Best Supplements for Muscle Recovery: The One That Came First Is Not a Supplement
In 2018 a French research group did the thing nobody selling recovery products wants done. They pooled 99 studies of the common post-exercise recovery methods — cold water, contrast baths, cryotherapy, compression garments, electrical stimulation, stretching, massage — and set the results side by side. One came out on top for soreness and fatigue. It was massage, and you cannot buy it in a tub. It was a line-up, not a contest: each method was measured against doing nothing, never against the others, and compression boots were not in it. [the full verdict]
You came here for a supplement. We have read the trials behind 15 of the things sold for recovery — powders, capsules, boots, tubs and one checklist — one at a time. This page is the running scoreboard.
Nothing here is for sale and nothing links to a shop. Every row goes to the verdict it came from, so you can check the working.
| What | Verdict | What we measured |
| Protein Two-thirds of a pound of extra lean mass in people who lift, over about thirteen weeks. Without lifting or dieting, no clear gain has been shown. | established | 0.66 lb of extra fat-free mass in people who lift: 26 comparisons, 1,089 people, in a 2018 review |
| Creatine One of two supplements out of 250 that beat lifting alone. What that screen measured was lean mass and strength. | established | 2 of 250 supplements that beat resistance training alone. Creatine was one. |
| What | Verdict | What we measured |
| Massage First for soreness and fatigue when a 2018 review set the methods in 99 studies side by side. A 2020 review of 29 studies found no clear effect on strength, jump, sprint or endurance, which is not the same as none. Both are on the page. | supported | 1,012 people in 29 studies in the 2020 sports massage review: small gains in soreness and flexibility, no clear effect on strength, jump, sprint or endurance |
| Ice baths Pooled trials find soreness eased for up to four days. Cold has not clearly beaten warm water, and used straight after lifting it can blunt the muscle growth you did the lifting for. | supported | 96 hours after muscle-damaging exercise, soreness was still lower with cold water, across 8 pooled studies |
| What | Verdict | What we measured |
| Magnesium Supported — for blood pressure, about 2 mmHg across 34 double-blind trials. For older adults’ night leg cramps, Cochrane pooled five trials and found no effect; it found no trial at all of cramps from exercise. | supported | 2.0 mmHg off systolic blood pressure, across 34 double-blind trials |
| Cold plunges, for fat The same tub as the ice-bath row above, sold for a second job. At the very most 188 calories a day, and only while you are cold. | unsupported | 188 calories a day, at most |
| Lymphatic drainage massage Unsupported — for detox, debloating and a slimmer face. After hard exercise it has been tried in seven small experiments: lower blood markers than rest in five, and no better than an ordinary massage in the one that tried both. In lymphedema the bandaging did 30 to 38.6% and the massage added 7.11% in 83 women; six later trials found it added nothing. | unsupported | 7.11% added in 83 women. Six trials since found none |
| Tart cherry juice Unsupported — for sleep: the famous 84-minute result came from eight people, and ten pooled trials show no clear effect. On recovery, five reviews each found some part of it helped, split on which part, on low-certainty evidence; rated alone, that half would be Preliminary. | unsupported | 8 people the 84-minute result came from |
| What | Verdict | What we measured |
| Compression boots In pooled trials, a little less soreness and nothing significant for strength or power, an effect on muscle function the 2024 review calls trivial to small. Most of the trials had no dummy treatment to compare against. | preliminary | 319 people in the 17 trials one 2024 review accepted: boots against rest or a dummy treatment, published to November 2022 |
| Blue light, and deep sleep The mechanism from evening light to disrupted sleep is real and measurable. The step from there to your recovery is the one nobody has shown. | preliminary | 1 in 5 studies finding sleep quality decreased, in the cited review |
| Your cortisol rhythm Flatter daily cortisol went with poorer health in 10 of 12 kinds of outcome across 80 studies, at an average correlation of 0.147 — small, and its authors leave open whether it is a marker, a cause or both. | preliminary | 0.147 the average effect size behind the cortisol-health link |
| What | Verdict | What we measured |
| Electrolytes Nine studies pooled in 2026 found no consistent link between cramp and blood sodium. In the one small trial that worked a muscle until it cramped, no fewer men cramped with the drink than with no fluid. | unsupported | 9 studies pooled in 2026 — no consistent link between cramp and blood sodium or fluid loss |
| Magnesium glycinate Sold on absorption. The trial built to compare it with the cheap oxide came out even, in twelve people with surgically shortened bowels, and two single-dose studies in healthy adults, each from the maker of a rival product, did not put it ahead either. | unsupported | 4 patients the absorption advantage came from |
| Sleep hygiene The sleep advice everyone knows scored 1.01 inside the treatment that works, where 1.00 is nothing. The insomnia guideline advises doctors against using it on its own. | unsupported | 1.01 what sleep hygiene adds inside CBT-I. 1.00 is nothing |
| Overtraining, as a score Not a product — the question underneath the whole aisle. The two professional bodies who wrote the consensus statement found no marker good enough to be generally accepted. So any quiz that hands you a number invented the number. | unsupported | 0 accepted markers, out of every one reviewed |
Established means multiple good studies agree: treat it as settled. Supported means the evidence is consistent, with known limits. Preliminary means the evidence is early, small, conflicting or from animals only: not enough to act on alone. Unsupported means we looked and the evidence does not back the claim: either it was tested and did not hold up, or nobody has tested it and nothing that was tested supports it.
Now look at the shape of it. Nothing you swallow reached the top section on a recovery claim. The two Established rows got there for building muscle, which is a different question asked at a different time. The method that came first in a 2018 line-up of 99 studies is a pair of hands, and the other Supported row is a tub of cold water that, used straight after lifting, can blunt the muscle growth you trained for.
That is not us being difficult. It is what a recovery shelf looks like once you sort it by evidence instead of by how confidently it is advertised.
Worth saying plainly: if soreness is severe, does not fade after a few days, or arrives with symptoms that feel wrong to you, none of this is your answer. That is a clinician, not a shelf.
The same method applied elsewhere: supplements for muscle growth, for men, for women and for gut health, and for weight loss, where nothing reaches Established for weight loss, and for sleep, where exactly one entry is Supported and what it buys is measured in minutes, and for sweeteners, where every claim is a worry and the ratings run backwards.
The four ratings are the same ones used on every verdict here, and we wrote down how we read a study so you can see what earns each one.