Alcohol in fight camp

The dose behind the number everyone quotes

The trial that generated the figure is a randomised, counterbalanced crossover in eight physically active men, aged 21.4 ± 4.8, with a VO₂peak of 48.1 ± 4.8 mL/kg/min and a fourteen-day washout between conditions. Each man performed eight sets of five leg extensions at 80% of one-rep maximum, then continuous and interval cycling — a deliberately hard concurrent session — and then, one hour later, drank.

The drinking protocol is the part that gets lost. Six equal serves of 60 mL of vodka with 240 mL of orange juice, spread across three hours, totalling 1.5 grams of alcohol per kilogram of body mass. The paper itself translates that as 12 ± 2 standard drinks. Blood alcohol was measured, not assumed: it peaked four hours post-exercise at 0.059 ± 0.017 g/100 mL in the alcohol-plus-carbohydrate condition and 0.056 ± 0.019 in the alcohol-plus-protein condition, and was still elevated at eight hours.

Against protein alone, myofibrillar protein synthesis over the 2–8 hour recovery window was 37% lower when alcohol was taken with carbohydrate. That is the number. It belongs to that dose, that cohort, that timing and that session.

It is worth resisting the arithmetic temptation here. A reader who knows their own body mass can convert 0.059 g/100 mL into a drink count and decide the study was closer to their Saturday than it sounds. It is not. The BAC in that trial was produced by a specific bolus schedule in a post-exercise, largely glycogen-depleted state, and reverse-engineering a drink count from a peak BAC measured under those conditions does not transfer to a different body in a different state on a different timeline. The dose in grams per kilogram is the measurement. The BAC is what that dose did in those eight men.

Protein does not buy it back

The same eight men, at the same dose, were also run through an alcohol-plus-protein condition: 25 grams of whey immediately post-exercise and 25 more at four hours, which is an amount the literature treats as sufficient to maximally stimulate the synthetic response in men of that size.

Protein synthesis in that condition was still 24% below protein alone. The authors describe it in their own text as "only a partial 'rescue' of the anabolic response compared with protein alone," and state that "even when protein was consumed in amounts shown to be optimally effective to stimulate MPS during post-exercise recovery, the intake of alcohol reduced MPS by ∼24%."

This matters more than the headline figure, because the headline figure is not actionable and this one contests something people believe. The gym answer to drinking is usually a shake — the idea that alcohol's cost is a missed feeding opportunity, and that filling the window closes it. At 1.5 g/kg it did not. Roughly a third of the suppression came back; the rest stayed.

The absolute rates are worth carrying because they set the scale. Rest was 0.025 ± 0.002 %/h. Protein alone ran at 0.052 ± 0.008 %/h, a 109% rise over rest. Alcohol plus protein ran at 0.039 ± 0.008, a 57% rise. Alcohol plus carbohydrate ran at 0.032 ± 0.005, a 29% rise. Every alcohol condition still sat above resting. Nothing was switched off. A signal was turned down, in eight men, at a dose most of them would describe as a heavy night.

Sleep: the place where someone measured two drinks

For every other outcome in this article, the honest answer is that the doses studied are not the dose asked about. Sleep is the exception, and it is the reason this article does not end at "nobody knows."

A 2025 systematic review and meta-analysis in Sleep Medicine Reviews pooled 27 studies of alcohol taken before sleep in healthy adults, and stratified them by dose. Its low stratum is defined as ≤0.50 g/kg, which the authors describe as approximately two standard drinks. Its high stratum is ≥0.85 g/kg, approximately five.

The finding at the low dose is that REM sleep is reduced. In the abstract's own words, a dose-response relationship was identified such that disruptions to REM sleep occurred following consumption of a low dose and progressively worsened with increasing doses. The effects that most people associate with drinking and sleep — falling asleep faster, dropping into deep sleep sooner — were observed only at the high dose. Two drinks did not, in the pooled data, buy the sedation. They cost the REM.

The meta-analysis is explicit about what it could not settle. Effects on total sleep time, sleep efficiency and nocturnal wakefulness remained uncertain across the pooled studies. And the authors call directly for research into sex differences and individual metabolic variation, which is another way of saying these pooled numbers are not sex-resolved and should not be read as though they were.

