> you're spiting your own [preferences] in favor of voters elsewhere I think that's a beautiful way to describe democracy, though I know you didn't quite inten

> you're spiting your own [preferences] in favor of voters elsewhere I think that's a beautiful way to describe democracy, though I know you didn't quite intend it that way. You're right that these proposals seem to run against everything game theory and rational voter models would suggest. But I think there's a way to couch this in terms of rational preferences, if you extend this beyond any one particular race. The setup: The fear is that in future elections, most of your state citizens might vote for the "popular" but not "electoral" winner. Flipping this would come at some other state's expense. Moreover, in every future contest with an electoral and popular vote split, the new system flips every state either from "winner" to "loser" or vice versa. You are essentially trading away all future elections where your state would back the unpopular electoral winner, and gaining all the future elections where your state would back the popular candidate with a bad map. That's not an even trade. Popular winners have more political capital. They can get more done than contentious winners. So swapping out all of your unpopular winners for popular winners in other elections nets your electorate more of its preferred policies in the long run. * * Slight tangent -- Former Mexican President Ernesto Zedillo recently explained why Mexico agreed to international dispute resolution, even though that constrained its government. Zedillo said, roughly, "We wanted to be part of a rules-based system because it works both ways. If you don't want to be subject to arbitrary actions, then you have to give assurances that you won't impose arbitrary actions on others." I know that's a really rough analogy, but this discussion reminded me of it, and I like the quote. https://www.tradetalkspodcast.com/podcast/79-a-president-explains-mexican-trade/ https://www.tradetalkspodcast.com/podcast/79-a-president-exp... Setup around ~9'50" EDIT: Some key assumptions required to make this work- a) your state is about equally or more likely to support unpopular electoral winners as it is to support popular losers, b) passing policies you support is better than avoiding policies you oppose. Those are pretty fragile assumptions, but they wouldn't be crazy to entertain for some fair chunk of the electorate. I suspect the more likely explanation is just that people think a national popular vote is more "fair" and aren't really methodically calculating self-interest in these contests, rational voter models be damned.