Sweatr
All articles
Hydration Science8 min read15 April 2026

Drink to Thirst or Follow a Plan? Warm-Weather Marathon Hydration Without the Cramp or the Hospital

How to avoid both cramping and hyponatremia when race day is hotter than you trained in — the middle path between drink-to-thirst and a rigid plan.

Drink to Thirst or Follow a Plan? Warm-Weather Marathon Hydration Without the Cramp or the Hospital

In Summary

  • Drink-to-thirst and "follow a plan" are both wrong at the extremes — especially in warm weather.
  • Thirst lags dehydration by 10–15 minutes, but drinking past your sweat rate causes hyponatremia.
  • Your safe range has a floor (your sweat rate) and a ceiling (never more than you lose).
  • Sweatr sets both numbers from your Apple Watch, Garmin, and Strava data and adjusts them for the forecast.

You're in the taper. You've checked the Boston or London forecast four times today. It's warmer than the 50°F you trained in all winter. And every marathon sub-Reddit thread you open tells you two opposite things:

  • "Just drink to thirst. Anything else risks hyponatremia."
  • "You can't trust thirst — by the time you feel it you're already dehydrated. Stick to the plan."

Both camps are right. Both camps are also wrong. The truth is that drink-to-thirst and follow-the-plan aren't opposite strategies — they're the floor and the ceiling of a safe range. In warm weather, that range gets narrower, and the cost of getting it wrong gets higher.

Here is the middle path, why generic advice fails in the heat, and exactly what to do at mile 6, mile 13, and mile 20 when the forecast turns.

Why both extremes fail in warm weather

The case against rigid plans

The old-school approach — "drink 200 ml every aid station, no exceptions" — was built on a dangerous assumption: that fluid loss is predictable. It isn't. Two runners of the same weight at the same pace can differ in sweat rate by a factor of two. Your body will also pull less water out of your stomach when you're working hard, so forcing fluid in can cause bloating, nausea, and in the worst case exercise-associated hyponatremia (EAH) — blood sodium dropping low enough to cause confusion, seizures, or death.

The 2002 Boston Marathon study that seeded this whole debate found that 13% of finishers were hyponatremic, and the strongest predictor was simply drinking too much water. Small, slower runners in warm conditions are at highest risk — the exact profile of a lot of first-time marathoners racing London or Boston this year.

The case against pure drink-to-thirst

Drink-to-thirst works beautifully when three conditions are true: the weather is moderate, your event is under three hours, and your sweat losses aren't dramatic. In warm weather, none of those are guaranteed.

Thirst is a lagging indicator. In the heat, by the time you feel thirsty you've often lost 2% of body mass in fluid — the threshold where pace drops and perceived effort climbs. Worse, as core temperature rises, the thirst signal itself gets blunted. Experienced runners know the feeling: you realise at mile 18 you haven't wanted a drink in an hour, and then the wheels come off.

The middle path

Use thirst as your permission to drink, but set hard numbers as your floor and ceiling:

  • Floor: the minimum fluid per hour that keeps you within 2% body-mass loss at your projected finish time in the forecast conditions.
  • Ceiling: an absolute cap equal to your hourly sweat rate. Never drink more than you lose per hour. This is the single rule that prevents hyponatremia.

If thirst shows up inside that range, drink. If you hit the ceiling and you're still thirsty, don't keep pushing water — take sodium instead (more on this below). If thirst doesn't show up at all and you're approaching the floor, drink anyway, small sips.

Why warm weather changes everything

A 70–75°F race is not just "a bit warmer." The physiology shifts materially:

  1. Sweat rate jumps. Typical fit runners sweat 0.5–1.0 L/hour in 50°F. Raise the temperature to 72°F with sun and that can climb to 1.2–1.8 L/hour. Some heavy sweaters push past 2 L/hour.
  2. Sodium loss climbs with it. Sweat sodium concentration is individual — ranging from about 400 mg/L for light-salt sweaters to over 1,500 mg/L for heavy-salt sweaters. Multiply that by a higher hourly sweat rate and the sodium delta between a cool race and a warm one is big.
  3. Gastric emptying slows. Blood is being diverted to the skin to cool you, so less is going to your gut. Your stomach will tolerate less fluid per hour than it does in training. This is why runners who "drank more to compensate" have worse races, not better.
  4. Thirst gets blunter, not sharper. Exactly when you need the signal to be accurate, it becomes less reliable.

The practical consequence: in warm weather, the safe range between your floor and your ceiling gets narrower. You need better numbers, not more willpower.

How to work out your floor and ceiling

You can do this with a scale and a bottle, or you can use your wearable data and an app that does it for you. Let's do it the manual way first so you understand what's happening.

