How to Rig Polls Through Oversampling

RUSH: I’m holding here in my formerly nicotine-stained fingers a story on how pollsters manipulate polls. How do you manufacture a lead for your chosen candidate? How do you do this? And it’s very crucial to do it. You publish a poll that shows your guy up with a 10- or 12-point lead. It can be dispiriting to your opponents’ voters. It can be dispiriting to your opponent. And it can chill, it can chill enthusiasm.

“Oh, my God. My guy’s down 12. Aw, jeez. All is lost. Ah.” And that’s how it works. How do you do it? Hey, this is no great secret. It’s just an admission. This story is from 2016 and it was about John Podesta, how New Podesta Email Exposes Playbook For Rigging Polls Through “Oversamples”. It’s October of 2016 at Zero Hedge.

And, by the way, I have people, “Hey, Rush, did you see what Tyler Durden said?” Tyler Durden is the name associated here. There is no Tyler Durden. It’s a made up name so nobody knows who the guy is. So is the picture next to it. He looks like a geek. Don’t know who he is but there is no Tyler Durden. Just a little information. Could be worthless or worth something to you regardless.

This is from 2016: “Earlier this morning we wrote about the obvious sampling bias in the latest ABC / Washington Post poll that showed a 12-point national advantage for Hillary. Like many of the recent polls from Reuters, ABC and The Washington Post, this latest poll included a 9-point sampling bias toward registered Democrats.”

And if you read the ABC News poll sampling and methodology note, it says, “This ABC News poll was conducted by landline and cellular telephone Oct. 20-22, 2016, in English and Spanish, among a random national sample of 874 likely voters. Results have a margin of sampling error of 3.5 points, including the design effect. Partisan divisions are 36-27-31 percent, Democrats – Republicans – Independents.”

Bingo! You have a phony sample. And we know that these are constant. How many times have you seen a poll, it looks bad for whoever you’re believing in or whatever you’re believing in, and then in a couple of days or maybe later the same day, wait a minute, we just went to the crosstabs here and we just analyzed it and look at the oversampling of Democrats.

It’s a common trick. And the newsworthiness of this story at Zero Hedge was that there was a John Podesta email detailing how to do it and what the advantages of it were. I’s not detailing how to do it. It’s easy to do. You just talk to more Democrats than Republicans. And you make sure your Republican sample is much less than — Twenty-seven percent of any random voter sample Republican versus 36% randomly gonna be Democrat?

And this is a common — it’s especially used in polls this far out all the way up to three months out. That’s why it’s another reason to just ignore these things, especially if you’re able to dig in and look at the sample. If you find out it’s vastly overweighted for Democrats, you know exactly what’s going on here. It is a poll designed to make news, to dispirit, depress, and make despondent Republicans and their candidates.

RUSH: I’m told that the name Tyler Durden is the character name for Brad Pitt in a movie called Fight Club, which I never saw. Did you see Fight Club? I’ve never seen Fight Club. All I know is that there’s no Tyler Durden at Zero Hedge.

It’s a common trick. And the newsworthiness of this story at Zero Hedge was that there was a John Podesta email detailing how to do it and what the advantages of it were. I’s not detailing how to do it. It’s easy to do.
You just talk to more Democrats than Republicans. And you make sure your Republican sample is much less than — Twenty-seven percent of any random voter sample Republican versus 36% randomly gonna be Democrat?

Another component is a press that plays brain-dead - as if they don't know the scam. Truth is the press is IN on the scam... they publish the results without question... Shame on them...shame on them.

I'm reposting my analysis of last week's Quinnipiac poll to demonstrate the point.

Here is my hunch on the Quinnipiac Poll Limbaugh is talking about.

The poll is of "registered" voters, not "likely" voters. Polls of likely voters have traditionally been considered the most accurate.
The poll weighting is D/R/I/DK 33/31/28/8. Usually, these polls are D+8, but now they've been closer to D+2. I suspect that Democrats are hiding as Independents in polls now, in order to skew results while claiming that the samples are more balanced.
In the Trump-Biden matchup, the R/D split for Trump voters is 91/3, and 6/95 for Biden voters. This means that twice as many (3%) Republicans are willing to cross over to Biden than Democrats are willing to cross over to Trump.
The interesting thing is in the Independents. Independents are breaking 2:1 for Biden in this poll. The Trump/Biden split is 28/58. Is this the true sentiment of so-called Independents or are they really dark-horse Democrats? Let's see how the Independents break for the downline Democrat candidates.
First, the R/D splits are consistent for the other top candidates:

Trump 94/3 Sanders 3/94
Trump 94/3 Warren 2/94
Trump 94/4 Harris 2/93
Trump 94/4 Buttigieg 2/91
Trump 94/4 Booker 2/91
There is clearly a 4% "Never Trump" faction amongst Republicans in this poll.

For Independents...

Trump 33 Sanders 57
Trump 34 Warren 51
Trump 31 Harris 53
Trump 32 Buttigieg 47
Trump 33 Booker 49
The Independents break about the same for Democrats regardless of who the candidate is. Is this normal for Independents, or should we expect some personalities to pull or push more than others?

When we move to the non-candidate questions...

President's Job Excellent+Good (R/D/I) 92/5/31
Economy Excellent+Good (R/D/I) 95/54/65

Independents generally think the economy is doing well, but the President is not doing a good job. Are these Democrats speaking?

Good Economy Trump Credit Yes (R/D/I) 85/8/34

Here we see that Independents who think the economy is doing well are not willing to credit Trump. Why, unless they are not really Independents?

Economy Getting Better (R/D/I) 76/11/32
Financially Doing Well (R/D/I) 89/68/76

Here, we see that Independents are doing better than Democrats, but don't think the economy is getting better. That tracks much closer to the Democrat answer, enhancing my belief that these Independents are really Democrats answering polls as Independents.

Approve of Mexico Policy (R/D/I) 89/4/32

I don't know why Independents would break so low on the Mexico policy unless they're really Democrats.

My conclusion is that this is a D+2 poll with Democrats hiding as Independents to give the 2020 polls the illusion of balance.
We're going to have to keep an eye out on the Independent poll results to see if this becomes a trend. Could it be the liberal response to anecdotal stories of Republicans lying to pollsters? Instead of stories of Republicans lying to pollsters that they will vote for Clinton, Democrats are going to lie to pollsters saying they are Independents and then answer as Democrats?

-PJ

Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.