Yeah, it’s basically a dead heat in national polls; it is pointless given the incredible number of issues that make it more complex to look at than just “which number in the national polling is higher” to get any level of heated about swings up or down by single percentage points in the national poll, much less “should Biden drop out?” or similar questions.
None of this is to say there is no problem. 538’s overall bottom line estimate which takes into account factors way more complex than I have dug into, arrives at an ultimate output of Biden having about a 25% chance of winning the election. That’s probably the closest thing we have to an overall “what’s the likely election outcome” polling barometer. I’m just sort of talking about poking holes in the lazy and partisan way it tends to be analyzed when the media talks about polling.
Edit: Wait WHAT THE SHIT
I went looking for the summary model to send that said 25%, and found this - Nate Silver gives Biden overall a 52% chance to win. WHEN THE FUCK DID THIS HAPPEN
Was I just looking at the 2020 model before and not realizing it, or something? Have I become so dyslexic that I read 52 as 25 even when there was a whole chart and everything?
My world is upended and my morning is somewhat happier now. @[email protected] hey dude I love polls again
Got it. That makes it make sense. This is Nate Silver’s, and I am almost sure that it predicts 25%. I think it was un paywalled at one point and that’s when I saw it.
Yeah, it’s basically a dead heat in national polls; it is pointless given the incredible number of issues that make it more complex to look at than just “which number in the national polling is higher” to get any level of heated about swings up or down by single percentage points in the national poll, much less “should Biden drop out?” or similar questions.
None of this is to say there is no problem. 538’s overall bottom line estimate which takes into account factors way more complex than I have dug into, arrives at an ultimate output of Biden having about a 25% chance of winning the election. That’s probably the closest thing we have to an overall “what’s the likely election outcome” polling barometer. I’m just sort of talking about poking holes in the lazy and partisan way it tends to be analyzed when the media talks about polling.
Edit: Wait WHAT THE SHIT
I went looking for the summary model to send that said 25%, and found this - Nate Silver gives Biden overall a 52% chance to win. WHEN THE FUCK DID THIS HAPPEN
Was I just looking at the 2020 model before and not realizing it, or something? Have I become so dyslexic that I read 52 as 25 even when there was a whole chart and everything?
My world is upended and my morning is somewhat happier now. @[email protected] hey dude I love polls again
The national poll isn’t going to decide this election though. It really is about a 100,000 people in 3 states.
Nate Silver is not at 538 any more. I believe his model has gives Biden 35% or so, but it’s paywalled so I’m not entirely sure
Got it. That makes it make sense. This is Nate Silver’s, and I am almost sure that it predicts 25%. I think it was un paywalled at one point and that’s when I saw it.
Thanks fur the link