So the asymmetry is this. For the anabolic question, the only data is a heavy night in eight men. For the sleep question, someone measured approximately the two beers. That is a more useful shape than a blanket "nobody knows," and it points the cost at a specific place: not Saturday, but the quality of the night that follows it, and therefore Sunday.

What the night actually looks like

The pattern that the older review literature describes is biphasic, and it is the part a drinker is structurally unable to notice.

At essentially all doses, sleep onset latency shortens and the first half of the night is more consolidated. Then the second half fragments. Total-night REM as a percentage decreases at moderate and high doses, with no clear trend at low doses; the reduction in first-half REM reaches significance only at high doses. The subjective experience is falling asleep quickly and easily. The measured experience is a good first half paid for by a worse second half — in the hours a sleeper is least likely to remember.

For a fighter this is the most transferable item on the list, because it does not require any of the mechanism studies to transfer. It does not need the twelve-drink cohort, the eccentric-contraction protocol, or a male-only sample. It needs only that a person slept after drinking, which is the actual scenario.

It also lands somewhere specific. The cost of Saturday night is not Saturday. It is the session on Sunday, and if the night is bad enough it is the one after. That is worth pairing with what the broader sleep literature establishes about camp: sleep is already the least-counted variable in a fight camp, already degraded by an energy deficit, and already the thing that failures get attributed to last.

Damaged muscle is where the penalty attaches

The second measured cost is conditional, and the condition is the interesting part.

Ten healthy men performed 300 maximal eccentric quadriceps contractions on one leg — a protocol designed to produce real muscle damage — and then received either 1 gram per kilogram of ethanol or a control drink immediately afterwards. Two weeks later they repeated it on the opposite leg with the alternate beverage.

At 36 hours, the strength losses were markedly worse under alcohol: isometric 40.9% versus 28.7%, concentric 42.8% versus 31.9%, eccentric 44.8% versus 25.9%.

And the non-exercised leg showed no change under either drink. That is the finding that shapes how the result should be read. This is not a systemic weakening of an athlete who drank. It is an interaction with tissue that had already been beaten up. The penalty attached to the damaged leg and nowhere else.

Which relocates the question. It is less about Saturday than about what Friday and Saturday's training was and what Monday's session demands. A hard eccentric-loaded session, a sparring day that left real soreness, a first session back after time off — those are the states in which this result has something to say. A light technical week is a different context, and the study does not speak to it.

Ten men, one leg each, at 1 g/kg. It is a clean design and a small one, and like the protein-synthesis work it is male-only.

The counterweight: ten weeks, 73 people, and nothing

The literature does not lean the way gym consensus assumes, and the honest version of this article has to say so at the same volume.

The BEER-HIIT study randomised 73 healthy adults, 35 of them women, mean age 24 ± 6, into five arms: beer at 5.4% (n=13), vodka with sparkling water matched at 5.4% (n=14), sparkling water (n=16), alcohol-free beer (n=15), and a non-training control (n=15). The alcohol arms drank 660 mL per day for men and 330 mL per day for women, Monday to Friday, for ten weeks, while training high-intensity intervals twice a week.

All four training groups improved VO₂max in both absolute and relative terms, and improved total test duration. Hand-grip strength improved. Vertical jump did not change in any group. The conclusion is printed plainly: "A moderate beer or alcohol intake does not mitigate the positive effect of a 10-week HIIT on physical fitness in young healthy adults."

Its limits are real and should be stated in the same breath. These were not combat athletes. Nobody was in an energy deficit, nobody was making weight, and there was no weigh-in. VO₂max and hand-grip are not striking output or grappling. And the women were dosed at half the men's volume with no body-mass rationale given for the split, which makes the sex comparison inside the trial hard to interpret.

But it is a ten-week randomised controlled trial at a genuinely everyday intake, and it found no blunting of the adaptation it measured. Set that against a review of the acute-ingestion literature which states that post-exercise recovery studies "have produced inconclusive results that fail to demonstrate a dose-dependency or critical threshold," and that "the notion that alcohol consumption effects performance has not received enough consistent validation to advance beyond being anecdotal." That review is fifteen years old and the field has not resolved it since.

The confident version of this topic — alcohol wrecks training — is confident past what has been measured.

Chasing "one night undoes two weeks"

This is the claim the question usually arrives wrapped in, so it deserves to be chased in public.

