The terms of the bet are such:
I have staked 25k mana against David's (@dlin007) 90k mana.
David wins the bet if Trump outperforms his final (morning of election day) RCP polling averages (Trump vote share, not margin) in all three of MI, WI, and PA, and does so by an average of at least 2.5%. I win otherwise.
TLDR: therefore, what you're betting on in this market is the outcome of this bet; resolves YES if I win (Trump doesn't significantly outperform) and NO if David wins (Trump significantly outperforms).
@Joshua @SemioticRivalry and @shankypanky are arbitrating and provided escrow.
I MAY bet in this market. If the resolution is at all controversial, then the team of 3 mods listed above will determine the resolution! If they decide to resolve the bet to some PROB or cancel it, then this market will resolve accordingly (although I really don't see why or how that would happen). I would ask those 3 mods to therefore not bet in this market to avoid the appearance of impropriety!
I've set the close date well after the election in case it comes down to a recount or something stupid, but hopefully this should resolve shortly after the election (as soon as the bet resolves itself).
It does look increasingly likely that I will end up surviving and winning this bet (although we will wait for vote counts to be finalized, etc). I thank @dlin007 for being a great counter-party! It was a pretty close one, and looks like the polling error from the RCP vote share will end up being about 1.8%.
I'll note that as the results stand now, it looks like this bet would have resolved the same way if we'd bet based on margin of victory (1.6%) instead of vote share, and also if we'd used Silver Bulletin or 538 instead of RCP (with either vote share or margin, all about 2.1-2.3%) though it would've been a bit closer! There's surprisingly only about a 0.7% difference between the most and least favorable (to me) of these selections.
@benshindel yeah, it was looking like I was going to win on election night, but the later counts have doomed me. You can resolve in your favor. I was directionally right, but got undone by hubris, so I can still claim a moral victory
@benshindel I hate this market. The vote share thing is such a trap. I also overlooked the undecided/third party number in polls vs results
@PlainBG ya, I’m sorry, but the resolution criteria for this market had to match the exact terms of the bet, and those were the terms
RCP forecasts Trump vote share at 47.8 in MI, 48.2 in WI, and 48.5 in PA.
He needs to outperform all 3 and by an average of 2.5 points to win!
@TimothyJohnson5c16 Yes indeed! (I've been selling my hedged shares for the last 20 hours or so)
Even G Elliott Morris is commenting on this market:
https://x.com/gelliottmorris/status/1839658178293780813?s=46&t=62uT9IruD1-YP-SHFkVEPg
@nikki I mean currently there's a 0.2% difference between RCP and Silver Bulletin in PA, 0.6 in WI, and 0.5 in MI, and I'd expect those to narrow as we get closer to election because RCP is pure polling and Silver Bulletin is modeling predicted final vote share
@benshindel RCP is a hack poll aggregator, not "pure polling"
They make stupid assumptions like polling error being correlated across years
@nikki I’m not trying to defend them tbh, but I do think they’re quite in line with other polling aggregators this cycle
@benshindel just to clarify we'll be using the 2-person average, so:
https://www.realclearpolling.com/polls/president/general/2024/pennsylvania/trump-vs-harris
https://www.realclearpolling.com/polls/president/general/2024/michigan/trump-vs-harris
https://www.realclearpolling.com/polls/president/general/2024/wisconsin/trump-vs-harris
@dlin007 The 2 candidate average doesn't still sum to 100 or even 98/99. The actual third party should be lower than polled, and undecided voters will split. I'm curious why this bet is determined based on vote share but not Trump - Harris margin?
@PlainBG yeah, my original idea was for Trump to beat the margin, but it was misunderstood by Ben which we discovered well into the whole thing, so this was the compromise
@PlainBG @dlin007 well, in 2016 and 2020 there was about 5% and 2% third party votes, so I’m not sure the final RCP vote share will be that too far off from that!
Currently in swing states in the RCP average it appears that there’s less than 5% who aren’t either Trump or Harris.
But this phenomenon should benefit David, since it implies both candidates should slightly out-perform their final RCP average. However, the two-way margin is more swingy (as taking off 1% from Harris and giving that to Trump, for example, would lead to a 2% swing towards Trump) and this would benefit me in the bet. However it’s not entirely that simple, as turnout is also a factor, and some ppl who would’ve voted for one candidate might just not vote. All in all I think it’s close to a wash but might benefit me slightly, which is why I was willing to adjust the odds from 4:1 to 3.6:1