Daily hydration in fight camp

1. The absence at the centre of this article

Start with what should exist and does not.

The most frequently cited synthesis of fluid balance in team sports is a 2017 review by Nuccio and colleagues, all four of whom work for the Gatorade Sports Science Institute, a PepsiCo subsidiary. It is a good review. It pools field measurements across studies and presents sweat rates sport by sport, each with the temperature and humidity range they were recorded in: soccer 0.3–2.5 L/h across 5–43 °C and 7–96% relative humidity; American football 0.6–2.9 L/h at 22–35 °C and 43–92%; basketball 0.7–2.7 L/h at 17–30 °C; tennis 0.6–2.6 L/h at 17–37 °C; rugby 0.4–2.0 L/h at 7–27 °C; ice hockey 0.7–1.8 L/h at 3–14 °C.

That is the table. There is no judo row. No wrestling row, no boxing row, no muay thai row, no jiu-jitsu row. A sport contested in a heavy cotton jacket, in a room often kept deliberately warm, with two people's body heat pressed together for five-minute rounds, does not appear.

This is not a criticism of the review, which never claimed to cover combat sports. It is a statement about the field. When a coach tells a fighter what their sweat rate is, the number came from somewhere else — and in the most common case, it came from a company selling hydration products.

The consequence is worth being blunt about. Every generalisation in this article about what happens to a fighter's fluid balance during ordinary training rests on either a handful of small studies in adolescents, or on transfer from sports with different clothing, different work-to-rest ratios and different environments. Both are weak grounds. Saying so is more useful than papering over it with a confident range.

2. The judo number that came from a marketing page

The figure a fighter is most likely to have heard is that judo produces sweat rates of 1.0 to 2.5 litres per hour, and that a 90-minute session therefore costs 1.5 to 3.5 litres of fluid.

We chased it. It traces to a commercial hydration app's marketing page. There is no study behind it, no cohort, no environment, no measurement method, no sample size. It is a number written to sell a product that tells you how much to drink.

It is also contradicted by both measured datasets we located.

Ceylan and Santos measured 22 under-15 judoka — 14 male, 7 female, mean age 12 ± 0.7 years, with about three and a half years of training experience, in the preparation phase — across a 90-minute technical-tactical session at 28.3 °C and 56% relative humidity, with fluid freely available throughout. Mean body-mass loss was 0.64 ± 0.66%, roughly 0.27 kg.

Demirkan and colleagues measured 14 male wrestlers aged 15.07 ± 0.73 years across three session types. Wrestling-specific training, indoors, produced a body-mass loss of 0.96%.

Neither of those is a sweat rate in the technical sense — body-mass change and sweat rate are different quantities, and body-mass change over a session with fluid available understates sweat loss by whatever was drunk. But they are the closest measurements that exist, and they sit far below what a 1.0–2.5 L/h figure would predict for a 90-minute session in a warm room. The vendor number and the measured data point in opposite directions.

If you take one thing from this section, take the general form of it: a hydration figure with no cohort, no temperature and no method attached is not a finding. It is a claim. Ask which people, in what room, doing what, measured how.

3. What was measured on ordinary training days

Two datasets. Both adolescents. Here is all of what they found, with their conditions attached, because the conditions are the content.

The judo study. Ceylan and Santos, 2022. Twenty-two U-15 judoka, single 90-minute technical-tactical session, 28.3 °C, 56% relative humidity, fluid freely available. Mean body-mass loss 0.64 ± 0.66%. Hydration status by urine specific gravity, euhydration threshold ≤1.020 g/mL.

The striking result is not the mass loss. It is that 81.2% of the athletes were dehydrated on the first morning void, before training began. By the immediate pre-training measurement, 63.64% were classified dehydrated. After the session, despite fluid being freely available throughout, 77.27% finished dehydrated.

The authors' own summary sentence is that "81.2% of the athletes were dehydrated while only 18.8% of the athletes were euhydrated," and that "77.27% of the athletes completed the training in dehydrated condition despite fluid availability."

Read that carefully. The problem, in this cohort, was not what happened during the session. The athletes arrived at the gym already below threshold. A water bottle on the mat does not fix a deficit accumulated in the previous twenty hours.