We could not find an origin for it. The nearest traceable form is a fitness-media assertion that alcohol "has the potential of negating up to 14 days worth of training in some athletes," which appears in secondary blog posts and university-wellness pages with no cited study, no cohort and no measurement attached. It propagates by being repeated.

Now look at what the underlying literature could support. The protein-synthesis work measures a 2–8 hour window in eight men at twelve drinks. The eccentric-damage work measures strength at 36 hours in ten men at 1 g/kg. Neither study ran for fourteen days. Neither measured anything at fourteen days. There is no measurement anywhere in this evidence base from which a two-week figure could be derived, and the ten-week randomised trial at a daily dose points the other way on the outcome it measured.

So the figure is unprintable as fact, and it is worth saying why the demolition matters rather than just performing it. A claim that large, repeated often enough, does one of two things to a reader. It either produces a rule nobody can keep, or it gets disbelieved along with everything attached to it — including the REM finding, which is real, modest, and measured at the dose being asked about. Inflated claims are expensive in both directions.

Energy, displacement, and a deficit

Inside a camp the argument is arithmetic, not character.

Ethanol yields 29.7 kJ, or 7.1 kcal, per gram. That is per gram of ethanol, not per drink, and the conversion to any particular glass depends entirely on its strength and its size. It sits between carbohydrate and fat on energy density, and unlike either it arrives with essentially no nutrient attached.

The metabolic wrinkle is that ethanol cannot be stored and is toxic, so the body oxidises it as a priority fuel. While it is being cleared, it displaces fat oxidation. In one overnight indirect-calorimetry study, adding alcohol to meals suppressed fat oxidation to an extent equivalent to storing 74% of the alcohol's energy as fat.

That figure invites over-reading, so take the same literature's caveat with it: short-term studies that do not capture the later readjustment of macronutrient balance can mislead, because fat balance can re-establish itself over longer measurement windows. The defensible statement is about fuel priority and energy, not about a metabolic catastrophe. Alcohol carries energy, contributes nothing else, and while it is present the body burns it first.

There is a mechanism most fighters in a deficit would rank above both of those, which is what happens to eating after drinking. We could not secure a primary study with a stated cohort for it in this pass, so it is not asserted here. The displacement argument in this article rests on energy density and fuel partitioning alone.

For anyone running a deficit across a camp, the relevant framing is the one that applies to any energy that arrives without nutrition: it comes out of the same budget as everything else, and the budget is what low energy availability is a failure of.

The cut interacts, and nobody has measured it

Here is the most important gap in the article, stated as a gap.

No study has measured alcohol intake during an actual weight cut. The interaction between drinking, acute dehydration and rapid weight loss — which is the specific thing a fighter in fight week is asking about — is unmeasured. That is an absence of data during the most dangerous week of camp. It is not reassurance.

What exists nearby runs slightly against intuition. Alcohol inhibits antidiuretic hormone, and the review-level figure is roughly 10 mL of excess urine per gram of ethanol consumed; it is also a peripheral vasodilator, which increases fluid loss and lowers core temperature, reducing work tolerance in hot and in cold environments. But when the rehydration question was tested directly — exercise-induced dehydration, then drinks of 0, 1, 2 or 4% alcohol over an hour starting half an hour post-exercise — total six-hour urine volume was not significantly different between them, although it trended upward with dose and peak urine flow came later on the 4% drink. The authors' own reading is that alcohol's diuretic action is blunted when the body is already hypohydrated.

The naive arithmetic — a drink costs you this many millilitres — therefore overstates it in exactly the athlete who is already dry. That is a finding about a mechanism, not a reason to relax. What it actually says is that the model in people's heads is wrong, and that the right study has never been run.

One thing is worth stating flatly because the framing is common and it is dangerous here. There is no mechanism by which sweating clears alcohol, and treating a sauna as a way to deal with a night's drinking combines two dehydration pressures in the week when fluid balance is already the thing most likely to hurt a fighter. The elimination happens in the liver at a rate no behaviour meaningfully changes.

