Hydration
Evidence: strong
Drink to thirst. Overdrinking, not dehydration, is the real danger: it causes exercise-associated hyponatraemia, which can be fatal. The “2% dehydration impairs performance” rule comes from fixed-pace lab tests and largely disappears in self-paced racing.
Not medical advice
This is a general knowledge base, not medical or dietary advice. If you are injured, unwell or weighing up a change to how you drink or fuel, speak to a doctor or registered dietitian who knows your situation.
The evidence-based position on hydration overturns much of what is sold to runners. Drink to thirst rather than to a fixed schedule. Overdrinking does not prevent fatigue, cramps or heat illness, and it raises the risk of a dangerous fall in blood sodium (Hew-Butler et al. 2017). The guidance shifted from the “drink as much as possible” advice of the 1990s precisely because that advice was harming people. The separate question of whether to add salt and other electrolytes to that fluid is much weaker, and is covered on its own page.
Hyponatraemia is the real danger
Exercise-associated hyponatraemia, a blood sodium below 135 mmol/L, is caused primarily by drinking too much hypotonic fluid combined with inappropriate fluid retention, not by losing too much sodium (Hew-Butler et al. 2017). Final blood sodium is governed by fluid volume, not sodium intake, which is why salt supplements cannot prevent it if fluid intake is excessive. It can be fatal, and symptomatic cases are treated with hypertonic saline, never more water. Risk rises with overdrinking and weight gain during an event, low body mass, female sex, slow pace, events over four hours, and freely available fluid on course.
Drink to thirst
For almost everyone, thirst is a good enough guide. It responds to the body’s actual need, it cannot be gamed by a schedule, and following it avoids both meaningful dehydration and the bigger danger of overdrinking. Practical signs back it up: pale-straw urine is a reasonable marker of being adequately hydrated, while drinking so much that body mass rises during a run is the warning sign to heed. There is no need to pre-load fluid or to drink on a fixed clock; start a session normally hydrated and drink when thirsty.
The 2% rule is contested
The familiar claim that losing 2% of body mass in fluid impairs performance comes from fixed-intensity laboratory tests. In self-paced time-trials that mimic real racing, exercise-induced dehydration changed performance negligibly, because thirst-guided drinking suffices (Goulet 2011). The 2% figure may still matter in fixed-pace efforts in the heat, but for self-paced racing it is not the hard limit it is often presented as.
How much to carry
Because the need is set by thirst rather than a schedule, most short road sessions need no fluid carried at all. A carrier earns its place when refills are scarce and losses are high: long runs, ultra and trail events, and hot conditions. The longer and hotter the effort, the more fluid (and the conditional case for added sodium) matters; the shorter and cooler, the less.