ES·EN·PT Members Access
Heat, humidity and performance

Racing in the tropics without losing twice what you had to

You train where it is cold and you race where it is hot. This guide tells you what that costs, what you can do with the time you have left, and what is not worth doing at all.

This photo is Mar del Plata, Ironman Argentina 2022, on the run. I went to a race beside the sea expecting cool weather and found 35 °C. I had planned a marathon of 3:30 to 3:45 and crossed the line in 4:20.

That is between 16 and 24% over plan: more than this same guide predicts for a standalone marathon, and exactly what it predicts for the run leg of an Ironman, which is the scenario with the widest margin of all. I was not short of fitness. I was short of heat-specific preparation, and I never adjusted the target before the gun.

I wrote this guide so it does not happen to you.

See what it will cost you
Iván Koch running the marathon leg of Ironman Argentina 2022 in Mar del Plata, hair soaked and face set, under direct sun.
IRONMAN Argentina 2022 · 35 °C · 4h20
Somewhere on the run, midday, no shade at all. Behind me, two people walking in the shade in short sleeves: same day, different sport.

How long until your race?

Start here. Each card takes you to what you can actually do in the time you have.

If you have no time to read all of it, read how to start and how to cool. It is free, it needs no preparation, and it is what will change your race most.

01 — The mechanism

Why does heat make you slower?

There is one idea here, and if it is the only thing you take away the guide has paid for itself:

In a hot race you do not run out of distance. You run out of thermal margin.

In a classic study, cyclists were ridden to exhaustion in three different states: starting cold, starting normal, or pre-warmed. The times to exhaustion were very different. The core temperature they stopped at was not.

63 minStarting cold (35.9 °C core)
46 minNormal start (37.4 °C core)
28 minPre-warmed (38.2 °C core)

All three groups stopped at the same core temperature: 40.1 °C. It does not matter how you get there. When you arrive, it is over.

That reorders everything else. Anything that lowers your starting temperature, slows the climb, or raises the ceiling buys you minutes. Starting slower, cooling before the gun and acclimating are not three separate tricks: they are three ways of doing the same thing.

Why is your heart rate higher at the same pace when it is hot?

It is the question I get most, and the answer is mechanical, not a fitness problem.

When your temperature rises, your body sends blood to the skin to shed heat. That blood pools in the skin's veins and does not return to the heart as efficiently. Less blood returning means less volume per beat. To hold the same cardiac output, the heart compensates with the only thing it can: more beats per minute.

It is called cardiovascular drift. You are not in worse shape than you were last week. You are paying a cooling tax.

What to do
  • Stop reading a high heart rate on a hot day as lost fitness. Your heart rate will climb during the race even if you do not speed up, and that is normal.
  • If you acclimate, that tax falls: between 4 and 12 beats at the same pace in trained athletes, up to 17 in less trained ones. It is the fastest and most reliable adaptation there is.
02 — The number

How much slower will you run?

Here is the number, and it comes with its limits, because most of the calculators you will find do not give you those.

The rule: performance falls between 0.3% and 0.4% per degree of WBGT above the optimal range, which sits between 7.5 and 15 °C WBGT. It comes from an analysis of 1,258 races and 7,867 elite athletes between 1936 and 2019.

WBGT is a measure of heat stress that combines humidity, solar radiation and air temperature. I will explain it once, properly, because you will use it through the whole guide.

How it is calculated

WBGT = 0.7 × wet bulb + 0.2 × black globe + 0.1 × air temperature

Read the weights. Air temperature — the one thing everybody checks — carries 10%. Humidity carries 70%. That is why the thermometer lies to you on the coast.

WBGTDegrees over optimalTrained athleteMid-pack
20 °C+51.5 – 2.0%3 – 6%
24 °C+92.7 – 3.6%5 – 11%
27 °C+123.6 – 4.8%7 – 14%
30 °C+154.5 – 6.0%9 – 18%

Derived table, not measured. The columns apply the published coefficient; the mid-pack column multiplies by 2–3 from large-marathon data, where 300th place lost 7.9% while 25th place lost 2.6%. Valid from the half marathon upward. For 5K and 10K no published equivalent exists: distrust anyone who gives you a number there.