There is a second finding in the same study that cuts against the usual advice. Across the timepoints there was no significant change in urine specific gravity (p = 0.22), and no significant correlation between how much an athlete drank and either their body-mass change or their change in specific gravity. That is an uncomfortable result for anyone treating a refractometer as a session-level scoreboard.

The wrestling study. Demirkan and colleagues, 2025, fourteen male wrestlers, mean age 15.07 years, one squad, three session types measured separately.

• Endurance: 10 km at 70–75% of maximum heart rate on a 400 m track. Body-mass loss −1.89 ± 0.4%.
• Circuit strength, indoors: −1.39%.
• Wrestling-specific training, indoors: −0.96%.

The order is the finding. The mat session was the least dehydrating of the three, and the roadwork was the worst. If you assume the hot room with the heavy jacket is the fluid problem and the run is the easy day, this dataset says the opposite — in this squad, in these conditions.

After the endurance session, urine specific gravity rose from 1.022 ± 0.005 to 1.038 ± 0.016 g/cm³ (Δ 0.016 ± 0.005, p < 0.05). Note the pre-session value: 1.022 already exceeds the 1.020 threshold used elsewhere. These wrestlers, like the judoka, started the day outside the line.

And the part that matters most for a camp: in the authors' words, "the hydration levels in the endurance group did not fully recover by the next day."

That is one small study, fourteen adolescent boys, one session, one next-morning measurement. It is not a camp-long claim and we will not inflate it into one. But it is the only measurement we found that speaks to carry-over at all, and in a period where a fighter is training eight to twelve times a week, carry-over is the entire question.

4. The performance decrements, and why we are not attributing them to dehydration

The same wrestling study measured performance before and after the endurance session. Back strength fell 7.02 ± 1.2%. Right-hand grip strength fell 8.79 ± 2.1%. Vertical jump fell 7.26 ± 1.8%. Maximal inspiratory pressure fell 9.01 ± 2.3%. All at p < 0.001.

Those are large numbers and they are tempting to use. We are going to decline, and it is worth explaining why, because the same reasoning applies to a great deal of what gets said about hydration and performance.

These decrements were measured after the whole session. The athletes ran 10 km. They were tired. Fatigue and fluid loss were not separated by the study design, and cannot be separated after the fact. A 7% drop in vertical jump after a 10 km run is exactly what you would expect from the run alone. Attributing it to the 1.89% body-mass loss requires a comparison condition that does not exist in this dataset.

What the study does establish is that an ordinary hard session leaves an adolescent wrestler measurably weaker for a period afterwards, and measurably drier, and that by the next morning the dryness had not fully resolved. Those are two co-occurring facts, not a causal chain.

The habit of collapsing co-occurrence into causation is how "2% dehydration impairs performance" became something everyone knows.

5. The 2% threshold does not survive being chased

This is the sentence a fighter has heard most: lose 2% of your body mass in fluid and your performance falls off.

We chased it to origin, and it does not hold as a bare claim. Every defensible version of it carries a condition, and the conditions disagree with each other.

The NATA figure is a target, not a cliff. The 2017 National Athletic Trainers' Association position statement defines euhydrated pre-exercise as within +1 to −1% of baseline, and states an aim of keeping end-of-exercise body-mass loss under 2%. It also states that athletes should not gain mass during exercise. That is a bound on a range, written for the physically active generally. It is not a report of a measured performance discontinuity at 2%.

The team-sport version carries a conditional clause that retellings drop. The GSSI-funded Nuccio review states that greater than 2% body-mass loss "can impair endurance performance, particularly in hot/humid environments." The clause about heat and humidity is load-bearing and is almost always removed in transmission.

A 2025 review puts the threshold higher. Armstrong and colleagues' biomarker reference review places 1–2% as where thirst first appears, and gives 3–4% as the threshold of performance degradation. That is a recent paper putting the line at roughly double the familiar figure.

The sceptical position goes higher still. Goulet's work — which we could not obtain in full, the publisher returned an access error, so we describe his position only as it is reported by Kenefick rather than quoting him — holds that body-weight loss up to 4% did not decrease cycling time-trial performance, and that drinking to thirst maximised endurance performance. The cohort is competitive endurance cyclists in time trials. That is not a fighter, and we are not going to pretend it is.

