Share this page!

Enter URL or ID to Unroll

You can paste full URL like: https://x.com/threadreaderapp/status/1644127596119195649
or just the ID like: 1644127596119195649

How to get URL link on X (Twitter) App

  1. On the Twitter thread, click on or icon on the bottom
  2. Click again on or Share Via icon
  3. Click on Copy Link to Tweet
  4. Paste it above and click "Unroll Thread"!
  5. More info at Twitter Help
dogemperor (https://tinyurl.com/socials314)🐀 Profile picture
dogemperor (https://tinyurl.com/socials314)🐀
@dogemperor
age 18+/gender is wtf/I write on politics and cults with side helpings of fandom/FUB free/I now have a carrd like real tumblr users! :D Socials @ link below!
Nov 4 • 22 tweets • 6 min read
Some gentle reminders for the coming week, because I know people are going to be nervous as hell

My first reminder: IGNORE THE POLLS, NO MATTER WHAT THEY SAY, AND *VOTE* IF YOU HAVE NOT ALREADY DONE SO

1/x Polls don't vote. People do. Today and tomorrow is where we seal the deal and bring it home and that only happens when we vote in NUMBERS.

It looks good so far, but it's always better to get MOAR and make it OVERWHELMING--important in light of GOP lawfare threats

2/x
Nov 3 • 38 tweets • 8 min read
@Politics_Polls @MorningConsult And here, I'm going to remind folks of my First Law of Political Polling:

Never, ever, *ever* trust any political polling group that refuses to publish its crosstabs freely and not locked up behind a paywall (including a paid subscription wall).

1/x @Politics_Polls @MorningConsult Yes, there's a VERY SPECIFIC reason I say Never Trust Anyone Who Won't Share Their Crosstabs Publicly And For Free.

Almost without *exception*, any group refusing to publish a breakout of basic things like "size of pool", "composition of pool", and so on is Hiding Something

2/x
Nov 1 • 87 tweets • 18 min read
So digging through the state-level Marist polls, and we are seeing more evidence (already indicated by both the CES national poll with 78k respondents AND previous Marist polls) of a "Dobbs Effect" that is reaching strongly even into white evangelical Christian pops

1/x So Marist is one of two polls that tend to be less hot garbage than most, and tend to go into some deeper details like sub-surveying of self-identified white evangelical Christians. And this is a MAJOR barometer for a Dobbs Effect, as I'll explain below

2/x
Oct 30 • 42 tweets • 13 min read
Potentially controversial take:

Even IF Biden had outright called MAGAts "garbage" he'd not have been wrong and he would have arguably been CHARITABLE to *merely* call them "garbage" after that "Daily Scheisser Live" three-sheets-short-of-a-cross-burning Hate Rally at MSG

1/x Like, literally, let's JUST go over what was pushed THIS WEEKEND AT AFOREMENTIONED HATE RALLY, and I will remind EVERYONE here that *all of this was explicitly approved to the point it showed up on teleprompters and in rehearsals* so they damn well KNEW what was up

2/x
Oct 25 • 13 tweets • 3 min read
@garymauney @matthewjdowd THIS THIS THISSITY THIS

And this is where I talk about a particular effect of WHY polling has gone to hell (and about all that you can get that's useful is things from crosstabs)

I have dubbed this the Straw Market Problem, and it's an issue even beyond political polls

1/x @garymauney @matthewjdowd More properly, I should refer to this as the Nassau Straw Market Problem, because it relates to a particularly infamous straw market (which does not exist anymore in its older form) and is an example of how aggressive capitalism and marketing can shit up *everything*

2/x
Oct 24 • 40 tweets • 7 min read
And just so people know I don't just look at GOP-friendly polls with a hairy eye, I look at ALL polls with a hairy eye: Big Village CARAVAN, which *is* looking good for Harris but STILL has methodology issues that can skew it rightward

A thread:

1/x So looking at their polling they appear to gather from "opt-in online panels", which is a real fancy term for "Used paid-survey portals".