An example with your own numbers

You run a 3:30 marathon, which is 4:58 per kilometre. You go to Cozumel, which by mid-morning is close to 27 °C WBGT.

3.6 – 4.8%Estimated penalty
+7 to +10 minOn your goal time
11 – 14 s/kmSlower per kilometre

The same calculation for someone running 4:30 (6:24/km), in the same race: 7 – 14%, which is +19 to +38 minutes.

That is not a flaw in the table. The slower you are, the harder heat hits you, because you spend longer exposed, accumulating heat at a rate you cannot shed.

Two things the table does not say

One. Acclimating compresses the penalty. It does not remove it. Nobody races Cartagena the way they race a mild winter. The goal is not to cancel the heat: it is to stop losing twice what you had to.

Two. The risk of not finishing climbs faster than the pace does. Across six large marathons studied over ten years, the drop-out rate grows quadratically with temperature. The worst case on record was Chicago 2007, at 25 °C: 30.7% did not finish.

What to do
  • Adjust the target before the race, not at kilometre 25.
  • Look up the day's WBGT, not the temperature. If you cannot get it, look at the dew point.
  • Run your own numbers: the calculator does this arithmetic with whatever conditions and goal you give it, and comes preloaded with the seven races in section 12.
03 — The deciding variable

Why does humidity hit harder than heat?

Because heat has only one way out, and humidity blocks it.

When air temperature approaches your skin temperature — around 33–35 °C — you stop being able to shed heat by convection and radiation. One route is left: evaporating sweat. And evaporation needs air with room to take that vapour. If the air is already loaded, the sweat does not evaporate. It runs off. And sweat that runs off costs you water and sodium without cooling you at all.

Relative humidityTime to exhaustionSweat rate
24%68 min22 mL/min
40%60 min23 mL/min
60%54 min27 mL/min
80%46 min30 mL/min

Eight active men cycling to exhaustion at the same intensity, with air held at 30.2 °C across all four conditions. Endurance capacity fell 32% on humidity alone.

Look at the last column: they sweated 36% more and cooled less. That is exactly the problem with the coast.

Dew point is the number you should be reading

Dew point tells you how much water is actually in the air, in degrees. It is more useful than relative humidity, because 80% humidity at 20 °C and 80% at 30 °C are nowhere near the same thing.

12 °C 16 °C 20 °C 24 °C
Below 16 °CYou will not notice it
16 – 20 °CStarts to bother you
20 – 24 °CHeavy. Adjust the pace
Above 24 °CDifferent sport. Your sweat barely evaporates

And here is the figure that orders every race in this guide. Dew point at the start:

RaceDew pointReading
Cartagena, November25.0 °COver the line
Barranquilla, June, 7am25.0 °COver the line
Cozumel, Ironman start21.1 °CHard, manageable
Miami, January marathon20.6 °CHard, manageable
Rio, June marathon17.5 °CNot a heat race

Cartagena and Cozumel look alike on a thermometer. They are separated by almost 4 degrees of dew point, and that difference is the whole race.

Which adaptation helps in humidity, and which does not

Almost nobody says this, and it changes how you train.

Helps: starting to sweat sooner, lower core temperature at rest and in exercise, more plasma volume, fewer beats at the same pace.

Helps far less: sweating more. In high humidity, extra litres of sweat that never evaporate are dehydration and sodium loss with no thermal benefit.

If somebody sells you a heat preparation whose main goal is to make you sweat more, that preparation was designed for the desert, not the Caribbean.

What to do
  • Before any coastal race, check the forecast dew point, not the temperature.
  • If it is over 24 °C, go into the race with the target already adjusted and the cooling plan ready.
04 — What changes

What changes in your body when you acclimate, and how fast?

There is no such thing as "acclimation" as a single event. Each system runs on its own clock, and knowing which is which tells you what to expect from a short block.

