Short answer
Adjust before the gun, not after the wall. Add the forecast air temperature and dew point: below a combined 38°C there is no meaningful cost; at 24°C with an 18°C dew point plan on roughly 3% slower, which is about seven minutes across a marathon; in oppressive air the cost reaches 8–10% or more. Slow the opening kilometres by the full amount, run the middle by effort or heart rate rather than pace, and treat anything faster in the final third as a bonus. Pre-cool with an ice slurry and cold water in the hour before, drink to thirst, and if the race is longer than two hours in heat, keep some sodium in the bottle.
The most common way a good marathon becomes a bad one is not injury or illness. It is a cool-weather goal pace, run from the gun, on a morning that turned out to be 24°C and sticky. The fitness was real. The number was for a different day.
Why the goal is a ceiling, not a plan
Every race prediction — from a tune-up race, from a training block, from your watch — assumes the conditions it was made in. Heat changes the physiology underneath it. Blood is diverted to the skin for cooling, so less reaches the muscles; heart rate rises at any given pace; and when the dew point is high enough that sweat cannot evaporate, core temperature climbs regardless of what pace you are holding. The result is that the effort your goal pace requires goes up, and effort is the thing that is actually limited on race day.
Ely and colleagues quantified it across seven marathons: performance declined progressively with the heat index, and the effect was largest for slower runners — a 3-hour runner lost a few percent, a 5-hour runner considerably more, because they are on course longer and running in the pack where airflow is worst. The cost is not optional and it is not a sign of weakness. It is the air.
Repricing the goal
Add the forecast temperature and dew point for the hours you will be running. Then read the cost off the curve — the same one behind the heat-adjusted pace calculator, which will do the sums for your own pace and distance.
| Forecast (temp / dew point) | Cost | 4:00 marathon becomes | 1:50 half becomes |
|---|---|---|---|
| 14°C / 8°C — cool | 0% | 4:00 | 1:50 |
| 18°C / 12°C — mild | ~1% | 4:02 | 1:51 |
| 21°C / 16°C — warm, humid | ~2% | 4:05 | 1:52 |
| 24°C / 18°C — sticky | ~3% | 4:07 | 1:53 |
| 27°C / 21°C — hot, humid | ~4.5% | 4:11 | 1:55 |
| 30°C / 23°C — oppressive | ~6% | 4:14 | 1:57 |
The reflex is to argue with the table — to treat 4:07 as giving up on 4:00. Look at it the other way. The runner who starts at 4:00 pace in sticky air and blows up at 30km finishes in 4:25. The runner who starts at 4:07 pace and holds it finishes in 4:07, and quite often a little faster, because the final third in heat rewards the people who have something left. The adjusted target is not the pessimistic plan. It is the fast one.
Nobody has ever run a hot marathon well by pretending it was cool. The good ones repriced the goal on Friday.
The tactics
Slow the first 5km by the full amount
Heat cost compounds. Every second you bank early in hot conditions is borrowed against a much higher rate later. Start at the adjusted pace, or slightly slower, and let the race come to you.
Run the middle by effort, not pace
Heart rate or breathing, not the pace field. Cardiac drift means the same pace costs progressively more; holding the number is how the last 10km disappears. Racinais and colleagues’ consensus guidance says the same: pace to thermal strain, not to the clock.
Pre-cool in the final hour
Siegel and colleagues showed that drinking an ice slurry before running in the heat lowered core temperature at the start and extended time to exhaustion. Cold drinks, ice under the cap, shade at the start line and a short warm-up all lower the temperature you begin at — and it is the temperature you begin at that sets how long you have.
Drink to thirst; use the aid stations to cool
Losing up to about 2% of body mass is normal and tolerated; overdrinking is the dangerous error. Take what thirst asks for and pour the rest over your head and neck, where it does more good than in your stomach. Past two hours in heat, keep some sodium in what you drink.
Treat the final third as the bonus
If you feel good at 30km, that is the adjusted pacing working. Now you can spend it. If you feel ordinary, you are still on for the adjusted time, which was the honest goal all along.
What you can do before race week
If the race is likely to be warm, ten to fourteen days of consistent training in heat — or an overdressed easy run, or a hot bath after sessions — expands plasma volume and brings sweating on earlier, so the same air costs less. Périard’s review puts most of the adaptation in the first week and the rest in the second. It shrinks the tax; it does not repeal it, and it does not replace the adjustment above. Even the fastest runners in the world slowed in the Doha World Championships marathon, run at midnight in 32°C air and heavy humidity.
Know the number that matters, too. The forecast will give you temperature; you want the dew point, because 26°C and dry is a different race from 22°C and saturated. And know your own sweat rate from a warm long run, so that “drink to thirst” is backed by knowing roughly what a hot hour costs you.
