When you’re a two-ply man on a one-ply budget, life gets messy. You empathize with Lady Macbeth’s hands.
Sure, you’re extra-careful. Maybe even methodical. (Everyone has their own technique.) But, alas, most of the time you still get [expletive] results.
Which is a perfect analogy for modern-day polling, because that entire industry is one-ply, too. Pollsters lack the budget to poll tens of thousands of people in million-plus communities, so they use guesswork, statistical projections, and sweeping assumptions about strangers.
Thirty years ago, when the internet was a novelty and everyone answered their landlines, cold calls had a very high success rate. Voters were willing (and often excited) to talk. A decade later, it took three calls to reach one respondent.
And today?
Check out this 2024 article from the Washington Post:
[N]obody answers their phone anymore.
Pew Research Center’s Courtney Kennedy is one of the nation’s most-cited polling methods experts. When she started in this business two decades ago, you had to call about three random numbers to get one to answer, a low response rate that already was sending tremors through the polling industry.
By last year, Pew had to hit up about 140 random numbers to get that one response. [emphasis added]
Garbage in, garbage out. When only one out of 140 people is willing to participate, there’s just no way it’ll produce reliable results.
Because you’ve filtered out the normies.
And instead, your data is top-heavy with lonely weirdos with nothing better to do than respond to political polls — often because they’re WAY more ideologically driven than the Average Joe. It’s probably why so many polls greatly exaggerated the strength of DSA candidates, projecting that Abdul El-Sayed would win the Democratic primary by 10 points (he won in a one-point squeaker) and Francesca Hong would win by 22 (she lost by half a point).
Assuming, of course, that the polls are real. Which isn’t always true.
But the weirdest, strangest, most utterly inexplicable question of the 2026 midterm polling is this: Why are the Democrats overperforming in red states — and BADLY underperforming in blue states?
This is supposed to be a blue wave election. Donald Trump is supposed to be the most unpopular president since Harry Truman. There have been countless news stories about the Democrats pulling jaw-dropping upsets in ruby-red states like Texas, Mississippi, Iowa, South Carolina, Montana, Kansas, and more.
Yet in blue states like Michigan, Minnesota, Maine, and New Hampshire, the Republican Senate candidates are all within striking distance — which is more consistent with a RED WAVE election!
Possible explanations: Candidate quality (or lack thereof) is the real story. Maybe the Democrats are overperforming in South Carolina because the Republican Senate candidate is unusually weak — and maybe the Republicans are overperforming in Michigan for a similar reason.
I’m sure it’s a factor. Quality matters.
When voters can choose between sane and Cypress Hill, they usually side with sane.
And it’s also true that each state is unique. Maine voters are more concerned about the Canadian border; Texans care more about Mexico’s. Somali fraud is a huge issue in Minnesota; less so in Kansas. What plays in Peoria might bomb in Biloxi.
Perhaps drawing national conclusions from state-specific elections is a fool’s errand.
But I doubt it. Because millions of Americans will be voting in every state. (More than 14 million already have.) The two national political parties are battling for control of Capitol Hill — and the contrast between the two has never been more glaring.
To paraphrase Barry Goldwater, this isn’t an “echo” election. It’s a “choice” election.
And it’s a helluva choice: MAGA versus the DSA. The battle lines are crystal clear.
Furthermore, the biggest issues of the 2026 midterms are all national issues: affordability, inflation, gas prices, AI/data centers, immigration, the Iran War, Israel/Gaza, trans, and (of course) President Trump. Because the sitting president is ALWAYS on the ballot in midterm elections.
If this isn’t a national election, I don’t know what is!
We all know about the historic headwinds: The party in power almost always takes it on the chin. That should be twice as true for a president with the lowest approval ratings since Harry Truman.
So if these numbers are right, the Democrats ought to be dominating.
Yet that’s simply not happening in blue states. DSA-aligned candidates are struggling. There’s a very real risk the Dems will LOSE Senate seats in Michigan, Minnesota, and New Hampshire! And even though Kamala Harris won Maine by seven points, Republican incumbent Susan Collins is still within the margin of error against her left-wing opponent.
Something is undermining their coalition.
It’s a group that was once loyal, but is now wandering in the wilderness. A group that’s bleeding Democratic support — because the party is no longer their home.
