Polls-plus for NC is at 49.7%. C’mon, NC, we can do it!
As for me, Election Day is a teacher work day, which is fine. I’m trying to figure out if I should take the day after as a personal day, or should I try to go in to teach kids on four hours of sleep?
Also, Iowa is moving from Trump to swing again, and Pennsylvania is pretty solid for Clinton.
Here’s my current 270towin projection map (80% chance or better for Hillary turns it blue, 60% or better for Trump turns it red, everything else is swing). NH and CO are both within 5 percentage points of moving to blue and giving Hillary the 270 she needs.
That being said, we’re still more than a month out from the final horn, and if it gets too lopsided, it could cause complacency on Hillary’s side.
ETA-- Also interesting: AZ is a horse fart away from moving into the swing list. It’s currently at 62.6% for Trump, and since Oct. 1, it’s moved almost 7 percentage points toward Clinton.
I’m considering taking the day after off so I won’t have to listen…I live in Alabama. Either way it goes, there will probably be talk I’d rather not hear. And I really don’t like to discuss politics or religion at work (though I’ve slipped on that a time or two).
Also we look good for kicking out our spineless weasel of a governor, and our chances of getting rid of the Burr in our butts are looking better and better as well.
NC now 58.7 in polls-only, 51.1 in polls-plus, and 64.5 in Nowcast.
Iowa is the next potential one to go. 45.9 in polls-only but movement there is based only on national data and whatever correlational magic 538 works with. Very little polling.
Due to one of those Google state poll bombs (which always include some weird results – in this case, HRC being ahead in Kansas and Indiana and tied in Kentucky).
The Google thing has shown her ahead in Kansas every time. Clearly there are some issues with their approach. It also shows her as way ahead in Alaska. And Trump as way ahead in North Carolina. So I’d take it with a pound of salt.
On calling North Carolina the night of the election - I seem to recall that NC has been a bit of a late call for the networks the last couple of elections, not just because it was a close race both times (natch), but also that a number of the polling places in cities seemed to be quite pokey in getting their counts in.
All right! Clinton cracked 70% on Polls-plus again! Granted, she was just below 80% when Trump was at his lowest point after the Khan fiasco. Does the Polls-plus model become more accurate as we approach Election Day?
If I understand correctly, it converges with the other two as we approach Election Day. “Now-cast” is a weighted average of polls. “Polls” adds a model accounting for the fact that the polls are now and not on Election Day. “Polls Plus” adds a model accounting for economic conditions and other non-poll models.
I don’t think there’s an easy way to evaluate accuracy.
While we’re all enjoying the Nate roller-coaster as it goes up again, I’ll add that there is now visible whitespace between Sam Wang’s shaded 95% EV confidence interval and a Trump win.