Once again, the pollsters got it wrong. They undercounted Donald Trump's support as they did in 2016 and 2020.

At least this time, they came a little closer. In 2016, the pollsters had Hillary Clinton ahead by as much as seven points on election day. She won the popular vote by less than three points and lost the electoral college.

In 2020, Biden was up by seven points in most polls on election day. He won by less than four. Even the state polls ended up being wrong by more than the margin of error.

In 2024, the polls were all over the map but it appears that the pollsters missed again by more than the margin of error in several states. The national polls, however, showed Trump or Harris with a one or two-point lead. It's going to be a tight one, everyone was saying. Here at PJ Media, we prepared for a long night and a day or two after before getting a result. 

It wasn't that close. But the final weekend saw some real drama out of Iowa when a shock Des Moines Register poll showed Kamala Harris leading Trump by three points in a state Trump carried by eight in 2020.

Ann Selzer has been fashioning and conducting polls for the Des Moines Register since 1987. She's hands down one of the most respected pollsters in America. Selzer  is uniformly “the great” or “most respected” or "Iowa’s polling queen.” 

When the Des Moines Register released its final poll before the 2024 election for Iowa, Selzer's work set off a political firestorm. Democrats, reeling from Joe Biden's comment that Trump supporters were "garbage," leaped on the poll like starving dogs. The New York Times led the way: "Sometimes outliers are early indicators, capturing something other polls just haven’t measured yet. Other times, they’re a fluke."

You can guess which category this poll falls into.

Trump ended up winning Iowa by a landslide, 56%-42%. On Tuesday night, Selzer, a grandmotherly woman who looks like she would offer you milk and cookies if you visited, had to face the music about a poll that was 17 points off.

"Tonight, I’m of course thinking about how we got where we are," Selzer, president of Selzer & Co., which conducts the Iowa Poll, said in a statement.

"The poll findings we produced for The Des Moines Register and Mediacom did not match what the Iowa electorate ultimately decided in the voting booth today. I’ll be reviewing data from multiple sources with hopes of learning why that happened. And, I welcome what that process might teach me."

So, what happened Grandma?

Des Moines Register:

Although Selzer said she planned to do a deeper look into the data, there were a few things she was eyeing Tuesday night.

"Technically, the poll had some 'give' in that neither candidate reached 50%," she said. "So, the people who said they had voted/would vote for Robert F. Kennedy, Jr. could easily have switched to Donald Trump. The late deciders could have opted for Trump in the final days of the campaign after interviewing was complete. The people who had already voted but opted not to tell our interviewers for whom they voted could have given Trump an edge."

The Iowa Poll showed Kennedy, who had ended his presidential bid but was still on the ballot, got 3% of the Iowa vote. Fewer than 1% said they would vote for Libertarian presidential candidate Chase Oliver, 1% said they would vote for someone else, 3% weren't sure and 2% didn't want to say for whom they already cast a ballot.  

Yes, but 17 points off target?

"Maybe I can gain clarity on that 9% and an underlying disposition toward the presidential race," Selzer said.

Related: The 'Normie' Realignment

We're seeing a phenomenon in Trump supporters either lying to pollsters or not responding. It's probably a lot more of the latter, since pollsters have several methods to determine if a subject is telling the truth. 

But what really troubles pollsters is that, after eight years of fiddling with the methodology, they still can't get a good handle on who is backing the former president. Trump broke polling and it's not clear when or if anyone can fix it.