It's a Numbers Game: DEMOCRAT TSUNAMI?! Ryan Girdusky RIPS APART Final Midterm Polls!
The Clay Travis and Buck Sexton Show
November 2, 2026
With the 2026 midterm elections just hours away, Ryan Girdusky tears into shocking New York Times and AtlasIntel polls predicting a massive Democratic tsunami, exposing questionable voter models, Republican response rates, and numbers he says simply don't add up.
Speakers Ryan Girdusky
TopicsDaily NewsNewsNews Commentary
Ryan Girdusky (0:00)
Welcome back to A Numbers Game with Ryan Girdusky. Thank you guys for being here. We are one day away from the 2026 midterm elections. If you haven't made a plan to get out and vote, make sure you go and get out and vote tomorrow. Every Sunday, I read the physical copy of The New York Times, basically cover to cover, because I'm looking for stories to say, what are we going to talk about this week? What am I going to work with? What kind of hidden information am I going to find on, you know, page D37 paragraph 6 that will unlock information for my audience? And this Sunday, I did not even finish the paper, because we had breaking news story after breaking news story after breaking news story. Everything from Haley Stevens' endorsement of Mike Rogers, Roger Marshall's insane video out of Kansas. New census data came out that literally no one is even bothered covering. But before we get through all that, we got to talk with the latest polls, because the New York Times and AtlasIntel released their last poll before the midterm election. The poll finds that Democrat El-Sayed is beating Mike Rogers in Michigan by 11 points. Democrat James Tallarico is beating Ken Paxton by 7 points in Texas. Democrat Troy Jackson is beating Susan Collins in Maine by 6 points. Democrat Sherrod Brown is beating Republican Johnny Husted by 5 points. Democrat Adam Hamilton is beating Republican Mike Rogers in Kansas by 5 points. Democrat Mario Patola is beating Republican Dan Sullivan in Alaska by 2 points. Democrat Josh Turek is beating Republican Ashley Henson by a point in Iowa. A clean sweep for Democrats. They are predicting not a Democrat wave, not a Democrat megastorm, but a Democrat tsunami. The poll also finds that Democrats are winning Governor's Mansion in Maine by 15 points, Alaska by 8 points, Iowa by 8 points, Kansas by 1 point, Michigan by 13 points, Ohio by 6 points, and Texas by 1 point. It is wall to wall. Democrat wins is what the New York Times is saying in their last poll. And I know what you're going to say, Ryan, it's the New York Times. They hate Trump. They hate Republicans. They're trying to convince us not to show up.
OK, I understand what you're saying. However, making matters worse is that AtlasIntel, who predicted that Trump would win the 2024 election, also has James Talerico beating Ken Pax in by 6 points in their final poll. They have Governor Greg Abbott losing in Texas by a point matching nearly identical to the New York Times poll. And they have the generic ballot at D plus 14, which would make it the worst year for Republicans since the Watergate scandal of 1974
I have to just say one thing that really drives you crazy, but AtlasIntel, they like to say, oh, we are the most accurate pollster of the 2024 election, which is true. However, the following year in 2025, they were one of the least accurate pollsters. They teamed up, I think, with Bloomberg News, by the way, in December of 2024, and things started taking a shift. They got the 2025 Governor's race in Virginia, and the Attorney General's race completely wrong. They also got the New York Mayor's race totally wrong. I think they were the only poll saying that Andrew Cuomo was going to beat Mondami by double digits or near double digits after all the votes were counted. So at this point, the 2024 election being correct, a broken clock, my friend. That's what I, at this point, have to say about that, Senator, considering how badly you were the year after. So what do I think about these polls? What do I think about the New York Times and the Senate poll saying, it's over, Republicans are going to lose in this massive landslide loss? Well, first, the New York Times' Nate Cohen admitted two things that I think are very important when analyzing these polls. First, across all states, registered white Democrats were about 30% more likely to take the poll than registered white Republicans, which was worse than in 2020, the year that we had a massive, massive polling error. I...
That's not unique to the New York Times. Multiple pollsters have been saying, hey, we can't get white Republicans on the phone. We can't get white Republicans to respond to our texts. We're kind of guessing here, but here's the problem.
If you're not getting the white Republicans, how do you think you're getting the white independents who typically vote Republican? Like, if no one's willing to trust you, how do you think your numbers for the independents are even somewhat in the same category as being accurate? And so what they do is they've decided to re-weight all these polls. That's their answer. Their answer to Republicans not answering the polls, hey, we're going to re-weight them. So they also provided the unweighted data. The unweighted data said that L. Saeed was not going to win by 11 He was going to win by 16, which would have been or would be one of the biggest landslides in Michigan US Senate history, recent history. They said Patola was up six, not two in the unweighted polling. They have Kansas going Democrat by 10, Ohio by 11, Texas by 9 in the unweighted polls. I want to scream at these pollsters.
Read the other 24 minutes of this episode here.
$15 gets you this episode and 99 more. Credits never expire. After checkout, one link brings you back here to read it.
Not what you expected? Full refund within 14 days.
Thousands of transcripts fetched by people building searchable podcast archives
Fetch transcripts with code
The same credits work from a script or an AI agent. The demo key returns a sample episode in full, no card needed:
curl -H "x-api-key: pt_demo" \
https://spoken.md/transcripts/1000651996090Markdown with the speakers named, for your notes, your knowledge base, or anything that makes HTTP calls.
From $0.10 per transcript. No subscription. Credits never expire. Prices exclude VAT, added at checkout for EU customers. Not what you expected? Email us within 14 days with 20 or fewer credits used and we refund the pack in full.
Using your own key:
curl -H "x-api-key: YOUR_KEY" \
https://spoken.md/transcripts/1000799421143