Totem polls

Sometimes a homophone is more than a homophone.

I’ll be the first to admit I know very little about political polling. And what I do know, I don’t trust. Every presidential election season, the closer we move toward election day, the more the polls seem to move — with almost precise regularity — in favor of the Democratic Party candidate for the presidency, where they perform as emblems of the perceived popular will.

I used to believe that this dubious movement was merely an attempt by a left-wing activist media to create or maintain a preference cascade — the idea being that, because we as voters like to back a winner, were the legacy media and their favorite polling arms able to convince us that their “fortified” candidate was the inevitable victor, they could sway a not insubstantial number of us to move in their favor. Additionally, such polls aim to dispirit potential opposition voters, the goal being to keep those voters at home, because a trip to the polling place is an exercise in futility, and nobody wants to waste time fighting the inevitable.

To accomplish these goals, captured media and their pollsters would selectively present polling data so that the results would show exactly what they wanted shown, even if this meant pollsters would have to bracket a number of important internals that belied their ostensible conclusions.

Now, however, I believe something even more sinister is at play. Namely, that the media and many pollsters not only wish to create a preference cascade, but that they aim to maintain polling margins that will, in a pinch, create plausible deniability should certain “unconstitutionally realized’ votes be necessary to add to the tally in order to reach the numbers needed to declare victory. By creating the illusion of a significant Democrat advantage entering election night — which is nearly always the case, and has been since at least 2000 — progressive partisans are able to delegitimate GOP victories (in the worst case), or find and add the requisite number of late-arriving votes (in the best case) to pull out narrow state victories.

— and if and when they do this, they can, having done the work beforehand, point back to pre-election polls to suggest nothing untoward has transpired, that in fact the polling indicated all along that the D candidate was the likely winner, and so what turned out to be a narrow Democrat victory really would have been much larger had not racist Republicans, e.g., kept blacks from voting by cruelly insisting voters must have the legal right to cast a ballot.

Oh. And also? Exit polling is notoriously unreliable, so it tells you nothing. Except when it shows a clear D preference. In which case, exit polling is sacrosanct, so cope harder, conspiracy theorist!

The whole thing is a scam — so much so that certain niche pollsters are able to predict when and how the more established pollsters, whose job it seems to be to “frame” election preferences, will begin moving the poll numbers in a transparent attempt to manipulate the electorate.

Case in point: Sunday, NBC came out with its latest national polling, which shows Kamala Harris with a 5 point lead over Donald Trump.

Meanwhile, certain swing state polls have also dropped, showing Harris with a similar lead of 5 points.

And yet with the top issues in the country being a faltering economy and a crisis of illegal immigration — at least among Republicans and Independents — it doesn’t seem quite right, somehow, that Harris, part of the ticket already in power, would represent the preferred change agent the electorate is looking for in order to address its most pressing concerns. So how to account for such polling?

I’ve tried to explain the why — to the limited extent I can — but now let’s look at the how. First, take the NBC poll. NBC surveyed 1000 registered voters nationally, and from that sample was able to garner nearly as many headlines hyping Harris’s “huge momentum swing” as voters actually surveyed — with scores of outlets both national and local parroting the same “conclusion.” But with a margin of error of 3.1 percent and NBC’s history of oversampling Democrat voters, how real are the numbers they are using to gin up this “tremendous” and “unprecedented” Harris momentum?

Most experienced pollsters will tell you that these are essentially garbage polls with no concrete insights into statewide races or electoral college results. Which means they’re produced and released merely as a means to garner the headlines themselves — headlines which, shockingly, are repeated by like-minded outlets and happily benefit their preferred candidate. Kismet! Why, it’s almost as if they’re hoping to create a preference cascade and an atmosphere of inevitability with an otherwise largely worthless poll! Imagine!

But let’s look more closely at a swing state poll, this one released by the Minnesota Star Tribune. With Harris at +5, the poll seems to align with the NBC nationwide poll quite neatly. That is, until you look more closely at how it was conducted, who was polled, and how its outcome is framed. For instance, PollFair, which describes its mission as “ensuring fair and accurate political polls by reweighting polls based on historical exit poll data,” has Minnesota as a statistical tie, which would seem to give Trump — who notoriously carries a hidden voter contingent who fail to show up in pre-election polling — an advantage; and indeed, Harris is polling well behind Biden in 2020, not just in Minnesota, but in every swing state. Which is something you seldom see reflected in newer polling.

