect of cryotherapy on soreness. The 2018 meta-analysis found one. The difference probably lies in how long the cold was applied and how soon after the session.

But there is a reason for caution that matters more than whether the cold relieves anything. A review of cryotherapy notes that the primary demonstrated benefit in humans is reduced pain, while the effects on inflammation and tissue damage shown in animal models lack comparable evidence in humans. The same review notes that chronic use of cryotherapy during resistance training blunts the anabolic training effect, meaning the very adaptation you trained for. Its conclusion is that cold is indicated when rapid recovery is required between bouts, as opposed to after routine training (Kwiecien and McHugh, European Journal of Applied Physiology 2021, PMID 33877402).

So the ice bath after Monday's strength session may well make Tuesday feel better, at the cost of some of what you trained for. For a 45-year-old training to get stronger, that is a bad trade. For someone who has to compete again in 18 hours, it is a good one.

Painkillers: less obvious than it sounds

The question always comes up: can you just take an ibuprofen? The research review is surprisingly lukewarm. Non-steroidal anti-inflammatory drugs have shown dose-dependent effects on soreness, effects that may also be influenced by the timing of administration (Cheung, Hume and Maxwell 2003). It is not a reliable tool, then, but one that works sometimes.

For a reader over 40, something else weighs more. Swedish health service guidance lists a series of situations in which you should not take an NSAID, meaning ibuprofen, naproxen or aspirin, without speaking to a doctor first:

• Illness. Kidney or liver problems, heart failure, a previous heart attack or stroke, lupus, inflammatory bowel disease, asthma or chronic nasal congestion.
• Increased cardiovascular risk from high blood pressure, high blood lipids, diabetes or smoking.
• Being over 75. The risk of stomach ulcers, bleeding, heart failure and kidney failure rises with age.
• Other medicines. Blood thinners, or medication for depression and anxiety such as an SSRI. NSAIDs slow blood clotting, which raises the risk of bleeding.
• Pregnancy. NSAIDs are described as unsuitable during pregnancy. In the first six months you should not take them without speaking to a doctor first, and in the final three months you should not take them at all. Nor should you take them if you are trying to conceive.

If you have or have had a stomach ulcer, you should not take NSAIDs at all. The same applies if you are hypersensitive to aspirin: these drugs closely resemble one another, so ibuprofen or naproxen can set off the same reaction. The guidance says that reaction can be severe, with hives or difficulty breathing, and that it is more common in people who have asthma. Alcohol irritates the stomach lining the same way NSAIDs do, so do not drink while taking them. The list above is a selection from that guidance rather than the whole of it. Ask a pharmacist or check with a clinician before taking an NSAID if you are at all unsure, and read the leaflet in the pack.

An occasional tablet so you can sleep is one thing, provided you know NSAIDs are right for you. Routinely medicating soreness away week after week is another, and that is not a road to take without checking with a pharmacist or a clinician. It is also worth remembering that no tablet makes the muscle ready to be loaded hard again. It removes the signal, not the state.

The one thing that genuinely prevents it

None of the above prevents soreness. Only one thing does, and it is free: having done the movement before.

The phenomenon has its own name in the literature, the repeated bout effect. A single bout of eccentric exercise protects against muscle damage from subsequent eccentric bouts, and the protection is substantial (McHugh, Scandinavian Journal of Medicine and Science in Sports 2003, PMID 12641640). The mechanism is not fully understood, but proposed explanations include the muscle fibre adding sarcomeres in series so that each unit is stretched less, along with adaptations in the inflammatory response. That the effect can be produced even with electrically stimulated contractions suggests most of it happens in the muscle itself rather than in how the nervous system drives it.

The practical translation is simple, and it appears in two entirely different sources. The review of eccentric work concludes that the most useful preventive strategy consists of repeating sessions involving submaximal eccentric contractions whose intensity is progressively increased over the training (Hody and colleagues, Frontiers in Physiology 2019, PMID 31130877). 1177 says it without the jargon: increase training time and load in steps, and train regularly.

It is the same rule that keeps you from breaking down in general. What it looks like in practice when you are starting over is in the guide to starting to run after 40, and for eccentric strength work specifically in the guide to strength training for runners.

What changes after 40

The received wisdom is that soreness gets worse with age. That picture does not hold, and the evidence against it is unusually good.

In 2025 a systematic review with meta-analysis screened 1,092 titles and included 36 studies comparing a younger group against an older one after the same kind of exercise (Fernandes and colleagues, Journal of Aging and Physical Activity 2025, PMID 40174882). The result ran opposite to expectation. Muscle function was affected no differently in older and younger adults. Soreness was consistently greater in the younger group across every comparison. Creatine kinase in the blood was likewise higher in the younger group, both at 24 hours and at peak change. The authors conclude that advancing age is not associated with greater symptoms of exercise-induced muscle damage, and that older adults can therefore pursue regular physical activity without concern for suffering more.

But there is an asymmetry that matters more than the soreness itself, and it only shows up when you look at recovery rather than at symptoms. Chapman and colleagues had ten older men around 64 and ten younger men around 25 perform 30 maximal eccentric arm contractions at high speed (European Journal of Applied Physiology 2008, PMID 18584196). The older men developed less soreness, exactly as in the meta-analysis. But they recovered their strength significantly more slowly. On that point the sources part ways somewhat: the meta-analysis found no difference in muscle function between the age groups at the time points it pooled. The slower return of strength is therefore a finding from individual studies rather than from the pooled evidence.

