Hydration Guide: How Much Water You Actually Need

Nearly everyone has heard that you should drink eight glasses of water a day. Almost no one can say where it comes from, and the most-cited origin is a 1945 US Food and Nutrition Board recommendation which suggested roughly 2.5 litres a day — and then added, in a sentence that got dropped somewhere along the way, that most of this quantity is contained in prepared foods.

The advice survived; the qualification did not. The result is a widely repeated figure that ignores body size, activity, climate, diet and the substantial contribution of food and other drinks.

That does not make hydration unimportant. It makes a single universal number the wrong way to think about it.

A more sensible starting point scales with body size and adds for the things that actually increase losses. The Water Intake Calculator uses roughly 33 ml per kilogram of body weight as a baseline, then adds about 350 ml per 30 minutes of exercise and about 500 ml for hot or humid conditions.

For a 70 kg person that baseline is about 2.3 litres of total fluid a day, rising toward 3 litres with an hour of training, and higher again in heat.

Like every estimate on this site, that is a starting point rather than a prescription. Individual sweat rates vary by a factor of several between people doing identical work in identical conditions.

Total fluid intake includes far more than what you drink from a water bottle, and this is where the eight-glasses framing does the most damage — it implies a quota of plain water on top of everything else.

The simplest everyday indicator is urine colour. Pale straw suggests you are well hydrated; consistently dark yellow suggests you are not. It is imperfect — B vitamins and some medications colour urine independently — but it is free and immediate.

For most people in ordinary conditions, thirst is a reasonably reliable guide. The thirst mechanism is not, as often claimed, hopelessly delayed; it responds to changes in blood osmolality well before meaningful dehydration develops.

Thirst becomes less reliable in a few specific situations: in older adults, where the response blunts with age; during prolonged intense exercise, where losses can outpace the signal; and in illness with fever, vomiting or diarrhoea, where losses are both large and easy to underestimate.

Sweat rates during exercise commonly range from 0.5 to 2 litres per hour, depending on intensity, heat, humidity, clothing and individual physiology. That is a wide enough range that generic advice is nearly useless.

The practical way to find your own rate is to weigh yourself before and after a training session, without clothes, having noted what you drank. Each kilogram of weight lost represents roughly a litre of sweat. Do this a couple of times in typical conditions and you will know your own figure better than any calculator can estimate it.

Performance impairment becomes measurable at around 2% of body mass lost as fluid — for a 70 kg person, about 1.4 kg. Below that, the effect on most training is small.

Sweat contains sodium, and to a lesser extent potassium, magnesium and chloride. For sessions under about an hour, replacing plain water is entirely sufficient and electrolyte products are a marketing proposition rather than a physiological need.

They become genuinely useful in longer or hotter sessions — beyond roughly 60–90 minutes, particularly in heat, and especially for people who sweat heavily or salt-crust visibly on their skin and clothing.

The failure mode worth knowing about is exercise-associated hyponatraemia: drinking large volumes of plain water during prolonged endurance events dilutes blood sodium, and in severe cases it is dangerous. It is more common than dehydration in marathon finishers, which surprises most people.

Yes. Healthy kidneys can excrete roughly 0.8 to 1 litre per hour, and consistently exceeding that dilutes blood sodium. Outside endurance events this is rare, but it does happen — usually as a result of forcing intake to hit an arbitrary target rather than drinking to thirst.

The practical rule is unglamorous: sip regularly, drink to thirst, drink more in heat and around training, and do not force large volumes to satisfy a number.

There are also people for whom fluid intake needs to be actively restricted — those with certain kidney, heart or liver conditions. For them, generic hydration advice is not merely unhelpful but potentially harmful, and their target should come from a clinician.

Modestly, and less magically than the claims suggest. Drinking water before meals has been shown in some trials to reduce intake slightly, presumably through gastric distension, with effects most apparent in middle-aged and older adults.

There is also a small thermogenic effect from warming ingested water to body temperature, but it amounts to a few calories per glass — nowhere near enough to matter against daily variation.

The larger and more reliable benefit is substitution. Replacing sugar-sweetened drinks with water removes a meaningful number of calories that most people do not register as food at all. A litre of soft drink is roughly 400 kcal consumed with essentially no satiety return, which is about as inefficient as calories get.

Most people who are chronically under-hydrated are not lacking information; they are lacking cues. Attaching drinking to existing routines works better than resolving to drink more.

A glass with each meal, one on waking, and a bottle within reach at your desk covers most of a typical requirement without any conscious tracking. If you train, drinking around sessions handles most of the rest.

Beyond that, the honest advice is to stop worrying about it. For a healthy person with access to fluids and a functioning thirst mechanism, hydration is one of the more self-regulating aspects of health.