And a skill test found nothing above the threshold. In a two-hour simulated tennis match, post-match shot accuracy on a ball-machine test showed no difference between players who drank water and finished at 1.1% body-mass loss and players who drank nothing and finished at 2.7%. Male and female players. A clean null at a magnitude above the famous number.

So what is left? A gradient, in some conditions, whose location nobody agrees on, measured almost entirely in sports that are not combat sports. That is a genuinely different thing from a cliff at 2%, and a fighter who has been told the cliff version has been told something the literature does not support.

6. The cognitive question, which is the one a technical sport should care about

A fighter does not primarily need to run further. They need to see a shot coming, judge distance, and time a counter. So the cognitive and skill literature is more relevant here than the endurance literature, and it is messier.

Wittbrodt and Millard-Stafford's 2018 meta-analysis pooled 33 studies, 280 effect-size estimates and 413 subjects, across laboratory dehydration protocols producing 1–6% body-mass loss. Overall cognitive impairment was ES = −0.21 (95% CI −0.31 to −0.11, p < 0.0001) — statistically clear, and small.

By domain, the pattern is interesting for this audience. Attention was worst hit at ES = −0.52. Motor coordination followed at −0.40. Executive function came in at −0.24. Reaction time, at −0.10, was not significantly impaired. Attention and motor coordination being the two hardest-hit domains is the relevant fact for a sport judged on timing and technique rather than raw speed of response.

Now the dose relationship, and here we are correcting a figure we were initially given in a stronger form than the data supports. Impairment was greater for studies reporting greater than 2% body-mass loss (ES = −0.28, 95% CI −0.41 to −0.16) than for studies at or below 2% (ES = −0.14, 95% CI −0.27 to 0.00), with the difference between strata significant at p = 0.04.

The confidence interval at or below 2% touches zero. That means the pooled effect below 2% is not statistically significant. It would be wrong to describe it as "small but real," which is how this comparison is usually reported. The honest framing is a gradient whose lower end is indistinguishable from nothing and whose upper end is small — which is neither a cliff nor a clean slope.

And the field data disagrees with the laboratory data anyway. The GSSI-funded team-sport review looked at hypohydration at field-relevant magnitudes — most studies around 1 to 2.5% body-mass loss — and concluded that the effect on cognitive performance in team-sport athletes is equivocal. Its authors put it directly: "hypohydration consistently increased ratings of thirst, perceived exertion, and fatigue, but subsequent effects on cognitive performance were equivocal."

That is a useful distinction to hold. Being dry reliably makes work feel harder. Whether it reliably makes you think worse, at the magnitudes a training session actually produces, is not settled.

On skill specifically: across soccer, field hockey and tennis, roughly 2–3% hypohydration had minimal impact on skill performance. At roughly 2–4%, basketball shooting and cricket bowling and throwing were impaired. No combat-sport skill test exists in this literature at all. Nobody has measured whether a hypohydrated fighter's distance management, grip or timing degrades. Basketball shooting is the nearest analogue anyone has, and it is not very near.

7. Two traps in the sources themselves

If you go looking for this material yourself, there are two things you will walk into. Both are worth knowing before you get there.

The NATA website serves the wrong statement. The National Athletic Trainers' Association hosts a fluid-replacement position statement PDF at a URL whose folder is dated 2025. The document it serves is the 2000 statement — "Fluid Replacement for Athletes," Casa et al., Journal of Athletic Training 2000;35(2):212–224. The string "2017" does not appear in it. Neither does "McDermott," the lead author of the current version.

This matters because the 2000 statement's advice was substantively reversed. It instructed athletes to drink "beyond thirst satiation," and it gave specific volumes: approximately 500–600 mL two to three hours before exercise, 200–300 mL ten to twenty minutes before, and 200–300 mL every ten to twenty minutes during. The 2017 statement, by contrast, records that thirst typically increases at about 2% hypohydration — which is to say, thirst carries information — and the field has moved decisively away from scheduled volume prescriptions.

Those 2000 volumes are still circulating in gyms. They are a protocol from a document that has been replaced, and we are printing them here only so that the sentence "they were superseded a quarter of a century ago" is louder than the numbers. Do not run them. And while we are here: the range 400–800 mL per hour, which is often attributed to NATA, is not in the NATA document at all — it belongs to the American College of Sports Medicine's 2007 position stand.

