jerseyhoya wrote:Polling situation for Trump and down ballot Republicans is worse than it was in June, which was worse than April/May, which was worse than pre-COVID.

Over the next week or two there will probably be a public poll a day that comes out from some random red state that has Biden within the margin or a lean R state that has Biden up maybe outside the margin. The race isn't close at all right now. It's hard to describe how bad (or good, depending on one's perspective) some of these numbers are. Clearly worse than 2018 for Republicans.

CalvinBall wrote:

jerseyhoya wrote:Polling situation for Trump and down ballot Republicans is worse than it was in June, which was worse than April/May, which was worse than pre-COVID.

Over the next week or two there will probably be a public poll a day that comes out from some random red state that has Biden within the margin or a lean R state that has Biden up maybe outside the margin. The race isn't close at all right now. It's hard to describe how bad (or good, depending on one's perspective) some of these numbers are. Clearly worse than 2018 for Republicans.

Number seem impossible. Plus double digits in all these "swing states." I cannot comprehend how good this may be for Democrats.

I know there is a debate about online/phone polls. Biden doing several points higher on the phone ones. I read that HRC actually did better on online polls in 2016 compared to phone? Is that correct? Any insight there?

jerseyhoya wrote:Polling situation for Trump and down ballot Republicans is worse than it was in June, which was worse than April/May, which was worse than pre-COVID.

Over the next week or two there will probably be a public poll a day that comes out from some random red state that has Biden within the margin or a lean R state that has Biden up maybe outside the margin. The race isn't close at all right now. It's hard to describe how bad (or good, depending on one's perspective) some of these numbers are. Clearly worse than 2018 for Republicans.

thephan wrote:finally got to the Person woman man camera TV video, that was something

traderdave wrote:This was Trump's 9/11. He would have guaranteed himself another four years if he had treated this like the real problem it is and showed any kind of leadership, compassion, empathy. But, thankfully for non-Trumpees, he #$!&@ it all up.

TomatoPie wrote:

thephan wrote:finally got to the Person woman man camera TV video, that was something

https://twitter.com/MelissaStetten/stat ... 62176?s=20

pacino wrote:

swishnicholson wrote:Posted in the mlb thread too Korean stadiums to open to fans within the week

As a reminder, our countries both had their first case on the same day. We diverged slightly since then.

jerseyhoya wrote:

CalvinBall wrote:

jerseyhoya wrote:Polling situation for Trump and down ballot Republicans is worse than it was in June, which was worse than April/May, which was worse than pre-COVID.

Over the next week or two there will probably be a public poll a day that comes out from some random red state that has Biden within the margin or a lean R state that has Biden up maybe outside the margin. The race isn't close at all right now. It's hard to describe how bad (or good, depending on one's perspective) some of these numbers are. Clearly worse than 2018 for Republicans.

Number seem impossible. Plus double digits in all these "swing states." I cannot comprehend how good this may be for Democrats.

I know there is a debate about online/phone polls. Biden doing several points higher on the phone ones. I read that HRC actually did better on online polls in 2016 compared to phone? Is that correct? Any insight there?

Online polls - or at least the methodologically sound ones - are going to be less noisy/more stable than most public phone polls. They are more stringent about maintaining the same demographics poll to poll, which means there's going to be less variance. An advantage of online polling - especially compared to random digit dialing (RDD), which is what most public polling is - is you know who you are interviewing. So if you're short on white non-college people (or African American women or Independent men) in your completes, you can target them with sample. However, if the changes happening in opinion is due to a material change in the electorate - more young people or more minorities or fewer people identifying as Republicans - but the weights don't change, then the online will probably miss that.

On the other hand, I think a lot of swings in public polls, especially RDD ones like Quinnipiac in Florida today, might be down to non-response bias. When something is going bad for your side, you're less likely to take a poll. You could follow the ups and downs in 2016 pretty well on that. The Quinnipiac survey today had a D+6 electorate, which is outside the realm of what is likely in Florida. Though even if it was D+2 or whatever, Biden would still be up a million. I mean, Joe Biden probably isn't up 13 points in Florida. Like, I think he'd need to be up like 18-20 nationally to be up 13 in Florida. It's a swing state, but the electorate itself is less swingy/pretty inelastic, which is why it always has close elections.

Not really an answer to the question, and I don't know who is more right or more wrong or what's responsible for the shifts. While choosing any input without regard for how it was collected and tossing it into the average is probably not a good idea, you're going to be right more often than not averaging things out, and even more so if you do it carefully like the 538 folks.