6 ms·
> often see people that are above average old, smoking. it's quite a sample bias, because those who you see smoking are likely quite healthy and thus is _able_
by chii 3y ago
> often see people that are above average old, smoking.
it's quite a sample bias, because those who you see smoking are likely quite healthy and thus is _able_ to smoke to old age (or won a genetic lottery to be able to withstand smoking).
What you need to count is how many people smoke, and then the % of them dying earlier than those who don't. Otherwise, sample bias will give you the wrong conclusions, esp. if this bias is affirming your own world view.
- zeteo 3y agoIsn't it the same with exercise, though? > Those who you see exercising are likely quite healthy and thus _able_ to exercise to old age (or won a genetic lottery to be able to withstand a lot of exercising). > What you need to count is how many people try to exercise, and then the % of them dying earlier than those who don't
- chii 3y agoIt's not exactly the same, because there's actually studies that tries to answer this exact question - https://youtu.be/-3dt7rpvz4g?t=294 https://youtu.be/-3dt7rpvz4g?t=294 TLDW; the curve of risk/mortality vs amount of exercise is a u-shaped curve, which means if you _don't_ exercise, you also die earlier, and if exercised _too much_ , you also die earlier.