Why your Garmin VO2 max still drops at 100% heat acclimation

You did the work. Four, five, six weeks of running through a summer that has been genuinely unpleasant, and your watch rewarded you for it — heat acclimation ticked up to 100%. That is supposed to be the fix. That is the number that says your body has adapted, you can stop worrying about the conditions.

Then you finish a decent Tuesday session, sync the watch, and your VO2 max has dropped two points.

This is one of the most common complaints on the Garmin forums every summer, and it comes with a particular kind of frustration attached, because the athletes reporting it have done everything the watch asked. One runner describes sitting at 100% acclimation since early June while training at a dew point above 19°C. Another reports the same size of drop as the previous summer, when they had no acclimation at all. The adaptation happened. The number fell anyway.

The reason is that acclimation status and the VO2 max estimate are two different systems, and only one of them is looking at the thing that actually matters.

What Garmin is actually measuring

Your watch is not measuring your VO2 max. It is estimating it, and the estimate rests on a single relationship: how fast you are moving versus how hard your heart is working to move you.

Run 5:00/km at 140bpm and the algorithm reads an efficient engine. Run that same 5:00/km at 152bpm and it reads a less efficient one. That is essentially the whole model. It draws on steady-effort segments from outdoor runs on relatively flat terrain, which is why treadmill sessions and technical trails often produce no estimate at all.

As a model of fitness, this is reasonable. Improved aerobic fitness genuinely does show up as a lower heart rate at a given pace. The problem is that it is not the only thing that does.

In the heat, a meaningful share of your cardiac output is diverted to the skin for thermoregulation rather than to the working muscle. Blood plasma volume falls as you sweat. Heart rate at a given pace typically rises somewhere in the range of 5 to 15 beats per minute for the same effort. Feed that into a model built on pace versus heart rate, and it produces exactly one conclusion: less efficient, lower estimate.

Which is the same mechanism as cardiac drift within a single run, just measured across weeks instead of minutes.

Why acclimation doesn’t rescue the estimate

Here is the part that catches people out. Heat acclimation on a Garmin is largely an informational status. It tells you the watch believes your body has adapted to training in heat. It does not meaningfully re-price the VO2 max estimate to compensate for the conditions of a given run.

Acclimation begins to accumulate above roughly 22°C, builds over several days of exposure, and decays after about three days without it. Useful to know. But the VO2 max calculation underneath is still doing the same arithmetic on pace and heart rate that it always was.

There is a second wrinkle. The temperature the watch uses generally comes from forecast weather data synced through your phone, not from the sensor on your wrist — which is sensible, because a wrist sensor sitting against a warm arm is a poor thermometer, but it does mean the conditions the algorithm believes you ran in can differ from the ones you actually experienced.

So you can be fully adapted, physiologically better at running in heat than you were in April, and still watch the number fall. Your body adapted. The estimate did not.

Dew point, not temperature, is the variable

This is where the model runs out of road, and it is the part almost nobody explains.

Sweating does not cool you. Evaporating sweat cools you. Sweat that rolls off your forearm and lands on the footpath has cost you fluid and delivered no cooling whatsoever. Whether evaporation can happen depends on how much water the surrounding air is already carrying — and the honest measure of that is dew point, not relative humidity.

Relative humidity is a percentage of capacity, and capacity itself changes with temperature, which makes it close to useless for comparing days. Ninety percent humidity at 10°C is a pleasant morning. Sixty percent at 30°C is brutal. Dew point is an absolute figure: the temperature air would need to fall to before it saturates. It answers the only question your thermoregulation cares about.

Which is why a runner on the Garmin forums put the case better than any marketing copy could, asking for dew point to be factored into acclimation and VO2 max at all: an 80°F day at a 50°F dew point and an 80°F day at a 75°F dew point are, physiologically, different days. The thermometer reads the same. Your cooling system does not.

The widely used coaching method sums air temperature and dew point (in Fahrenheit) and reads the cost off that combined figure. It is the model behind the heat-adjusted pace calculator, and it looks like this.

Conditions (temp / dew point)	°F sum	Pace cost	At 5:30/km
13°C / 7°C — cool, dry	≤100	0%	5:30 (no change)
18°C / 12°C — mild	~118	~1%	5:33
21°C / 16°C — warm, humid	~131	~2%	5:37
24°C / 18°C — sticky	~140	~3%	5:40
27°C / 21°C — hot, humid	~150	~4.5%	5:45
30°C / 23°C — oppressive	~160	~6%	5:50
32°C / 26°C — extreme	~170+	8–10%+	5:56–6:03

Read the two 30°C rows against each other and the point lands. Hold the air temperature and move only the dew point, and the cost moves several percent. Acclimating to heat does not solve a problem caused by saturation. You cannot adapt your way out of the fact that the air has nowhere left to put your sweat.

What a 3% pace cost does to the estimate

Three percent sounds trivial until you put it through the algorithm.

On a 5:30/km easy run, a 3% cost is about ten seconds per kilometre. But you probably did not run ten seconds per kilometre slower — most runners hold roughly the pace they planned and absorb the difference as extra cardiovascular load. The watch then sees your normal pace paired with an elevated heart rate, which is precisely the signature it was built to interpret as declining efficiency.

Do that across a fortnight of humid mornings and the estimate has a consistent, one-directional bias applied to every eligible run. A one to two point drop through a sustained hot spell is a completely ordinary result, and it says nothing about your fitness.

The same distortion is why race predictors disagree with each other and with race day — they are all built on top of an estimate that has been quietly re-priced by the weather.

How to tell an artifact from real fitness loss

The drop is not always weather. Detraining, illness and accumulated fatigue all produce a genuine decline, and dismissing every fall as heat is its own mistake. Four checks, in order, will usually settle it inside a minute.

The third check is the one that does the most work, and it is the same logic as separating training fatigue from genuine fitness loss: one metric moving alone is usually the metric being wrong. Several metrics moving together is your body telling you something.

What to do instead

The genuinely damaging response to a summer VO2 max drop is to train harder to fix it. You add intensity to chase a number that fell for reasons intensity cannot address, in conditions that already carry a higher physiological cost than the same session in April. That is how a July block turns into an August injury.

Better: run to effort rather than pace through the hot months, treat the dew point as a known tax rather than a personal failure, and set your expectations from adjusted paces instead of your spring bests. Separating weather from fitness is a skill, and it is largely what summer training is for.

And take the long view on the adaptation itself. Heat training genuinely works — plasma volume expands, sweating starts earlier, sweat electrolyte concentration falls. Those changes cash out in autumn, when the air cools and the fitness you built in miserable conditions finally has somewhere to show up. The estimate will catch up. It is just measuring the wrong thing until then.