What holds regardless of which source carries more weight is simpler: less soreness does not mean a faster return. The meta-analysis alone gets you there, because the older group hurt less without their muscle function being any less affected. If the individual studies on strength are right, the picture is sharper still: less pain, longer to come back. Either way, soreness is a poor gauge of how much recovery you need, and it becomes a worse gauge the older you get.

Protection against the next session also appears to work somewhat differently. In one direct comparison, eight older men around 70 and ten younger men around 20 performed the same unaccustomed eccentric arm exercise twice, four weeks apart (Lavender and Nosaka, European Journal of Applied Physiology 2006, PMID 16767435). The younger group showed clear protection going into the second bout across every measure. The older group showed no protection in strength, circumference or creatine kinase, only in range of motion, myoglobin and soreness. In two of those three, range of motion and myoglobin, the protection was also significantly weaker than in the younger men. The authors reason that the protection was partly smaller because the first bout did less damage to begin with, but they also raise the possibility that protection in older adults does not last as long.

That last study is small: eight older participants around 70, all of them men, an isolated arm exercise. The meta-analysis above carries more weight. Read the single study as a possible nuance, not as a rule.

In practice none of this means you should expect more soreness. It means that the protection you have built is the perishable thing to watch. That is why January hurts more than November: your age has not moved in six weeks, your protection has run out. The same logic applies after a holiday or a week with a cold. How quickly the different parts of your fitness fade during a break is covered in the guide to how fast you lose fitness.

What to do after a session that made you sore

1. Do not write off the week. Complete rest is rarely the answer. Movement relieves it, if only temporarily.
2. Switch body part instead of resting. If your legs are wrecked from the lunges, you can still train your upper body perfectly well.
3. Reduce intensity for one to two days for the muscles that hurt, rather than pushing through.
4. Do not stack two heavy days on the same muscles during a sore spell. Force is reduced and the pattern is altered.
5. Next time, take half the dose of anything new. Half the lunges on the first outing. Protection is built by having done the movement, not by how much of it you did.

Fitting strength and running into the same week without one eating the other is its own question, with three ready-made weekly templates in the guide to strength training and running.

When to see a doctor

Ordinary soreness is harmless and resolves on its own. There are still a few situations where you should check with a clinician rather than wait it out.

• The problem has worsened or has not cleared within a week. Contact a doctor. This is 1177's explicit threshold for exercise soreness.
• You have significant pain and feel swollen in, for example, an arm or a leg after training. Contact a doctor. 1177 gives this advice in the context of compartment syndrome, where a swollen muscle runs out of room inside its compartment.
• Dark or reddish-brown urine together with severe muscle pain and weakness after an unusually hard session. This is uncommon, but it can indicate muscle tissue breaking down on a scale that burdens the kidneys. Seek care the same day.
• The pain sits in a joint or tendon, arrived during the movement itself, is on one side only, or comes with numbness or weakness. Then it is not ordinary soreness.

This page is general health information, not medical advice, and it does not replace an assessment by a clinician.

Frequently asked questions

Is muscle soreness caused by lactic acid?

No. This was tested head-on in 1983 by Schwane and colleagues. The same people ran for 45 minutes on a treadmill twice: once on the level, once on a 10 percent downhill grade. Level running raised blood lactic acid significantly, yet produced no meaningful soreness. Downhill running never raised lactic acid at all, yet produced significant delayed onset soreness. The conclusion was that lactic acid is not related to delayed onset muscle soreness. Lactic acid does belong to a different kind of pain: the burn you feel during the effort itself, not the stiffness that arrives a day later.

How long does muscle soreness last?

It usually peaks one to two days after the muscles were loaded and is normally gone within a week, according to Sweden's national health service, 1177. The research literature describes the same course: soreness develops 24 to 48 hours after a session with a lot of braking, or eccentric, muscle work. If the problem gets worse or has not cleared within a week, 1177 advises contacting a doctor.

Does soreness mean the workout was good?

No. Soreness tells you something was unfamiliar, not that something was effective. In a study following ten men through ten weeks of resistance training, muscle damage was highest in the first week, and neither the damage nor the protein synthesis at that point was related to how much muscle the participants actually built. Only once the damage had subsided, at three and ten weeks, did protein synthesis start to correlate with growth. The muscle grew most once it stopped hurting. The study is small and was done in young men, so read it as a direction rather than a verdict.

Can I train with muscle soreness?

Usually yes, but not hard with the same muscles. Both 1177 and the research reviews say the same thing: light, gentle movement relieves it. The recommendation in the literature is to reduce intensity and duration for one to two days, or to train body parts that are not affected. The reason not to go hard is concrete: while you are sore, force production is reduced and your movement pattern changes, which places unaccustomed stress on tendons and ligaments. If the pain is sharp, one-sided and sits in a joint or tendon, it is not ordinary soreness, and different rules apply.

Does muscle soreness get worse as you age?

No. A 2025 systematic review with meta-analysis went through 36 studies comparing younger with older adults after the same kind of exercise. Soreness was consistently greater in the younger group across every comparison, muscle function was affected no differently, and creatine kinase in the blood was likewise higher in the younger group. The conclusion is that advancing age is not associated with greater symptoms of exercise-induced muscle damage. Individual studies do suggest, however, that the recovery of strength slows with age: in a comparison between older and younger men, the older group developed less soreness but recovered their strength significantly more slowly. Less pain, longer to come back.