*Brand New A + Rated Marist Likely Voter National Poll* Biden:52% Klansman 43%

[DemocratSinceBirth]

http://maristpoll.marist.edu/npr-pbs-newshour-marist-poll-results-election-2020-president-trump/#sthash.tsiS1feL.0dmSXdSF.dpbs

Was Closest To The Mark In 2016


https://www.mcclatchydc.com/news/politics-government/election/article112635048.html

[PTWB]

1. National polls really don't matter.

Joe could win nationally by 20 points and still lose the election if the swing states go for Trump by just a few tenths as they did in 2016.

We must GOTV in the states that actually matter.

[Johnny2X2X]

2. Yes they do matter!

He wins by 9 points and there's zero chance he doesn't win the EV.

And this poll is great, best pollster out there.

[DemocratSinceBirth]

5. It is hard to believe an otherwise serious person can believe the inflection point is 20.

It's more like 3.

[Drunken Irishman]

10. It might even be 2.5.

[DemocratSinceBirth]

11. You are correct, sir. The inflection point in 016 was about 2.3 or so.

[Buckeye_Democrat]

3. True, but the states tend to follow national trends.

State polls are not independent variables. When national polls go one way, state polls generally go the same direction to varying degrees.

[Demsrule86]

4. Yes they do...and no Joe could not win by 20 points and still lose the EC.

[Chichiri]

7. Somewhere around here there's a Nate Silver tweet...

...laying out the odds of Biden winning as a function of net popular vote victory. If he wins the popular vote by 3 points it's a toss up, at 4 points it's 80%, and at 5 points it's 98%. (Our system is really broken.)

[DemocratSinceBirth]

13. It's bad enough we have to win by 3 o 4.

But not 20!!!

[Drunken Irishman]

9. Mathematically improbable, if not impossible, for that to happen.

[obamanut2012]

14. That is literally impossible

He wins just by 6%, and he wins. 20% and he wouldn't win lol.

[PTWB]

15. No it isn't "literally impossible"

Extremely unlikely and exaggerated to make a point, to be sure, but it is absolutely within the scope of “possible.”

As blue states get bluer but battleground states stay ultra competitive, our national margin grows with no real advantage.

Biden could win by 10 million votes and still lose the election if those votes are in the wrong states.

[Awsi Dooger]

17. Yes it is impossible

Shifts don't happen along the extremes you suggest. That's because the national ideological numbers do not change appreciably, and the ideological numbers dictate how the nation votes in federal races. If moderates/independents are slanting appreciably one way or another there's going to be a reason for that which attaches everywhere, not merely in the blue states.

Nate Silver properly gave Trump more chance than other compilers in 2016 because he understood that dynamic, that an unforeseen shift would be felt everywhere. The dummy analysts who fixated on state polls alone looked at each state as rigid 85% Hillary likelihood here and 81% Hillary likelihood there, without grasping that they'd all fall together upon a shared variable.

State polls are junior varsity polls. They are poorer funded, less sophisticated, devote less time, more prone to flawed approach, and so forth. It is silly to rely on state polls and proclaim national polls don't matter. Somehow it became conventional wisdom to boost state polls while scoffing at national polls. It should be the other way around. That's why you are receiving justified blowback.

[PTWB]

20. You must not be familiar with the definition of impossible.

To quote your aforementioned Nate Silver:

SEPT. 18, 2020
Joe Biden’s lead in national polls has narrowed to 7 percentage points, but he remains favored to win the election because we have a number of high-quality state polls that contain mostly good news for him.

[BannonsLiver]

18. They actually do

And moreover, they matter to people who like to be informed, and who are capable of discussing data points rationally. Give it a whirl sometime.

[uponit7771]

21. above 5% they do

[mahina]

23. There's a lot of very detailed information in there

It’s worth checking out. It’s not one number

[honest.abe]

6. This is a nice solid lead that should translate into a huge EC victory.

[Ohiogal]

8. I love how you always call him the Klansman

Cause that’s what he is!

[Yeehah]

12. 43 percent support a traitorous conman

Boggles the mind.

[RhodeIslandOne]

16. He hates who they hate

All that matters.

[Buckeye_Democrat]

19. Right-wing authoritarianism is found 'round the world.

However, it's really bad in this country!

I don't consider it a "personality trait" like some scholars who have studied it. I'm convinced that it's awful behavior that can be learned, and the decades of right-wing hate-media helped bring us here.

[OregonBlue]

22. Johnson 61.1, Goldwater 38.5. At this rate, Trump may be in Goldwater territory by the election.

[Capt. America]

24. Not even. People who are polled may feel "ashamed" to admit they are going to vote for Dear Leader.

[SunSeeker]

25. So 52% Biden, 42% Trump, 3% Other, 3% undecided.

I'll take that!