Race nutrition
Hydration for runners: how much to drink and the hyponatremia risk
For decades runners were told to drink early and often, before thirst could set in. That advice was wrong. For slower marathoners it was dangerous. Here is what the evidence actually supports.
Drink to thirst rather than on a schedule: that is the short answer, and the bigger danger is drinking too much, not too little. Hydration is one of the few areas of running science where the official advice has been turned almost completely on its head. The slogan many of us grew up with, “drink before you are thirsty,” encouraged exactly the behaviour that has put runners in hospital and, on rare occasions, killed them. The villain is not dehydration. It is overdrinking.
This article walks through what the research really shows: how much you need to drink, why thirst is a better guide than a bottle schedule, what exercise associated hyponatremia is and why slower marathoners are most at risk, and how to handle electrolytes and your own individual sweat rate.
The old advice, and why it changed
Through the 1980s and 1990s, runners were urged to drink as much as possible to stay ahead of fluid losses. The fear was dehydration and heat illness. But as field marshals and finish line doctors collected real data, a different and more sinister pattern emerged: a small but serious number of runners were collapsing not from too little water, but from far too much.
Timothy Noakes, the South African physiologist who did more than anyone to expose this, argued bluntly that advising athletes to overdrink could cause fatal hyponatraemic encephalopathy, and that thirst alone is sufficient to protect hydration during exercise (Noakes, 2003). The era of drinking as much as you can was, in his words, a public health mistake.
What hyponatremia is, and why it is the bigger hazard
Exercise associated hyponatremia (EAH) is an abnormally low concentration of sodium in the blood, defined as below 135 mmol/L, occurring during or up to 24 hours after physical activity. In almost every case the cause is the same: the runner has consumed more fluid than they have lost, and the excess water has diluted the blood. Sodium falls, water shifts into cells, and in the brain that swelling can be catastrophic.
The landmark dataset comes from the Boston Marathon. Almond and colleagues (2005) drew blood from 488 finishers and found that 13 percent had hyponatremia, and 0.6 percent had critical, potentially fatal levels. The strongest predictors were substantial weight gain during the race, a marker of overdrinking, and a finishing time greater than four hours. The runners most at risk were not the fast, lean elites but the slower participants who had plenty of time to drink at every aid station.
The asymmetry that matters most.
Modest dehydration is uncomfortable and may slow you slightly. Hyponatremia can kill you. When you are unsure whether to take that extra cup, the safer error is almost always to drink a little less, not a little more.
The 3rd International Exercise Associated Hyponatremia Consensus statement (Hew-Butler et al., 2015) makes the headline message explicit: the most important preventive measure is to avoid overdrinking, and drinking to thirst is the recommended strategy for almost all athletes. The consensus also warns that sodium supplements do not reliably prevent EAH, because the underlying problem is fluid overload, not salt shortage.
Does dehydration really wreck your performance?
For years it was taken as gospel that losing even two percent of body mass to sweat would measurably impair endurance. The reality is more nuanced. Goulet (2011) pooled the controlled time trial studies and found that exercise induced body mass losses of up to around four percent did not impair real cycling time trial performance, and that drinking to thirst optimised performance. The catch is important: much of the laboratory evidence for dehydration harming performance came from protocols where participants rode at a fixed workload with no chance to slow down, or were artificially dehydrated beforehand, conditions that do not reflect a real race.
This does not mean dehydration is harmless. Heavy fluid deficits, especially in extreme heat, genuinely degrade performance and raise heat illness risk. The honest summary is that a modest, self paced fluid deficit of around two percent is normal, well tolerated and not something to panic about. If you want to go deeper on heat specifically, see our guide to running in the heat, where fluid and temperature management interact.
Drinking to thirst, in practice
The current consensus, echoed by the American College of Sports Medicine position stand (Sawka et al., 2007), is to replace enough fluid to keep total body mass loss within roughly two percent, while explicitly avoiding overdrinking. For almost all recreational runners, the simplest and safest way to hit that target is to drink to thirst.
A sane field protocol.
- Runs under 60 minutes in mild weather: usually no fluid needed at all.
- Longer or hotter runs: carry fluid and sip when thirsty, not by the clock.
- Never force fluid down to hit a number, and be especially cautious at the back of the pack in a marathon, where the temptation to drink at every station is greatest.
- If you finish a long race heavier than you started, you drank too much. Aim to finish a little lighter.
Hoffman and Stuempfle (2014) followed runners through a 161 km ultramarathon in extreme heat and found that drinking to thirst was the most common strategy, and that those who drank to thirst maintained hydration just as well as those who drank more deliberately. Drinking beyond thirst, they concluded, is not required to stay hydrated even in brutal conditions.
Electrolytes and your individual sweat rate
Sodium is the electrolyte that matters most for endurance, because it is the one you lose in meaningful quantities through sweat. But the popular fear that you must replace every milligram of lost salt to avoid cramp or hyponatremia is not well supported. As noted above, sodium supplements do not reliably prevent hyponatremia; managing total fluid intake does.