The strongest methodological sources are Gatorade-funded. Four of the best sources on this subject are products of the Gatorade Sports Science Institute, a PepsiCo subsidiary: Baker's 2017 methodological review of sweat measurement, whose author is employed by GSSI and whose supplement was GSSI-funded; the Nuccio team-sport review, all four of whose authors are GSSI or PepsiCo; and Kenefick's 2018 review of planned drinking versus drinking to thirst, which was GSSI-funded and whose author took a GSSI honorarium while the paper carries a declaration of "no potential conflicts of interest." That discrepancy is worth one dry sentence, and this is it.

The sports-drink industry funded much of the literature on whether athletes should drink more. That is a reason for care, not automatic dismissal — and the reason we say so rather than quietly dropping these sources is that the honest reading cuts the other way. Both GSSI authors argue against over-precision. Baker's own recommendation is to categorise athletes' sweat water and sodium losses as low, moderate or high and to offer "a range of fluid replacement options (rather than attempting to pinpoint exact values)" — a direct rejection of the precision that commercial sweat testing is sold on. Kenefick's closing rule is that one should never consume so much fluid that weight is gained. Neither of those positions helps sell more fluid. When a funded author argues against their funder's commercial interest, that is evidence about the argument, and it belongs in the open.

8. Thirst: what it can and cannot do

The live argument in this field is whether thirst is a sufficient guide. Kenefick's review is the best summary of both sides we could obtain in full, and it is GSSI-funded, as above.

The case for thirst. Ad libitum drinking — drinking when thirsty, as much as wanted — replaces about half of fluid losses. That is precisely why it protects against hyponatraemia: it is self-limiting. Kenefick states that since drinking to thirst "appears to result in fluid replacement of about half of fluid losses, this strategy would appear to be successful in the prevention of hyponatremia." The exercise-associated hyponatraemia consensus is blunter still: "The most individualized hydration strategy before, during, and immediately following exercise is to drink fluids when thirsty."

The case against. Kenefick also reports that "during periods of high sweat rates (>1.0 L/h) humans practicing ad libitum drinking have been reported to markedly under-consume fluid," and notes historical work in which subjects felt fully recovered and unthirsty while carrying a four-to-five-litre water deficit. We flag that those historical subjects are not described as athletes in the text we read; the figure comes from older military and desert work.

Where the line falls. Kenefick's own framing is that thirst suffices under roughly one to two hours, at lower intensity, in cool or temperate conditions, with losses staying under 2%. It does not suffice beyond ninety minutes to two hours, at higher intensity, in warm or hot conditions.

Look at that and then look at a fight camp Tuesday: two sessions, one of them ninety minutes or longer, in a room kept warm, at high intensity. A fight camp day straddles the line exactly. That is the honest answer to the question this article was asked, and it is not a satisfying one. The conditions under which the drink-to-thirst consensus is confident are not quite the conditions a fighter trains in, and nobody has measured the conditions a fighter trains in.

One more thing worth holding onto, because it changes what a scale reading means. Work in soldiers exercising in cool conditions found that total body water was protected despite roughly 2% body-mass loss under ad libitum drinking, suggesting "voluntary dehydration" may be partly mislabelled — body-mass loss during exercise includes substrate oxidation and metabolic water, not only body water. That was soldiers, in the cold, and we only have the abstract. But the mechanism transfers: the number on the scale after a session is not a direct readout of how much water you lost.

9. What a refractometer can and cannot tell you

Urine specific gravity is the marker combat sports actually use, largely because it is the marker commissions use at weigh-ins. Its limits in this population are worse than most people assume.

Start with the thresholds, which do not agree.

• The NATA 2017 statement's operative figure is 1.025 on a first morning void, used as a gate on other measurements: "If USG is not <1.025 (first morning void), as quantified by a digital or clinical refractometer, body mass and body fat should not be measured."
• Armstrong and colleagues' 2025 reference review gives 1.015–1.025 as a euhydrated baseline, below 1.010 as suggesting overhydration, and above 1.020 as indicating dehydration. Note that those last two overlap: the band from 1.020 to 1.025 is simultaneously "euhydrated baseline" and "dehydrated."
• The research studies above used 1.020 as the dehydration cut-off.
• And Zubac and colleagues, writing specifically about Olympic combat-sport athletes, argue there is no defined limit value at all.

