If you’ve spent any time refreshing election maps or doom-scrolling through political Twitter over the last twenty years, you’ve heard the name. J. Ann Selzer. To some, she was the "Queen of Polling." To others, she was the only person in the Midwest who actually knew what was going to happen before it happened. For decades, her Iowa Poll was the one everyone waited for—the one that could move markets and make seasoned campaign managers sweat through their suits.
But then came November 2024.
The shockwave was almost physical. Just days before the presidential election, Selzer released a poll showing Kamala Harris leading Donald Trump by 3 points in Iowa. In a state Trump had won by 8 points in 2020 and 9 points in 2016, this wasn't just an outlier. It was a statistical explosion. When the dust settled on election night and Trump carried the state by 13 points, the "gold standard" didn't just miss; it missed by 16 points.
Shortly after, Selzer announced she was done with election polling. But to understand why that exit was so dramatic, you have to look at how she became the most feared and respected name in the business in the first place.
The Woman Behind the Iowa Poll
Ann Selzer didn't just fall into polling. She built it. Born in 1956 and raised in Topeka, Kansas, she initially thought she’d be a doctor. She was a pre-med student at the University of Kansas before realizing that drama and speech were much more her speed. She eventually landed at the University of Iowa, where she earned a PhD in political communication.
In 1987, she started working at the Des Moines Register. That’s where the legend began. By 1996, she’d founded her own firm, Selzer & Company, but she kept running the Register’s Iowa Poll.
✨ Don't miss: Who Has Trump Pardoned So Far: What Really Happened with the 47th President's List
For a long time, she was basically a prophet.
Think back to 2008. Everyone "knew" Hillary Clinton was going to win the Iowa Caucuses. Every poll said so. Except Selzer's. Her data showed a massive surge for a relatively unknown senator named Barack Obama. People thought she was crazy. She wasn't. Obama won, and Selzer became a household name for anyone who cared about power.
Why Her Method Was Different (and Why It "Blew Up")
Most pollsters today use a lot of "weighting." Basically, if they don't get enough young people or enough rural voters on the phone, they use math to boost those voices to match what they think the electorate looks like.
Selzer hated that.
She was a purist. Her philosophy was simple: "I assumed nothing. My data told me." She used random digit dialing and spoke to whoever picked up. She didn't try to force the data to look like the 2020 results or the 2016 results. If her poll showed a weird result, she published it anyway. This "no-herding" approach gave her an A+ rating from Nate Silver’s FiveThirtyEight for years.
🔗 Read more: Why the 2013 Moore Oklahoma Tornado Changed Everything We Knew About Survival
Then came the 2024 miss.
What happened? Some experts think her sample just happened to catch a weird "pocket" of voters. Others suggest that the "shy Trump voter" or the "couch vote" (people who don't usually vote but showed up for Trump) just weren't being captured by traditional phone calls anymore. Selzer herself was humble about it. She wrote in a Des Moines Register op-ed that "science has a way of periodically humbling the scientist."
The Retirement and the Lawsuits
The timing of her retirement felt like a movie ending, but the reality was a bit more bureaucratic. Selzer clarified that she had actually told the Register a year earlier that she wouldn't renew her contract after 2024. She wanted to move on to "other ventures."
Still, the exit wasn't exactly quiet.
In early 2025, legal drama followed. Donald Trump filed a lawsuit against the Des Moines Register and Selzer & Company. The claim? That the poll was a "fraudulent" attempt to deceive donors and voters. Most legal experts see this as a massive uphill battle for the Trump team because of First Amendment protections for the press, but it turned what should have been a quiet retirement into a courtroom spectacle.
💡 You might also like: Ethics in the News: What Most People Get Wrong
Ann Selzer spent 30 years being right when everyone else was wrong. In the end, being wrong once—at the highest possible stakes—is what people might remember most. It’s a tough way to go out for someone who dedicated her life to the "science of estimation."
What You Can Learn from the Selzer Era
If you’re looking at polls in the future, don't just look at the headline number.
- Check the Methodology: Does the pollster use "weighting" to match past elections, or are they a "minimalist" like Selzer? Both have risks.
- Look for Outliers: Selzer was famous for not "herding" (adjusting her results to match other polls). Outliers can be a sign of a massive shift, or they can just be a "bad sample."
- The Human Factor: No matter how good the math is, humans have to answer the phone. If a certain type of person stops answering, the poll stops working.
The era of the "God-tier" pollster might be over. As the way we communicate changes, the tools we use to measure what we're thinking are struggling to keep up. Ann Selzer’s career is a masterclass in how to build a reputation—and a cautionary tale about how quickly the world can move under your feet.
For those interested in the future of public opinion, the next step is to watch how the Des Moines Register "reimagines" the Iowa Poll. They are currently looking for new ways to capture the "Iowan voice" without relying solely on the traditional phone-call methods that defined Selzer's reign. Keep an eye on the 2026 midterm cycles to see which new firms try to claim her vacant throne.