Quinnipiac-PA: Biden +11, Sanders +7, Waren +3, Harris Tied, Butti +1, Beto -2

[Alben Barkley]

50-43=7, not 9.

Sanders is +7

[Alben Barkley]

If Sanders is winning Pennsylvania by 9, he's certainly winning the election as a whole, easily.

This doesn’t follow necessarily. First of all he’s actually leading by 7. Second of all he might be less likely to win Virginia and (especially) Nevada compared to Hillary. Third of all I maintain that Sanders will be destroyed by a general election campaign. The GOP has more than enough ammunition from his past to scare the hell out of moderates/independents and suburban voters who have been leaning D. It could just as easily end up as McGovern 2.0 after he’s put through the ringer.

Not so for Biden, who everyone knows and whose support is therefore more likely to be stable and predictable. And whose ties to PA almost guarantee a win in that state, unlike Bernie who couldn’t even beat Hillary there.

Trump approval is -12 which is certainly too low especially considering his national approval rate which is around -8

Trump's approval in the 538 average is currently -10.9, which is where it's roughly been for a long time.  The RCP average has it closer, but the 538 methodology is arguably better.  It's also not implausible for Trump to have lost significant ground in PA in 2016.  

Having said that, I do agree that the poll looks a little too unfriendly for Trump (but not ridiculously so).

In 2016 Trump did three points better in PA than in the rest of the country, so if his approval rate is somewhere around -9 at the moment, you would expect a approval rate of -6 in PA which means that this poll is overestimating Biden by six points, in other words Biden is probably leading by a middle single digit margin in PA rather than a double digits one (at least as of now)

This post is hereby nominated for Poll Unskewer of the Month.

Before we had Bernie Math, now we have Trump Math.

As you can see, that’s frightening stuff:

https://www.youtube.com/watch?v=DHFB40WOMOo

[Alben Barkley]

Trump approval is -12 which is certainly too low especially considering his national approval rate which is around -8

Trump's approval in the 538 average is currently -10.9, which is where it's roughly been for a long time.  The RCP average has it closer, but the 538 methodology is arguably better.  It's also not implausible for Trump to have lost significant ground in PA in 2016.  

Having said that, I do agree that the poll looks a little too unfriendly for Trump (but not ridiculously so).

In 2016 Trump did three points better in PA than in the rest of the country, so if his approval rate is somewhere around -9 at the moment, you would expect a approval rate of -6 in PA which means that this poll is overestimating Biden by six points, in other words Biden is probably leading by a middle single digit margin in PA rather than a double digits one (at least as of now)

This post is hereby nominated for Poll Unskewer of the Month.

Biden leading by 11 is almost as realistic than Trump leading by 4 in WI

Biden is from Pennsylvania and a ticket with Biden on it won the state easily twice. Trump only won it by a fraction of a percentage point due to depressed Democratic turnout that would not exist with Biden as the nominee. Many more whites, suburban voters, blue collar voters, moderates, and independents — especially in that state — would likely break for Biden in a head-to-head with Trump. Trump currently has a -7 approval in PA according to Morning Consult, but it does not follow that means the poll is overestimating Biden, as “Who would you vote for” and “Who do you approve of” questions are different, and you can’t just add Trump’s 2016 margin in an entirely different race against an entirely different candidate on top like that. That’s now how math works. That’s not how anything works...