Thread by @dogemperor

So, again, I'm going to point out some reminders re those "scary polls" (and a friendly reminder that polls don't vote, people do, and a lot of the people that vote are simply *not being picked up in the polling*)

So sit your asses down and read the goddamn thread:

1/x

https://twitter.com/IanSams/status/1845472556322635799

One: These polls are based on EXTREMELY LOW POOLS OF VOTERS.

The NBC pool that people are flipping their shit about is 1000 people NATIONWIDE. The NYT/Sienna cross-tab polls are 800 people on average per state

These pools are actually *shrinking*, too

2/x

I've written a rather extensive thread on the methodology problems in the NYT polls (including the fact they *STRONGLY* skew to pops over 45 and include people who hang up mid-survey)

And even in "high quality" polling these are common issues

3/xthreadreaderapp.com/thread/1845207…

The NBC poll only has *16%* of respondents who are under 30, and only 41% total under 45; this is also the demographic that tends to be smartphone-only, that tends to *heavily* use either callblocking apps or whitelisting due to telemarketing/phishing calls

4/x

If a pollster calls a number that either never picks up, or gives a message that the caller is not accepting calls or is not in service (which is how pretty much all callblocker apps work except for Robokiller in its "Robocall Revenge" mode) that gets marked as uncallable

5/x

These apps also block any survey calls (and younger persons under 30 often tend to flat-out whitelist), meaning the only callers that get reached are the few smartphone users that DON'T use something like Hiya or Robokiller. And this WILL skew calls, as I'll note

6/x

This means that 59% of the calls hit people over 45, and 45% of the calls in the NBC poll reached people 55+ (who are also more likely to have landlines, are less likely to screen calls using a callblocker app or even CID, and trend more conservative).

7/x

This is not quite reflective of US demographics. Over 50% of Americans are under the age of 40, and in Denver alone over 45% of people are between 15-45

This and knowing patterns people use to protect privacy and how they impact polling is Important

8/x

People who will accept a random telephone call, especially on a cellphone (and who do not use SOME form of whitelisting or callblocking app) are also in a cohort that is more susceptible to various scams, to affinity fraud, to telemarketers and phishing schemes

9/x

And this cohort ALSO tends to vote more conservatively than the general population, and is notably more likely to not just believe conspiracy theories but *actively propagate them* because they are worse at "reality testing"

10/xthreadreaderapp.com/thread/1842984…

And younger folks who ARE more privacy savvy (who are also considerably more likely to be Harris/Walz voters!) aren't even going to be counted at all, because their numbers don't get counted as refusals, it gets counted as "phone number invalid" in phone bank data.

11/x

10% of households contacted were Latino, 58% of Mexican origin--meaning 41% were non-Mexican Latinos, which can be Boriquens (who have been trending more D) to Cubanos (a fairly solid R bloc, especially in Florida) to Venezuelans (solid D)

They don't break out non-MX tho

12/x

So you've already got a population in phone polls (by the actual nature of a phone poll itself) that is going to lean R, and the polling involved 72% non-Hispanic whites and 12% African-Americans

And there's ANOTHER major thing that pops up in that poll

13/x

37% of the people who answered the survey are unemployed, with fully a fourth of them being retired. Only 3% students in college, similar numbers of SAHPs and unemployeddocumentcloud.org/documents/2520…

There are also SUBSTANTIAL skews towards professional managers (25%), a population that includes CEOs, and which trends more R than the general population; blue-collar workers (20%) as well, which in non-union states tend to *also* skew more R, esp. in the South

Also, something like 47% of the survey are people who have not earned at least an Associate's Degree, and *62 percent* who haven't earned a Bachelorate's Degree; populations who haven't completed college, or earned at least a BA/BS, are more likely to vote R

So you already have a pool that's pretty weighted R compared to the average population, and even worse, this is a *nationwide* poll that does not break out its respondents by state or by urban/rural/suburban--which is a pretty serious issue in this poll.

And there's also some other stuff in demographic trends that I had to scroll to the end that also indicated this swung towards rural and suburban voters and probably hit at least ONE major hotspot of full-on New Apostolic Reformation sorts.

I'll explain.

70% of people who answered the survey say they "own their own home or residence", which skews pretty heavily against urban/suburban people under 30 (and also skews even against older people in cities)

Nationwide, it's more like 65% own, 35% rent *overall*

For people who are Millennials and younger, the number of renters is SHARPLY higher--only something like 42% of millennials (including millenials in cities who are more likely to vote Harris/Walz) own their own homes metonic.net/millennials-co…

And most of the Millennials and younger who own their own homes either tend to be rural (trends R), or tend to be CEOs/managers (trend R).

That's not even the thing that stands out most in demographics though, and not even what makes me raise an eyebrow

The *religion* data--combined with "Do you identify as MAGA?" question--makes me STRONGLY suspect that a good portion of their calling pool was NAR, including NAR groups targeting Hispanics and some African-American populations, and that will STRONGLY skew the survey

So MSNBC's polling has been getting something like a near consistent 28-31% of respondents who identify as evangelicals--and in the last two polls, an *identical* number who identify as MAGAts (before July it's more like 20-25%).

