Okay, I’ll bite the lottery bullet. If I happen to win the lottery, then this does not make it more likely that something unusual happened (e.g. the lottery being rigged). The hypothesis that I naturally won the lottery is unlikely, true; but the hypothesis that the lottery was rigged so that I would win is equally unlikely for the exact same reasons: why was it rigged for me to win, and not someone else?
Similarly, in the haunted house situation, you should in fact consider it unlikely that these events would happen to you. However, this penalty should apply equally to just about any explanation: it’s equally unlikely that psychopaths would choose you, specifically, to play tricks on, as it is that you, specifically, would happen to find yourself in a haunted house.
The only way to break symmetry is if you have reason to believe that some kind of rare event is more likely to happen to you. Is there something about you that makes you more likely to find yourself in a haunted house? (e.g. are you a ditzy blonde girl between the ages of 16 and 19.5?) Then you should update on that. Similarly, if you’re related to someone running a lottery, and you find yourself winning it, you should suspect that foul play was involved.
Note that all these explanations carry over: if the lottery winner is related to the lottery organizer, you should suspect foul play whether or not the lottery winner is you.
You might argue that supposing you live in a simulation breaks this: specifically, if we consider simulations of the “everyone you don’t know is fake” flavor. In that case, the hypothesis “the simulation is rigged so one of the people being simulated in detail wins” means you have maybe a 1:70 chance of winning, compared to a much smaller chance if the world is big.
But here the improbability is just redirected to a higher level: what are the odds that someone would simulate you, specifically? Once we factor this in, the simulation hypothesis is back to “odds that an arbitrary person finds themselves in a lottery-based simulation”. Calculating these odds is beyond the scope of this comment.
Okay, I’ll bite the lottery bullet. If I happen to win the lottery, then this does not make it more likely that something unusual happened (e.g. the lottery being rigged). The hypothesis that I naturally won the lottery is unlikely, true; but the hypothesis that the lottery was rigged so that I would win is equally unlikely for the exact same reasons: why was it rigged for me to win, and not someone else?
Similarly, in the haunted house situation, you should in fact consider it unlikely that these events would happen to you. However, this penalty should apply equally to just about any explanation: it’s equally unlikely that psychopaths would choose you, specifically, to play tricks on, as it is that you, specifically, would happen to find yourself in a haunted house.
The only way to break symmetry is if you have reason to believe that some kind of rare event is more likely to happen to you. Is there something about you that makes you more likely to find yourself in a haunted house? (e.g. are you a ditzy blonde girl between the ages of 16 and 19.5?) Then you should update on that. Similarly, if you’re related to someone running a lottery, and you find yourself winning it, you should suspect that foul play was involved.
Note that all these explanations carry over: if the lottery winner is related to the lottery organizer, you should suspect foul play whether or not the lottery winner is you.
You might argue that supposing you live in a simulation breaks this: specifically, if we consider simulations of the “everyone you don’t know is fake” flavor. In that case, the hypothesis “the simulation is rigged so one of the people being simulated in detail wins” means you have maybe a 1:70 chance of winning, compared to a much smaller chance if the world is big.
But here the improbability is just redirected to a higher level: what are the odds that someone would simulate you, specifically? Once we factor this in, the simulation hypothesis is back to “odds that an arbitrary person finds themselves in a lottery-based simulation”. Calculating these odds is beyond the scope of this comment.