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

Imagine someone asks you: "If I flip a fair coin 99 times and it comes up heads each time, what are the odds it comes up tails on the 100th toss?"

Do you say 50%, or do you dismiss their prior? At what point do you dismiss their prior or them, entirely? You don't have to dispute the mathematics, you just have to get bored with the impracticality of the question: it will never happen. Maybe they should flip that allegedly fair coin until it comes up heads (or tails) 99 times in a row and get back to us.

It's reasonable to ask what confounding factors are at play. It's reasonable to have a null hypoothesis.



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

Search: