Nc Republicans going hard for low turnout with a purge of inactive voters (such an American thing it's absurd there should just be a national database that if someone enrols to vote in another state their previous state enrolment is cancelled and otherwise enrolment to vote is just permanent)

Robinson so toxic to their brand they may lose anyway in NC though with people turning out to vote against him for governor

I think I'm going to get a bit of the -64.5 line sometime in the next few weeks when a good price is up, saw some +400 yesterday but didn't take it yet, do like it though that allows Harris to lose either NC or GA or Az+NV and still cover. Don't think the latter if likely if she wins NC and GA though she will be winning Nevada. Same applies to Michigan.

I think the chances of Harris winning a super close election by one tipping point state are lower than the market seems to think, it's super likely to be a multiple state victory if she gets over the line with some outright losses too if her polling gets worse

PA is probably the state that gets either candidate to 270 but it's pretty unlikely it's decided by exactly the tipping point state

never thought i'd agree with swoop but here we are. there's no way you can treat each state as independent events. they're absolutely correlated and not in a small way. not to say they'll all go one way or another... 5-2 or 4-3 seems most likely... but you're a fish if you're calculating each state's odds as independent events.

that said, swoop's argument that half of elections last half century have been 100+ electoral margins is unconvincing because it ignores polling distributions for each of those elections relative to today's polls, plus the inherent outperformance of trump against polls in his last two campaigns. polling/sampling bias is a huge issue today not just for politics but all jobs/sentiment/economic indicators we've relied on in the past.

To be fair the pollsters claim to have changed their methodology after previous misses and Trump isn't the unknown quantity he was in 2016 or the incumbent he was in 2020, he's been a different candidate each time - since the last election he's had Jan 6, the various indictments and conviction and is now 78 etc against a younger candidate who again has factors neither previous opponent had (significantly younger than Trump, is black, etc). Assuming he'll outperform polling because of 2016 or 2020 just because he did last time is flawed thinking. He absolutely could, but it is absolutely not guaranteed either, polling can miss in either direction and a significant polling miss in either direction could easily lead to either candidate sweeping the swing states 7-0 or coming close at 6-1 etc

I'd go as far as to say there's probably a 50%+ chance one of the two candidates wins o5.5 of the swing states regardless of who wins. I'd be surprised if the winning candidate doesn't sweep PA, WI, MI for example.

Granted we don't know what the polling will look like on election day, but the market is for sure overestimating the probability the swing states go 4-3 either way etc

interesting... how so?

https://www.pewresearch.org/short-re...lling-in-2024/

General article about some of it, obviously impossible to know how accurate the polls will be this cycle and past performance is not always an indication of future performance although obviously pollsters (the non partisan ones anyway) want accurate results or their product is worthless. Partisan polls for either party are obviously serving a different purpose (to push whatever narrative the party wants) so should be discounted

2022 polling was pretty good, previous few cycles were a miss, polling had Romney closer than he was, Clinton and Biden both up more than they ended up being in national popular vote but again pollsters adjust to try and get their samples to better match the electorate as a whole over time if they realise they're over or under sampling certain likely voters etc in hindsight so who knows if they'll be accurate or off this cycle and if they're off, by how much as there's a huge difference between a 1% polling error and a 4-5% polling error

Hypothetically if we go into election day with Harris up 3-4% or whatever, she's definitely the favourite, but it could be anything from a landslide (Harris wins national vote by 7% and sweeps the swing states or similar) or if there's a 3% gap in the other direction and Harris wins the popular vote by 1% and the electoral college is close or Trump edges it

If we go into election day with polls tied nationally, Trump's probably a fav due to electoral college math and the last two cycles polling errors but again past performance does not guarantee future performance in that the last time there was a black candidate for President in 2012, polling actually under-sampled Obama's support and over-sampled Romney's.

A lot will depend on turnout too. If turnout is low, conservatives tend to do well, if turnout is high, progressives tend to do well. The right needs to try and disillusion the left over Palestine etc and try and paint Harris as 'both parties are the same/bad' to try and get people that will never vote for a Republican to stay home. If youth turnout (or single female or black turnout) is super high, Republicans are ****ed. If they're super low, Democrats are ****ed.

I think it's going to come down to turnout again. Trump got 74m to Bidens 81m votes last time. USA has about 6m more citizens than last time, let's say 4m are eligible voters and 3m of them actually vote - i'd expect Trump gets about 74-75m votes again (i'm expecting him to be slightly less popular as a non incumbent and not running against a super old dude). If Harris matches Biden's number turnout wise, Trump won't win. Now turnout could be higher or lower with Trump out of office but i'd say if Trump gets to 80 million votes he's a favourite, by 70 million he loses in a massive landslide and loses all of the swing states. In 2016 it was 66m for Clinton 63m for Trump approximately, but Trump ran well expectation with where the votes landed even with Republicans EC advantage. If he loses the popular vote by 3 million or so he's probably a tiny dog, if he loses the popular vote by closer to the 7 million he lost to Biden by he isn't going to win but it could be anything from a 1-3 state victory to Harris sweeps the swing states depending on the distribution. By 8-10 million votes i'd expect Harris covers the -99.5 electoral votes line and wins 7/7 swing states. If Trump wins the popular vote outright I don't see a scenario where Harris wins the electoral college she would have to run way, way above expectation in the exact 270 electoral votes she needs (probably the midwest plus Nevada or ME district route) and massively underperform in the sun belt or similar

Anyway, none of this is particularly helpful but there are a ton of similar articles to the one I linked at the start of the post that discuss how pollsters are trying to change their methodology to get accurate samples. Polling was quite good in 2022 overall in that people expected the polls to underestimate Republican support and form a red wave and it never eventuated.

It's more likely that polling under-samples Trump voters than over-samples them but it's not impossible that the reverse is true or that the polls are accurate within a percentage point both of those are very possible as well.

If black or female turnout vs white male turnout is significantly higher or lower than expected or if young people show up in force to vote or don't show up at all it'll obviously have a huge impact