I trust these people when they say it is huge. They were right on the taxes story. It might be the China story if tonight was a lead-in to something much bigger. But once the Judge in the Maxwell case set the date to Thursday, given the time frame that was explained to me, my alarm bells went off. Also apparently Trump had something whispered to him before he went bonko on Lesley Stahl right about the time the judge released that order.

Who if you know the name, is not the kind of person that makes one go bonko.

I think it is the Maxwell depo. Edit: can't do a poll. Not Star Member.

I swear that on my Grandfather and Grandma's grave, the latter of which died of dementia related heart failure two years ago.

On this very date.

So I am not fucking with you.

But I will not retaliate with counter ad hominem posts.

If I got bad info then so be it. Life is too short to waste it arguing over the trivial.

https://twitter.com/matthewjdowd/status/1318731054954459139?s=20

?fit=scale

https://twitter.com/Rasmussen_Poll/status/1318628054894223361?s=20

Not plausible 100 percent of undecideds go to incumbent.

The actual number though is 50-45.

Of those certain to vote, 52-44. Spin to win, eh Ras?

a 7-10 point solid Biden lead. Votes are being banked right now at plus 7-10. The polls will only get better too, because "already voted" is a likely voter.

Duh.

Somebody is way wrong. NYTimes has Joe up 9 with only 50 percent support with Hispanics.

Tipp has Joe up 2 with 65 percent of them.

So he would either be tied in TIPP at the NYTimes Hispanic number or up by 11 in the NYTimes poll with the TIPP number.

NY Times has him 44 with whites. Good number. But even TIPP is at 41 with whites, not as good, but projects to a 53-45 national vote tally.

I do believe the race has tightened to what I call baseline, everyone in their partisan corners so to speak.

But the issue with TIPP is their turnout model looks to be something like 75 percent white.

It won't be higher than 70. Might be 68. NY Times had it at 67. But because of the odd Latino split it would only drop him to plus 8 if I set NYT model at 70 white.

So you tell me. Polls taken 10-15-19 TIPP, 10-15-18 NYT. Did I miss something on news last night?

Hillary got 599k votes out of Philadelphia CO.

There are 851,483 Democratic voters alone. Not to mention independents that in an urban area probably tilt blue.

Do our jobs, get this vote out, and a 44k loss turns into a sizable win. (Side note-half projected statewide early vote is in in MI, perhaps more, and almost half in WI and MN.)

Obama got 557k out of Philly in 2012. Somewhere around 730k came out in 2016.

About 15 percent of the statewide projected turnout has already been turned in.

191k of that is from Philly.

21 percent of all early vote is from Philly CO.

197k from Allegheny.

43 percent of ALL statewide ballots returned come from two Democratic counties. About half of the ballots requested in those counties.

It is October 20th.

it is bad, BUT, it was only based on an application to grant a stay not a full ruling on the merits.

Which would be much scarier.

https://www.supremecourt.gov/orders/courtorders/101920zr1_ebfi.pdf

Landline numbers and cell phone numbers are scheduled for 4+ call attempts. When calling
landlines interviewers ask to speak with the adult member of the household having the
next birthday. Interviews are conducted on cell phones with both cell only and dual owner
respondents. The complete land and cell sample is weighted to National Health Interview
Survey estimates for [land only/cell only/dual owner] households.

https://poll.qu.edu/images/polling/pa/pa10072020_demos_bgny70.pdf

https://nationalinterest.org/blog/buzz/study-only-65-percent-us-households-have-landline-135212

6.5 percent.

Yes it matters. It explains "tight" polls vs. blowouts. Who will be right? The ones that get the correct sample of white college vs. non.

But this I do know-there won't be as much "missing" white non-college using only cell phones or online as "missing" white college with no landlines and a 36.7 percent sample.

So on balance a high landline sample could lean a poll several points red.

TIPP uses 20 percent. They are the outlier now; were off in 2016 by 5 points.

State polling is still doing a lot of IVR and landline. This might explain the disconnect in some state polls to national.

Polls can be different. Ipsos is an online poll, which today in PA is about where Joe's own internals are.

But there were 50-45 a few weeks back when Q and Monmouth showed more than double digits there for Joe.

This much I believe-an emerging story from this election however it turns out will be the zero sum game played amongst pollsters and which pollsters can move forward versus have to completely re-evaluate their methodologies.

Polling will be almost exclusively online in four years in my opinion.