Step 1: Know your sweat rate — for this weather

You probably did a sweat test in winter. That number is not your warm-weather number. Redo it in a temperature-controlled session that matches your forecast as closely as you can. The method:

  1. Weigh yourself naked before a 60-minute run at goal marathon pace.
  2. Run the full hour in the warmest conditions you can safely simulate. Note what you drank.
  3. Towel off, weigh naked again.
  4. Sweat loss (L) ≈ (pre-weight − post-weight) + fluid consumed.

If you lost 1.4 kg and drank 500 ml, your sweat rate in those conditions is roughly 1.9 L/hour. That's your ceiling. Do not drink more than 1.9 L in any hour of the race.

Step 2: Set your floor

Aim to finish the race having lost no more than 2% of your body mass. For a 70 kg runner that's 1.4 kg of net fluid deficit across the whole race.

  • Projected finish time: 4 hours
  • Sweat rate in race conditions: 1.5 L/hour → total loss 6 L
  • Allowed deficit: 1.4 L
  • Floor intake over the race: 6 − 1.4 = 4.6 L total, or ~1.15 L/hour

That's your floor. Below that line you drift past 2% loss and your finish time suffers.

Step 3: Set your sodium per hour

This is the piece most plans skip. Using your sweat rate and a sodium concentration estimate (most runners land in 600–1,000 mg/L; salty sweaters go higher):

  • Sweat rate 1.5 L/h × 800 mg/L = 1,200 mg sodium/hour

That is much more than any gel, and most sports drinks only cover 300–600 mg/L. This is why runners who rely on gels and plain water cramp in the heat — they're hitting their carb target but missing sodium by half.

Step 4: Build a simple in-race rule

Three cues, easy to remember on a course:

  • Every 15 minutes: 3–4 sips of your sports drink. That lands you between your floor and ceiling at typical sweat rates.
  • Every hour: one electrolyte tab or a sodium shot on top of the sports drink, targeting your per-hour sodium number.
  • If thirsty but at your ceiling: stop adding water. Take sodium instead, and pour water on your head to cool down — cooling reduces sweat rate and thirst within minutes.

The mile-10 check: am I on track?

Most runners don't realise they're off-plan until mile 20, when it's too late. Do this single check at mile 10:

  • Colour of the last time you urinated before the start: pale straw is fine; dark means you started already down. Adjust fluid up slightly early.
  • Mouth feel: sticky mouth at mile 10 is early dehydration; drink a little more per aid station.
  • Effort vs pace: if your HR is 8–10 bpm higher than it was on your long runs at the same pace, you're either dehydrated or your core temperature is climbing. Drink, cool with water over your head, slow 10 sec/km if needed.
  • Cramp twinges: early calf or hamstring twitches are almost always sodium, not fluid. Take an extra sodium tab; don't pile in more water.

The mile-10 check exists because the cost of correcting at mile 10 is small, and the cost of correcting at mile 20 is the difference between finishing and a DNF.

The three warm-weather mistakes we see most

  1. "I felt fine in training so I'll just wing it." Training plans rarely simulate your race-day conditions. If your longest run at goal pace was in 55°F and race day is 72°F, your entire fluid and sodium plan needs to scale up. Winging it is a version of follow-the-plan with the wrong plan.
  2. "I drank more to be safe." More is not safer — it's the single biggest risk factor for hyponatremia. The safer move is drinking to your number, not drinking more.
  3. "I didn't want to stop at the aid stations so I skipped them." Every aid station you skip in warm weather compounds. Carry a soft flask if you need to, but don't trade a 4-second stop for a mile-22 wall.

Where Sweatr fits

Doing this by hand works — and you should understand the logic whether or not you use an app to execute it. What an app gives you is continuity: your sweat rate updates as your fitness changes, the forecast re-enters the equation automatically, and the floor and ceiling are recalculated for your goal time on the day.

Sweatr pulls your Apple Watch, Garmin, and Strava data, personalises your sweat rate from your training history, layers the race-day forecast on top, and gives you a simple per-hour number for both fluid and sodium — with a ceiling you can't accidentally push past. It's the middle path, built for you, updated weekly.

If your race is inside the next four weeks and the forecast looks warmer than what you trained in, you don't need to pick a side in the drink-to-thirst debate. You need your own numbers.

One-line summary

Drink to thirst, but never past your sweat rate — and don't forget that the gap between your fluid plan and your sodium plan is where warm-weather marathons are won or lost.

Ready to stop guessing?

Sweatr builds your fueling plan automatically

Connect your Garmin, Apple Watch, or Strava and get a personalised hydration and fueling plan before your next long run. Set up in 5 minutes.