Which that introduces a risk of pool-poisoning, as we've seen from sites that use YouGov in particular

2/x
Oct 23 • 69 tweets • 11 min read
Yup, kids, it's that time again--in which I am yet again am educating folks on the functional limitations of polling in this day and age (and what you SHOULD focus on with a poll, and what IS usable versus what is Not) because I know you are Panicky on Main

Today: Monmouth

1/x Now, firstly, Monmouth is a TINY poll--all of 802 registered voters, for the ENTIRE US. It's also got a fairly wide margin of error as a result (plus or minus 4.1% on really any given question)

Also seems to have been *entirely* phone and MMS based

2/x
Oct 23 • 31 tweets • 9 min read
@exelephant2017 @ForecasterEnten This is where I remind folks:
a) CNN was taken over in a *hostile takeover* (like this hellsite) by a Trumper CEO who actively promoted incestuous Quiverfull families as entertainment
b) This does *not* match polling especially for under-45s (and polling oversamples over-55s) @exelephant2017 @ForecasterEnten And just so I can bring ALL my receipts:

Even New Republic has noted the takeover of media by right-wing companies newrepublic.com/post/178256/ba…
theguardian.com/commentisfree/…
thehill.com/homenews/media…
vox.com/2022/8/26/2332…

(That's for CNN and its hostile takeover)

1/x
Oct 22 • 30 tweets • 6 min read
@RobertVenditti4 @schlagteslinks Fox, HarrisX are known to be right-wing; NBC is known to have issues (oversampling of older *and* particularly "white evangelical" pops); Quinnipiac is known right-aligned; Morning Consult is known to have issues in methodology; Suffolk & Emerson have similar issues w/ pools

1/x Image @RobertVenditti4 @schlagteslinks Re the Suffolk poll (yes I am about to do one of my Deconstructions), it's another poll with a *massive* skew towards South (33.6%), which *INHERENTLY fucks a poll because that's right in the Christian Nationalist Belt

2/xsuffolk.edu/academics/rese…
Oct 21 • 76 tweets • 16 min read
Yes, this is my usual Don't Rely On Polls and Know The Limitations Of Polling post, this time focusing on Emerson (and I'll compare with observations with Marist--a decent poll--and polls with other methodology issues)

A thread:

1/x So again, we're dealing with another small sample pool of 1000 polled, and tries a mix of online-polling (via CINT, which goes through paid-survey portals), landlines (robocalls), and cells (MMS), and is English-only (so misses a chunk of Hispanic vote)

2/x
Oct 20 • 35 tweets • 9 min read
THIS is the way. We can WIN if we come out in NUMBERS

Check reg! vote.gov
Learn options! iwillvote.com
Make a plan! threadreaderapp.com/thread/1763057…
Make a guide! threadreaderapp.com/thread/1800169…

AND IF WE TURN OUT IN NUMBERS *WE GOT THIS*. It's BASIC MATH

1/x x.com/UnrealBluegras… It's important you come out in NUMBERS even if you think you are in a DEEP red state, because there are ALWAYS Congressional races in even years, there are ALWAYS state and local races of SOME kind (and we need to hold the seats we have and gain as MANY as possible)

2/x
Oct 18 • 61 tweets • 13 min read
The fact there is a new Dunnibg-Krugerrand related scam targeting MAGAts is a GREAT opportunity to discuss what I mean by "easy marks", Affinity Fraud, and how this impacts polling (especially with a candidate that has heavily engaged in affinity fraud).

1/x So (as the post notes) this is not only a classic scamcoin (even more so than other Dunning-Krugerrands) but is also affinity fraud.

The t/o/ken is locked to a wallet, and can't be sold until TFG allows sale (meaning at least 80% will lose their money).

