← Longevity & Supplement Guides
Is Soreness a Good Sign? What Aching Muscles Actually Measure
There is a particular kind of self-assessment that happens on the second morning after a hard session, somewhere around the third stair. If the legs complain, the session counted. If they feel fine, something must have been missing.
It is one of the most durable beliefs in training culture, and it is close to backwards. Soreness is real and measurable, with a published timecourse. It is simply not a scoreboard. What it mostly reports is novelty — how unfamiliar the movement was, not how much good it did.
What soreness actually is
Delayed onset muscle soreness — DOMS, in the literature — turns up 12 to 24 hours after unaccustomed work, peaks inside the first two days, and fades over the following three. It is provoked most reliably by eccentric contractions: the lowering half of a lift, the downhill mile, any moment when a muscle is lengthening while still bearing load. That is also, awkwardly, the half of the movement that does a disproportionate share of the adaptive work.
It is not lactic acid. That explanation died decades ago and refuses to stay buried. Lactate clears within roughly an hour of finishing; the ache arrives long after it has gone. A 2025 scoping review in the Journal of Functional Morphology and Kinesiology, which gathered 25 clinical studies published between 2020 and 2025, describes what higher-resolution imaging now shows instead: microstructural disturbance in the muscle and connective tissue, alongside a nervous system that has turned up the gain on ordinary sensation. Mechanical disruption plus sensitisation — not a chemical hangover.
Which raises the obvious question. If soreness marks disruption, and disruption is where adaptation starts, isn't the ache a reasonable proxy for progress?
The study that pulled soreness and growth apart
The cleanest answer came from Felipe Damas, Stuart Phillips, Cleiton Libardi and colleagues in The Journal of Physiology in 2016 — a small study that quietly reorganised how the field thinks about muscle damage.
Ten untrained young men began a ten-week resistance programme. Biopsies were taken before, 24 hours after and 48 hours after a session, at three points: the first week, three weeks, and ten weeks. Protein synthesis was measured with heavy water, and the tissue itself was inspected for the signature of damage.
The pattern is the interesting part. Damage was highest in week one, lower at three weeks, minimal by ten. Protein synthesis was also highest in week one — the most brutal phase produced the biggest apparent response. And yet fibre size had not budged. Growth showed up only at ten weeks.
More pointedly: in week one, neither damage nor the protein-synthesis spike was related to how much muscle a man eventually gained. At three and ten weeks, once damage had faded, protein synthesis tracked growth closely (correlations around 0.9). The early, sorest response was mostly repair. The later, quieter one was construction.
Ten men is ten men, and this is mechanism rather than outcome. But it inverts the folk model rather neatly: the period when you are most wrecked is the period when the least building is going on.
What soreness does measure: how new the movement was
Do the same damaging session again a few weeks later and it hurts markedly less. This is the repeated bout effect, one of the most reliable phenomena in exercise physiology, and it is the reason your first week back always feels like evidence of decline.
In January 2026, Oliver Hayman, Paul Ansdell, Glyn Howatson and colleagues asked in the Journal of Applied Physiology where that protection comes from. Twenty-three participants performed ten sets of ten eccentric contractions with the muscles that lift the foot, then repeated the identical session three weeks later — the same total work both times, around 1,300 joules. The second time, strength loss was smaller and soreness lower.
What had changed was partly in the nervous system. Using high-density electromyography to track individual motor units, the researchers found less variability in how those units fired during the second bout, and a smaller rise in firing rate. The muscle had not simply grown a thicker skin; the instructions sent to it had become steadier.
The authors are careful about what this can and cannot show — the neural changes may protect the muscle, or may simply reflect a muscle that was less disturbed in the first place. Both readings fit. Either way, the practical reading is the same: a large part of "unaccustomed" lives in coordination, and it is spent after one or two exposures. Which means soreness is loudest exactly when your training is newest, and goes quiet as the training gets good.
The ache and the weakness are not on the same clock
Soreness does carry a genuine cost, but it is not the cost people track. A 2026 meta-analysis in the European Journal of Pain by Jacques Abboud's group at Trois-Rivières pooled 18 studies and 452 healthy adults in whom soreness had been deliberately induced in the low-back muscles — not as a training study, but to use soreness as a controlled, non-invasive model of movement-evoked pain.
