Politico/Harvard/SSRS: Biden +10%

[TiltsAreUnderrated]

https://www.politico.com/f/?id=00000174-bb97-dc18-a57d-bbf77dd00000

August 25-September 6
1459 likely voters
MoE: 3%

Biden 52%
Trump 42%
Neither 2%
Other 1%
Don't know/refused 3%

[GeneralMacArthur]

National polls don't matter, state polls are all that counts.

[TiltsAreUnderrated]

National polls don't matter, state polls are all that counts.

National polls - at least, their aggregates - tend to be a little more accurate.

[ON Progressive]

I don't see the topline anywhere at all.

[Interlocutor]

"Pretty good polling day for Trump."

[tagimaucia]

National polls don't matter, state polls are all that counts.

If you knew that national and state polls were of equal quality, sure. But in different cycles, one
or the other has sometimes painted a more accurate picture. National polls were a lot better in 2016 and 2018.  State polls were much better in 2012.

[TiltsAreUnderrated]

I don't see the topline anywhere at all.

Page 17

[ηєω ƒяσηтιєя]

Pretty old poll but good poll for Biden.

[CellarDoor]

Pretty old poll but good poll for Biden.

[MT Treasurer]

This suggests that the takes about how Biden's support is way more effectively distributed in the EC than Clinton's were absurd. FL, GA, and NC being this close in a Biden +7-10 environment while PA is looking more R than the nation than in 2016 suggests that Trump's EV advantage relative to the popular vote has grown since 2016.

(I’m not saying that it matters because he’s certainly not winning the EV if those Biden +7-10 national polls are accurate [and there’s little reason to believe they aren’t], but it wasn’t that long ago that some people here unironically entertained the idea of the tipping-point state voting more D than the "nation as a whole.")

[Dr. Arch]

This suggests that the takes about how Biden's support is way more effectively distributed in the EC than Clinton's were absurd. FL, GA, and NC being this close in a Biden +7-10 environment while PA is looking more R than the nation than in 2016 suggests that Trump's EV advantage relative to the popular vote has grown since 2016.

(I’m not saying that it matters because he’s certainly not winning the EV if those Biden +7-10 national polls are accurate [and there’s little reason to believe they aren’t], but it wasn’t that long ago that some people here unironically entertained the idea of the tipping-point state voting more D than the "nation as a whole.")

Alternatively, state polls could be underestimating his support in certain big ticket states, like Texas, while national polls are capturing that support.

[Terry the Fat Shark]

This suggests that the takes about how Biden's support is way more effectively distributed in the EC than Clinton's were absurd. FL, GA, and NC being this close in a Biden +7-10 environment while PA is looking more R than the nation than in 2016 suggests that Trump's EV advantage relative to the popular vote has grown since 2016.

(I’m not saying that it matters because he’s certainly not winning the EV if those Biden +7-10 national polls are accurate [and there’s little reason to believe they aren’t], but it wasn’t that long ago that some people here unironically entertained the idea of the tipping-point state voting more D than the "nation as a whole.")

Maybe? We don't have a lot of good quality polling from PA lately

[Splash]

This suggests that the takes about how Biden's support is way more effectively distributed in the EC than Clinton's were absurd. FL, GA, and NC being this close in a Biden +7-10 environment while PA is looking more R than the nation than in 2016 suggests that Trump's EV advantage relative to the popular vote has grown since 2016.

(I’m not saying that it matters because he’s certainly not winning the EV if those Biden +7-10 national polls are accurate [and there’s little reason to believe they aren’t], but it wasn’t that long ago that some people here unironically entertained the idea of the tipping-point state voting more D than the "nation as a whole.")

Alternatively, state polls could be underestimating his support in certain big ticket states, like Texas, while national polls are capturing that support.

Also Ohio. Which, by the way, has been absurdly underpolled this cycle.

[republican1993]

Biden can win it by 10 points all i care about is pennsylvania, arizona, and florida if trump wins it by 10k votes idc

[Penn_Quaker_Girl]

Biden can win it by 10 points all i care about is pennsylvania, arizona, and florida if trump wins it by 10k votes idc

I mean, if Biden's PV margin is 10 pts., he's not losing the election (even if Cali and New York max-out for the Democrats).  

There's no concession to be made here: anything over Biden +5-6 is a pretty certain victory for him. 

