@[email protected] @[email protected] I swear one day I’m going to sit down and do the actual math to prove that voting systems are broken by having a majority of voters factor their perception of “electoral math” into their preferences even when their perceptions are accurate. Arrow’s impossibility theorem is already pretty discouraging without all this meta stuff.
@[email protected] That’d be such a great thing to see in data. I was alluding more to the theory of voting systems, like rational choice theory. The setup in those is something like you have a set of people, and there’s a choice they need to make collectively. Each person can have a different preference about what the choice should be. Arrow’s impossibility theorem states, roughly, that in most cases no matter what system you use to take account of the people’s preferences and make the final choice, at least one person’s preferences will be violated (they won’t like the choice).
What I was imagining was, in the same setup, everybody modifies their preferences based on what they think the other people’s preferences are. So now the choice isn’t being made based on their preferences, it’s being made on the modification of their preferences. Arrow’s impossibility theorem still holds, so no matter how the final choice is made some people will still be unhappy with it. But, I think it’s possible that even more people will be unhappy than if they’d just stuck with their original preferences. Or, maybe the people who’d already have been unhappy are even more unhappy. I’d have to actually sit down and work it out though, which I haven’t.
The example of your dad talking himself out of voting for Buttigieg because he thinks other people won’t vote for Buttigieg is exactly the kind of case I was thinking of! Except I was thinking more theoretically than data-wise. It’d be great to see data on it too, for sure.