I think for most of us, the anxiety isn’t that Trump will formally win this election, but that it’ll be close enough to be grist to the mill to an aggressive litigation against democratic results.
Yeah, if Biden were racking these numbers up against a Mitt Romney, there’d be much less to worry about, and I wouldn’t have preemptively taken Wednesday off work.
Trump has already said that he plans to do what he has done all of his life when things don’t go his way, litigate the crap out of it. There will be a fight in every swing state/close state to invalidate ballots and stop counting at a time that is opportune for Trump. Luckily so far it looks like even the most conservative of Judges are ruling against GOP attempts to disenfranchise. It think that it would take a razor thin difference like Bush/Gore for the Courts to change things.
Brett Kavanaugh is on the record that the nation has a valid interest in the results being known the night of Nov. 3. Which is complete bullshit (said Kagan nicely in rebuttal) but get ready for a lot of that in the years to come.
Yeah, not one state certifies their results on election night.
What I noticed from Trump is that he does that, but with others footing the bill. If he can’t do that, he will take the ball and go home.
He can use his campaign funds — he won’t be out a cent.
Not sure about that either.
So Biden is asking donors to help pay for defense against Trump’s BS
WHAT campaign funds??
(unless of coure the reason that his campaign has already run out of (gobs of) money was that Orange Julius Caesar purloined it to run lawsuits in case he lost the election.
Talk about cutting off your nose to spite your face…
We can never know what the odds “really” are for any event of any kind (aside from certain microscopic physical events). But we can apply modern analysis to historical data and come up with a number similar to the current predictions. In fact, I’d be curious if 538’s 2020 model, applied to 2016, would give a somewhat different answer.
For most elections in American history, there is no historical data to go by. There wasn’t proper polling in 1836 or 1916; there just isn’t enough to put 538’s model onto.
He’s pissed away an ungodly amount of money, but as of mid October he still had over $43 million cash on hand.
538 doesn’t just use polling data, although that’s a significant part. Other data is available, such as whether it’s an incumbent election or not. The predictions are certainly going to get worse as you go farther back, but that’s just likely to push the predictions to the center. In any case, Mike_Mabes was curious about the most recent election that went against a biased prediction. Maybe the answer to that is farther back than we have good data for, but maybe not. That is very different from “there’s no way to know.”
Not at the very end (now). From their prediction page:
We’re now a day away from Election Day, and that means two big things for our forecast. First, the forecast is now totally polls-based; that is, any advantage our forecast gave Trump for the economy or for being an incumbent is no longer factored in. Second, the uncertainty around how much the polls will change between today and Election Day is also no longer an issue.
Interesting. Actually the Twitter thread it links to had all kinds of interesting tidbits. Nevertheless–their model clearly supports additional data in a variable way, accounting for drift and other effects. So it has the capability to account for polls with lower reliability, using the auxiliary data to fill in the gaps to some extent. So maybe one couldn’t build an exact equivalent to a “day before the election” prediction with historical data, but one could standardize on a weaker prediction and compare that number vs. past elections.
It looks like we are done with changes to the model. Nothing else is due that could move the needle much.
Darn, I never got the page to read “clearly favored.” 
Me neither. I’ll find solace if Biden wins. 
No we won’t.
The idea that white supremacists were going to shoot up polling stations like a number of people predicted here on this very forum never materialized.