General Discussion
Related: Editorials & Other Articles, Issue Forums, Alliance Forums, Region ForumsRosenberg was right.
Link to tweet
MeidasTouch
@MeidasTouch
·
Follow
Yikes. In a FiveThirtyEight podcast released Nov 2, hosts mocked Simon Rosenberg for suggesting partisan pollsters were cooking the polling averages: "Oh no, not this partisan pollster bullshit. Not this bullshit dude. I've never seen so much hopium smoked."
Rosenberg was right.
Watch on Twitter
5:40 PM · Nov 9, 2022
Fiendish Thingy
(15,644 posts)Nate got played big time by obvious rat-fucking partisan polls.
uponit7771
(90,347 posts)Fiendish Thingy
(15,644 posts)They were the most accurate in 2016, giving Trump a 1-in-3 chance of winning.
I guess they decided narrative driving click bait was more important than integrity this time.
uponit7771
(90,347 posts)... for this long tilted towards one party and that system broke hard this election cycle.
538 needs to do something different no doubt.
DestinyIsles
(263 posts)and a lot of leaning left leaning podcast where trashing the same fact that Republicans were flooding averages with garbage polls.
Nevilledog
(51,157 posts)Indykatie
(3,697 posts)That's not what 538 does. They're an aggregator site that uses other people's polls to populate their data engines for their analyses. Shitty polls results in shitty analyses. I fault 538 for not being more selective about the pollsters and polls they include in their analyses. I don't think any pollster that hasn't done a specified number of polls of the state or area they're working in should ever be included. At the end we had marketing firms and others firms releasing crappy polls to boost the GOP averages and 538 included them in the mix. That's a big part of why they ended up being so off the mark on their projections.