Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Don't be faked out by BMI; it's no different from the Dow Jones Index (something easy to calculate with pen and paper, or in your head), that gives you a low-resolution signal at best. If you have a BMI of 50 (or 16) you have problems, but you already knew that.

It's bizarre that people worship it, but then again, simplicity has always beat having to think.



Good thing the post you're replying to isn't worshipping it and in fact is based on a study using DEXA scans instead!


> It's bizarre that people worship it, but then again, simplicity has always beat having to think.

It's pretty good over populations, and is pretty good even on lots of individuals. Body fat percentage and lean mass might be better but they're much harder and more expensive to measure (particularly accurately).


Very nice analogy. What's the equivalent of the S&P 500 then? :-)


The equivalent is probably a 2D DXA scan with body composition report. Supposedly it can measure the amount of bone, muscle, and fat in your body to within 1% of actual values. Those scans are widely available for <$50.

BMI is still a pretty good proxy measurement. There's a very high correlation between BMI and body fat content. People like to complain that BMI is useless because a few weightlifters show up as "obese" but those are so rare as to be statistically insignificant.


> People like to complain that BMI is useless because a few weightlifters show up as "obese" but those are so rare as to be statistically insignificant.

A good explanation I saw is that if you care what your BMI is you probably aren't a weightlifter


IIRC the largest problem comes from BMI under-reporting high body fat.


You can be a bit healthier by just doing a whole variety of things and accepting the average, than by letting an expert choose what things you should do. ;-)




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: