Hacker News Reader

Comment by meowkit | Hacker News Reader

Anecdotal results are highly dependent on the provider.

The scientific method is supposed to allow someone with no credibility to provide signal through >process< and then gain credibility.

Anecdotes from >a credible< person are very meaningful as they are often a high indicator of signal. A social heuristic outside of rigorous research.

Your results as an anon? Basically useless. On HN? Slightly better than on reddit or facebook. Provide your methodology? Better still, and now youre approaching a research proposal.

The meta discussion around anecdata being useful or not is silly.

View on HN

If you want the protocol? Sure water temp 50-55f for 15 minutes 4-6 days a week (generally 5) for 3 months. Water chest height, arms generally above water.

How exactly does credibility matter here? Science is heavily reputation based because there’s so much at stake and so few experiments take place, but with anecdotal evidence I don’t see the credibility of individual datapoints mattering much.

I can’t tell if I minimally influenced my upper respiratory tract infection risk etc, all I can tell is the hassle wasn’t worth a benefit too small for me to notice. And that’s the issue, more than credibility if effect size was large enough to be noticeable to individuals you’d get extremely strong results from scientific studies. Conversely if it isn’t showing up as extremely effective in the literature that’s strong evidence individuals are misattributing their experiences.

how often did you get sick? weight? what season did you do it in?

what was your goal in the first place?

> you’d get extremely strong results from scientific studies

check population health data in countries where cold exposure is traditional.

science is cool, scientists are just regular ol' people and enough regular ol' people do a lot to look good, to not be unemployed and having to write applications again, to get their tap on the head and or a nice bonus. **regular in their little slices under the bell curve

It's strange to question the work of scientists in this way. Precisely because they are only human the studies are peer reviewed and the meta studies have the most explanatory value and weight.

Unless you are assuming there is some global conspiracy of scientists to have jobs the system it selfs manages it by it's transparency.

Simplifying it in a "check population health data in countries where cold exposure is traditional" way is exactly the wrong approach. How do you know what kind of diet, genome and culture they have? There are so many variables that the mere fact of whether one diverse group does something different than another doesn't really say anything. It's so difficult to try to isolate individual problems. That's why we have the scientists.

Don’t both things have value, when measured on different axes?

I think it’s just a bit more gray.

You haven't heard of the replication crisis?

Not till now. Thanks. But as weird as it can sounds. The failing and iterationg over and over is integral part of science. Which includes method itself.

The "method" was never fixed. It was identified but we have no way of correcting it. A ton of science is still completely dubious.

> check population health data in countries where cold exposure is traditional.

That tells you about those countries but very little about cold exposure specifically. One inherent difference is they lack a lot of tropical diseases because they aren’t in the tropics. You could still try and remove all those differences, but directly studying cold exposure itself is vastly more practical.

There are many things (I frankly don't think there's a more specific word for this) like diets and psychedelics where many of us (rigorously minded) are constantly seeking proofs within systems containing sufficient complexity to render conclusions nearly impossible.

> constantly seeking proofs within systems containing sufficient complexity to render conclusions nearly impossible

Language. Nothing is impossible.

Diet interventions aren't terribly hard to test, they just consistently tend to result in "as bad or worse than no intervention" (short of fixing the sort of malnutrition you rarely see in the contemporary developed world)

The rigorous result is right in front of us we just don't like the answer

Reply on news.ycombinator.com