That last paper is the most important one for this audience, and we could only reach its abstract; the full text is paywalled. It records a "growing debate regarding the diagnostic accuracy and the applicability of USG in characterizing whole-body fluid status and fluctuations," notes that even a maximum measurable reading of 1.030 g/mL can be physiological, and argues that specific gravity is far better suited as a progression parameter — a trend tracked over weeks — than as a pass-or-fail line on a given morning.

Then the confounders, which Armstrong's review quantifies. A large water bolus — above 1.4 litres in thirty minutes — drives specific gravity down rapidly. A high-protein diet may acutely alter values. Body size and body composition influence readings. Dietary osmolar load moves the number. Supplements, for once, are unlikely to matter measurably.

Put plainly: if you drink on the drive to the gym, your reading is about the drive, not about you. And Armstrong's review is direct about the general case — "no single hydration biomarker allows a conclusive clinical diagnosis of dehydration for all athletes in all sport scenarios," and "hydration assessment via urine specific gravity has limitations when used alone."

One more caveat specific to the datasets in this article: Armstrong notes that thirst was not correlated with hydration status in youth populations. Since the two cleanest training-day datasets we have are twelve-year-olds and fifteen-year-olds, the adolescent data may not even validate the adult advice about thirst.

If you use a refractometer in camp, the defensible use is the one Zubac describes: same time each morning, first void, recorded across weeks, read as a line rather than a verdict. Whether that line means anything for a given athlete is a question for someone qualified who can see them — not for a web page. Commission hydration testing at a weigh-in is a separate matter with separate stakes, covered in hydration testing at weigh-ins.

10. Two Tuesdays, same fighter, same gym

Here is how to read your own numbers without being handed a target. It is a paired comparison, not a calculation, and deliberately so.

Tuesday A. A fighter steps on the scale before a ninety-minute evening session and again after. The loss is some percentage. What should they expect? The only measured combat figures we have say: 0.64% for U-15 judoka over ninety minutes at 28.3 °C with fluid available; 0.96% for adolescent wrestlers doing mat work indoors; 1.89% for the same wrestlers running 10 km. The unexpected part is the ordering — the mat session was the least dehydrating and the roadwork was the worst. The hot room is not automatically the problem, and the easy-looking run may be.

Tuesday B. Same fighter, same session, but they drank a large bottle in the car on the way in. The morning refractometer reading said one thing; every subsequent reading is now confounded, in the direction and roughly the magnitude Armstrong's review describes. Body size, body composition and the previous day's dietary osmolar load are all moving that number too, independent of hydration status.

The trap to show, not the target to hit. The judoka were already below threshold on the first morning void, before training started, and a full day of freely available fluid did not fix it. The wrestlers had not fully recovered by the next morning after one session. So the question a training day poses is not "did I drink enough during the session." It is "what did I bring in, and what did I carry out" — and both of those are measured before and after the gym, not during it.

Close both doors. Whatever a fighter concludes from their own trend line, two bounds apply at once. Kenefick's closing rule is that one should never consume so much fluid that weight is gained. NATA says the same. And the reason both say it is the subject of the next section.

Then hand it over. The two numbers a fighter is most likely to have been given — 2% and eight glasses — do not survive contact with their sources. The position statements deliberately decline to give a litres-per-hour figure. Nobody has measured a fighter in a gi on a two-session day. What is left is a trend on a scale and in a refractometer, read over weeks, interpreted by a sports dietitian or a physician who knows the athlete. Fight week is a different article and a different set of risks — rehydrating after a weigh-in is where that material lives.

11. Hyponatraemia: the failure nobody warns fighters about

Everything above is about not drinking enough. The opposite failure is worse, and it is the one no gym poster mentions.

Exercise-associated hyponatraemia is defined as blood sodium below 135 mmol/L during or immediately after physical activity, regardless of symptoms. NATA extends the window to within twenty-four hours and notes that it typically becomes symptomatic below 130 mmol/L.

The mechanism, per the 2017 consensus update, is "the overconsumption of hypotonic fluids likely in combination with non-osmotic stimulation of AVP secretion." In plain terms: taking in more dilute fluid than sweat, urine and breathing are removing.

