Has it occurred to you that if you’re really that confident you could be making money on bets
It occurred to me that I might be making money from Pringlescan on this issue, if I was willing to bet with him, and if he was willing to bet with me. But I don’t even know whether he’s a legal adult—and either way he was too obviously biased in favour of his idea. I arrived at the same conclusion as pedanterrific that it might be unethical to bet with him.
Assigning a high probability to a complicated hypothesis and assigning a high probability to the negation of a complicated hypothesis are two very different things. I’m more than 99% sure that my neighbor is not a bank robber, even though I don’t know all that much about him and would consider a 90% certainty that he was a bank robber with comparable information massively overconfident (the example isn’t supposed to be an exact equivalent, just to illustrate the general principle, and I don’t necessarily share ArisKatsaris’ specific positions).
The hypothesis he’s negating with 99% probability is pretty simple though: he will find Pringles’ story better or Eliezer will. Which is a disjunction, not a conjunction.
I already knew Pringlescan’s idea and I was sufficiently aware of Eliezer’s level of writing skill. My estimate on the probability didn’t originate primarily from the complexity of Pringlescan’s idea, but from the related fact that its complexity would make it a bad idea for Harry to have or for Eliezer to write.
...
It occurred to me that I might be making money from Pringlescan on this issue, if I was willing to bet with him, and if he was willing to bet with me. But I don’t even know whether he’s a legal adult—and either way he was too obviously biased in favour of his idea. I arrived at the same conclusion as pedanterrific that it might be unethical to bet with him.
i’m not sure if this is the prediction you are referring to, but he did make and win a bet on the last page.
That particular bet would be hard to settle, because who wins depends on the bettors’ beliefs.
Assigning a high probability to a complicated hypothesis and assigning a high probability to the negation of a complicated hypothesis are two very different things. I’m more than 99% sure that my neighbor is not a bank robber, even though I don’t know all that much about him and would consider a 90% certainty that he was a bank robber with comparable information massively overconfident (the example isn’t supposed to be an exact equivalent, just to illustrate the general principle, and I don’t necessarily share ArisKatsaris’ specific positions).
The hypothesis he’s negating with 99% probability is pretty simple though: he will find Pringles’ story better or Eliezer will. Which is a disjunction, not a conjunction.
I already knew Pringlescan’s idea and I was sufficiently aware of Eliezer’s level of writing skill. My estimate on the probability didn’t originate primarily from the complexity of Pringlescan’s idea, but from the related fact that its complexity would make it a bad idea for Harry to have or for Eliezer to write.