The pooled timecourse is worth knowing. Soreness was worst on day one and had largely receded by day four. Maximal force, however, bottomed out on day two and was still measurably down on day three. Range of motion barely changed at all.
Figure — recovery
Soreness peaks on day one. Your strength bottoms out on day two.
Pooled standardised mean differences from Ducas et al., European Journal of Pain (2026): 18 studies, 452 healthy adults with experimentally induced soreness in the low-back muscles. Both series come from the same meta-analysis and the same scale, so they are comparable; force values are reductions in maximal voluntary contraction, plotted as magnitude. Range of motion showed no significant change on any day.
Two useful things fall out of that gap. The first is that feeling recovered and being recovered part company around day two — the ache is fading while your force production is at its lowest. The second is that the honest limitation matters: this was low-back soreness induced for pain research, in a lab, in healthy volunteers. It describes the shape of the response, not a personal prescription. But the shape is consistent with what every coach eventually learns by accident.
What helps, and how much
The market for soreness relief is enormous; the evidence behind it is smaller and more specific than the marketing suggests. A 2025 Bayesian network meta-analysis in the Journal of Pain Research compared physical modalities across randomised trials and found one consistent thing: whatever benefit exists lives almost entirely inside the first 48 hours. Light-based therapy and heat produced the clearest short-term effects on reported soreness; past two days the advantage largely dissolved. The scoping review reached a similar verdict — vibration, percussive massage, heat and cold, and some polyphenol-rich foods each showed modest symptom benefit, and combinations beat any single method.
Note what is being measured, though. Every one of these outcomes is comfort: pain scores, perceived recovery, sometimes short-term force. None of them demonstrates faster adaptation. And in at least one case there is a real trade-off in the other direction — cold water immersion after training is good at making you feel better and, when used habitually after strength work, appears to blunt some of the response you were training for. If you have a match tomorrow, comfort is the goal. If you have a training block, it may not be.
The unglamorous options remain the strongest: sleep, food, easy movement, and time. Heat is a reasonable and pleasant addition — see what the sauna studies actually measured — as long as you are honest that it is buying comfort rather than adaptation. Our rest and recovery range covers the nutrition side of the same window, and the longevity protocol hub shows where recovery sits in the rest of the system.
How to use soreness this week
Soreness is information. It is just answering a different question than the one people ask it.
- Read it as a novelty signal, not a quality signal. Very sore after a session usually means the movement, the range or the eccentric load was new. That is worth knowing, and it will fade whether or not the training was any good.
- Expect day two to be your weakest, not your sorest. If you plan a heavy or technical session, put it on day one or day three — not on the day your force production is quietly at the bottom.
- Do not chase the ache. Adding volume purely to feel it tomorrow buys repair, not growth. The Damas data suggests you would be trading building for patching.
- Introduce new movements at a fraction of the dose. One or two exposures buys you most of the protection. A short, easy first session of anything unfamiliar costs almost nothing and saves the week.
- Judge a programme on strength, capacity and consistency — the kind of thing a sensible strength dose is designed around — rather than on how the stairs feel on Tuesday.
- Keep moving gently while sore. Range of motion holds up even when force does not, so walking, easy cycling and mobility work are usually available.
- Watch the trend, not the day. A steadily elevated resting heart rate and a flat or falling heart-rate variability across a week say more about accumulated fatigue than soreness ever will. The Agen Band and the Agen app exist for exactly that timescale, and deciding when to push or rest is a weekly judgement, not a morning one.
Two flags worth taking seriously: soreness that is severe, unusually prolonged, accompanied by dark urine or significant swelling is not ordinary DOMS and warrants a doctor rather than a foam roller. And sharp, localised pain during a movement is a different animal from diffuse ache the next day — tendons in particular adapt on a slower clock than muscle and complain in their own idiom.
The bottom line
Soreness measures unfamiliarity. It peaks on the first day, costs you the most force on the second, and shrinks on repeat exposure to the same session — partly because your nervous system learns the movement. The training phase in which you ache the most is, on the best mechanistic evidence available, the phase in which you are building the least.
Nothing on the relief menu has been shown to speed up adaptation; the good options buy comfort inside a 48-hour window, which is a legitimate thing to buy and a poor thing to confuse with progress. Sleep, food, easy movement and time still do the heavy lifting, and consistency over months does more than any single session that hurt.
Use the numbers to correct the fantasy, not to replace the experience. The stairs are telling you something. They are just not keeping score.
