538 ain't perfect, who is, but they own up to it and also bring the receipts. Its the receipts I always like about em. But thats statistical work, their punditry fucking blows, or maybe its just nates punditry in particular.
One of their writers, Claire Malone, is bae tho.
Sent from my SM-G975U using Tapatalk
No one has a viable likely voter model for the current conditions:
1) huge increase in mail-in voting
2) mail-in voting becomes partisan
3) COVID might keep people from the polls (olds and city folk)
4) Republican (il)legal chicanery
5) Trump forces the issue on election night and declares himself the winner before mail-in ballots are counted
Elections these days are mostly about who votes. We have a shit tier model of who is going to vote because a lot of voting factors have changed at once.
Nate predicts a 6-7 point win is 99% odds of Biden victory, but before his tweet caused a bunch of tweetnalysis articles that I know I saw other estimations before this up near an 8% necessary margin.
Confidence in statistical models this election cycle should be a lot lower than in 2016.
rasmussen was the closest in 2016, so they're who i've been following recently. The trump "era" has made taking polls more difficult than ever before. I could be wrong about rasmussen but i'm following them this time.
Yeah you're wrong about Rasmussen lol. Theyre precise, but not accurate. Theyre consistently heavier on the R side, so their polls are still useful, but its understood that they weigh it too much to the right when doing their behind the scenes math.
Sent from my SM-G975U using Tapatalk
when evaluating predictors it's important not to place too much emphasis on a single data point, and Rasmussen has a, charitably, spotty record when broadening one's scope. their 2018 midterm polling was calamitously bad (claiming congressional Repubs had a generic ballot advantage of +1 when they lost by -8, a horrific almost double digit misfire), they had Romney over Obama by a point in 2012, et cetera.
that said they did poll both 2008 and 2016 very accurately. they are not totally incompetent but when they are wrong they are terribly wrong (they use 'likely voter' sampling that is sometimes way out of wack with the voting population), and they are a partisan outfit as their inaccuracies are always in favor of Republicans.
and THAT SAID, Rasmussen currently has Biden +8 over Trump which is a little above but largely in line with the aggregate. they are currently seeing what everyone else is seeing.
https://www.nytimes.com/video/opinio...WuHwgjO_bugLdY
Weird Al managed to knock out a pretty good video in one day.
They're hacks when it benefits them and lie about it after the fact.
Has anyone here ever taken one of these? Who TF are these people calling that are still undecided?
matt gaetz is big Cs get degrees energy
do you guys remember when matt gaetz unveiled his adopted mexican son/probably gay lover
feel like we moved on from that pretty quickly
I mean that's a given. Those attitudes go hand in hand.