Women are not a scaled-down man

The mechanism studies in this article are male. Parr's protein-synthesis trial was eight men. The eccentric damage trial was ten men. The main review of alcohol and recovery is explicitly a male-athletes review. The glycogen work's sex composition could not be confirmed. The sleep meta-analysis is not sex-resolved and says so. The only well-sexed dataset is the ten-week trial, with 35 women of 73 — and it dosed the women at half the men's daily volume without giving a body-mass rationale.

The instinct at this point is to apply a coefficient. It does not work, and the reason is pharmacokinetic rather than arithmetic.

The 1990 measurement that established this found first-pass metabolism in women at 23% of men's and gastric alcohol dehydrogenase activity at 59% of men's, with higher blood ethanol after comparable intakes even after allowing for body size. A dose in grams per kilogram does not equalise exposure between sexes, because the difference is in how much alcohol is metabolised before it reaches the bloodstream at all. Differences in body composition and total body water compound it.

So a female fighter reading the 1.5 g/kg figure cannot scale it, and neither can anyone reading it on her behalf. The correct statement is that these are different exposures at the same nominal dose, and that the mechanism studies in this field have not been run in women.

Two other populations are simply absent. There is no data here on adolescents, which matters because wrestling and judo are heavily adolescent sports and nothing in this article transfers to an under-18 athlete — in most jurisdictions the question is settled by law before it reaches physiology. And there is no amateur fight-camp cohort anywhere in this evidence base; the participants are college athletes, recreationally active young men, and healthy adults.

The rules, separated cleanly

Two different questions get collapsed into one, and they are answered by different bodies.

The first is anti-doping. Alcohol is not on the WADA 2026 Prohibited List. That List was agreed on 11 September 2025 and came into force on 1 January 2026, and alcohol appears nowhere in it. It was removed from the List effective 1 January 2018, with the stated intent that the affected International Federations would set and enforce their own alcohol rules and thresholds. Those federations were Air Sports, Archery, Automobile and Powerboating — four of them, and not one a combat sport.

That last detail is worth sitting with, because it settles more than it first appears to. Alcohol was not a combat-sport anti-doping matter even before 2018. No combat federation lost an alcohol threshold in the removal, because none of them had one on the WADA instrument to lose. Checking the Prohibited List for alcohol has never been the way a fighter would find out what applies to them.

The second question is what the body licensing the bout permits, and that is where the binding rule lives. It is jurisdiction-specific: a commission, an athletic board, or a national federation writes it, and the rule that applies to a fighter is the one written by whoever sanctions that fight.

We could not verify any current combat-commission alcohol or weigh-in rule today. The Association of Boxing Commissions' site and Nevada's administrative code both refused our requests, and the Nevada provision that older articles cite for this appears to have been repealed or reorganised, so it is not cited here. That means this article states nothing about what any commission currently requires. The honest instruction is the unsatisfying one: read your own commission's current rule, or ask the person who holds your licence, because we could not read it for you.

What the surveys say, and what they cannot

Prevalence data in this area is thinner than the confidence around it.

The largest relevant dataset is the NCAA's Student-Athlete Health and Wellness Study, surveyed between October 2022 and June 2023 across more than 23,000 athletes and released in January 2024. Self-reported past-year binge drinking fell from 55% in 2009 to 35% in 2023, with binge defined as four or more drinks for women or five or more for men in one sitting. Athletes reported less alcohol, tobacco and recreational drug use than the general college student population.

What it does not give is any division- or sex-level alcohol breakdown, or a wrestling-specific figure. And these are US college athletes, not professionals and not a camp population.

For combat sports specifically there is almost nothing. One cross-sectional study of 53 Brazilian jiu-jitsu black belts — all male, all with at least ten years' practice, data collected in São Paulo between 2014 and 2016 — records that 32 of 53, or 60.4%, reported drinking alcohol socially. No frequency was captured, no quantity, and alcohol was an incidental descriptor in a study about shoulder function. It is a weak signal, not a rate.

There is also a tension in the literature worth leaving unresolved. Review material about male athletes often describes them as consuming greater volumes in binge patterns than the general population even while drinking less often. The NCAA data has athletes reporting less use than their peers. Both statements are in print. They are measuring different things in different populations and neither settles the other.

One further association circulates and is not repeated here as fact: an injury incidence of 54.8% in drinkers against 23.5% in non-drinkers, reported at review level. The originating cohort, country and sport mix could not be recovered, the ascertainment is self-report, and drinkers and non-drinkers differ in age, sport, competitive exposure, training volume, risk-taking and social environment. The review's own causal framing is stronger than the design supports. It is named here so a reader who meets it elsewhere knows what it is made of.

