I appreciate your take on the interview but I am so tired of people worrying about how Biden will do over the next 4 years if reelected. The crucial issue is that he gets reelected.
That is all.
Might as well lean into it 
After Lester Holt: 538 sez 525-472 Biden.
Seriously, why do we have Vice Presidents? Not that I think he’ll drop dead, or resign, or get tossed as unfit. However, that is what the VP is for.
It definitely changed my opinion a bit. I’ll admit being relieved. He wasn’t perfect, but it was very good in the ways he needed to be. He was quick enough, and he was energized enough.
And this is how the MSM has been failing. Just because they hold much higher standards for Biden than Trump doesn’t mean they have to skew so wildy into bagging on Biden.
67% of Americans think Biden should drop out. He is trailing in the polls. He’s a known quantity, having been in the White House twelve of the last sixteen years. The question I am asking is what people think Biden is going to do to close the gap and change the opinions people have already formed about him. Seems to me that Democrats desperately need to shake this race up.
And yet I’d wager that a majority of that 67% will still vote for Biden. Could even be a strong majority (say, 3 of 4 in that 67% group).
EDIT: Well, not the committed Trump/MAGA/never-liberal voters – remove them from the 67% first.
Asking people if they think Biden should drop out without also asking who should replace him is not really useful. I feel a lot of people don’t really think through complicated situations like these. They are basing their answer on a fantasy that Biden drops out, a magical Super Candidate replaces him, and gets 99% of the vote. The reality is that if Biden drops out, his replacement may have a tougher time winning than Biden does.
Also, asking all American’s isn’t all that useful, either. If I was asked if Trump should drop out, I’d say yes. It’s not because I think someone has a better chance. Rather, it’s because I don’t like him. Also, if he drops out, the Dems would probably have a better chance. So if all Americans are being asked if Biden should drop out, some of the answers are coming from Republicans who might think they would have a better chance with Biden out of the race.
The choice of Vance helps.
There are very different views of what America should be, who we as people should be, on display.
It may sadden me that given the stark choice half … critically more or less… of American voters are choosing a vision that I would not choose … but it is a clear choice. Biden can present that choice.
Leaning into it some more …
534 - 462 Biden
Every Simulation CountsTM
538 numbers make no sense. If you look at say Wisconsin they have Trump up 2.6 points in their adjusted polling average and the fundamentals as Biden up .1. Somehow that leads them to project Biden winning the state by 1.4.
One thing you’ll hear analysts say is that “poll responses aren’t votes”. That can be taken broadly or narrowly – either way allowing for significant adjustment of figures. In this case and for the sake of an example of a number-affecting concern: 538 may think conservative voters are oversampled in some of Wisconsin’s recent Presidential election feeder polls.
Additionally: There seems to be something of a gentlemen’s agreement not to heap too much dung on fellow pollsters … but I can’t help but wonder if there, indeed, are some serious factually-supported doubts** preventing some election analysts from accepting poll results at face value.
** to which the public is not privy
Sure, but isn’t that the point of having an adjusted polling average? Their raw polling avg is Trump up like 1.4, so they already are doing some weighting that is in Trump’s favor.
Looking at 538’s Wisconsin analysis somewhat more closely … I think 538 is tacitly admitting “Yeah, a bunch of input polls are definitely exaggerating Trump’s support”.
From the header of the poll listing:
What are the latest polls in Wisconsin?
We adjust polls for house effects, mode, partisanship, voter likelihood and third parties and weight them based on their firm’s 538 pollster rating and how often it polls.
Then scroll down to the ‘What do the polls and fundamentals alone say?’ section:
Polling average
The state of the polls today according to 538’s polling average only. This does not account for the chance that polls systematically underestimate one candidate.
Full forecast
Our final forecast of the popular vote, based on both polls and fundamentals and accounting for the chance that polls systematically underestimate one candidate.
I dunno. I can only surmise that the following phrase is doing a lot of lifting:
This does not account for the chance that polls systematically underestimate one candidate.
I would think that would increase the certainty of a number but not swing the it. Like I can understand 538 thinking polls are more uncertain and thus Biden has a 40% chance of winning or something, but it really shouldn’t change the favorite. At any rate at this point, I’m inclined to ignore 538 and rely on Nate Silver’s new model which gives Biden roughly a 1 in 4 shot.
One thing I’ve learned from Nate Silver, though, is that in doing that kind of analysis you have to assume that the polls are equally likely to underestimate either candidate. So ignoring that possibility might cause you to think the race isn’t as close as it actually is, but it shouldn’t affect your chances of correctly picking the winner.
If election analysts actually have doubts about the credibility of specific polls, but choose not to divulge those in public, they’re not doing their jobs.
Edit: Yeah, what hawkeye said.
Not necessarily for a single given poll though IF you have some confounding information revealed, say, in the poll’s published data set. A ready example is that NYT/Siena’s polling oversampled self-identified conservatives in their 2023-24 general-election polls – but this was revealed openly in their data sets.
As an aside: I have not checked the very latest NYT/Siena poll (July 2nd) to see whether that is still happening.
It’s because they think the fundamentals favor Biden.
IMO this is dangerous in general and especially in this election. In general, there are just way more variables with some of the fundamentals than the polls. And you only get to test the exact match (did the fundamentals help forecast the presidential election results) once every 4 years, which is a pretty tiny sample. It’s worse in this election because everything about it is so different. Not to mention I don’t think they factor something like voter concern about Biden’s age (not to mention that is probably something they can’t factor because each case is going to be very different and there are only 1 or 2 analogous cases in recent history with presidential elections.
But they have a fundamentals number. Right now they show the fundamentals as even. For Wisconsin:
As of now they show.
Polls: 1.1 Trump
Adjusted Polls 2.6 Trump
Fundamentals: Even
Explain to me how you get Biden plus 1.1 from those inputs. At the very least 538 needs to be much more forthcoming on what factor is showing Biden ahead that offsets the pollings.