Stretching for DOMS is one of the few claims in this subject that has been settled properly, and the answer is that it does almost nothing. Cochrane pooled the trials of stretching for delayed onset muscle soreness and found an effect of the order of one to two points on a hundred-point soreness scale. That is not a small positive result to be built on. It is a difference nobody could feel.
This page gives the review, what delayed soreness appears to be, why the habit still feels worth keeping, and what a mobility app is entitled to say about recovery — which in our case is nothing.
The short answer
- The pooled effect of stretching on post-exercise soreness is about one to two points on a hundred.
- The evidence is good, not thin. This is a clear answer, not an uncertain one.
- Stretching before exercise does not help either.
- Soreness and stiffness are different complaints and only one of them responds to stretching.
- Limber has no recovery feature, no soreness tracking and no way to see that you trained.
- Stretch after a hard session if you like it. Just do not do it for the soreness.
What DOMS actually is
Delayed onset muscle soreness is the ache that arrives a day or two after unfamiliar work, peaks somewhere around 24 to 72 hours, and fades on its own. It follows eccentric loading in particular — the lowering half of a movement, downhill running, a long descent, the first session back after a break.
What it is not is an injury, and what it is not is stiffness. Those two get conflated constantly, and the conflation is the reason the stretching claim sounded plausible in the first place. A sore muscle also feels tight, so stretching it feels like the obvious response. Feeling tight and being short are different states, and the research on the second one does not transfer to the first.
Soreness that is sharp rather than dull, that is on one side only after a specific movement, that comes with swelling or bruising, that does not settle within about a week, or that comes with weakness is a different matter and belongs with a clinician who can examine you. Nothing here diagnoses anything.
Stretching for DOMS, reviewed
Herbert, de Noronha and Kamper produced the Cochrane review, updated in 2011, pooling the randomised and quasi-randomised trials of stretching before, after, or before and after exercise, with muscle soreness as the outcome.
The effect was tiny. On a hundred-point soreness scale, the pooled difference sat at one to two points.
Herbert and Gabriel had reached the same conclusion in the BMJ in 2002, reviewing stretching around exercise for both soreness and injury risk and finding no useful effect on either.
Grade: solid, and disappointing. This is a well-conducted review of a well-studied question, and it produced a clear answer rather than an ambiguous one. It is not a case of "more research needed" — the research was done, and it came back close to zero.
Print that at full strength before any softening, because it is against the interest of a page like this one and the category tends to bury it in a paragraph about how everyone is different.
One to two points on a hundred
It is worth sitting with the size of that number.
A hundred-point visual scale is the standard way soreness is measured: mark on a line where your legs sit between nothing at all and the worst you can imagine. The day-to-day wobble in how any one person marks that line, on the same amount of soreness, is far larger than two points. Two points is inside the noise of the instrument.
So the honest translation is not "stretching helps a little". It is "stretching does not measurably change how sore you are, and the trials were large enough and numerous enough to say so".
The same applies to the timing dodge. Stretching before exercise, after exercise, or both were all covered, and none of them produced a useful effect. Moving the routine to a different point in the day is not the missing variable. Stretching before or after a workout covers what the ordering does change.
Why it still feels better
Two things are true at once, and the second one is not a consolation prize.
The feeling is real. Lying on the floor at the end of a hard session with a long breath out is pleasant, and it lowers something, and a person who does it usually reports feeling better afterwards. Nobody is imagining that.
The measurement does not move. What the trials show is that when you come back the next morning and mark the line, the stretchers and the non-stretchers mark it in the same place.
Both can be true because the immediate feeling and the next-day soreness are different outcomes. A stretch changes how a muscle feels to move through its range in the moment — Konrad and Tilp confirmed in Clinical Biomechanics in 2014 that six weeks of stretching raises range with no detectable change in muscle or tendon, and Weppler and Magnusson set out in Physical Therapy in 2010 the case that what changes first is tolerance rather than tissue. Grade: reasonably solid. Tolerance is a real thing to change. It is not the same thing as being less sore tomorrow.
So keep the habit if you like it. Just stop attaching a claim to it that the trials will not carry.
What we do not know
Three honest gaps, none of which rescues the finding.
Nobody has tested a serious dose. The trials used ordinary stretching around a session. Whether a much larger volume would do something is unstudied rather than refuted, and there is no reason from the mechanism to expect it would.