[MT Treasurer]

Alternatively, state polls could be underestimating his support in certain big ticket states, like Texas, while national polls are capturing that support.

It’s possible, but TX polling has been fairly reliable historically (usually slightly underestimating Ds) while FL/PA had some of the more accurate polling compared to other battleground states (esp. IA/OH/WI/MI) in 2016. This is also why I’m very skeptical of the "PA and IA are competitive while WI/MI are Likely D" takes, which are eerily reminiscent of 2016.

It’s weird and I have no explanation for the discrepancy, but I do think people are underestimating how much ground Trump has lost in CA and especially TX (that doesn’t explain the entire gap, but it’s a big part of the story). It bears repeating: Trump is in real danger of losing TX.

[Illiniwek]

I don't care to see polls this old.

[Landslide Lyndon]

Why do they sit on these polls for 2-3 weeks?
And what's the purpose of releasing them now?

[Sbane]

This suggests that the takes about how Biden's support is way more effectively distributed in the EC than Clinton's were absurd. FL, GA, and NC being this close in a Biden +7-10 environment while PA is looking more R than the nation than in 2016 suggests that Trump's EV advantage relative to the popular vote has grown since 2016.

(I’m not saying that it matters because he’s certainly not winning the EV if those Biden +7-10 national polls are accurate [and there’s little reason to believe they aren’t], but it wasn’t that long ago that some people here unironically entertained the idea of the tipping-point state voting more D than the "nation as a whole.")

I think it's possible a lot of the battleground states in the sunbelt and the midwest will vote right around each other such that Biden could lose if he only wins the PV by 5 but could get 400+ EV's if he wins it by more than 9 points.

[wbrocks67]

This suggests that the takes about how Biden's support is way more effectively distributed in the EC than Clinton's were absurd. FL, GA, and NC being this close in a Biden +7-10 environment while PA is looking more R than the nation than in 2016 suggests that Trump's EV advantage relative to the popular vote has grown since 2016.

(I’m not saying that it matters because he’s certainly not winning the EV if those Biden +7-10 national polls are accurate [and there’s little reason to believe they aren’t], but it wasn’t that long ago that some people here unironically entertained the idea of the tipping-point state voting more D than the "nation as a whole.")

Maybe? We don't have a lot of good quality polling from PA lately

Exactly, we've seen polls of PA 7-9, AZ 5-10, FL 3-5, etc. - if those are actually the true totals that Biden +10 nationally makes sense.

[MT Treasurer]

I think it's possible a lot of the battleground states in the sunbelt and the midwest will vote right around each other such that Biden could lose if he only wins the PV by 5 but could get 400+ EV's if he wins it by more than 9 points.

This has generally been my belief as well, which is why it’s weird to see the Sun Belt states polling way to the right of the Upper Midwest/PA. I know there are different dynamics at play in those regions and that there might be fewer persuadable voters/higher Republican floors in the Southern states, but I have a very hard time believing that FL is as competitive as IA and ME-02. Biden's strength in places like ME suggests a blowout of historic proportions, whereas his numbers in FL/GA/NC suggest a tilt/lean D race at best.

[Tekken_Guy]

I think it's possible a lot of the battleground states in the sunbelt and the midwest will vote right around each other such that Biden could lose if he only wins the PV by 5 but could get 400+ EV's if he wins it by more than 9 points.

This has generally been my belief as well, which is why it’s weird to see the Sun Belt states polling way to the right of the Upper Midwest/PA. I know there are different dynamics at play in those regions and that there might be fewer persuadable voters/higher Republican floors in the Southern states, but I have a very hard time believing that FL is as competitive as IA and ME-02. Biden's strength in places like ME suggests a blowout of historic proportions, whereas his numbers in FL/GA/NC suggest a tilt/lean D race at best.

The Sun Belt states are less white than the ones in the Rust Belt, but whites are way more Republican in the former.

[Alben Barkley]

10 is the magic number today.

[Mr.Bakari-Sellers]

Daines, Ernst and Marshall, in a plus 10 will lose, sorry Indy Rep, in plus 10 election Daines will lose. Bullock is only 1 pt behind Daines.

[Alben Barkley]

National polls don't matter, state polls are all that counts.

No universe in which Biden wins nationally by 10, as THREE polls today have found, and does not win the EC. So I’d say it’s significant.