US 2024 Election

More of a case "there's always one." 😀

Not sure why I should comment...but since you asked so nicely...

Is it legal? No idea, some say yes, others no - I'm sure the courts will decide and decide quickly. Is it unprecedented? In this specific way yes but plenty of other people and organisations have funded plenty of schemes to get people to register to vote and even helping them to physically cast a postal or in-person ballot - where do you draw the line?

Listened to that whole podcast today, not sure what your point is...? Should I critique every error they made with citations?


In regards to the part I assume you're referencing, 538/Nate Silver and his team produce multiple forecasts and explain in detail how they calculate their predictions based on the quantity and quality of polls involved. Alex Andreou is correct that several low-quality pollsters dump lots of low-quality polls. But the 538 model weights them massively down based on both historical accuracy and the large volume of results in a short space of time meaning the more accurate polls still play the largest part in their forecasts. 


Nationally to win, a Democratic candidate needs at least a 2% lead because of the way the states/Electoral College works. Early Oct high-quality polls have Harris up by 2-3% (ABC News/Ipsos +2%, Marquette University Law School+3%, YouGov/CBS News 3%, Fairleigh Dickinson University 3%). Most of these polls have been giving very similar results since mid-August and the deep dive (gender, race, age) also show no major moves with any block of voters. When the polls give you the same numbers week after week after week that normally means they have got their methodology right and in reality little is changing. Their latest forecast (the poll of polls if you like) has Harris up by 1.8%...

They do the same thing at state level as well, adjusting all polls (not just the overtly partisan ones) so that the data can be used but allowances are made for known issues. Hence why they get closer than most predictions most of the time...

The point was and as you point out…. The polls are not always to be trusted.
It’s going to be mightily close and id bet now we don’t get to know the result for at least a month after the election. Trump is already planning on litigating all results that he loses .

...did you just congratulate yourself...?!?

The way I look at polls (and this is just my view not any particular polling company or polling analyst now) is that if you watch them closely for a couple of months you get a good feel for where the various candidates are and then over time you watch for patterns (mostly general trends up or down). 


I've been watching as much as I can since June/July (when it was still Biden/Trump). Back then the Electoral College was pretty even and the swing states broke evenly. Harris got a bump up when she was confirmed as the nominee, Trump got a boost after the Republican Convention/assassination attempt and Harris got a boost after the Democratic convention. 


Aside from those events, Trump is gaining ground and Harris is losing it in nearly every poll over recent weeks - national or state, reliable or not. That's why unless Harris does something significant to shift momentum in her direction (or Trump does something with the same effect) Trump will win and the margin will probably be quite large.

Nah

You lost one bet to me and bottled on the other . 😜