just noodling: suppose you have a coin which, when flipped to predict some outcome, comes up 51% of the time with a correct prediction of the future. you can flip it as many times as you want. would that be useful?
When you say "I can flip it as many times as I want" do you mean "I can flip it once per question for as many questions as I want" or do you mean "I can flip the coin as many times as I want on the same question" because while both are good, the second one is *crazy* good
That's what I'm saying, if you can do 1000 coin flips on major decisions that's basically omniscience. 51% chance of accurate results with only a single flip is still good (you could easily beat the stock market or the casino with those odds) but not nearly as good as the second option
i actually meant this interpretation. like nearly half of what you get will be wrong, and yet as you put more flips in, you approach within epsilon of the right prediction.
that coin is wrong 49% of the time, yet statistically, in the aggregate, you're still better off than if you left things to pure chance. you've culled a small part of the uncertainty in your favor. it's the casino's edge, but you're the house. casinos built an industry worth billions on that edge.
so that I understand the mechanics: you say/think "heads if X happens", flip, and if it comes up heads then X you can assume is more likely than not to happen? definitely sounds like it'd be useful for winning at gambling/stocks/etc, so long as you keep enough to float your losses.