Putting the timing back into it

The measured costs in this article do not float free. Each one lands somewhere.

The sleep cost lands on the night itself and therefore on the next morning's session. It is measured at roughly the dose being asked about, and it arrives without a subjective signal — the drinker falls asleep faster and remembers the good half.

The damaged-muscle cost lands 36 hours out, and only on tissue that was already damaged. Its relevance depends entirely on what the training either side of the drink was.

The protein-synthesis cost, at the dose it was measured at, lands inside an eight-hour window post-exercise and is not fully rescued by feeding. At any lower dose it is unmeasured.

The energy cost lands on whatever budget the camp is running, immediately, and is the only one of the four that is simple arithmetic.

And the interaction with a cut — the question a fighter in fight week is really asking — lands nowhere in the literature, because nobody has run it.

That is the accounting. It is not a verdict, and the shape of it is deliberately not a rule. A reader in the tenth week of a camp with a hard Sunday planned is looking at a different set of those four than a reader in week two with a rest day. One of the arguments for logging what actually went in rather than estimating it later is that alcohol is the single most commonly under-recorded energy source in a food diary, and a camp budget that silently excludes it is not a budget.

What we could not verify

Stated plainly, because an article this insistent about other people's conditions owes you its own gaps.

• The reader's dose, for everything except sleep. The protein-synthesis trial used 1.5 g/kg, the eccentric damage trial 1 g/kg, the ten-week trial a moderate daily intake. A couple of beers on a Saturday sits between those designs for muscle and has not been studied directly. For sleep it has: the pooled low-dose stratum is ≤0.50 g/kg, approximately two standard drinks. We report the asymmetry rather than flattening it either way.
• Anything about alcohol during a weight cut. No study has measured it. The interaction with acute dehydration and rapid weight loss is unmeasured — an absence of data during the most dangerous week of camp, not reassurance.
• Women, in the mechanism studies. The protein-synthesis and eccentric-damage trials are male-only, the main recovery review is explicitly male, and the sleep meta-analysis is not sex-resolved by its authors' own statement. Female first-pass alcohol metabolism was measured at 23% of men's, so a g/kg figure cannot be scaled across sexes with a coefficient. There is no correct adjustment to offer.
• Any combat-sport performance measurement. There is none. Nothing has been measured against striking output, grappling, reaction time or sparring performance. Every performance figure in this article comes from cycling, isokinetic dynamometry or HIIT.
• Adolescents and amateurs. No data at all in either case.
• "One night of drinking undoes two weeks of training." We chased it and found no origin. The nearest traceable form is an uncited fitness-media claim about "up to 14 days." We do not print it as fact and we do not have a replacement figure, because no study measured at that horizon.
• Any percentage change in testosterone. We secured no primary study for it in this pass, so no figure appears.
• Every "safe" number. No amount of alcohol is presented here as acceptable, permitted or harmless — and that includes a dose that appears in one review as the level below which most aspects of recovery are unlikely to be affected. That is one reviewer's inference from a male-only literature, and reproducing it as a limit would convert a measurement into a permission. It is not printed.
• The injury association. The 54.8% versus 23.5% figures could not be traced to their originating cohort, and we do not repeat the source's causal framing.
• The "10 mL of urine per gram of ethanol" figure. It is review-level and we did not chase it to its primary.
• Any current commission rule. The ABC's site and Nevada's administrative code both refused our requests today, and the Nevada section usually cited appears repealed. We state nothing about commission alcohol or weigh-in rules.
• The WADA List itself, read directly. The wada-ama.org pages returned empty bodies to us. Alcohol's absence from the 2026 List is corroborated from the List's published summary and the International Testing Agency's record, and from the documented 2018 removal, rather than read off the primary PDF.
• Nothing here is specific to you — not your training age, your medical history, your medication, or any condition that changes how you handle alcohol. If you have been told by a physician not to drink, this article does not override that, and nothing in it is a reason to revisit it.