AdaptationHow much it movesWhen it arrives
Heart rate at fixed pace−4 to −17 bpmDays 4 – 7
Core temperature in exercise−0.3 to −0.43 °CDay 5
Plasma volume+4 to +10%Days 3 – 5
Perceived exertion−1 pointFirst week
Sweat onset (earlier)−0.28 °C thresholdDays 8 – 14
Sweat rate+5% to +33%Days 8 – 14
Sodium in sweat−22 mmol/LDays 8 – 21
Performance in heat+7% to +21%Days 7 – 14

Read it this way: the cardiovascular adaptations arrive fast and the sweat ones arrive late. A one-week block gives you heart rate and temperature. Sweating better and losing less sodium needs two or three.

Two warnings that change decisions:

If you are already fast, expect less. The −17 beat headline mixes in untrained people, who have the most room to gain. In endurance athletes the real change was −4 beats. If you expect the headline, you will conclude the protocol failed when in fact it worked.

Acclimating makes you less thirsty. Thirst perception drops and voluntary fluid intake falls by around 20%. That is not a benefit: it is a risk, and it is the reason race hydration gets planned rather than improvised.

05 — The protocols

How do you acclimate if you train where it is cold?

Ordered by how practical they are for somebody with a job, not by laboratory elegance. All three work. The first one does not touch your training at all.

Option 1 — the one I would pick

Hot bath after training

You train normally, in your normal climate. The thermal stimulus is added afterwards, without touching a single session.

Dose 40 °C, up to 40 min, immersed to the neck, starting within 10 min of finishing. 6 days in a row.
  • 4.9% improvement over 5 km in the heat
  • Adaptations still present 2 weeks later
  • Works the same in trained and recreational athletes
  • You need a bathtub and a thermometer. It is uncomfortable, not dangerous
Option 2 — fewest sessions

Sauna after training

The protocol with the fewest sessions of anything that works. If your gym has a sauna, it is the cheapest in time.

Dose 28 min at ~100 °C, 3 times a week, 9 sessions across 3 weeks. Get in within 5 min of finishing.
  • Peak core temperature −0.2 °C, peak heart rate −11 bpm
  • Stretching it to 7 weeks added 0.1 °C and nothing else
  • Infrared saunas do not work for this: they do not reach the temperature the studies used
Option 3 — the most demanding

Indoor trainer in a hot room

The home version of the climate chamber. It is the protocol with the most evidence behind it, and the one that costs you most time.

Dose Room above 30 °C, no fan, extra clothing. 60 to 90 min at low intensity. 5 days in a row.
  • No fan matters more than the number on the thermostat
  • Nobody has tested 30-minute trainer sessions: the studied dose is 60–90
  • The point is for core temperature to rise and stay up, not to hit a number on the wall

The minimum dose, and the trap inside it

5 to 7 exposures of 30 minutes or more, on consecutive or near-consecutive days. That is the floor where heart rate and core temperature changes become reliably measurable. Three independent sources agree.

The trap: that floor is physiology, not performance. Two studies found the short dose was not enough to move a time trial. With 4 days nothing happened in female athletes; with 9 days performance rose 8.1% in power. Another got +9% plasma volume in 5 days and a non-significant improvement over 20 km.

Translated: 5 to 7 sessions as an absolute minimum, 9 to 14 if what you want is a better time. The "5 days is enough" that circulates is the physiological floor, not the performance one.

What is not worth doing

Overdressing on your normal session. Two weeks of training in a jacket in mild weather produced no adaptation at all, even though the athletes felt hotter. The version that did work was a protocol designed specifically to reach a target core temperature: 30 min of jogging and 60 of walking in standardised kit. That is 90 minutes that are not training. The version most people will actually do is precisely the one shown not to work.

Infrared sauna and hot yoga. Not a single acclimation study exists on either.

Training at the hottest hour of the day, if you live somewhere cold or at altitude. There is no dose available: your hottest hour never arrives.

06 — Durability

You did the block two weeks ago. Is it all gone?

No. And the reason it is not gone is the most useful part of this section.

The figure you will read everywhere is that you lose around 2.5% of the adaptation per day without heat. It is real, it comes from a meta-analysis, and there are studies contradicting it in both directions: one found no decay in rectal temperature at 26 days; another, in elite cyclists, found performance back at baseline within two weeks.