Two things follow that fighters routinely get wrong.

A sports drink does not protect you. Both water and commercial sports drinks are hypotonic relative to plasma. Switching from one to the other does not change the direction of the problem.

Adding sodium does not buy permission to drink more. The consensus position is that sodium ingested during exercise may attenuate the fall in blood sodium but cannot prevent hyponatraemia if fluid intake is excessive. It is the fluid volume, not the sodium, that sets the final concentration. The reasoning "I'll just add electrolytes and drink more" is precisely the reasoning this finding kills.

And it is not confined to marathons. The case literature includes 33% of ten rugby players after an eighty-minute match — a third of a very small sample, and we say so because the denominator matters — high-school American football players who died following practice, a college student doing calisthenics during a fraternity initiation, a soldier on the first day of Ranger training, and a policeman on a nineteen-kilometre bike ride.

Look at those durations. An eighty-minute match and a calisthenics session are ordinary training lengths. They are indistinguishable from a Tuesday at your gym. The risk factors identified by the 2015 consensus include overdrinking water or hypotonic beverages, weight gain during exercise, event inexperience, slow pace, high or low BMI, and readily available fluids.

Weight gain during exercise is on that list, which is why both position statements converge on the same single bound: never drink so much that you gain weight during a session. It is not a schedule. It is a ceiling, and it is the only quantitative instruction in this entire article that we are comfortable repeating.

We should note the limit of our own evidence here: no exercise-associated hyponatraemia case in a combat athlete was located. The rugby, football, calisthenics, military and cycling cases establish that it occurs in non-endurance, ordinary-session contexts. Nobody should claim a documented fighter case without one, and we are not.

If an athlete becomes confused, disoriented, headachy or nauseated during or after training and has been drinking heavily, that is an emergency and a hospital question, not a coaching question. Nothing in this article substitutes for a physician, and nothing in it justifies overriding one.

12. Where the eight-glasses rule came from

The other number every fighter has heard is eight glasses of eight ounces a day.

Heinz Valtin, professor emeritus at Dartmouth Medical School and a renal physiologist, went looking for its evidence base in 2002 and reported finding no scientific studies supporting it. He further reported that peer-reviewed surveys of fluid intake in healthy adults of both sexes strongly suggest such volumes are not needed.

His proposed origin — and he presents it as a reconstruction, not a documented paper trail — is a Food and Nutrition Board recommendation of roughly one millilitre of water per calorie of food, whose following sentence noted that most of that quantity is contained in prepared foods. The second sentence appears to have dropped out in transmission, leaving the first as a standalone instruction to drink.

There is an honest coda that usually gets left off. Valtin's review concerned ordinary healthy adults. A fighter training twice a day in a warm room is outside the population he reviewed, so his paper does not tell that fighter that eight glasses is too much for them. What it tells them is that the number was never evidence in the first place — for anyone. It is a slogan with a plausible clerical origin, and it has never been a finding about athletes, because it has never been a finding at all.

13. What a camp can do with none of this

Given all of the above, a reasonable question is what is actually left to act on. Four things, none of which is a volume.

Arrive, rather than catch up. The single most transferable result in this article is that the judoka were already dehydrated on the first morning void. Whatever is true about drinking during a session, the state an athlete walks in with is the larger variable, and it is set across the preceding day — which makes it a question about the ordinary rhythm of eating and drinking between sessions, not about the water bottle on the mat. Fitting food around two sessions covers the adjacent half of that problem.

Track a trend, not a verdict. Morning body mass and, if you use one, first-void specific gravity, recorded at the same time under the same conditions, read across weeks. That is what Zubac's paper argues for and what the confounders in Armstrong's review permit. A single reading carries too much noise to mean anything. Whatever you use to record it — a notebook, a spreadsheet, or the log in this app — the value is in the line, not the point.

Respect the two bounds. Do not gain weight during a session. Do not treat electrolytes as permission to drink more. Those are the only two quantitative rules with consensus behind them, and both are ceilings rather than targets.

Take it to someone who can see you. Everything above is population data measured in people who are not you, mostly in adolescents, mostly male, mostly not in combat sports at all. A sports dietitian or a physician who knows your camp, your environment and your weight history can do something with a trend line that a web page cannot. If you are also managing a cut, that conversation is not optional. The broader camp context sits in the fight camp hub.

