The short version
- A 2025 umbrella review of 29 systematic reviews and 863 randomised trials on muscle soreness rated 17 of those reviews critically low quality, so most soreness advice rests on weak evidence.
- Massage was the most effective single technique for reducing delayed onset muscle soreness and perceived fatigue in a 2018 meta-analysis of 99 studies.
- Post-exercise cold water immersion reduces soreness, but in a 12-week trial of 21 men, strength and muscle mass increased more with active recovery than with cold, and the authors advised reconsidering routine use.
- Protein around training preserved maximal voluntary contraction for up to 96 hours in a meta-analysis of 29 studies but had no effect on muscle soreness.
- The 2025 umbrella review classed the evidence for stretching, exercise and electrical stimulation as weak for reducing delayed onset muscle soreness.
Soreness is a sensation. Muscle damage is an event. Being ready to train again is a third thing. We collapse all three into one word — sore — and then wonder why the research keeps producing results that look contradictory. Usually they are not contradictory. They just measured different ones.
Hold that distinction and the evidence on the four things people actually reach for gets a lot easier to read.
The size of the literature, and then the quality of it
In 2025, Wiecha and colleagues published an umbrella review in Sports Medicine — a systematic review of systematic reviews — on physical therapies for delayed onset muscle soreness. It swept up 29 systematic reviews containing 863 unique randomised controlled trials across 24 distinct treatments.
Then it graded them, and this is the number to sit with: 17 of the 29 reviews were of critically low methodological quality. Two were rated high. Pooled effect sizes ran from about 0.36 for cooling therapy to 1.82 for heat therapy — small to large. The review found some strong evidence for cooling, cryostimulation, contrast therapy, massage, phototherapy and kinesiotaping at various follow-up points, and described the evidence for stretching, exercise and electrical stimulation as weak.
Stretching after training is the most widely repeated soreness advice in the gym. It is in the weak column.
Massage keeps winning, which is inconvenient
Dupuy and colleagues pooled 99 studies in Frontiers in Physiology in 2018. Active recovery, massage, compression garments, immersion, contrast water therapy and cryotherapy all produced a small-to-large reduction in soreness (Hedges' g between −2.26 and −0.40); nothing else moved it. Massage came out as the most powerful single technique for both soreness and perceived fatigue, and alongside cold exposure it was the most effective for inflammation. Across the whole pooled dataset, creatine kinase fell by a standardised mean difference of −0.37, interleukin-6 by −0.36 and C-reactive protein by −0.38.
It is the least scalable answer on the list and the best supported one. We are not going to pretend otherwise because we do not sell it.
The cold plunge works. That is exactly the problem.
Cold water immersion reliably reduces how sore you feel. The question is what else it does.
Roberts and colleagues ran two studies in The Journal of Physiology in 2015. In the first, 21 physically active men strength trained twice a week for 12 weeks, each session followed by either 10 minutes of cold water immersion or active recovery. Strength and muscle mass increased more in the active recovery group. Isokinetic work rose 19%, type II fibre cross-sectional area 17% and myonuclei per fibre 26% in the active recovery arm — and not in the cold arm. In the second study, satellite cell numbers and p70S6 kinase phosphorylation after exercise were both greater with active recovery than with cold. The authors' own closing line is that using cold water immersion as a regular post-exercise recovery strategy should be reconsidered.
Note the word regular. If you have a second match in six hours, feeling less wrecked is the goal and cold is a reasonable tool. If you are in a block whose entire purpose is to make muscle, you are buying comfort with the adaptation you came for.
Protein does something real, and it is not the ache
Pearson, Hind and Macnaughton pooled 29 studies and 40 trials in the European Journal of Clinical Nutrition in 2023. Protein around training preserved isometric maximal voluntary contraction at 96 hours (effect size 0.563, 95% CI 0.232 to 0.894) and isokinetic contraction at 24, 48 and 72 hours. Creatine kinase was lower at 72 hours (1.335, 95% CI 0.294 to 2.376). And then their flat conclusion: protein had no effect on muscle soreness compared with control.
That is the dissociation in a single paper. Strength came back faster. The ache did not care.