But the number that decides your calendar is a different one:

8 – 12×Faster to re-acclimate than to lose it
2 daysTo get it all back after 12 days without heat
4 daysAfter 26 days without heat

Building it the first time costs 10 to 14 days. Getting it back costs 2 to 4. A block that ends one or two weeks before you travel is not wasted: it turns you from a first-timer into a re-acclimator, and that is exactly what makes a short trip work.

Maintenance: 2 sessions a week hold or improve performance; 1 a week maintains it; zero lets it fall by around 11% over eight weeks. Never leave a gap longer than 5 or 6 days.

The mistake to avoid
  • Finishing the block 10 to 14 days out, doing nothing in the gap, and arriving 48 hours before the race. The block was not wasted — it made getting it back cheap — but the getting it back never happened.
07 — The plan

What do you do with the time you have left?

Including the honest row: do nothing. It is what most people choose, and it deserves a real answer rather than a telling-off.

Time leftWhat is achievableWhat to doWhat to expect on race day
8+ weeks Everything, including the slow adaptations: sweat rate and sodium. Block of 10 to 14 sessions, 3 to 5 weeks out. Then 2 maintenance sessions a week. Then 2 to 4 easy exposures in the final week. The full effect. The heat penalty comes down by close to half. You are still slower than in cool conditions.
4 weeks All of the cardiovascular and most of the sweat adaptations. 10 sessions across 2 weeks starting now, finishing 2 weeks out. Then maintenance. The best effort-to-benefit ratio in the whole table.
2 weeks Full cardiovascular adaptation, partial sweat adaptation. 6 to 10 post-training hot bath sessions. Add no intensity. Finish 2 to 4 days before you travel. Real but partial. You will notice heart rate and thermal comfort more than the clock.
1 week Heart rate, core temperature, plasma volume, comfort. Not sweat rate. 5 to 7 consecutive days, 30 min or more of exposure after training. Low intensity only. The markers move; the clock may not. This is the window where people conclude "it did not work" because they measured the wrong thing.
3 days Almost nothing new. But if you ever did a block, this is your re-induction window. If you have acclimated before: 2 or 3 short, easy exposures. If you never have: do not start now. Spend the 3 days on pacing, cooling and sleeping with air conditioning. Re-acclimator: you get much of it back. First-timer: effectively the "do nothing" row, but with better tactics.
You arrive the day before Nothing physiological. Everything tactical. Sleep cool, do not train hard. Ice before the gun, your own cooling plan, and a target pace calculated from the WBGT. Full penalty, moderated only by pacing and cooling. It is a valid option, as long as the acclimation happened at home.
Do nothing Nothing. — The full penalty: 3.6–6.4% if you are strong, 8–15% mid-pack. On a 4:30 marathon that is 20 to 40 minutes. On a 12-hour Ironman, more than an hour. And a higher chance of not finishing.

On arriving early: nobody has ever studied what the best day to arrive at a hot race is. No consensus statement says it, no trial tested it. The closest thing in the literature is "4 to 5 days of re-acclimation on arrival". If somebody gives you an exact number, they are making it up — including me, if I did.

What is clear: expose yourself during training, sleep and recover cool. Putting air conditioning in the room is not cheating.

08 — Race day

How do you start, and how do you cool?

This section is free, needs no preparation, and is the one that will change your result most if you arrived with three days.

Pacing

Three possible anchors, and choosing which one you use is the pacing decision:

Pace per kilometre is the worst anchor. It is the only variable that does not respond to thermal stress. Holding goal pace at 28 °C WBGT is exactly the behaviour that produces the collapses.

Heart rate is a conservative anchor. Holding a heart rate ceiling forces you to slow continuously, because cardiovascular drift pushes it upward. In a long, humid race that error runs in the safe direction. Over 5 or 10 km it will leave time on the road.

Perceived exertion sits in between, and it is the channel every cooling strategy that works operates through.

I see two options. From the half marathon upward, I would use a heart rate ceiling for the first third and perceived exertion after that. For 10 km or less, perceived exertion from the gun, with the goal pace already corrected for WBGT before you start.

Why short races are the dangerous ones

An 11.3 km race at 23 °C and 70% humidity — objectively milder than a morning in Cartagena — produces ten times the heat stroke rate of a marathon. That is 18 years of medical records and 274 cases.