This already makes my Spidey-sense ping

Actual estimates of "how many evangelical Christian in the US" vary wildly, from 5% to 35% depending on how the question is framed, but MOST higher-quality surveys estimate it's something like 15%
prri.org/research/censu…
news.gallup.com/opinion/pollin…

And depending on HOW a surveyor defines it, it can be anything from "Church part of the Wesleyan and Baptist traditions" to "Catholic with a 'charismatic' service" to full-on Christian Nationalist

And there's some interesting data on religious affiliation

In part BECAUSE of Christian Nationalism and because of an entire generation or three of people who have become #exvangelicals, the age cohort from 18-50 is STRONGLY "religiously unaffiliated" and trending ever more in that direction axios.com/2024/09/13/rel…

And--interestingly enough--also this hits RIGHT in the polling demographics least likely to be pollable, including the "Oregon Trial Generation" section of GenX.

There ARE a few areas of the country that are exceptions, however, primarily in the South.

There are in fact rural counties in the Southeastern US where a strong majority of people identify as "white evangelical Protestants" (including in Marion County, AL and Johnson County, TN)

These also tend to be MAGAt strongholds of "so red it's infrared" counties.

Evangelicals in these areas are STRONGLY Christian Nationalist, and often of a specifically neo-Confederate or New Apostolic Reformation flavor; there is also a STRONG social stigma against being religiously unaffiliated or even part of a non-evangelical church

And with the percentage being *EXACTLY* 30% "self identified evangelicals" AND 30% "identifies as MAGA", that STRONGLY points to "likely 30% of more of the people who answered this are full on members of the New Apostolic Reformation or otherwise Christian Nationalist".

Multiple studies have shown that the SINGLE strongest factor that exists on whether someone is likely to vote for TFG is if they are specifically New Apostolic Reformation

@andre_gagne1 has noted 76-81% of white evangelicals voted for TFG in 2020 concordia.ca/cunews/artsci/…

49% of surveyed white Protestants regularly attending church showed specifically NAR beliefs in TFG being "anointed by God" as President: religionwatch.com/prophecy-voter…

And Lance Wallnau, NAR preacher, has been stumping heavily for Trump & Vance archive.is/05Jl9

And there are issues with NARasitized mainline Christian churches, but--again--it looks like the surveys before September might have captured non-Trump Evangelicals...and it is SUBSTANTIALLY more likely self-identified evangelicals are NAR religionnews.com/2024/05/06/how…

And this becomes very obvious if you ask them on specific distinctives of NAR theology that are almost completely exclusive to "Seven Mountains" theology, including the idea that demons and "principalities" can control actual meatspace and functionally "possess" it.

This is in fact such a striking correlation that PRRI has conducted a survey regarding Christian nationalism itself

And there are very specific regions where Christian Nationalism is strongest, and likewise, MAGAt support.prri.org/research/suppo…

In *BATTLEGROUND STATES* specifically (including the *same* battleground states that Lance Wallnau is now trying a revival-cum-political-stumping-call as noted below) something like 10% of people are full-blown Christian Nationalists and 29% "Sympathisers" peoplefor.org/rightwingwatch…

BUT there is a specific "Christian Nationalist Belt" from the Midwest to the Deep South up through Appalachia, and North Dakota, Mississippi, Alabama, West Virginia, and Louisiana all have over 45% Christian Nationalist or "sympathiser" populations.

And while the national average is more like 25-30%, in the "Christian Nationalist Belt" often 4 out of every 10 residents are full-blown Christian Nationalists or sympathetic to Christian Nationalism

(THIS is why we tell you the South *is* really different)

The same survey also found (again) that the single strongest predictive factor for a potential Trump voter IS in fact "Are they a Christian Nationalist"--AND showed, conversely, a strong correlation between BEING a Christian Nationalist and voting GOP

And this correlation holds for exactly two demographics--white "Pentecostal/Charismatic" Christian Nationalists, and "Hispanic Protestants" (who are LARGELY members of New Apostolic Reformation churches)

But NOT African-Americans

And there's actually a historic reason for that (in part because the movement that ultimately rebranded itself as the NAR was descended from KKK-linked "white Pentecostal" churches, and in part because African-American Evangelicalism has had a long separate history)

Also of particular note (and this might give the Dems an actual wedge issue) is that the same survey shows white Christian Nationalists are *far* more likely to use isolating news media (Fox/RSBN/Christian Nationalist media etc.) AND to go for racist talking points re immigration

So I am STRONGLY suspecting that the NBC survey hit a fair pocket in heavily-Christian-Nationalist states which is going to skew the survey BADLY.

Dear NBC: THIS is why you put data in on what states you called!

This is ESPECIALLY true with battleground states, because there is actually a slightly *higher* percentage of people who overtly reject Christian Nationalism and a slightly smaller number of sympathizers versus the US in general.

So yeah, THAT and the fact they're not breaking out *non Mexican Latinos* (many of whom are in populations *explicitly targeted* by the NAR--I'm thinking explicitly Cubanos and Guatemalans in particular here) makes me think some of the methodology here is Fucky

So again, your reminder: Polls don't vote, people do, and WE DON'T WIN IF WE DON'T COME OUT IN *NUMBERS*.

Check reg! vote.gov
Learn voting options! iwillvote.com
Make a vote plan! threadreaderapp.com/thread/1763057…
AND GET READY TO VOTE IN *NUMBERS*!

/end

@threadreaderapp unroll