So speak in your language and back up your assertion that the polling "margin of error" properly estimates the vote probability distribution.
Sure - any poll that doesn't properly sample the population (i.e. all of them) could lead not only to systematic errors in the mean, but in the margin of error.
Exactly right. Union members that vote for Trump value culture war issues over union support. In my opinion they're suckers.
Too choices: one side isn't pro-labor enough, the other is strongly anti-labor. Choose wisely.
You're asserting they're directly equal to one another. If there's just one source of polling error that doesn't translate into the voting probability distribution, it would be enough to prove you wrong. Can you think of any?
Reposting this excellent meme from four years ago (be sure to click to expand) because of all the blatant lies and dangerously batshit conspiracy theories about the Biden administration’s preparations for and response to Hurricane Helene:
And Kamala is awesome!
Abortion bounties, book bans, forcing families with trans kids to flee their homes: Republicans are building a model of govt that forces people to live and act how conservatives want, enforced by the fear that anyone can be an informant. A Snitch State (gift link) www.theatlantic.com/magazine/arc...
In Texas and elsewhere, new laws and policies have encouraged neighbors to report neighbors to the government.
You don't need to know how everyone will vote to have a probability distribution. The assumption that the polling error distribution is equal to the voting probability distribution is silly in my opinion.
How do you know the model includes coupling? It doesn't sound like it does: "Our model simulates ... by adding randomly generated polling errors to our current forecast vote margins in each state." IMO a silly model assuming probability distribution is the same as error distribution