After the race
A warm race that came in three percent slower than the cool goal is a race that went exactly to plan. Write it down that way. The mistake runners make in the days after is to log the slower time as a fitness ceiling and train the next block chasing a number the weather set. The fitness that produced the cool-weather goal is still there; it was priced for a day you did not get.
Kovr prices your race prediction for the weather you will actually get.
Every run syncs with the temperature and dew point it happened in, so your predictions are built from sessions that were themselves heat-adjusted. Set a race and Kovr shows what the forecast costs, moves your fuel targets through the loading days, and after the race tells you whether the air or the pacing was the difference — so a slow hot marathon reads as the air, not as a step back.
Download Kovr on the App StoreFree with a 14-day trial · Garmin, Apple Watch & Oura
Sources cited
- Ely, M.R., Cheuvront, S.N., Roberts, W.O. & Montain, S.J. (2007). Impact of weather on marathon-running performance. Medicine & Science in Sports & Exercise — performance decline versus heat index across seven marathons, largest for slower runners
- Siegel, R. et al. (2010). Ice slurry ingestion increases core temperature capacity and running endurance in the heat. Medicine & Science in Sports & Exercise — pre-race ice slurry lowered starting core temperature and extended running time in the heat
- Racinais, S. et al. (2015). Consensus recommendations on training and competing in the heat. British Journal of Sports Medicine — pacing to thermal strain, cooling strategies and competition preparation
- Périard, J.D., Racinais, S. & Sawka, M.N. (2015). Adaptations and mechanisms of human heat acclimation: applications for competitive athletes and sports. Scandinavian Journal of Medicine & Science in Sports — the timeline and mechanisms of heat acclimatisation
Frequently asked questions
Is it normal for HRV to drop during taper?
Yes, particularly in the first taper week. Heavy training defers repair, and reducing volume is when the body finally carries it out — muscle repair, glycogen restoration and plasma volume shifts are all metabolically demanding. The pattern typically recovers toward baseline during the second week.
Why do I feel terrible during taper when I'm running less?
Because the taper is when recovery gets paid for, not when it pays out. Heavy legs, sluggishness and phantom niggles are the repair process being felt from the inside. In one study of 11 runners, 5km performance was at its worst at the end of the heavy block and improved through the taper — fitness was already built, fatigue was still masking it.
When is a taper HRV drop actually a problem?
Three cases. If the trend is still falling five or six days out, the taper likely started too late or is too shallow. If HRV is down and resting heart rate is up with no session to explain it, treat it as a possible early illness signal. And if a niggle has a specific location and reproduces on a specific movement, it is an injury, not a taper phantom.
Should I do a hard session to check my fitness during taper?
No. It cannot answer the question — you cannot outrun taper fatigue in a week — and it re-adds the exact load you are trying to shed. Keep some short race-pace work to hold the neuromuscular pattern, but cut overall volume substantially.
Why did my weight go up during race week?
Glycogen and water. Every gram of stored glycogen brings water with it, so 1–2kg of gain across the final loading days is the carb load working rather than a nutrition error. Do not try to optimise HRV during a carb load.
How long does the taper dip last?
Most commonly the first week, with metrics climbing back toward baseline through the second week and sitting at or above baseline by race week. Individual variation is large, so the shape matters more than the schedule — what you are looking for is direction across several days, not any single morning.
Frequently asked questions
How much slower should I run a marathon in the heat?
Add the forecast temperature and dew point. Around 24°C with an 18°C dew point plan on about 3% slower — roughly seven minutes for a four-hour marathon; at 27°C with a 21°C dew point about 4.5%; in oppressive 30°C and 23°C air about 6%. Slower runners lose more because they are on course longer.
Should I still try for my goal time if race day is hot?
Treat it as a ceiling. Start at the heat-adjusted pace, run the middle by effort, and if you feel strong in the final third, spend it then. Starting at the cool-weather pace in heat is the most common cause of late-race blow-ups, and the blow-up costs far more than the adjustment.
Does pre-cooling before a race actually work?
Yes. Drinking an ice slurry in the hour before running in the heat lowered core temperature at the start and extended time to exhaustion in a controlled study. Cold drinks, ice under the cap, shade at the start and a short warm-up all lower your starting temperature, which sets how long you have before heat limits you.
Should I run by heart rate or pace in a hot race?
Effort or heart rate. Heat raises heart rate at any given pace and the gap widens as the race goes on, so holding the pace number means running progressively harder. Set a heart rate ceiling from your training and let the pace float.
How much should I drink during a hot marathon?
To thirst. Losing up to about 2% of body mass during the race is normal and well tolerated, and drinking beyond thirst is the error that causes hyponatraemia. Pour extra water over your head and neck rather than drinking it, and past two hours in heat keep some sodium in what you drink.
Does heat acclimatisation replace adjusting my pace?
No. Ten to fourteen days of training in heat expands plasma volume and starts sweating earlier, which shrinks the cost of hot air, but never removes it. Acclimatise if you can, and still reprice the goal for the forecast.