The eight-glasses rule is a misquotation, and we can point at the sentence
Almost every water calculator on the internet is built on top of a piece of advice that nobody can source. Follow it back far enough and the trail ends in 1945, at the Food and Nutrition Board of the US National Research Council, which wrote that a suitable allowance of water for adults is about 2.5 litres a day. Then came the sentence that everyone forgot: most of this quantity is contained in prepared foods.
Two and a half litres is roughly 84 fluid ounces, or about ten eight-ounce glasses. Strip off the qualifier and you have "drink ten glasses of water a day"; round it down and you have the number that ended up on refrigerator magnets. In 2002 Heinz Valtin, a kidney physiologist at Dartmouth Medical School, went looking for the evidence behind the rule in a paper whose title was simply a question — Drink at least eight glasses of water a day. Really? — and reported finding none. No trial, no committee, no publication. The most likely origin was that 1945 paragraph with its final sentence amputated.
The reason this matters is not pedantry. Eight eight-ounce glasses is 1.89 litres. Run the calculator above for a 60 kg sedentary woman in a temperate climate and the drinking target comes out around 1.5 to 2.0 litres — eight glasses sits at the top of her range. Run it for a 70 kg sedentary man and it lands at 2.0 to 2.8 litres, where eight glasses is the floor. One rule, two people, opposite meanings. That is what a number without a body attached to it is worth.
Total water and drinking water are two different numbers
The figures people quote as authoritative come from the 2004 Dietary Reference Intakes published by the US Institute of Medicine, now the National Academies (NASEM): an adequate intake of 3.7 litres a day for adult men and 2.7 litres for adult women. Both of those are total water — everything you swallow, food included. The same report estimated that about 20% of the water in a typical diet arrives in food, which puts the beverage share nearer 3.0 litres for men and 2.2 for women.
Handing someone 3.7 litres as a drinking goal counts the water on their plate twice, and that is the standard failure of competing calculators. It is a 20% error in the direction that costs you bathroom trips and, in the tail, safety. This tool keeps the two figures in separate cards and shows the subtraction.
Food water is not a rounding term. Watermelon, cucumber and lettuce run above 90% water. Cooked rice is about two-thirds water and cooked beans about seven-tenths, which is why a person eating rice, beans, salad and fruit every day genuinely needs to drink less than someone living on bread, crackers and cured meat.
| Food | Water by weight | Per 200 g serving |
|---|---|---|
| Lettuce, cucumber | ≈ 96% | ≈ 190 mL |
| Watermelon, strawberry | ≈ 92% | ≈ 185 mL |
| Orange, apple, tomato | ≈ 86% | ≈ 172 mL |
| Cooked beans, yoghurt | ≈ 70% | ≈ 140 mL |
| Cooked rice, pasta | ≈ 68% | ≈ 136 mL |
| Bread | ≈ 37% | ≈ 74 mL |
| Crackers, nuts, cured meat | ≈ 5% | ≈ 10 mL |
One more thing about the reference values: an adequate intake is a median of what a healthy population was observed to consume, not a physiological minimum and not a target you fail by missing. The European authority EFSA, looking at European populations, landed on 2.5 litres of total water for men and 2.0 for women — 1.2 litres below the US figure for men. Two expert panels, two continents, a gap the size of a large bottle. The honest conclusion is that the true requirement is a range, which is why this calculator never prints a single number.
Sweat is the term that actually moves the answer
Body weight sets the baseline, but sweat sets the day. Measured sweat rates during exercise run from about 0.5 to 2.0 litres per hour, and trained athletes in the heat have been recorded above 3 L/h. That is a fourfold spread between two people doing the same session in the same room, and it dwarfs every other input on this page. An hour of hard running on a hot afternoon can add more fluid to your day than your entire resting requirement.
Heat and humidity act separately, and humidity is the underrated one. In dry heat sweat evaporates and cools you efficiently. In humid heat it drips off without evaporating, so the body sweats harder to achieve the same cooling and you lose more water for less benefit. Fitness works the same way, counter-intuitively: acclimatised, well-trained people start sweating sooner and sweat more than untrained ones. Being fit raises your fluid needs, it does not lower them.
If you want your own number instead of a table's, weigh yourself naked before and after a typical session, allowing for anything you drank during it. One kilogram of weight lost is about one litre of sweat. Two or three sessions is enough to learn your personal rate, and that number beats any calculator, including this one.
There is a ceiling, and it is not theoretical
A water calculator that only ever pushes you upward is an unsafe tool. Water intoxication is real. The healthy kidney can only clear so much free water per hour — commonly cited as 0.8 to 1.0 litres per hour — and drinking past that rate dilutes blood sodium. Cells take on water to correct the gradient, and because the brain sits inside a rigid skull, the swelling has nowhere to go. This is dilutional hyponatremia, and it has killed endurance athletes, hazing victims and radio-contest entrants.
The cruel part is the symptom overlap. Early hyponatremia presents as nausea, headache, bloating and confusion — which is very close to how dehydration presents. Someone who has been told all day to drink more feels awful and drinks more. Screening at a large city marathon found a substantial minority of finishers hyponatremic, and the strongest predictors were slow finishing times, weight gained during the race, and heavy fluid intake, not heat.
Two practical rules come out of it. First, spread intake: the calculator's daily plan exists because the same volume taken as two enormous bottles is how people get near the hourly ceiling, and any serving above roughly 800 mL gets flagged to be split. Second, when sweat losses are large, replace sodium along with the fluid. Sweat is salty. Above about 1.5 litres of loss, plain water alone starts to work against you, and salted food, a sports drink or an electrolyte tablet solves it.
