Protein and Recovery: What Actually Helps After Training

For twenty years the advice was to get protein in within thirty minutes of finishing training or waste the session. That idea has not survived closer examination, and the correction is worth understanding — because the things that do matter got overlooked while everyone was rushing to the changing room with a shaker.

First, decide what you mean by “recovery”

Almost every argument about recovery is really three different questions wearing one word:

Recovery of function — can you produce force again tomorrow?
Recovery of feeling — how sore are you?
Adaptation — do you get bigger, stronger or fitter over months?

These come apart, and a 2014 systematic review in Sports Medicine put it about as bluntly as academic writing gets: high quality and consistent data demonstrate no apparent relationship between recovery of muscle function and ratings of muscle soreness.

Soreness is not a readout of readiness. And, as the cold-water section below shows, an intervention can improve how you feel while actively degrading what you trained for. Keep the three apart and most of the confusion in this field disappears.

The window is much wider than advertised

Muscle remains sensitised to protein for considerably longer than thirty minutes after a training session — on the order of 24 hours, with the strongest response in the first few. A single meal missed at the “wrong” moment does not undo a session.

The direct test was done in 2013: a meta-analysis of 23 studies found a small-to-moderate hypertrophy effect from protein timing that disappeared once total daily protein intake was controlled for. The authors’ conclusion was that the results refute the common belief that timing in the immediate pre- and post-workout period is critical to muscular adaptation.

We sell whey, so it is worth saying plainly that we can retire the anabolic window and still recommend whey without contradiction. Whey’s genuine advantages are convenience, leucine density and fast digestion. None of them require a thirty-minute deadline, and the deadline was never real.

There is one exception worth keeping. If you train fasted — first thing, nothing since dinner — then you have been in a net breakdown state for a while, and eating reasonably soon afterwards does matter more.

What protein does after a session, and what it does not

Training does two things at once: it damages muscle fibres slightly and it increases their sensitivity to amino acids. Protein afterwards supplies the material for repair and, over months, adaptation.

A dose of roughly 0.3 to 0.4 grams per kilogram of body weight post-training is sufficient — about 25 to 35 grams for most adults. Beyond that, the extra protein is not building more muscle; it is being used for something else. There is no benefit to a 60-gram post-workout shake. Our piece on how much protein you actually need per day covers the daily side of this.

Here is the part that will annoy the marketing department, ours included: protein does not reliably reduce muscle soreness.

That 2014 review of 27 studies found insufficient experimental data that post-exercise protein attenuates soreness or lowers markers of muscle damage, and concluded that acute increases in muscle protein synthesis have not translated into measurable reductions in damage or enhanced recovery of function.

A 2018 meta-analysis in Nutrients looked specifically at whey across 13 randomised trials and 152 subjects. Recovery of contractile function did favour whey at every time point, with small-to-moderate effect sizes of 0.4 to 0.7 out to 96 hours. But it stated there was no definitive evidence that whey reduced markers of exercise-induced muscle damage or soreness. That study was supported by a dairy-industry-linked research centre, which we mention because you should know — and because a dairy-funded analysis reporting a null on soreness is more persuasive, not less.

So the honest synthesis: protein modestly speeds the return of force production. It does not reliably reduce soreness and it does not reliably lower creatine kinase. And total daily intake matters more than any single post-workout dose.

While we are here: creatine kinase is a poor readout for an individual. Between-person variability is enormous, and a high CK tells you much less about your own muscle damage than the number’s precision implies.

How much protein, across the whole day

The best dose-response evidence comes from a 2018 meta-analysis in the British Journal of Sports Medicine covering 49 randomised trials and 1,863 participants across 17 countries. Protein supplementation added 0.30kg of fat-free mass and 2.49kg of one-rep-max strength on average, with benefits plateauing at about 1.62 grams per kilogram of body weight per day.

Print the uncertainty with the number, because it is large: the confidence interval on that plateau ran from 1.03 to 2.20 g/kg/day. Anyone quoting 1.6 as a precise threshold is quoting the midpoint of a wide band. The analysis also found efficacy declined with age, which is one reason requirements go up rather than down as you get older.

Where carbohydrate fits

Carbohydrate matters for recovery in proportion to how much glycogen you depleted, and this is where general advice goes wrong by treating all training the same.

The cleanest answer comes from a 2010 Sports Medicine review. When carbohydrate intake is optimal — at least a gram per kilogram per hour — adding protein provides minimal additional benefit for glycogen resynthesis. When carbohydrate is below that, adding at least 0.3g/kg/h of protein does become meaningful, through a synergistic insulin response.

The same review makes a methodological point most articles skip: much of the apparent performance benefit of adding protein to a recovery drink may simply be that it added calories, or that the carbohydrate fraction was dosed below the recommended rate to begin with. That is a calorie-matching artefact, not a protein effect.

And the practical filter: rapid glycogen restoration only matters when the next hard session is within roughly eight to twenty-four hours. If you lift four times a week, glycogen is a non-issue for you and the protein-plus-carbohydrate recovery drink pitch is not aimed at your situation.

Ice baths: they work, and that is the problem

This is where the three-outcomes distinction earns its keep.

