Frank Luntz predicted Hillary Clinton would win on Election Day 2016. He admitted it publicly, took the credibility hit, and spent the better part of a decade trying to rebuild his reputation as a serious analyst. So when the same guy goes on CNN and tells Democrats their favorable midterm polling might be garbage, the moment has a certain weight to it.
The man who got it wrong is warning them they're about to get it wrong.
Luntz, a veteran pollster who teaches at West Point, appeared on CNN on October 18 and delivered a cold shower for every Democrat counting on a blue wave in November. His core message: Trump voters don't talk to pollsters, the current models aren't measuring turnout correctly, and the party betting its midterm strategy on favorable numbers is playing a dangerous game.
"We know that they don't respond to phone calls," Luntz said of Trump's base. "I got 2016 wrong. I believed in the exit polls. Republicans are less likely to participate."
That's not a Republican operative spinning. That's a pollster who lost credibility with conservatives precisely because he spent years criticizing Trump's rhetoric and distancing himself from the GOP. Luntz isn't cheerleading. He's reading the data and seeing the same gap that burned him a decade ago.
The numbers Democrats are clinging to look solid on paper. RealClear Polling averages show Democrats leading the Senate 53-47 and holding a 235-226 House majority. Comfortable margins. The kind of numbers that get cable news anchors excited.
But Luntz pointed out that the Democrat turnout advantage is only about two points. For context, the party out of power in a president's first midterm historically enjoys a massive edge. Two points isn't a wave. It's a rounding error.
A three-point polling miss — the same size error that plagued 2016 — would flip the Senate to 54-46 Republican and turn the House into a coin toss. The math is straightforward: if the polls are off by the same margin they've been off before, every "safe" Democrat seat in Texas, Iowa, Alaska, and Ohio becomes a dogfight.
"Donald Trump does have more influence on more races than any other Republican, even than Ronald Reagan," Luntz told CNN. "I've never seen a Republican as engaged and involved as this guy. And it matters."
The engagement question is the whole ballgame. Trump's coalition in 2024 pulled in Latino voters at levels that stunned both parties, shifted male union voters toward the GOP at historic rates, and gained ground with voters under 30. Those voters showed up when Trump was on the ballot. Whether they show up when he's not is the only question that matters — and it's the one the polls can't answer.
"If he turns out his people who still like him, still believe in the Trump message and the Trump agenda, we can have an upset," Luntz said.
Democrats could take that warning seriously. They could look at the structural turnout problem Luntz is describing and adjust their ground game. They could acknowledge that a two-point advantage in a midterm year where voter frustration over affordability and war is running hot isn't exactly a fortress.
Instead, they're doing what they did in 2016. They're reading the top-line numbers and planning the victory party.
Luntz admitted he's "not sure that they're doing it correctly" when it comes to measuring likely voter turnout. The man is essentially standing in front of the same building that burned down last time, pointing at smoke, and saying he smells something. The question isn't whether Luntz is right. The question is whether anyone learned anything from the last time he was wrong.
The Senate is 53-47. The House is 235-226. The turnout gap is two points. And the guy who blew the call in 2016 is on CNN saying the numbers don't add up.
Those are the facts. The election is in fifteen days.