Here is some general advice about how to translate between “your life” and statistics.
Imagine that some new product gets released that increases your chance of chancer by 10x, so something extremely bad and very noticeable in studies. Now let’s say 2/10000 typically get this type of cancer normally and that it’s an extremely popular product so 50% of the world population use it. Start consuming it straight away.
That leads us to having 2/10000 cases before the product and 7/10000 afterwards, and you’ll have 4994/5000 of people who use the OBVIUSLY bad product stating confidently that there can’t be any issue, because they aren’t effected. Imagine they all have 2 kids who have two kids and suddenly the common sense attitude that “this can’t be dangerous because my grand parents…” is the overwhelming majority opinion.
You don’t have “a small dataset” you only have a bias, nothing more, nothing less. You can’t translate between data and your personal life how you’d expect to, it’s just not possible.
Also regarding your buddies girlfriend’s advice she’s completely wrong. It’s actually anything with less than 3 ingredients or more than 5 but less than 7 you shouldn’t eat. Everything else is safe. You can trust me because “I’m a huge health nut”.
Disclaimer: numbers might be slightly off, corrections are appreciated, but the point stands.
> Also regarding your buddies girlfriend’s advice she’s completely wrong. It’s actually anything with less than 3 ingredients or more than 5 but less than 7 you shouldn’t eat. Everything else is safe. You can trust me because “I’m a huge health nut”.
I think he's making fun of the arbitrary number that was originally given. Obviously healthy foods can have far more than four ingredients, such as dozens of spices (or really anything). I think many people want to be able to follow some sort of broad guideline in life and it frequently causes those to arrive at meaningless conclusions.
> I think many people want to be able to follow some sort of broad guideline in life and it frequently causes those to arrive at meaningless conclusions.
But it's just that, a broad guideline.
I think that only few people would not intuitively understand that seven types of fresh vegetables in a box is not the same as some meaty goo made from seven unpronouncable industrial ingredients with E-numbers attached.
HN types seem to like to nitpick a broader point by conjuring up edge cases.
I think the response is misunderstanding thd difference between a guideline and a rule. "Nothing from a bag or box, nothing with more than 4 ingredients" allows for bags of frozen veggies, allows for 5+ simple ingredients, but a good way to quickly rule out ultraprocessed foods. Saying "avoid ultraprocessed foods" would likely not be actionable to a great many people: look at a set of ingredients and is it ultraprocessed, processed, or just fancy names for everyday things (should I avoid sodium chlorite and dihydrogen monoxide?).
Imagine that some new product gets released that increases your chance of chancer by 10x, so something extremely bad and very noticeable in studies. Now let’s say 2/10000 typically get this type of cancer normally and that it’s an extremely popular product so 50% of the world population use it. Start consuming it straight away.
That leads us to having 2/10000 cases before the product and 7/10000 afterwards, and you’ll have 4994/5000 of people who use the OBVIUSLY bad product stating confidently that there can’t be any issue, because they aren’t effected. Imagine they all have 2 kids who have two kids and suddenly the common sense attitude that “this can’t be dangerous because my grand parents…” is the overwhelming majority opinion.
You don’t have “a small dataset” you only have a bias, nothing more, nothing less. You can’t translate between data and your personal life how you’d expect to, it’s just not possible.
Also regarding your buddies girlfriend’s advice she’s completely wrong. It’s actually anything with less than 3 ingredients or more than 5 but less than 7 you shouldn’t eat. Everything else is safe. You can trust me because “I’m a huge health nut”.
Disclaimer: numbers might be slightly off, corrections are appreciated, but the point stands.