2/x
Oct 17 • 24 tweets • 5 min read
This little demographic actually brings up a really interesting point I keep bringing up

Younger voters (who are more likely to vote Harris) are simply *not* being picked up in polling, and even media consumption between 55+ and under-55 is VERY different

1/x As amazing as it sounds, quite likely *87.6%* of the viewers of the Fox News interview with Kamala Harris were over 55 years of age. At the very least, a very good chance of over 80%

Only *12.4%* were between 25-54, and there's reasons for that

2/x
Oct 17 • 24 tweets • 6 min read
I really don't think people truly, deeply appreciate just HOW thoroughly The Former Guy not only shat the bed but the walls, the dresser-drawer, the vanity, the door, the shag rug, and even the ceiling fan in that disasterpiece of a Univision town hall

1/x So first off. Univision is THE largest Spanish language network in the US. There are cable systems where it's still the ONLY Spanish language network available

It's also one that *fairly recently* was bought out by conservatives

2/xwashingtonpost.com/politics/2023/…
Oct 16 • 56 tweets • 11 min read
@MaryBushard @SimonWDC I'm still trying to find out with the most recent ones, but with the polls people were panicking about last weekend, it's a mix of phone + "paid survey" online, and where age data was available it skewed STRONGLY 45+ @MaryBushard @SimonWDC Also, for the phone polls it was *entirely* registered-voters (only one out of these four sorted by "likely voters"), and at least two are based on completely opt-in "paid survey" portals (YouGov and Ipsos) which are susceptible to poll-bombing
Oct 16 • 46 tweets • 10 min read
Your daily reminder that literally ANYTHING from Polymarket needs to be treated as a junk poll:

a) Polymarket is owned by MAGAt brogliarch Peter Thiel, who is VERY close in with JD Vance (likely to actually become President if TFG is elected) AND with incentive to skew bets

1/ b) Polymarket is not allowed to operate in the US due to a consent decree with CFTC (a federal agency regulating futurities trades) relating to charges Polymarket has been operating an unlicensed, illegal futurities market in the US

2/cftc.gov/PressRoom/Pres…
Oct 15 • 43 tweets • 11 min read
@nerdzsaywhat @KamalaHQ Because of two or three primary factors:

a) A lot of skewed polls (yes, even the "high quality" polls are *significantly* skewed towards populations that tend to vote R):



1/x @nerdzsaywhat @KamalaHQ b) There is a distinct belt of states (ranging from parts of the Midwest through the South and Appalachia, particularly in the "Bible belt") that upwards of 45% or more of registered and active voters are Christian Nationalists (either adherents or sympathisers)

2/x
Oct 14 • 49 tweets • 13 min read
So it would in fact appear that the "third assassination attempt" is looking more and more like a deliberate false flag event, almost certainly meant to try to drive ammo (pun intended) for Trump to get that Personal Defense Army he's been demanding from DoD

A thread:

1/x So this gets Spicy, and this IS going to require a BIT of backstory on players.

Player 1 is Mike Flynn, NAR Christian Nationalist, Trump's likely appointee for Secretary of Defense, and purveyor of more QAnonsense than you can shake a bag of bats with diarrhea at

2/x
Oct 13 • 40 tweets • 8 min read
I am STILL. NOT. DONE in pointing out the methodology problems with these polls, and again this is ANOTHER poll (the ABC/Ipsos poll) that is based, again, on *paid survey pools* and which have the SAME methodolology issues as most online polling

Here we go

1/x So we have to, again, dive for the actual PDF and I should note a bit more on Ipsos as a polling group.

Ipsos gets its polling pools from two sources: Its own paid polling, and redirects from various "paid survey" portals like Qmee

2/xlangerresearch.com/wp-content/upl…
Oct 13 • 46 tweets • 11 min read
We're not done with me discussing how polling is absolute Hot Garbage, and that even the "high quality polls" have such serious issues in methodology that they are missing entire populations and are highly susceptible to skew (so stop panicking and work on GOTV)

1/x So first off, this CBS/YouGov poll *again* is A SMALL pool, 2,719 registered voters--and of note, YouGov is a *PAID SURVEY GROUP*. You have to ACTIVELY sign up to YouGov to get polled, people are PAID to complete surveys for YouGov.

This will ALREADY skew results

2/x
Oct 13 • 47 tweets • 11 min read
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 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