How to Calculate Your Sweat Rate and Build a Hydration Plan

You are already guessing, you just do not know the size of the guess

You finish a long ride, drink a bottle because the ride was long, and get on with your day. Maybe you were 400 ml short. Maybe you were two litres short. You have no idea which, and neither does the person who told you to drink 500 ml per hour.

That advice is not wrong so much as it is unaddressed to anyone. Sweat rates in trained athletes span roughly 0.5 to 2.0 litres per hour, and the top of that range is four times the bottom. A blanket recommendation lands correctly on the small fraction of people who happen to sit near the middle of the distribution and quietly fails everyone else in one of two directions: chronic under-drinking that costs you the back half of long sessions, or over-drinking that at best sloshes and at worst is a medical emergency.

The fix is a 90-minute test with a bathroom scale. It is one of the highest information-per-effort measurements available in endurance sport, and almost nobody does it.

The measurement

Sweat rate is body mass lost, corrected for what went in and what came out, divided by time.

Sweat rate (L/h) = (pre-mass kg - post-mass kg + fluid intake L - urine output L) / duration in hours

One kilogram of body mass lost during exercise is treated as one litre of sweat. That approximation is doing some work, because you also oxidise substrate and lose water vapour through breathing, but for sessions under about three hours the error sits under 5% and is not worth chasing.

The protocol:

1. Empty your bladder, strip to nothing or to dry underwear, and weigh yourself on a scale that reads to 0.1 kg. Write it down.
2. Train for 60 to 90 minutes at a steady, representative effort. Note the temperature and humidity.
3. Track every millilitre you drink. Weighing your bottle before and after is more reliable than guessing at how much of it you finished.
4. If you urinate, capture the volume or at least note that it happened. It matters more than people expect on long sessions.
5. Finish, towel off completely, remove wet kit, and weigh yourself again on the same scale in the same state.

Worked example. You start at 74.2 kg, finish at 72.7 kg, drink 500 ml, and ride for 90 minutes. That is (74.2 - 72.7 + 0.5) / 1.5 = 1.33 litres per hour. Our sweat rate calculator does the arithmetic and the unit conversions if you would rather not.

Three details ruin more tests than anything else. Towel off properly, because a litre of sweat sitting in your jersey reads as a litre you did not lose. Use the same scale both times, since absolute accuracy does not matter but consistency does. And do not eat during the test, because food mass shows up as retained weight and quietly deflates your result.

One number is not a plan

Here is where most people stop, and where the useful part actually starts.

Your sweat rate is not a property of you. It is a property of you under a specific set of conditions, and it moves substantially when those conditions move.

Temperature and humidity. These are the big ones, and humidity is the one athletes underrate. Sweat only cools you when it evaporates. In humid air it cannot evaporate efficiently, so your body produces more of it chasing a cooling effect it is not getting. A 28 degree day at 80% humidity will empty you faster than a 32 degree day in dry heat, which is why athletes who train in dry climates get ambushed at humid races.

Intensity. Metabolic heat production scales with work rate, so a threshold session produces meaningfully more sweat than a zone 2 ride of the same duration. If you test on an easy ride and plan a race off it, you will be short.

Airflow. This is why indoor trainer sessions produce the highest sweat rates most cyclists ever record. Outdoors at 30 km/h you have a permanent evaporative advantage that disappears the moment you come inside. Do not plan an outdoor race off an indoor number, or the reverse.

Heat acclimation. Over about 10 to 14 days of consistent heat exposure, your body starts sweating earlier and more, while the sweat itself gets more dilute as the glands reabsorb sodium more efficiently. Your April number is not your July number.

So run a small grid rather than a single test. Four sessions covers most of what you need: cool and easy, cool and hard, hot and easy, hot and hard. That gives you a range and, more importantly, a sense of how sharply your own rate responds to heat. Some athletes barely move between 15 and 30 degrees. Others double.

What to do with litres per hour

The number by itself does not tell you how much to drink. It tells you how much you are losing, and the target sits somewhere below that.

The ACSM position stand on exercise and fluid replacement (Sawka et al., 2007, Medicine & Science in Sports & Exercise) recommends limiting body mass loss to under 2% during exercise. For a 70 kg athlete that is 1.4 kg, and for a 55 kg athlete it is 1.1 kg. Note that this is a percentage rather than a volume, which is the first reason bottles-per-hour advice fails: a 90 kg cyclist and a 52 kg runner do not have the same allowance.

Work backwards from there. If you sweat 1.3 L/h and you are 70 kg, your 2% budget is 1.4 litres of allowable deficit. On a two-hour session you will lose 2.6 litres, so you need to replace at least 1.2 of them to stay inside budget, which is 600 ml per hour. On a five-hour ride you lose 6.5 litres, and the budget barely moves, so you need close to 1.0 to 1.1 L/h. The longer the event, the closer your intake has to track your actual sweat rate, because the fixed deficit allowance gets spread thinner and thinner.

There is a real scientific argument about how hard that 2% threshold is. Wall and colleagues (2015, British Journal of Sports Medicine) found that dehydration up to 3% did not impair cycling time-trial performance in the heat when athletes were blinded to their hydration status, and Goulet's earlier meta-analysis pointed the same way for exercise under two hours. The honest reading is that 2% is conservative for short efforts and much more defensible as events stretch past three hours, where thermoregulatory and cardiovascular drift compound. Treat it as a planning anchor rather than a cliff edge.

The ceiling matters more than the floor. Exercise-associated hyponatraemia is caused primarily by drinking more than you sweat, and the 2015 international consensus statement (Hew-Butler et al., Clinical Journal of Sport Medicine) is unambiguous that overdrinking, not sodium loss, is the driver. If you finish a long race heavier than you started, you drank too much. That is a rule worth holding onto: never plan to drink more than your measured sweat rate.

