Author here. Methodology upfront because I'd ask the same things:

Data: daily records from wearable users who logged sauna sessions via connected apps. Within-person design — each user is their own control, comparing their own sauna-day nights against their own non-sauna-day nights. No cross-user comparisons.

Stats: paired t-tests, FDR-corrected p < 0.05, Cohen's d > 0.2 threshold for "meaningful effect." Anything below d=0.2 we don't report as a finding.

What we measured: minimum nighttime HR, max and average HR, HRV, activity minutes and distance, menstrual cycle phase (for female subset).

What we found: - On sauna days, minimum nighttime HR drops ~3 bpm (~5%) vs. the same user's non-sauna days. - Effect survives controlling for activity level. It's not "sauna users just exercised more that day." - Strongest hypothesis: elevated parasympathetic tone from the post-sauna cooling phase carries into sleep. Consistent with heat-stress physiology literature. - Sex difference: for women, the nighttime HR effect only crosses the d > 0.2 threshold during the luteal phase. No meaningful effect during the follicular phase. We didn't expect this; worth replicating.

What we can't control for: - Sauna type (dry / infrared / steam), duration, temperature. Not captured. - Dose-response. We don't know session length per user. - Timing of sauna relative to sleep. - Reverse causation: people may sauna on days they already feel recovered. - Selection: wearable users who bother logging sauna are a health-conscious cohort.

What surprised us: the effect is larger than what we see for comparable-intensity exercise days. If you treat nighttime HR as a parasympathetic recovery signal, sauna beats a moderate workout on the same user. Not what I'd have predicted.

I can tell you wrote the article with ChatGPT. I’m out as soon as I pick up the smell. I don’t dislike the usage of AI, I just don’t trust. It if you haven’t written it yourself.

> Motivated to understand the immediate physiological response to saunas, we looked at the same-day effects across ~59,000 daily records from 256 users.

Editorialized title is wrong. n=256

n= traditionally refers to the number of participants, not the number of data points.

The headline claim is very misleading for anyone who thought there were 59,000 people in this data set.

The absolute difference is also small. Small enough that the effect might be attributable to something secondary, such as sauna users consuming more water in recovery and being more hydrated. Heart rate has a relationship with hydration status.

N=256

Website doesn’t load, it times out. Anyone have tl;dr?

N=1, but I started rowing (indoor, on an erg) an hour a day -- not hard, generally 120-140 bpm -- every day starting February 28, after rowing inconsistently for a year or more before that. My resting (not sleep) pulse has dropped by 10% over the past ~7 weeks, from 60 to 54.

Anecdotal, of course, but the biggest change I ever made in my life was right before bed: take a screaming hot shower with dim lighting. I'd say 95% of the time, I get in bed and just pass out and have no real memory of time passing before falling asleep.

Not to be glib, but being dead lowers your night time heart rate more then exercise as well.

Is having a lower night time heart rate the core goal of exercise? Is it even a goal at all? Or is it just an indicator of other goals being reached? I'm genuinely curious, I wasn't aware that the number mattered, more than what that number actually represents.

Well ...

Finland life expectancy for 2023 was 81.69.

Norway life expectancy for 2025 was 83.23.

Japan life expectancy for 2025 was 85.27.

Sumo wrestlers in Japan have a life expectancy between 60-65 years or so - significantly lower than the other japanese.

I am not saying that sauna has no positive effect at all, but I would reason that the number one risk factor is ... weight. And I'd also still say that exercise is correlated here, if only secondary, e. g. you may be able to maintain better bodily functions if you exercise, if you can avoid injury. I do not think that going into the sauna rather than e. g. light running for 5 to 10 minutes or so, is anywhere near on the same level.

Seems to me what we now know about neural networks, we should maybe weighted sum of inputs, that fire off the desired output. The human body/brain process all kinds of stimulus at once, and might only react to a combination of inputs.

I know that for myself exercise increases my resting heart rate in the short term. It only decreases after a day or two, sometimes more depending on how fatigued I am. I thought that was common, with recovery times obviously decreasing the fitter ones gets.

Can anyone suggest why after covid I can't do Finnish sauna anymore? Prior to that I used to do 1-2x a week a sequence of 5x(10 minutes in sauna + 5 minutes cold water immersion + 10 minutes rest) which was absolutely great for both stress reduction and blood flow. Now if I do 5 minutes in sauna I feel like my skin was burning and I am about to die, and I need to recover for 1 hour from that to be able to just walk away from sauna.

Why is this quackery front page?

A delta of 3 bpm on sauna days corresponds to around 4% delta if the baseline is 72 bpm. I've gone from a resting heart rate over a 7-day average of 64 bpm to 58 bpm by jumping 15 min. of rope a day, 4 times a week. I've lost weight, body fat, and I feel like my body is more efficient with corresponding lower heart rates throughout my active day. I like saunas for recovery and aches, they put me in a relaxed state after, and I believe the dilation is flushing my system. Like anything else, moderation. Perhaps I will add sauna to my weekly routine 1x per week or less.

This would not pass peer review for a journal as written.

Maybe the conclusion is correct, or maybe not, but as written the methodology is under specified, statistics are not supported, and there too many confounders not addressed. One should not take anything from this without a better write up. Just misunderstanding what n= means is a huge flag.

Since the author is here, I have to ask: Why a blog post and not an actual paper? Why spray this onto the internet without validating the work? Or, conversely, why not caveat the work as exploratory data science?

I didn’t see a reference to the amount of time in the sauna required to receive this effect. Was that measured as part of this research?