Widespread 😱 🙀 💩

It’s that time of year again: time to panic over polls showing your guy in trouble and the bad guy ascendant. Like the swallows at Capistrano, pollsters and pundits return to this pile of shit (okay, like flies, not swallows) every four years. Always within two years of the last presidential election.

And it’s always a stinking pile of crap because nobody is paying attention this far in advance. Not, at any rate, the vast majority of people who will vote in the next presidential election. But it keeps pundits employed. Remember that, because that’s all these polls are worth.

The analysis of failure changes, but the fundamental problem remains. Nobody cares this far out. And polls are bullshit anyway.

They had Republicans winning both houses of Congress at 57% based on polling averages; their most likely scenario. What actually happened, Democrats take the Senate and Republicans take the house, had a 27% likelihood based on their polling averages. 

And then there were the other elections the where the polls completely failed to predict what would happen. 

In Kansas, the much-hyped Coefficient poll said voters would vote to remove the constitutional right to an abortion in that state 47% to 43%. What actually happened: Voters rejected the measure 59% to 41%, roundly supporting keeping the right to have an abortion in place.

And all the polls, right down to Election Day, said Hillary would win in 2016. At least, nobody said Trump would.

The other, darker side, of polling? Trump uses them, in his stupidity and ignorance (it is neither feral nor low cunning; that’s 19th century nonsense we should have discarded by now), to claim he wins, when he loses. Because: polling. And we examine every poll like fresh chicken entrails in order to divine the future. The only problem is, MAGA takes us seriously, and doesn’t know we know it’s just chicken guts, not auguries. At least that’s what we tell ourselves, when we analyze the polls for the predictions we want.

Polls are bullshit. That is all you should know, and all you need to know. They can never be verified; they can only be justified. Which is how you polish a turd. Because: garbage in, garbage out;

First, pollsters are still primarily using land line telephones, not cell phones, when they conduct phone call polling. Most Millennial and Gen Z voters, who vote heavily Democratic, don't even have land line phones, so they aren't being reached by pollsters at the rate of older more conservative voters. 

Second, participation rates in polls are higher among older voters. Scientific polls don't take into account human factors, such as the introversion of younger Americans in their samples. Younger Americans screen their calls and don't answer calls from numbers they don't know. Younger Americans don't spend that much time on phone calls; they prefer to text. Most important, younger Americans have school and jobs and young families: they don't have time to spend an hour responding to a phone poll or taking a half-day off to appear in an in-person focus group. 

But what might be the most telling factor about why polls in the Trump era have been so wrong is that polls don't adequately capture new voters. 

Most scientific polls will rely on samples of frequent or reliable voters. In order to fit into those classifications, voters will have to have voted in at least one and typically in three previous election cycles. If a college student for example, who has been politically activated by the overturn of Roe v. Wade and is a guaranteed Democratic voter in the next election, has never voted before they will be missed by polling samplers. As Kansas reveals, the most recent elections in the U.S. have seen a ton of new voters that simply weren't polled. 

Gen Z proved to be the key factor in the 2022 mid-term election. They voted 65% for Democrats. The Gen Z share of Democratic votes was so large they essentially cancelled out Baby Boomer votes for Republicans, despite a higher turnout by Baby Boomers. Gen Z was the wall that stopped the red wave.

I’m not a Millennial, but I don’t have a landline (so much for the accuracy of that vague and glittering generalization). My cellphone won’t ring if you aren’t in my phone book. And yet I have voted, as regular as clockwork, since 1973. I’m an “older voter,” too. But I screened calls on my landline back when I had it. It has never struck me as my civic duty to answer a pollster’s questions.

And again: how is the poll “scientific” if it cannot be verified? Or worse, is not verified? Isn’t that the standard used to disqualify other speculative practices, like prayer?

I believe in prayer, but not as a practice that accesses cosmic powers. I don’t see how polling predicts the future; not even the immediate future just after the poll is taken. I certainly don’t see polling as even vaguely “scientific.”

And if you’re still not comforted, consider this:

"Donald Trump’s false claims that the 2020 election was rigged may come back to bite him in 2024," Psaki wrote on Saturday, pointing to a new USA Today/Suffolk University poll showing that "the former president’s rhetoric may be persuading some would-be Trump voters to sit out the 2024 presidential election." 

"The poll found that unlikely voters (that is, those who are eligible to vote but aren’t planning to cast a ballot) actually support Trump over President Joe Biden by nearly 20 percentage points. But many also don’t believe their vote will actually count, so why bother," according to the article.

Whether or not the poll numbers are right, anything that dissuades Trump voters from voting is fine with me; and a more likely cause of depressed voter turnout than bad polling. After all, all those Gen Z voters who keep voting don’t seem to give a wet snap for the polls.