I post all of this not to make the argument that Trump is going to win the national popular vote, or cruise to an easy electoral college victory in November. For the sake of my family, I certainly hope that’s the case, sure, but I’m no election expert, and I’ve never claimed to be one. Moreover, I believe if the election is close enough in several swing states, the Ds will almost certainly attempt to steal it — and this sobering distrust of our system leaves me less than sanguine about the ultimate results.

Still, I’ve been heartened of late by niche pollsters who have proven accurate over the last few election cycles — and who have really plumbed the data in polling internals to shine a light on just how the media and the legacy polling industry works to manipulate election outcomes.

One bit of push back I’ve been encountering when I question some of the current legacy polling showing Harris a clear favorite, though, comes courtesy of readers pointing me — often aggressively and with sheer disdain for my obtuseness — to the betting markets. Polling can get it wrong, their argument goes, but where money is involved, the bookies are always right. And the gamblers have Harris the current favorite. So it’s effectively over. “Cope,” Jeff. You dumb chump. You simply don’t get how the world works.

The above post was offered today in response to a reply I posted a couple weeks back answering this indictment against me. In that reply, I linked a video in which the presenter explained how the betting markets can be manipulated. Having no expertise in the matter, I offered up someone with expertise to make my case. Dismissed here as a “rando,” whose “theory” doesn’t hold up.

— Though if you’re thinking, “how precisely does pointing to the current betting odds disprove that the betting markets are often manipulated work, exactly, as substantive rebuttal,” I’m right there with you.

Still, I dug a bit further. For instance, Polymarket’s odds, which has Harris as a current 3 point favorite — though now showing a movement toward Trump, even after the NBC poll dropped — are determined mostly by overseas money, representing a bettor pool that may not have access to the kind of information that, say, certain domestic pollsters do.

And in fact, the timing couldn’t have been more perfect, because just yesterday, pollster Rich Baris and attorney Robert Barnes, as part of their series “What Are The Odds,” dropped a lengthy video explaining — among other things — just how betting markets work, how odds are set, and which odds are currently way out of pocket, in their estimation, based on the polling data they’re combing through. They point to other markets they believe are more accurate and explain why they believe so.

My critic had no time for this nonsense, which I passed along to him without comment. Instead he berated me. I must accept fate. Harris has this election locked up.

Well, no. I choose hope. And you aren’t the boss of me.

I highly recommend the episode (click the pic to listen), which digs deeply into poll internals from swing states and presents an entirely different — and far more rosy, if you’re a Trump supporter — polling picture than does, say, NBC, or Polymarket, or Nate Silver’s predictably malleable outfit, FiveThirtyEight, one of the key architects of polling as preference driver.

Among the key internals Baris and Barnes examine in the episode, for instance, are the margins for key demographic voters relied upon by Democrats, which they point out are down significantly from Biden’s 2020 totals, leaving Harris vulnerable, whether it be among young voters, Hispanics, traditionally non-engaged blacks, et al. — realities that aren’t being factored into a lot of traditional predictive polling data built around 2020 results.

Indeed, internal polling from the campaigns themselves are showing results far different from those most polling outfits are offering publicly. In Pennsylvania, for instance, Trump internal polling shows him up 3. Harris’s internal polling is even worse, showing her trailing Trump by 4. Yet the Real Clear Politics polling average for Pennsylvania shows Harris up by 0.6.

So who is most accurate?

Polling, it seems, is as much art as it is science. And for many legacy polling operations — at least, for those favored by legacy media and Democrat activists — the art it most closely approaches these days is that of writing fiction. For that reason, you might wish to choose pollsters based on recent performance. It may not lower your blood pressure, but at least you’ll have a better chance of not being played.

I don’t know what will happen in a little over 40 days. But I do know that what we’re being sold by most of the polling we have foisted upon us in the mainstream press is framed for purposes other than that of providing a clear snapshot of where the country is politically.

That’s unfortunate. But it’s not unexpected. Frankly, we deserve — and should demand — better.

Not that anyone will listen. But at least we’d have the satisfaction of showing our disgust for how we’re treated by those who seek to control us rather than inform us.

****

Buy me a cup of coffee? Lunch? A sports car?