dbaupp comments on 2011 Less Wrong Census / Survey - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (694)
Gah! A single data point tells you very little about over-/underconfidence! Please, please stop acting like getting a 60% certain thing wrong (or 20% certain thing right) is a mistake.
A die has an 18% chance of rolling a 6. And that still happens. A die has a 66% chance of rolling a number larger than 2, and that sometimes doesn't happen. There is nothing unusual about these things, and the same applies for these estimation-with-confidence exercises!
(This isn't directed at you specifically, but there have been a few instances of this in this thread, and your comment was the "final straw")
I don't mean to be pedantic, but I have seen numerous people make this mistake on Lesswrong, so I am pointing it out:
1 die
2 or more dice
Hahaha, Muphry's Law strikes again... Thanks.