Why are people putting these two so high? I don't see any specific reason for them to be correlated other than "people think both will be won by a Republican", but I think the chances of both being won by a Republican are a lot lower than the >80% probability people are putting this at.

@PlasmaBallin take my limit then :)

@SemioticRivalry I took part of it. Didn't want to take the whole thing since it's already at a lower probability than I was betting against before. I think there is a pretty decent change of D wins ME-2, R wins NC, given how strongly ME-2 breaks for Dems in House elections

@PlasmaBallin I actually think Biden has a better shot at NC. ME-2 is trending poorly for Dems

@SemioticRivalry I think either is a possibility. I'm not actually sure about ME-2 going badly for Dems, though. It has in the last two presidential elections, and I guess maybe that means it will in 2024 as well since it will probably be the same candidates. But they won the House race by a safe margin in 2022, despite the Republican-leaning environment.

@PlasmaBallin Golden is special

From 538's old 2020 model, the joint probabilities were

ME-2 D and NC D : 44.16%

ME-2 D and NC R : 12.96%

ME-2 R and NC D : 19.84%

ME-2 R and NC R : 23.04%

But back then ME-2 had a 57% chance to go D and NC had a 64%, whereas now, the main market gives 28% and 37%, which is in the other direction and more extreme.

This one shouldn't be higher than the other two. MI, WI, and PA usually vote together, but out of the three states, Michigan is the bluest and Wisconsin is the reddest. So if any one of them votes differently than the other two, it will probably be Wisconsin or Michigan.

I don't think we want a market on whether HI and NY will vote the same way here.

Actually, I think it could be valid to have ONE pair of "pretty far apart" in the market, to represent the probability of "there is a dramatic and sudden shift in the political landscape between now and the election". Like Washington+Idaho maybe. Or Kentucky+Illinois.

@Eliza I think that is far better handled by: /EvanDaniel/how-many-of-the-top-7-democratvotin or similar structures. Combine them all in one pot, not pairwise.

That market could probably be improved by adding another option or two. And I think the unlinked format (as per /EvanDaniel/how-many-of-the-original-13-colonie or similar) is probably a better structure. It could probably stand to be expanded to more like 10 states on the list, and it probably needs an R counterpart.

@EvanDaniel I was trying to think of some split pairs that are maybe not in the very top 5 most R or top 5 most D. And have some kind of geographical angle as neighbors. I don't know exactly what kind of regional issue could unite states like that within 1 year, though. If it's going to instantly go to 0.5% then it's not worth adding but if some far split pair has a chance of 5% to vote the same it could be nice to add. Someone might be able to think up something.

@Eliza Sure, that makes sense. I wouldn't mind one or two such on this list, but lots would be boring clutter.

@Eliza I think my major questions would be something like:

• Is this the most interesting such pair(s)? We have plenty of months to trade this, no need to rush in. Let's choose at least a little bit carefully.

• Does having it as a pair add value over the individual state result market? IMO that's the big one.

GA and NC would be an interesting pair. They're relatively similar states but the current markets gives them different leans. Some other ones like that are (OH and PA) and (OH and MI).

@Gabrielle On second thought, I don't think there's any need to have this be limited to creator only. Please feel free to add other pairs that look interesting! Thanks for the suggestions!