We should also name a boundary we are not crossing. There is a literature on repeated dehydration and heat exposure producing recurrent acute kidney injury, proposed as a pathway to chronic kidney disease of non-traditional origin, with marked prevalence reported in male outdoor workers chronically exposed to dehydration and heavy exertion. Exertional rhabdomyolysis is likewise increasingly recognised in athletes, military recruits and recreational gym users, with something like 10–30% of cases developing acute kidney injury. Those are occupational and clinical cohorts, not fighters. The mechanism is plausible for a hot gym across a long career. The epidemiology has not been done there, and we are not going to walk it across the gap. The gap is the finding.

What we could not verify

This article rests on a thin evidence base and it is worth being explicit about exactly how thin.

No adult-female combat-sport training-day hydration dataset exists anywhere we could find. The judo study included seven girls out of twenty-two, all aged about twelve. The wrestling study was entirely male. The only adult combat training-day report we located was all male. The GSSI team-sport review included female subjects in only six of twenty performance studies. Sweat rate, sweat sodium and the menstrual cycle's effects on fluid regulation are all live variables that this literature does not address for this population, and no coefficient closes that gap.

The two cleanest datasets are children. Twelve-year-old judoka and fifteen-year-old wrestlers. Every training-day figure in this article comes from adolescents. That is not a minor caveat — a professional adult's thermoregulation, sweat sodium concentration and drinking behaviour are different, and Armstrong's review specifically notes that thirst did not correlate with hydration status in youth populations, so the adolescent data may not even validate the adult advice.

No study covers a two-session day, which is precisely what a fight camp is and precisely what a reader wants to know. The wrestling study measured three single sessions of different types. The judo study measured one ninety-minute session. Whether deficits compound across a morning and an evening session is unmeasured. The only hint is that recovery was incomplete by the next morning after one session, in fourteen adolescent boys.

No measured sweat rate for training in a gi, under pads, or in a hot MMA gym exists. The gi hypothesis is entirely plausible and entirely unmeasured. Anyone stating a figure for it is extrapolating.

No combat-sport skill test appears in the hypohydration literature. Basketball shooting and cricket bowling are the nearest analogues.

No exercise-associated hyponatraemia case in a combat athlete was located. The condition is well documented in rugby, American football, calisthenics, military training and cycling. Nobody has published a fighter case we could find, which is an absence of evidence rather than evidence of absence.

The ACSM position stand is still the 2007 version. We checked ACSM's position-stand listing for a 2024–2026 replacement and found none. The field's most-cited hydration document is nineteen years old and predates most of the ad libitum debate it is routinely invoked to settle. NATA 2017 is the current statement to lean on.

Several sources we wanted, we could not open. A 2019 normative sweat-rate update that would be the obvious place to look for a combat-sport subgroup returned a publisher access error, so no number from it appears here. Goulet's primary paper returned an access error, so his position is described only as reported by Kenefick and is never quoted. Zubac's combat-sport specific-gravity paper is paywalled; we have the abstract only. A frequently second-cited finding that urine markers take roughly three hours to normalise after a dehydration bout could not be traced to its primary source, so we have used Armstrong's quantified confounders instead.

Things we refused outright. The vendor sweat figure of 1.0–2.5 L/h for judo, which has no study behind it. A commercial sodium-loss range of 500–1,500 mg per hour with no cohort or method. The eight-glasses rule. The superseded 2000 NATA drinking volumes, which we printed only to say loudly that they were replaced. And an unqualified "2% dehydration impairs performance," which is the claim this article exists to dismantle.

One correction to our own working notes. We had recorded the sub-2% cognitive effect size as "small but non-zero." Checking the published abstract, the confidence interval at or below 2% runs from −0.27 to 0.00 and therefore touches zero; the pooled effect at that magnitude is not statistically significant. The gradient across strata is real at p = 0.04; the lower end of it is not distinguishable from nothing. We have corrected the text accordingly rather than carrying the stronger version.

