If Trump picked up the right mix of 42,921 votes in Arizona (10,457), Georgia (11,779), and Wisconsin (20,682), the Electoral College would have been tied at 269 all.
Back in 2016, Clinton needed to pick up the right mix of 78,000 votes in Michigan, Pennsylvania, and Wisconsin to win the Electoral College.
After all the math is set and done, I'll calculate the least number of voters in the states that if they were to change, would give the election to the runner-up (or at least, make a tie). I plan to eventually calculate all the baserates for this for previous elections.
I won't bet on this market. I'll rely on the mainstream media and the Wikipedia to adjudicate this market. In case of recounts, this market will consider the official recount.
Other election markets:
1,000
3.00@MP I believe this can resolve. You can see Cash's reasoning on his market with a more strict 200k bar: https://manifold.markets/cash/will-the-2024-us-presidential-elect-e428fef56c68. This should be confirmed to be >100,000 voters.
https://www.nytimes.com/2023/09/11/upshot/electoral-college-trump-2024.html
Trump's electoral college advantage is fading
@MaybeNotDepends The title is misleading. 100k votes don't decide the election. You could say that Biden or Trump or one of tens of thousands of people who made key decisions determined the winner of the election, but that is overly deterministic.
Also if the vote margin is 100k votes, 100k people didn't determine the election. At a minimum 1000 sets of 100k voters determined the election.
@mistersplice I have been thinking about what is important about this distinction and I think I got it. In elections with a binary choice, switching a voter from Candidate A to Candidate B actually results in a net swing of TWO VOTES FOR EACH VOTER.
That means Trump actually lost to Biden by half of the absurdly tiny number you initially gave that I said was highly misleading!!
I was wrong it was very very close and I appreciate both @MP and @mistersplice in particular for giving me those three words to think about.
@MP No it’s all good. I lost the forest looking at the trees. The title and the market are great!
@BTE I was only taking a jab at the fact that some people claim there were more votes than voters in the last election
@mistersplice Haha! Well those people are illiterate because all they need to do is read the audits of each state and do the math. But I still like your comment, because it halved the difference from 42,921 to 21,460 in 2020 and from 78,000 to 39,000 in 2016 and that is really close validating the idea that it was a coin flip election.
Like I said yesterday, you are missing the big picture by saying Trump just lost by the numbers you gave. That basically assumes it was a toss up or neither was a favorite in those states you listed but the truth is Republicans are HEAVY favorites in those states. Georgia in particular is absolutely not supposed to be close, he should have won by 5 pts like in 2016. Georgia also should have elected 2 Senators in 2020 easily but Trump did his best to make that impossible. My point is elections can’t be analyzed as stand alone events, they are data points in a time series of events. The events of 2020 make it very clear that Donald Trump GOT CRUSHED by every metric and did not in fact lose by the 100,000 votes or whatever you reference. It was closer to 1.5 million votes Biden won that 9/10 elections vote GOP that were the huge difference.
Now if you think it will be close next time with Trump on the ticket, the burden is on you to explain why more people will turn out for Trump than did in 2020. How is he winning over new voters?? Serious question, because as of today based on polls from this week needs to makeup a 5 million vote deficit among general election voters to not be McGoverned into ultimate loser status.
@BTE This seems like an argument over phrasology to me - I agree that as a general statement 'Trump only lost by 43k votes' would be misleading, but in the context of the fairly clear resolution criteria the implied statement 'the lowest possible number of people whose votes could have changed the outcome of the election if flipped is 43k' is reasonable. It's not a direct comment on the 'real terms closeness' of the coming or previous election.
@AngolaMaldives I created this market because I am arguing BTE that the election is a coin-toss, in parts because in a 300M people country, so few of them will ultimately decide the election. BTE is arguing that the ex-ante probabilities of the election can be very high, even if ultimately the election is decided by not that many people. BTE is also arguing that 2020 election ex-ante probabilities were also way bigger than one would conclude given how close it was in the tipping off states
@MP I plan to create soon (when I get to the PC) a market on how the odds will be prior to the election, if Manifold is going to price a coin-toss, just like these markets.
@MP Ah, I see I was missing context. I won't claim the expertise to weigh in seriously, I just bought a bit of no since 80% seemed high given the last 2 elections failed to meet the criterion.