Soreness scales are crude. They measure one dimension of a complicated experience. A real effect on something else — how willing you feel to train the next day, say — could exist and would not appear.
The literature is mostly young healthy adults. As almost all of this literature is. Bandy and Irion's dose work in Physical Therapy in 1994, the 1997 follow-up on frequency, and Thomas and colleagues' 2018 pooling of total weekly time in the International Journal of Sports Medicine all share that population and that limit.
None of those gaps is a reason to keep making the claim. They are reasons to state it precisely: for the intervention that was tested, in the people who were tested, the effect was practically nothing.
What the app claims about it
Nothing, and the reason is worth stating.
Limber has no recovery feature. There is no soreness tracking, no readiness reading, no wearable input and no recovery score. The only place the app asks about how a body feels is an onboarding step that offers three sore areas — the lower back, the neck and the knees — and what it writes are contraindication flags. Those flags remove matching movements from the pool before a routine is generated rather than ranking them lower inside it. An exclusion rule is not a recovery protocol.
Limber cannot see the session that made you sore. Health data goes one direction: the app writes a finished session and never reads a workout back. The engine holds a designed rule that a hard session should lead the evening with the areas it loaded, inside an 18-hour window. It has never had an input.
And it does not claim otherwise. No screen, notification or page in Limber says a routine reduces soreness, speeds recovery or prevents injury. Lauersen, Bertelsen and Andersen's 2014 meta-analysis in the British Journal of Sports Medicine found strength training reduced sports injuries to under a third while stretching showed no significant protective effect, and does stretching prevent injury is the page that prints it in full.
What the app does is measure range and plan range work. The toe touch test reads the posterior chain in five bands, and the Range Score composes what you have measured across four of five stations, leaving the fifth blank because the test that would feed it is not built.
Where that leaves the habit
Somewhere honest and rather pleasant.
Stretch after a hard session because you want the range, because the end of a session is a slot with nothing competing for it, and because five minutes on the floor with a long breath out is a good way to finish. Those reasons stand on their own and do not need a trial.
Then set your expectations by the evidence. Thirty seconds a hold is the supported floor, most days beats one long session, and total weekly time is what tracks the gain. Soreness will do what it was always going to do, on its own schedule, which is one to three days.
Nuzzo argued in Sports Medicine in 2020, in The Case for Retiring Flexibility as a Major Component of Physical Fitness, that the category claims more than it can support. Grade: a serious minority position, well argued. Pages like this one exist partly to make sure he is arguing against an honest version of us.
Post-run stretches is the routine most people arrive here looking for, rest day mobility is the slot where holds cost nothing, and the cluster sits under mobility for training.
Questions
Does stretching help sore muscles?
Barely. Cochrane's pooled review of trials of stretching for post-exercise muscle soreness found an effect of about one to two points on a hundred-point scale, which is smaller than the day-to-day variation in how anyone marks that scale. The evidence here is good rather than thin, so this is a clear answer and not an open matter.
Should I stretch before or after exercise to reduce DOMS?
Neither works. The pooled review covered stretching before exercise, after exercise, and both, and none of the three produced a useful effect on soreness. If your reason for stretching is tomorrow's stairs, the honest answer is that this is not the lever, and the ordering of your routine will not rescue it.
Why do my muscles feel tighter when they are sore?
Because feeling tight and being short are different states. A sore muscle reports tightness through sensation rather than through any change in its length, which is why stretching it feels like the obvious response and why the trials found no benefit. What stretching reliably changes over weeks is how much range you tolerate, not how sore you are.
Is there anything that does help DOMS?
Not within what this site is willing to claim. The sanctioned evidence here covers stretching, and the answer for stretching is close to zero; recovery modalities in general are outside what these pages cover, so asserting anything about them would be borrowing authority we do not have. Delayed soreness resolves on its own in roughly one to three days.
Should I train again while I am still sore?
That is not something a web page can decide for you. What can be said is that ordinary delayed soreness is not an injury and usually settles within a few days, while soreness that is sharp, one-sided, swollen, weak or still there after a week belongs with a clinician who can examine you. Anything that hurts in a way you would not describe as a dull ache deserves a look rather than a routine.
Does Limber track soreness or recovery?
No. There is no soreness input, no readiness reading, no wearable connection and no recovery score anywhere in the app, and it writes a finished session to Health without ever reading a workout back. The only thing it does with how a body feels is an onboarding step for three sore areas, which writes flags that remove matching movements from a routine before it is built.