Coffee, tea, and the drinks people wrongly discount
The belief that coffee and tea dehydrate you is one of the most durable pieces of nutrition folklore, and at ordinary intakes it is false. Caffeine is a mild diuretic acutely, but habitual drinkers develop tolerance to that effect within days, and controlled trials comparing coffee against an equal volume of water find no meaningful difference in hydration markers at intakes up to roughly 400 mg of caffeine a day. The water in the cup comfortably exceeds the extra urine it provokes. Your coffee counts.
Alcohol is the genuine exception. It suppresses vasopressin, the hormone that tells the kidney to hold water back, and produces a net fluid deficit — which is why a beer belongs in a different mental column from an iced tea. Milk, juice, soup and sugar-free soft drinks all count normally; sweetened drinks count as fluid but arrive with calories you would not have chosen to drink.
Who should ignore this page entirely
Some people are told by a doctor to drink less, and for them generic advice to drink more is not merely useless, it is harmful. Fluid restriction is standard management in heart failure, in advanced chronic kidney disease and dialysis, in cirrhosis with ascites, and in the syndrome of inappropriate antidiuretic hormone secretion. Prescribed limits are often around 1.5 litres a day and sometimes lower. Certain medications — thiazide diuretics, some antidepressants, desmopressin — raise hyponatremia risk independently of how much you drink.
Age cuts both ways. After about 65 the thirst signal blunts and the kidney concentrates urine less effectively, so thirst arrives late and dehydration is common; drinking on a schedule works better than drinking on demand. Infants and small children are not covered by any of the arithmetic here and should never be given plain water in volumes worked out from an adult formula.
How this calculator builds its range
The baseline is millilitres per kilogram of body weight per day, banded by age, with a span rather than a point: the lower bound is close to clinical maintenance fluid, the upper bound approaches the adequate-intake territory of the reference figures. Men get a 10% uplift, standing in for greater lean mass rather than for sex itself. Above 100 kg only half of the excess weight counts, because adipose tissue is roughly 10 to 20% water against about 73% for lean tissue, and a flat per-kilogram rule asks a 150 kg person for almost seven litres a day.
Climate adds a resting increment for the extra evaporative loss you incur without doing anything, and separately multiplies the sweat rate for whatever exercise you logged. That multiplier is capped, because stacking heat, sun and an all-out session produced sweat rates above 4 L/h in testing — beyond anything recorded in a human being. Pregnancy adds about 300 mL and breastfeeding 700 to 1100 mL, reflecting the 750 to 800 mL of milk a day that has to come from somewhere. Finally the food fraction is subtracted from the baseline only, never from the sweat term, because your dinner does not get wetter when you run: extra sweat has to be replaced by drinking, which is why the food share falls from about a fifth on a rest day to well under an eighth on a hard one.
Signals that beat any calculated number
Use the output as a starting point and then let your body correct it. Pale straw-coloured urine through the day is the classic marker; consistently dark and scant suggests you are behind, and completely colourless all day long suggests you are drinking more than you need. B-vitamin supplements turn urine bright yellow and make the test useless for a few hours. In healthy adults under about 65, thirst is a reasonably good controller on its own — the number here is mostly useful for planning a hot day, a long session, or a life stage where thirst is not trustworthy.
What this tool does not do
It does not calculate electrolyte replacement doses, model a specific medical condition, or replace an individualised prescription from a clinician or a sports dietitian. It assumes a healthy adult with normally functioning kidneys and heart. It does not cover infants, and it does not account for fever, burns, vomiting, diarrhoea, high altitude or air travel, all of which raise losses in ways this arithmetic does not model.
Everything runs in your browser as plain arithmetic. Nothing is uploaded, nothing is stored, and there is no account. Disconnect the network and the calculator still works.
Frequently asked questions
Where did “8 glasses a day” actually come from?
The best-documented origin is a 1945 recommendation from the Food and Nutrition Board of the US National Research Council, which suggested about 2.5 litres of water a day for adults — and then added, in the very next sentence, that most of that quantity is contained in prepared foods. The qualifier was dropped and the number survived on its own. In 2002 the kidney physiologist Heinz Valtin searched the literature for evidence supporting the eight-glass rule and reported finding none: no study, no committee report, no trial. It is advice with a citation trail that ends in a misquotation.
Is the 3.7 litres for men how much I should drink?
No, and this is the single most common error in water calculators. The 3.7 L for men and 2.7 L for women in the US Institute of Medicine reference are TOTAL water: food and drink added together. Roughly 20% of a typical diet's water arrives in food, so the beverage share of those figures is closer to 3.0 L and 2.2 L. A tool that hands you 3.7 L as a drinking target has counted the water on your plate twice. This calculator reports the two numbers separately for exactly that reason.
Can you drink too much water?
Yes, and it can be fatal. Drinking faster than the kidney can excrete free water — somewhere around 0.8 to 1.0 litres per hour in a healthy adult — dilutes blood sodium, and cells, including brain cells, swell. It is called dilutional hyponatremia. It has killed marathon runners, fraternity pledges, and contestants in radio drinking competitions. Early symptoms are nausea, headache and confusion, which are also the symptoms of dehydration, which is exactly why people respond by drinking more. Spread the volume across the day and take sodium with fluid when sweat losses are large.
Do coffee and tea count toward my intake?
Yes. Caffeine is a mild diuretic acutely, but habitual drinkers develop tolerance and controlled studies comparing coffee against an equal volume of water find no meaningful difference in hydration status at ordinary intakes of up to roughly 400 mg of caffeine a day, about four cups. The fluid in the cup more than covers the extra urine. Alcohol is genuinely different: it suppresses vasopressin and produces a net fluid loss, so beer does not count the same way coffee does.
Is my data sent anywhere?
No. Every number here is arithmetic running in your browser. Nothing is uploaded, nothing is stored, there is no account, and the page keeps working with the network disconnected.