The authors' own explanation: the distance is short enough that nobody backs off. It is not the distance that protects you. It is slowing down.

And a figure that runs against intuition: in a hot sprint triathlon, amateurs and World Cup athletes reached the same peak core temperature. Going slower does not keep you cooler: it leaves you exposed for longer at lower air speed.

Cooling

Before the gun. Crushed ice or a slushie, 7.5 g per kilo of body weight, finished between 30 and 60 minutes before the start. The aim is to drop core temperature 0.5 °C — 0.3 °C is already useful. If you use a cooling vest, keep it on until the last possible moment, and wear it during the warm-up rather than sitting still.

Be honest about what it gives: in the two best trials run in humid heat with self-paced racing, pre-cooling left the athlete cooler and more comfortable, but not faster. It is still worth doing, because comfort on the start line changes the pacing decisions of the first third. Do not promise time.

During the race. This pays more than pre-cooling, and it is the part almost nobody plans. The rule for high humidity:

Cold water yes, warm water barely
  • Pouring water on yourself works two ways: evaporation and cold contact. Humidity cancels the first but not the second. That is why ice and properly cold water still help, and why room-temperature water on the skin does almost nothing.
  • Neck first, then head, then face. Combining zones adds nothing over cooling the neck alone.
  • Carry your own plan. In Cozumel it is documented that the ice runs out for anyone reaching the aid stations late. A plan that depends on the aid station is not a plan.

One point of order: cooling during the race pays around five times more in an already acclimated athlete. Acclimate first, then add ice. Not the other way round.

09 — Fluids

How much do you drink?

Deliberately short: hydration and sodium have a guide of their own. Here are only the two points that are safety, not performance.

One. Never gain weight during a race. Hyponatremia — blood sodium below 135 mmol/L — appears in up to 13% of marathon finishers and up to 25% in Ironman depending on the environment. The cause, at the highest grade of evidence in the international consensus, is drinking hypotonic fluid in excess of losses. The strongest predictor in the Boston study was gaining weight during the race, with a 4.2-fold risk; finishing in more than 4 hours multiplied it by 7.4.

The risk profile is specific: light athlete, first time at the distance, slow pace, a lot of heat, and instructions to "drink as much as you can". It is a common profile in the age-group field.

Two. Salt tablets do not make over-drinking safe. The consensus says so explicitly: supplemental sodium cannot prevent hyponatremia if fluid intake is excessive. A sports drink carries between 10 and 38 mmol/L of sodium against roughly 140 in your blood: dilution wins. They can help balance and taste. They do not authorise you to drink more.

And a detail that connects to section 04: acclimating lowers your thirst. Your fluid plan has to be written beforehand, not decided by feel during the race.

10 — Warning signs

When everything stops

The signal is the head, not the thermometer

Confusion, aggression, staggering, disorientation or strange behaviour during or just after hard effort in the heat: assume heat stroke until proven otherwise.

Do not wait to be sure. Do not wait for the person to appear to improve: the guidelines warn explicitly that there can be a lucid interval, after which deterioration is rapid.

Wet skin rules nothing out. The idea that heat stroke comes with hot dry skin comes from classic heat stroke, the heatwave kind in older people. In exertional cases the athlete is sweating until minutes before they go down.

What you do

  • Cool first, transport second. That is the explicit principle in the international guidelines, and the order matters.
  • Cold water immersion, if it is available. It is the fastest method and it is well ahead of the others.
  • Call emergency services, without stopping the cooling.
  • Do not force fluids into somebody who is confused or vomiting.

With rapid cooling, survival recorded across series of hundreds of cases is 100%. With inadequate cooling, 4.4% mortality. Time is the treatment.

Before starting any heat protocol

Talk to your doctor before using a sauna or hot baths if you have or have had: unstable angina, a recent heart attack, severe aortic stenosis, decompensated heart failure, uncontrolled arrhythmias, uncontrolled hypertension, or any cardiovascular disease. The same applies if you are pregnant, if you have a fever or an active infection, or if you have ever had heat stroke.

Several drug groups affect your ability to regulate temperature — among them diuretics, beta blockers, antihistamines, antidepressants, antipsychotics and stimulants. If you take any of them, ask whoever prescribed it. Do not change anything on your own.