Which is making mincemeat of polling predictions and predictive models: These voters aren’t behaving like WE think they’re supposed to!
I’m talking about black voters.
Sixteen years ago, black people voted 95% Democrat. By 2020, 77% of black adults identified as Democrats. And in 2023, the number fell to just 66%.
A generational and structural shift is decoupling Black identity from Democratic Party loyalty, transforming a once-reliable voting bloc into a cohort of "political free agents" that the GOP is uniquely positioned to exploit.
[…]
By the numbers: Republican identification among Black Americans climbed into the mid-to-high teens in recent Gallup data.
- Trump surged to an average near 20% approval among Black voters in the first quarter of 2025, nearly double his approval at the same point in his first term, according to Gallup.
- That was driven largely by Black men and those moving toward Republican affiliation.
Also relevant is religious affiliation: 73% of black people are Christian, compared to 62% of white people. Black people are also more likely than white people to attend church and engage in “religious practices.”
Just 39% of white people called the Bible “very important” — compared to 70% of black people. Black Americans are (roughly) twice as likely to read the Bible daily and pray daily.
Which might be why black voters are rejecting the anti-Christian, pro-Islam positioning of the DSA.
If Abdul El-Sayed had 2008-level black support, he’d win Michigan’s Senate seat in a cakewalk. The Republicans wouldn’t have a chance. This election would already be over.
But in the Michigan primary, black voters rejected El-Sayed, voting en masse for his (white) opponent, Haley Stevens.
[Haley Stevens] led El-Sayed by 8 percentage points in districts where more than 15 percent of the population is Black. Polling in July showed that just 21 percent of Black voters backed El-Sayed while 67 percent supported the House Democrat.
A Michigan Education Survey also showed El-Sayed struggling with Black voters. Stevens notched 69 percent from the bloc; El-Sayed garnered 37 percent support.
Stevens made no secret of her support for Israel. That’s probably one of the reasons why she did so well with black voters.
[A]ccording to the Fox News poll, African Americans were the racial group most likely to say the U.S. isn't supportive enough of Israel when asked the question along with White and Hispanic voters. That number was 25% compared with 19% for Hispanic voters and 15% for White voters.
African American voters were also the least likely of the three groups to agree that America was too supportive of Israel, with 40% disagreeing as compared to White voters at 44% and Hispanic voters at 56%.
The stat stands sharply against the ubiquitous notion that African Americans — owing to supposed shared experiences of injustice and oppression — organically align with the Palestinian experience.
Meanwhile, the DSA movement doesn’t exactly mirror the Rainbow Coalition: According to a 2021 survey, 85% of DSA members are white, and 80% have college degrees. It’s whiter than mayonnaise on a polar bear.
As the DSA wrests control of the Democratic Party, black voters are walking away.
This has flipped the dynamics of the 2026 midterms. It’s why the pollsters’ models are garbage: They’re relying on voting projections that no longer apply.
Black voters aren’t gonna vote 90% for El-Sayed. It ain’t happening. Millions will vote for his Republican opponent, Mike Rogers.
Many won’t vote at all.
And until a new voting pattern emerges that can be tracked, predicted, and quantified in polling models, none of these polls mean anything. They don’t have the budget to interview thousands of people, and their base model is simply wrong. We know this because they’ve been wrong consistently and repeatedly.
"This is our modern-day 'Dewey Defeats Truman'" -@ChuckTodd #ElectionNight https://t.co/sdij0Iaa1e
— TODAY (@TODAYshow) November 9, 2016
It’s now 2026. Not 2016, 2008, or 1992. If pollsters botched elections when people still answered phones — and when large demos still voted predictably — what do you think will happen when they don’t?
Because, very clearly, black people are voting differently. (And they’re not the only ones.) It’s a brave new world. Polling data hasn’t caught up.
Yet we’re still extrapolating group decisions on any weirdo who’s willing to take a poll that 139 other people have already rejected!
So OF COURSE these polling results are full of [expletive]. It’s one-ply technology in a two-ply world. Flush it away and start over.
And don’t look now, but black voters are the wildcard of the 2026 midterms. If El-Sayed and his DSA-aligned brethren go down in defeat, it’ll be because of black voters. They’re our wall between sanity and Cypress Hill.
God bless ‘em. They just might save America.