A 2021 Sports Medicine review of eight controlled trials examined regular cold-water immersion alongside training. For resistance training, the pooled effect on strength outcomes was negative — a standardised mean difference of −0.60, p < 0.0001 — with ballistic performance similarly impaired. For endurance training it was neutral: essentially zero effect on time-trial performance or maximal aerobic power.

The mechanism has been shown directly. A twelve-week study of cold-water immersion versus active recovery in trained men found blunted gains in muscle mass and strength, with attenuated satellite cell and anabolic signalling responses acutely. The inflammation that makes you sore is part of how you adapt, and cooling it down cools that too.

There is also a good placebo-controlled study in which a sham “recovery drink” in thermoneutral water matched cold water’s perceived-recovery benefit — so a substantial part of the subjective effect is belief.

The practical rule. Use cold water when performance tomorrow beats adaptation next month: in-season, during a tournament, in the heat, when you compete again in twenty-four hours. Do not use it after a hypertrophy session in a training block. You are paying for comfort with the adaptation you trained for.

The same logic is hypothesised to apply to high-dose antioxidant supplementation blunting endurance adaptation, which is worth keeping in mind whenever a supplement is marketed on its antioxidant content.

Tart cherry, curcumin and omega-3, honestly graded

Curcumin has the best soreness evidence of the three — notable, given that protein does not. A 2024 meta-analysis of 14 randomised trials and 349 subjects found reductions in soreness, creatine kinase and IL-6, and improved range of motion, with pre-emptive low-dose regimens working best for soreness and immediate post-exercise dosing best for the blood markers. Two caveats: 349 subjects across 14 trials is about 25 per trial, and curcumin bioavailability is poor enough that formulations differ enormously, so “curcumin” is not really one intervention.

Tart cherry is more interesting than its reputation and less consistent. A 2026 review of 19 trials found significantly improved maximal voluntary contraction recovery, with a large effect at 72 hours, and reduced CRP — but no significant effect on soreness, creatine kinase or range of motion. A separate network meta-analysis of 30 trials found the mirror image: a soreness benefit at 48 hours and no clear advantage for strength or the blood markers. Report that for what it is, an unstable evidence base of small trials with heterogeneous protocols. One protocol detail does matter: the positive studies started supplementation days before the damaging session and continued afterwards, at over 1,000mg of total phenolics daily. A single post-workout dose is not what was tested.

Omega-3 is the thinnest of the three. There is no high-quality meta-analysis on recovery specifically; what exists are small individual trials with heterogeneous doses and formulations, some reporting reduced soreness. There are good reasons to take fish oil; accelerating recovery is not a well-supported one, and the adaptation question raised above applies here too.

The things that matter more than any of this

Sleep is the recovery intervention with the largest plausible effect and the least commercial infrastructure behind it. A 2023 systematic review of 25 intervention studies found that increasing sleep duration, at night or through napping, was the most effective intervention for improving physical and cognitive performance. A six-week sleep-extension study in 25 professional rugby players improved sleep quality by around a quarter, added time in bed, cut cortisol by 18.7% and improved reaction time.

Two honest qualifications. Regularity may matter as much as duration — one study in collegiate players found no correlation between average sleep duration and performance, but a clear inverse correlation between sleep inconsistency and service accuracy. And the intervention literature is not uniformly positive: one trial found sleep extension increased total sleep time but also fragmentation, with neuromuscular function actually favouring active recovery, and another sleep-extension intervention failed to increase sleep duration at all. Sleep is the best lever available; it is not a solved intervention. More in our piece on sleep, appetite and body composition.

Total energy. Training hard while under-eating means protein gets diverted to fuel rather than repair. Notably, the 2014 review found protein’s benefits were greatest when participants were in negative nitrogen or energy balance — which points at the real lever. Chronic under-eating is the most common cause of poor recovery in recreational athletes, and no supplement fixes it.

Alcohol. A small but mechanistically clean study found that post-exercise alcohol at 1.5g per kilogram suppressed myofibrillar protein synthesis by roughly a quarter to a third — even when protein was consumed alongside it. If you are serious about a training block, that is a more consequential decision than which recovery powder you buy.

Fluid. Performance decrement becomes measurable at around 2% of body mass lost, more so in heat and endurance work than in a short lifting session. Drinking to thirst is adequate for most recreational training, and the old “drink as much as possible” advice is obsolete — hyponatremia from overdrinking is a real and occasionally fatal risk in long endurance events. To restore a deficit afterwards, roughly 125 to 150% of the fluid lost, with sodium. See how much water you actually need.

A sensible protocol

Hit your daily protein target — somewhere around 1.6 g/kg if you train regularly, remembering that the evidence supports a range rather than a threshold.
Eat 25 to 35g of protein within a couple of hours after training. A shake is convenient, not magic.
Match carbohydrate to the demand of the session, not to a rule.
Rehydrate to thirst, and a bit beyond it after a hot or long session.
Sleep seven to nine hours, at consistent times. This is the highest-return item on the list and the one most often skipped.
Save the ice bath for the days when tomorrow’s performance matters more than next month’s adaptation.
Judge recovery by whether you can produce force, not by whether you are sore.