If you have had heat stroke at any point, your risk of a repeat is higher for the following two years. Returning to competition after heat stroke is a medical decision, not a training decision: it needs rest, normal blood work, and clearance from your doctor. Once you have that clearance, the progression is mine to build.

11 — Special case

What if you train at altitude?

For anyone who lives and trains in Bogotá, Quito or Mexico City and races at sea level on the coast.

Coming down from altitude gives you less than you think. The "+4 to 6%" that circulates comes from studies of athletes who live high and train low. You live high and train high, which is a different model: its pooled effect on sea-level performance is +0.9%, with an interval that crosses zero. The realistic numbers are between +1 and +3%. And your haematological advantage is already baked into your current times: it is not a bonus that appears on race day.

Altitude does not give you heat tolerance. This is the most important thing in the section and it runs against intuition. The one study that tested that direction directly — ten days of hypoxia followed by a heat trial — found nothing: no better core temperature, no better sweating, no better performance. And there is a mechanical reason: hypoxia contracts plasma volume and heat expands it. You arrive on the coast with the opposite profile to an acclimated athlete.

Do the two cancel out? No. The altitude bonus is bounded and small; the heat penalty has no ceiling and grows with duration.

DistanceExpected net result
5 kmNeutral to slightly positive. The only distance where "at sea level I should be faster" holds up.
10 kmRoughly neutral. The start time decides it more than the athlete does.
Half marathonNegative, unless the start is before dawn.
MarathonClearly negative.
Long-course triathlonVery negative, and by the widest margin of all.

Nobody has ever studied the exact combination of dropping altitude and gaining heat and humidity at the same time. These rows are reasoning over two separately measured effects, not a published result.

What to do
  • Do the heat work at home, before you travel. You will not arrive with enough time to acclimate there.
  • Put it on easy days or in a separate block. Every study that combined heat and hypoxia in the same session returned a null or worse result.
  • Calculate the pace from the WBGT, not from "I am at sea level". The ones who get it most wrong are the strongest, because they carry the highest expectation.
12 — Reference

Typical conditions for each race

Conditions observed on race morning, at the hour each row states, at airport weather stations. Ordered from most to least demanding.

RaceHour of the readingAirDew pointEstimated WBGT
Cartagena — 70.3, November07:0026.1 °C25.0 °C31.2 °C
Barranquilla — 70.3, June07:0026.1 °C25.0 °C31.2 °C
San Andrés — November07:0025.0 °C23.9 °C29.8 °C
Cozumel — Ironman, November07:0025.0 °C21.1 °C28.0 °C
Miami — marathon, January06:0022.8 °C20.6 °C26.4 °C
Rio de Janeiro — marathon, June07:0017.8 °C17.5 °C21.9 °C
Mexico City — marathon, August06:0014.6 °C14.2 °C18.6 °C

The WBGT values are estimated, not measured. None of these races publishes a WBGT. The formula used carries no solar radiation and no wind term, so it approximates a shaded value: the real WBGT in the sun is higher than the table's. Treat them as a floor, not a central estimate, and confirm on race day. The hour is the hour of the reading, not of your start: several of these races go off in waves — Rio between 5:25 and 5:35, Mexico City in five groups from 5:45 — so if you start before the row's hour, your conditions are cooler than the table's.

Three practical readings.

Cartagena and Barranquilla are the worst combination on the list, and not by a little. By seven in the morning they are already in red-flag territory, before the gun. Barranquilla is also inland and heats up faster: by midday the air reaches 32.2 °C. In a 70.3, the decisive part is the run, not the bike.

Cozumel is the most variable of them all. The dew point at the start ranged between 17.9 and 23.9 °C across four years, depending on whether a cold front came through. A Cozumel race plan cannot assume last year's conditions.

Mexico City is not a heat race, but it is not a dry one either. The morning of 31 August 2025 was run at 14.6 °C and 97% humidity. That is not a heat problem — it is a problem of sweating cooling you very little at that humidity, and at 2,250 metres you are also losing water through breathing that you are not counting. The limiter is the altitude, not the thermometer, and almost everything in this guide is the wrong tool for that race.