The heat adaptation model, and why it tells you to distrust it

Every other number in this app describes something that happened. This one predicts something that might have. It is the app’s own model, it carries the theoretical tier, and its own page tells you which number to believe when it is wrong.

The model

A day with 30 minutes or more of sauna or hot bath counts in full. Shorter days count in proportion. Days under 5 minutes do not count at all.

Each counting day adds up to a quarter of what is left to gain. That is the standard shape for an adaptation curve: the first day moves you a long way, the tenth barely moves you, because there is less left to move.

After three days without heat, the estimate fades by 2.3% a day.

The output is one of three words — building, adapting, adapted — rather than a number, because a percentage would imply a precision the model does not have.

Where the numbers come from

The quarter-of-remaining rule produces a curve where roughly three quarters of the total arrives within four to six days, which matches Pandolf’s review of the heat acclimation time course: about three quarters of the heart-rate adaptation to heat arrives within four to six days of exposure.

The 2.3% daily decay is taken directly from a 2018 meta-analysis of heat acclimation decay and re-induction. That is a measured figure from the literature rather than a chosen one.

The 30-minute threshold comes from the passive heating review, which found that the studies using passive heat ran sessions of at least 30 minutes and needed about six to seven of them for adaptation.

So the shape, the decay rate and the daily dose all have citations. What has no citation is the combination — nobody has built this model and tested whether its output predicts anything.

Why “theoretical” is the honest label

The evidence tier exists for exactly this case. Three defensible parameters assembled into a model is not the same thing as a validated measure, and the difference matters to someone deciding whether to act on the output.

Two gaps in particular.

Passive heat is weaker than exercise in the heat. Almost all the acclimation timing research studies people exercising in hot conditions. Sitting in a sauna produces a smaller stimulus, and the 2025 meta-analysis of post-exercise heat exposure found a small effect with low certainty. A model calibrated on the stronger stimulus will run ahead of reality.

People adapt at different rates. The four-to-six-day figure is a population average with real spread around it. The model applies one curve to everyone.

The self-check, which is the most unusual thing about it

The page does not just show the estimate. It shows whether your measured heat strain agrees — whether your heart rate at the same heat is actually coming down — and the metric’s reading line gives an explicit instruction:

If it says adapted but your heat strain isn’t easing, trust your heart rate.

A model that tells you to override it is doing something most software will not. The reasoning is straightforward: heat strain is a measurement of you, and the adaptation estimate is a prediction about a generic person with your recent schedule. When a measurement and a prediction disagree, the measurement is not the thing that needs explaining.

The disagreement has diagnostic value too. The most common reason for “adapted” plus a flat heat strain is that the 30-minute days were shorter than they looked, or that the sauna you used got hotter over the same period, or that the humidity changed. The model has no way to see any of those.

What to do with it

As a planning tool it is reasonable. If you know a two-week break will cost you most of it, and that four to six consecutive days will rebuild most of it, you can plan around a holiday. Re-induction is also faster than the original induction, which the same meta-analysis covers.

As a status indicator it is weak. “Adapted” is a statement about your recent calendar, not about your body.

As a target it is wrong. There is no evidence that a maximally adapted state is a health goal. Heat acclimation research exists mostly to help athletes perform in hot conditions and to keep workers safe in hot environments, and neither of those is what most people are doing in a sauna.

The most useful thing on the page

The decay figure, which is the part people find genuinely surprising. Two weeks off costs most of what a month of consistent use built.

That reframes the consistency question in a way a weekly minutes total does not. For heat adaptation specifically, four sessions spread across four weeks and four sessions in one week are not the same stimulus, and the second one is what the acclimation literature is about.