Harry Enten is not a conservative, does not work for a conservative outlet, and has no incentive to tell Republicans anything encouraging. On CNN Wednesday, he went through the Midwest Senate map anyway and arrived somewhere his colleagues clearly didn’t enjoy.
His framing was the old proverb, updated. “Fool me once, shame on you. Fool me four times,” and you have to start asking whether it’s going to happen again.
Here’s the arithmetic behind it. In 2024, Republicans beat their September polling by nine points in Iowa, six in Ohio, and five in Michigan. “So, you look at this, and you go, you know what, maybe we should just hold on a second here because we’ve seen this dance before and it was not a dance that Democrats like when the votes were actually cast and counted,” Enten explained.
In 2018 — Trump’s first midterm, with Trump not on the ballot — they beat it by six in Iowa, six in Ohio, and eight in Michigan. Look across 2016, 2018, 2020, and 2024 and the GOP outran its September numbers by at least five points in all three states, every single time.
Now set that against today. Democrats have been posting good surveys out of Ohio, Michigan, and Iowa, and their leads in those states are smaller than the average size of the misses. Not smaller than the biggest miss — smaller than the routine one.
Apply the historical correction and the map rearranges itself. Michigan, Ohio, and Iowa go red. As Enten noted, Democrats could win Texas, Georgia, North Carolina, and Alaska — but that’s a sweep nobody seriously projects. If it were to happen, though, the Senate would still land at 50-50, with VP JD Vance holding the gavel and the tiebreaker.
Anyone want to put money on Texas electing freaky-deeky James Talarico?
CNN host John Berman raised the one objection worth raising: 2024 had Trump on the ballot, and 2026 doesn’t. Enten had already answered it.
In 2018, during Trump’s first midterm elections, the GOP exceeded its September polling numbers by six points in Iowa, six points in Ohio, and eight points in Michigan. When looking at the election years 2016, 2018, 2020, and 2024, Republicans consistently outperformed their September polls by at least five points in all three states each time.
Which brings us to what this actually means, because the wrong lesson is available and it’s seductive. Frank Luntz — another analyst with no affection for this president — told CNN on Monday that the Democratic turnout edge is running around two points. That’s remarkably thin for an out-party in a first midterm, when the opposition normally enjoys a wide enthusiasm advantage. And Luntz’s qualifier is the one that matters: the whole thing hinges on whether Trump’s voters show up when he isn’t on the ballot. They came out in force in 2016 and 2024. They stayed home in 2018 and 2022.
So the polls are probably understating Republicans in the Midwest again. That is not a prediction of victory. It’s a measurement of how much room there is for the people who have shown up before to show up once more.
Democrats have spent a year being told by the press that this is finished. They may spend November 3 learning what Enten just told their favorite network.
