Skip to main content

Jupiterimages/Getty Images/Brand X/Jupiterimages/Getty Images/Brand X

The question

How much should I drink during exercise to stay properly hydrated?

The answer

Story continues below advertisement

Here's a riddle posed recently by South African scientists: A group of soldiers undertook a gruelling 14.6-kilometre march during which they lost an average of 1.3 kilograms. But sophisticated measurements with isotope tracers showed the total amount in water in their bodies actually increased by 0.53 kilograms. Where did this "extra" water appear from?

Groups such as the American College of Sports Medicine have long advocated weighing yourself before and after exercise to determine how much fluid you lost. A litre of water weighs exactly one kilogram - so by this calculation, if you're a kilogram lighter, that means you sweated out one litre more fluid than you replaced by drinking. Lose more than about 2 per cent of your starting weight, the ACSM warns, and your performance will suffer due to dehydration.

But the South African study, published in the British Journal of Sports Medicine by researchers at the University of Pretoria, adds fuel to a simmering debate about whether weight loss during exercise corresponds to water loss. They argue that some of the weight loss is from the energy stores you burn, and that your body has "hidden" stores of water that are released during exercise - which may mean we need to rethink how we approach hydration.

No one disputes that weight loss and water loss could, in theory, be different, says Nicholas Tam, a doctoral student at the University of Cape Town, whose forthcoming study in the Clinical Journal of Sports Medicine found similar anomalies in half-marathon runners: a loss of 1.4 kilograms on average, but with no change in "total body water" measured with isotope tracers.

"You will use fat, and you will use carbohydrate," he says, "and once you've burned it up, it's not there any more."

Scientists agree that when you burn carbohydrate, fat or protein, the byproducts are carbon dioxide, which you breathe out, and water, which actually adds to the fluid available for hydration.

Perhaps more significantly, your body stores carbohydrate in a form that locks away about three grams of water for every gram of carbohydrate. This water doesn't contribute to essential cellular processes until you start unlocking the carbohydrate stores, so your body sees it as "new" water when it's released during exercise.

Story continues below advertisement

For decades, these factors were assumed to be insignificantly small. But a 2007 paper by British scientists at the University of Loughborough estimated that a marathoner could conceivably lose 1 to 3 per cent of his or her body mass without any loss of water.

These calculations were swiftly disputed. Last year, researchers at Pennsylvania State University published a study funded by the Gatorade Sports Science Institute that put runners through a two-hour intermittent treadmill test and used the isotope tracer technique to measure water in the body. "We found that there were no statistical differences between the runners' change in body mass and the change in their total body water from pre- to post-exercise," says Lindsay Baker, the study's lead author, who is now employed by Gatorade.

The reason for these conflicting findings isn't yet clear, but it may relate to the details of the experimental protocols, since the amount of water produced by factors such as fuel-burning depends on the intensity of the exercise and whether the subjects have loaded up on carbs.

There's more than just curiosity about the inner workings of the body at stake in this debate. We've known for a long time that drinking only when you're thirsty during vigorous exercise means you'll lose some weight; the new results suggest that might not be a problem when it comes to performance.

Mr. Tam believes that a better marker of hydration status is "plasma osmolality," a measure of the concentration of various substances, such as sodium and glucose, in the blood. In his half-marathon study, the subjects were allowed to drink as much (or as little) as they liked. They lost about 2 per cent of their starting weight - a potentially harmful level, according to the ACSM - but showed essentially no change in plasma osmolality and, more importantly, no noticeable signs of compromised function.

Despite the controversy, weighing yourself before and after exercise remains by far the simplest way to estimate fluid loss. But the new studies suggest that you should use weight loss as a way of comparing the effects of different workouts, rather than worrying about a specific threshold (such as 2 per cent).

Story continues below advertisement

Instead, Mr. Tam says, you can trust your body to tell you when it needs water - as long as you listen carefully to signals such as thirst.

Alex Hutchinson blogs about research on exercise at

Report an error Editorial code of conduct
As of December 20, 2017, we have temporarily removed commenting from our articles as we switch to a new provider. We are behind schedule, but we are still working hard to bring you a new commenting system as soon as possible. If you are looking to give feedback on our new site, please send it along to If you want to write a letter to the editor, please forward to