Sodium, and why the range is so wide

Fluid is only half of it. Sweat sodium concentration ranges from about 10 to 90 mmol per litre between individuals, roughly 230 to 2,070 mg per litre. Baker's 2017 review in Sports Medicine documents that the between-person variability dwarfs the within-person variability, which has a useful practical consequence: once you know whether you are a salty sweater, that classification tends to hold.

At the low end, an athlete losing 1.0 L/h at 250 mg/L sheds 250 mg of sodium per hour and can go a long way on plain water plus normal food. At the high end, 2.0 L/h at 1,800 mg/L is 3,600 mg per hour, and by hour four that is a genuinely large deficit that plain water will not cover.

Without a lab test, the crude indicators are more useful than nothing. White crusting on kit and skin after a hard session, sweat that stings your eyes badly, and a persistent taste of salt all point to the high end. Cramping is a weaker signal than it is usually treated as, since the evidence linking cramp specifically to sodium loss is far softer than the internet suggests, but it is worth noting alongside the rest.

The practical move is to start with something in the 300 to 700 mg of sodium per litre range for sessions over 90 minutes, then adjust based on how you feel late in long efforts and whether you are one of the visibly salty ones.

Reading it in your training data

Hydration status shows up in numbers you are already collecting, if you know which ones to watch.

Cardiac drift. As plasma volume falls, stroke volume drops and heart rate climbs to hold cardiac output. On a steady-state session, heart rate rising 5 to 10 bpm over the back half at unchanged pace and power is a textbook signature. It is not exclusively dehydration, since heat load and glycogen depletion produce the same pattern, but combined with a hot day and a light bottle it usually is.

Pace and power decoupling. The ratio between output and heart rate degrading through a session is the same phenomenon viewed from a different angle. This is exactly the pattern documented in our analysis of how much pace slows in the heat, where the effect is measurable across a whole population of athletes rather than just felt.

Next-morning resting heart rate. Meaningful fluid deficits carried overnight elevate resting heart rate and suppress HRV, which reads on your watch as a poor recovery score. Chronically under-drinking on long sessions produces a recovery signal that looks like overreaching, which is one of the more expensive misreads available, because the response to it is usually to train less rather than to drink more. The distinction is covered in more depth in our guide to HRV-guided training and the early signs of overtraining.

Morning body mass. The cheapest longitudinal measurement there is. Weigh yourself at the same point every morning, and a drop that persists across two or three days after hard training is usually incomplete fluid replacement rather than anything more interesting.

This is the kind of cross-referencing that is tedious by hand and trivial once your training and recovery data live in one place. If your rides, runs, weight and overnight recovery all flow into a single coach, connecting a hot session to the next morning's suppressed HRV stops being a thing you have to remember to do. That is the whole idea behind connecting your training data to an AI coach, and it is why the Garmin MCP server exists: so the numbers can be interrogated rather than merely stored.

Race day is a different problem

Training hydration is about consistency. Race hydration is about executing a plan under conditions you cannot fully control, and it deserves separate treatment.

Start topped up, not overloaded. Drink normally in the 24 hours before, and take 400 to 600 ml in the two hours before the start. Pale yellow urine is a good enough target. Loading beyond that just means you spend the first hour looking for a portaloo.

Plan against the forecast, not against last month. If you tested in 18 degrees and race day is 29, your sweat rate is going somewhere near the top of your measured range. This is the single most common way a hydration plan fails: it was correct, for different weather.

Rehearse the plan in training. Drinking 900 ml per hour is a skill, and gut tolerance for fluid and carbohydrate is trainable. Discovering on race morning that you cannot stomach your planned intake is a solved problem, and the solution is practising it on your long sessions. This is one more reason brick workouts earn their place in a triathlon build, since they are where transition-day fuelling actually gets tested.

Know your aid station spacing. Your plan is a function of what the course gives you. Litres per hour is meaningless if the stations are 12 km apart and you only carry one bottle.

Putting it together: a practical hydration protocol

1. Run a baseline sweat test on a 60 to 90 minute steady session in cool conditions. Weigh in and out, track fluid, record temperature and humidity.
2. Repeat it in three more conditions: cool and hard, hot and easy, hot and hard. Four data points give you a range rather than a single fragile number.
3. Convert the range into a personal rule of thumb, something like "1.0 L/h under 20 degrees, 1.6 L/h over 27 degrees."
4. Calculate your 2% body mass budget. Multiply your weight in kg by 0.02. That is your allowable deficit in kg, and therefore roughly in litres.
5. For any session, multiply expected sweat rate by duration, subtract the budget, and divide by duration. That is your minimum intake per hour. Never plan above your measured sweat rate.
6. Classify your sodium loss as low, moderate or high using kit residue and late-session symptoms. Start at 300 to 700 mg of sodium per litre for anything over 90 minutes, and adjust from what you observe.
7. Rehearse the exact race intake on at least two long sessions before the event, using the same bottles and products.
8. Retest after any 10 to 14 day block of heat exposure, because acclimation will have moved your numbers.
9. Watch cardiac drift within sessions and morning resting heart rate across days. Both are early evidence that the plan is under-delivering.
10. Weigh in after key sessions through the season. It takes fifteen seconds and it is the only direct feedback the whole system provides.

Everything above rests on one number that takes 90 minutes and a bathroom scale to obtain. Most athletes will spend more than that this week comparing carbon wheels.

Your body loses a specific, measurable amount of fluid per hour. You can either know that number or keep drinking on a schedule someone invented for a person who is not you. If you would like a coach that reads your sessions, your weather, and your recovery data and holds you to the plan you actually built, start a 7-day free trial.