Sources

Sourced to

1. National Athletic Trainers' Association Position Statement: Fluid Replacement for the Physically Active — McDermott BP, Anderson SA, Armstrong LE, et al., Journal of Athletic Training, 2017;52(9):877–895. PMC5634236
2. Exercise-Associated Hyponatremia: 2017 Update — Hew-Butler T, Loi V, Pani A, Rosner MH, Frontiers in Medicine, 2017;4:21. doi:10.3389/fmed.2017.00021
3. Statement of the Third International Exercise-Associated Hyponatremia Consensus Development Conference, Carlsbad, California, 2015 — Hew-Butler T, Rosner MH, Fowkes-Godek S, et al., Clinical Journal of Sport Medicine, 2015. PMID 26102445
4. Exercise and Fluid Replacement: ACSM Position Stand — Sawka MN, Burke LM, Eichner ER, Maughan RJ, Montain SJ, Stachenfeld NS, Medicine & Science in Sports & Exercise, 2007;39(2):377–390. PMID 17277604
5. Sweating Rate and Sweat Sodium Concentration in Athletes: A Review of Methodology and Intra/Interindividual Variability — Baker LB, Sports Medicine, 2017;47(Suppl 1):111–128. doi:10.1007/s40279-017-0691-5 (Gatorade Sports Science Institute)
6. Fluid Balance in Team Sport Athletes and the Effect of Hypohydration on Cognitive, Technical, and Physical Performance — Nuccio RP, Barnes KA, Carter JM, Baker LB, Sports Medicine, 2017;47(10):1951–1982. doi:10.1007/s40279-017-0738-7 (Gatorade Sports Science Institute)
7. Fluid intake, hydration status and body mass changes in U-15 judo athletes during a training day — Ceylan B, Santos L, Acta Scientiarum Health Sciences, 2022;44:e57233. doi:10.4025/actascihealthsci.v44i1.57233
8. Post-training breakdown: acute effects of different training types on body hydration status and performance — Demirkan E, et al., Frontiers in Psychology, 2025;15:1528840. doi:10.3389/fpsyg.2024.1528840
9. Hydration status in elite wrestlers, judokas, boxers, and taekwondo athletes on competition day — Pettersson S, Berg CM, International Journal of Sport Nutrition and Exercise Metabolism, 2014;24(3):267–273. PMID 24280038
10. Urine specific gravity as an indicator of dehydration in Olympic combat sport athletes; considerations for research and practice — Zubac D, Reale R, Karnincic H, Sivric A, Jelaska I, European Journal of Sport Science, 2018;18(7):920–929. doi:10.1080/17461391.2018.1468483
11. Drinking Strategies: Planned Drinking Versus Drinking to Thirst — Kenefick RW, Sports Medicine, 2018;48(Suppl 1):31–37. doi:10.1007/s40279-017-0844-6 (Gatorade Sports Science Institute)
12. Dehydration and endurance performance in competitive athletes — Goulet EDB, Nutrition Reviews, 2012;70(Suppl 2):S132–S136. doi:10.1111/j.1753-4887.2012.00530.x
13. Protection of total body water content and absence of hyperthermia despite 2% body mass loss during prolonged exercise in cool conditions — Ely BR, et al., Medicine & Science in Sports & Exercise, 2011. PMID 21047838
14. Dehydration Impairs Cognitive Performance: A Meta-analysis — Wittbrodt MT, Millard-Stafford M, Medicine & Science in Sports & Exercise, 2018;50(11):2360–2368. doi:10.1249/MSS.0000000000001682
15. Reference Values for Hydration Biomarkers: Optimizing Athletic Performance and Recovery — Armstrong LE, et al., Open Access Journal of Sports Medicine, 2025;16:31–50. doi:10.2147/OAJSM.S508656
16. "Drink at least eight glasses of water a day." Really? Is there scientific evidence for "8 × 8"? — Valtin H, American Journal of Physiology: Regulatory, Integrative and Comparative Physiology, 2002;283(5):R993–R1004. doi:10.1152/ajpregu.00365.2002
17. Strategies to mitigate acute kidney injury risk during physical work in the heat — Chapman CL, et al., American Journal of Physiology: Renal Physiology, 2024. doi:10.1152/ajprenal.00350.2023
18. National Athletic Trainers' Association Position Statement: Fluid Replacement for Athletes (2000 — superseded) — Casa DJ, Armstrong LE, Hillman SK, et al., Journal of Athletic Training, 2000;35(2):212–224. Cited here only to document that it has been replaced by source 1

Read next