Sleep, without the usual oversell
Bonnar and colleagues reviewed sleep interventions in competitive athletes in Sports Medicine in 2018. Ten studies. 218 participants, all aged 18 to 24. Of the approaches tried — sleep extension, napping, sleep hygiene, post-exercise strategies — sleep extension had the most beneficial effects on subsequent performance. Their summary sentence is the one worth keeping: sleep plays an important role in some, but not all, aspects of performance and recovery.
Ten studies in young athletes is not a mandate. It is also more evidence than most things in the recovery aisle have, and unlike the rest of this list it is free.
What we'd actually tell you
Soreness is not a scoreboard, and reducing it is not the same as recovering. If you want to feel better, massage has the best data and warmth is cheap. If you want to come back stronger, protein and sleep are doing more for you than anything you can sit in.
Save the ice for the days when the next performance matters more than the next adaptation, and keep it away from your hypertrophy blocks. Skip the post-session stretching if you are only doing it for soreness; do it if you want the range of motion.
And the boring safety line, which we mean: soreness that keeps getting worse after four or five days, severe swelling, or dark-coloured urine after an unusually hard session is not a recovery question. That is a same-day call to a clinician.
Good questions
Does stretching after a workout stop you getting sore?
No. The 2025 Sports Medicine umbrella review of 29 systematic reviews classed stretching among the treatments with weak evidence for delayed onset muscle soreness. Stretch if you want range of motion or because it feels good, but it is not a soreness treatment. Massage, compression and water immersion all have better data behind them for the ache itself.
Should I take an ice bath after lifting?
Not as a routine habit if muscle growth is the point. Cold water immersion does reduce soreness, but in a 12-week trial, strength and muscle mass increased more in the group doing active recovery instead, and type II fibre area rose only in that group. Save cold for days when the next performance matters more than the next adaptation.
How long is it normal to be sore after training?
Delayed onset soreness typically peaks a day or two after an unaccustomed session and eases over the following days. The research on it measures effects at 24, 48, 72 and 96 hours for that reason. Soreness that keeps worsening past four or five days, severe swelling, or dark urine after a hard session is a same-day clinician call, not a recovery question.
Is being sore a sign the workout worked?
No. Soreness, muscle damage and readiness to train again are three separate things that studies measure separately, and they often disagree. In one meta-analysis, protein supplementation restored strength faster while doing nothing at all to soreness. You can have a productive session with no soreness and a soreness-heavy session that mostly reflects doing something unfamiliar.
Are massage guns as good as an actual massage?
Unknown, and we would rather say that than sell you one. The meta-analysis that put massage at the top of the list pooled trials of hands-on massage, not percussive devices. A gun may well help; it has simply not been tested to the same standard. If the device makes you more likely to do it at all, that is a reasonable argument for it.
Will taking protein powder make me less sore?
No, and the evidence on that is fairly clear. A 2023 meta-analysis of 29 studies found peri-exercise protein preserved maximal voluntary contraction and lowered creatine kinase, but had no effect on soreness versus control. Protein helps you get your strength back sooner. It does not make the ache go away, and any brand promising that is going past the data.
Sources
- Wiecha S, Cieśliński I, Wiśniowski P, et al. Physical Therapies for Delayed-Onset Muscle Soreness: An Umbrella and Mapping Systematic Review with Meta-meta-analysis. Sports Med, 2025. View study
- Dupuy O, Douzi W, Theurot D, Bosquet L, Dugué B. An Evidence-Based Approach for Choosing Post-exercise Recovery Techniques to Reduce Markers of Muscle Damage, Soreness, Fatigue, and Inflammation: A Systematic Review With Meta-Analysis. Front Physiol, 2018. View study
- Roberts LA, Raastad T, Markworth JF, et al. Post-exercise cold water immersion attenuates acute anabolic signalling and long-term adaptations in muscle to strength training. J Physiol, 2015. View study
- Pearson AG, Hind K, Macnaughton LS. The impact of dietary protein supplementation on recovery from resistance exercise-induced muscle damage: A systematic review with meta-analysis. Eur J Clin Nutr, 2023. View study
- Bonnar D, Bartel K, Kakoschke N, Lang C. Sleep Interventions Designed to Improve Athletic Performance and Recovery: A Systematic Review of Current Approaches. Sports Med, 2018. View study
These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease.