Sources

Sourced to

1. Alcohol Ingestion Impairs Maximal Post-Exercise Rates of Myofibrillar Protein Synthesis following a Single Bout of Concurrent Training — Parr EB, Camera DM, Areta JL, Burke LM, Phillips SM, Hawley JA, Coffey VG, PLoS ONE, 2014;9(2):e88384. DOI 10.1371/journal.pone.0088384, PMID 24533082
2. The effect of alcohol on subsequent sleep in healthy adults: a systematic review and meta-analysis — Gardiner C, Weakley J, Burke LM, Roach GD, Sargent C, Maniar N, Huynh M, Miller DJ, Townshend A, Halson SL, Sleep Medicine Reviews, 2025;80:102030. DOI 10.1016/j.smrv.2024.102030, PMID 39631226
3. Influence of daily beer or ethanol consumption on physical fitness in response to a high-intensity interval training program: the BEER-HIIT study — Molina-Hidalgo C, De-la-O A, Dote-Montero M, Amaro-Gahete FJ, Castillo MJ, Journal of the International Society of Sports Nutrition, 2020. PMCID PMC7254771
4. Alcohol, Athletic Performance and Recovery — Vella LD, Cameron-Smith D, Nutrients, 2010;2(8):781–789. PMCID PMC3257708
5. Post-exercise alcohol ingestion exacerbates eccentric-exercise induced losses in performance — Barnes MJ, Mündel T, Stannard SR, European Journal of Applied Physiology, 2010. PMID 20012446
6. Alcohol: Impact on Sports Performance and Recovery in Male Athletes — Barnes MJ, Sports Medicine, 2014;44(7):909–919. DOI 10.1007/s40279-014-0192-8, PMID 24748461
7. Restoration of fluid balance after exercise-induced dehydration: effects of alcohol consumption — Shirreffs SM, Maughan RJ, Journal of Applied Physiology, 1997;83(4):1152–1158. DOI 10.1152/jappl.1997.83.4.1152, PMID 9338423
8. Effect of alcohol intake on muscle glycogen storage after prolonged exercise — Burke LM, Collier GR, Broad EM, et al., Journal of Applied Physiology, 2003;95(3):983–990. DOI 10.1152/japplphysiol.00115.2003, PMID 12740311
9. Alcohol and Sleep I: Effects on Normal Sleep — Ebrahim IO, Shapiro CM, Williams AJ, Fenwick PB, Alcoholism: Clinical and Experimental Research, 2013;37(4):539–549. DOI 10.1111/acer.12006. Read at indexing level; the publisher returned 403 to our requests
10. High Blood Alcohol Levels in Women: The Role of Decreased Gastric Alcohol Dehydrogenase Activity and First-Pass Metabolism — Frezza M, di Padova C, Pozzato G, Terpin M, Baraona E, Lieber CS, New England Journal of Medicine, 11 January 1990;322(2):95–99. Named without a link: the publisher's text is paywalled and we read this at abstract level only, so we do not print a link we have not opened
11. Alcohol and the Athlete — O'Brien CP, Lyons F, Sports Medicine, 2000;29(5):295–300. PMID 10840864
12. Alcohol and the regulation of energy balance: overnight effects on diet-induced thermogenesis and fuel storage — Sonko BJ et al., British Journal of Nutrition. Named without a link: we hold this at abstract/review level only and could not confirm its volume, pagination or DOI, so we do not print a link we have not opened
13. Alcohol, energy metabolism and body weight regulation — Suter PM, Proceedings of the Nutrition Society, 2000;59:511–517. Named without a link: the citation reached us at review level and its DOI was not verified
14. Student-athletes report drop in binge drinking, use of narcotic pain medication and spit tobacco — NCAA Student-Athlete Health and Wellness Study, surveyed October 2022–June 2023, released January 2024
15. Shoulder function in Brazilian jiu-jitsu black belts (cross-sectional study; alcohol reported as an incidental cohort descriptor) — Rodrigues-Menezes et al., data collected 2014–2016, São Paulo, Brazil. PMCID PMC10592395
16. 2026 WADA Prohibited List — USADA advisory on the World Anti-Doping Agency 2026 Prohibited List, agreed 11 September 2025, in force 1 January 2026. Note: wada-ama.org returned empty responses to our requests on 2026-09-23; alcohol's absence is corroborated from this advisory, the List's published summary and the International Testing Agency record rather than read from the primary PDF

Read next