Political Trade Secrets

🚨 NEW EPISODE: Who Wins the 2026 Midterms? A Pollster & Elections Forecaster Compare Notes Join @DustinOlson and Patrick Allocco (@psallocco) of the @ZoosePolitical Index for their 2026 midterm outlook, a closer look at polling and forecasts, and the stories behind Patrick’s career—from campaign computers in the 80s to Michael Jackson and an Angola hostage situation. The episode explores: • The Senate map, candidate quality, and Patrick’s House forecast • Pollster track records, public polling incentives, and AI-assisted analysis • Campaign computing in 1985, Michael Jackson and the Angola hostage story Watch the full episode and follow @PolTradeSecrets for more.

Yesterday’s Israeli election did not produce a clear governing majority, BUT it did produce a clear media polling miss. As Israeli political analyst @AmitSegal put it, “this is a Yom Kippur (Day of Atonement) for most pollsters." I agree, though I’d call them the “drive-by media pollsters.” I joined anchor @ShannonCake on @NEWSMAX’s News Now, alongside @Rabbi_Barclay, to explain why Likud and Netanyahu’s bloc performed much better than Israel’s media predicted. The problem looks familiar: Israeli media polls, like so many in the United States, appear to have systematically undercounted key groups, especially right-leaning voters and Arab voters. And when you rely on cheap methods like text-only surveys or online panels—as so many media polls do—how many ultra-Orthodox voters who eschew smartphones are you ever going to reach? And for an even harder challenge, how many Arab Bedouin voters without an address or electricity can you reach? Luckily, that’s one American pollsters don’t have to worry about! Bottom Line: It appears Trump’s late endorsement DID make a difference. More importantly, Israelis know what's at stake. Security and victory remained paramount—and turnout was the highest since the 1990s. Thanks to Shannon, Michael Barclay, and the NEWSMAX News Now team for the conversation. Get a Full Briefing on this and RSVP to Friday’s Noon EDT Midterm Strategy Session at PollingClub.com

🚨 NEW EPISODE: Who Wins the 2026 Midterms? A Pollster & Elections Forecaster Compare Notes Join @DustinOlson and Patrick Allocco (@psallocco) of the @ZoosePolitical Index for their 2026 midterm outlook, a closer look at polling and forecasts, and the stories behind Patrick’s career—from campaign computers in the 80s to Michael Jackson and an Angola hostage situation. The episode explores: • The Senate map, candidate quality, and Patrick’s House forecast • Pollster track records, public polling incentives, and AI-assisted analysis • Campaign computing in 1985, Michael Jackson and the Angola hostage story Watch the full episode and follow @PolTradeSecrets for more.

It doesn’t matter. Whether Muslim, Mormon, or a Moonie, Abdul El-Sayed’s faith is not the issue. His own signature is. I joined anchor Cary Dunst on @NTDNews Evening News to explain the question reporters should be asking El-Sayed—and keep failing to ask. Why did El-Sayed sign a 2012 statement endorsing Egypt’s Muslim Brotherhood president and calling on him to “purge the media and the police as soon as possible”? And why has he never directly explained or denounced it? In 2011, El-Sayed pushed back against concerns that the Muslim Brotherhood was gaining power, dismissing those fears as part of an “orientalist narrative” in op-eds for The Guardian and HuffPost. Forget guilt by association—even though there is A LOT of that. Hold El-Sayed accountable for his own words, actions, and political judgment. Thanks to Cary and the @NTDNews Evening News team for the conversation. > Watch the full segment below—don’t forget to like, share, and subscribe to my X for more teapawt.org/DustinOlson/status/211…

There’s an old saying in politics: You hunt where the ducks are. I joined hosts @SharlaMcBride and @marc_lotter on @NEWSMAX’s Wake Up America to shed light on Michigan’s tale of two rallies. Trump went to turn out his base in a part of the state he won with 66%. Meanwhile, Obama was in Detroit because, as our research shows, key parts of the Democratic coalition have real problems with Abdul El-Sayed. Our research found a 17-point candidate-specific drop for El-Sayed among Jewish Democrats, with similar warning signs among churchgoing Black voters. And across the Senate map, any potential blue wave seems to have crested a few weeks ago—and may not have the same energy when it hits the beach, as Republicans appear to be turning out. Thanks to fellow guest @MaryWalterRadio, Sharla, Marc, and the entire Wake Up America team at NEWSMAX.

Can you trust the poll behind the headline? Here are three simple tests. I joined anchor Cary Dunst on @NTDNews Evening News to unpack the crazy Wedgewood fake polling firm and similar revelations, and to shed light on how reporters—and the public—can separate credible polling from bad polling. 1. Three clicks or you’re out. The methodology, full toplines, crosstabs, sponsor, field dates, interviewing methods, and the percentage collected through each mode should be easy to find–-in three clicks or less from the headline. If they hide the details or make them hard to find, throw it out. 2. Find the #1 question: “Who did you vote for in the last presidential election?” That is one of the most important tools pollsters have developed for finding hidden Trump voters, shy Trump voters, working-class Republicans, and other working-class voters who are less likely to answer surveys. If they didn’t ask it, throw it out. If they asked it, but their sample describes a fantasy version of the last election, throw it out. 3. Examine how they collected the interviews. If a survey relies entirely on one cheap method—such as 100% paid online panel respondents or 100% text messages—throw it out. Remember the famous “Dewey Defeats Truman” polling failure. In 1948, Republican households were more likely to have telephones, so telephone polling oversampled Republicans. Today, text-only polling can create the opposite problem by producing too many Democrats. Serious polling reaches voters in multiple ways with a focus on real human-to-human telephone interviews. You can do much more, but these are 3 simple steps to get started.

Could some of today’s cheap media polling methods produce another “Dewey Defeats Truman” moment? I joined anchor @VeronicaDudo on @EWTN News Nightly to explain how cutting corners with 100% text-message surveys or cheap online panels of paid respondents could distort the 2026 midterm picture. In 1948, telephone polling oversampled Republicans because Republican households were more likely to have phones. Today, text-only polling typically produces too many Democrats—and too many Republicans who don’t like Trump. By contrast, higher-quality internal polls connect with voters in more than one way and still conduct a majority of interviews human-to-human by telephone. The technology changed. The sampling challenge didn’t. Thanks to @VeronicaDudo and the @EWTN News Nightly team. > Watch the full segment below—don’t forget to like, share, and subscribe to my X for more teapawt.org/DustinOlson/status/211…

Will Artificial—or Super—Intelligence shape the 2026 midterms? I joined anchor @VeronicaDudo on @EWTN @News Nightly to explain what our polling says about voters’ views of AI—and why research suggests a political AI arms race is already unfolding behind the scenes in campaigns. We also discussed why the Senate map may play out differently from the House, why it could buck the national media narrative, and the polling pitfalls to watch in 2026. It was a great conversation for EWTN’s worldwide Catholic news audience. Thanks to the talented @VeronicaDudo and the @EWTN News Nightly team.