Where electrolytes earn their place is in long, hot efforts and for people who sweat heavily and saltily. Here the key word is individual. Baker (2017) reviewed the field and found enormous variation between athletes in both sweating rate and sweat sodium concentration, driven by genetics, heat acclimatisation, fitness, body size and diet. Generic guidelines are therefore a starting point at best.
Estimate your own sweat rate.
Weigh yourself before and after a one hour run, ideally with minimal clothing. Add back any fluid you drank. Each kilogram of net loss is roughly one litre of sweat per hour. Repeat in different conditions and you will quickly learn whether you are a light or a heavy sweater, which tells you far more than any generic chart.
For races, hydration is only one leg of a fuelling plan that also includes carbohydrate. We cover the intake side in detail in how many carbs per hour running and the build up phase in carb loading for a marathon.
The bottom line
The science has converged on a refreshingly simple message. Drink to thirst. Do not force fluids on a schedule. Accept that finishing a long run or race a couple of percent lighter is normal and not harmful. Use electrolytes for long, hot efforts and tune them to your own sweat, not a generic table. And remember the asymmetry that underpins all of it: a slightly dry runner can finish the race, while an overhydrated one can end up in hospital.
Frequently asked questions
How much water should I drink while running?
The modern consensus is to drink to thirst rather than on a fixed schedule. Thirst is a reliable signal that protects you from both dehydration and overdrinking. For runs under an hour in mild conditions, most runners do not need to carry fluid at all. Beyond that, sip when thirsty and let conditions guide the amount.
What is hyponatremia in runners?
Exercise associated hyponatremia is a dangerously low blood sodium concentration, almost always caused by drinking more fluid than you lose. The excess water dilutes the blood. Symptoms range from nausea and confusion to seizures, brain swelling and, in rare cases, death. It is the more lethal of the two hydration errors and is largely preventable by not overdrinking.
Is it better to be slightly dehydrated or overhydrated when racing?
Modest dehydration of around two percent of body mass does not reliably impair real world endurance performance and is rarely dangerous. Overhydration, by contrast, can cause life threatening hyponatremia. When the choice is genuinely uncertain, the safer error is to drink a little less, not a little more.
Do I need electrolytes or sports drinks when running?
For runs up to about 60 to 90 minutes, plain water and a normal diet are usually enough. Electrolytes matter more for long efforts, hot conditions and heavy, salty sweaters. Importantly, sodium supplements do not reliably prevent hyponatremia; the primary defence is not drinking to excess in the first place.
How do I know my own sweat rate?
Weigh yourself naked before and after a one hour run in typical conditions, accounting for any fluid you drank. Each kilogram of weight lost is roughly one litre of sweat. Sweat rates and sweat sodium losses vary enormously between individuals, so a personal estimate is far more useful than a generic guideline.
Related reading: running in the heat: how to train and race when it is hot.
References
- Almond, C.S.D., Shin, A.Y., Fortescue, E.B. et al. (2005) ‘Hyponatremia among runners in the Boston Marathon’, New England Journal of Medicine, 352(15), pp. 1550 to 1556. PubMed.
- Baker, L.B. (2017) ‘Sweating rate and sweat sodium concentration in athletes: a review of methodology and intra/interindividual variability’, Sports Medicine, 47(Suppl 1), pp. 111 to 128. PubMed.
- Goulet, E.D.B. (2011) ‘Effect of exercise-induced dehydration on time-trial exercise performance: a meta-analysis’, British Journal of Sports Medicine, 45(14), pp. 1149 to 1156. PubMed.
- Hew-Butler, T., Rosner, M.H., Fowkes-Godek, S. et al. (2015) ‘Statement of the Third International Exercise-Associated Hyponatremia Consensus Development Conference, Carlsbad, California, 2015’, Clinical Journal of Sport Medicine, 25(4), pp. 303 to 320. PubMed.
- Hoffman, M.D. and Stuempfle, K.J. (2014) ‘Hydration strategies, weight change and performance in a 161 km ultramarathon’, Research in Sports Medicine, 22(3), pp. 213 to 225. PubMed.
- Noakes, T. (2003) ‘Overconsumption of fluids by athletes’, BMJ, 327(7407), pp. 113 to 114. PubMed.
- Sawka, M.N., Burke, L.M., Eichner, E.R., Maughan, R.J., Montain, S.J. and Stachenfeld, N.S. (2007) ‘American College of Sports Medicine position stand: exercise and fluid replacement’, Medicine & Science in Sports & Exercise, 39(2), pp. 377 to 390. PubMed.
All citations point to peer reviewed primary sources or consensus statements. Page numbers and volume details are presented per Harvard referencing convention.
Building a smarter race day plan?
PaceBrain reads your training data and gives you an honest race time prediction, so you can pace and fuel for the finish you are actually trained for. Free, no signup.
Try the predictor