Ep. 349 Can You Trust the Polls? Nate Silver’s 2026 Midterm Election Forecast

Can election forecasts really tell us what will happen in November—or do they simply give political junkies a false sense of certainty?

Chris and Sean break down Tuesday’s primary results, beginning with a massive polling miss in Wisconsin. Democratic socialist Francesca Hong had led Milwaukee County Executive David Crowley by double digits—and by roughly 20 points in some of the final polling averages. But when the votes were counted, Crowley won the Democratic nomination for governor by less than half a percentage point, 39.8% to 39.3%.

Then they turn to Connecticut, where former Hartford Mayor Luke Bronin defeated Rep. John Larson by roughly 54% to 33%. Larson had represented Connecticut’s 1st Congressional District for 28 years, but Bronin’s argument for generational change brought the congressman’s long tenure to an abrupt end.

Sean also reviews Nate Silver’s new FLIPR election model at Silver Bulletin. At launch, Silver’s forecast gave Democrats an 86.6% chance of winning control of the U.S. House and a 57.3% chance of capturing the U.S. Senate. The model runs 40,000 simulations and combines polling with factors such as fundraising, incumbency and expert race ratings. That sounds scientific—and it is—but a probability is not a prediction, much less a guarantee.

Chris and Sean discuss why polling and election models can struggle to capture late movement, candidate quality, campaign organization and what is actually happening on the ground. The lesson is simple: elections are decided by voters, not spreadsheets—and certainly not by the people yelling at each other on X.

If you care about the outcome, get involved. Donate. Volunteer. Knock on doors. Talk to your neighbors and friends. And remember: Twitter is not real life.

Chris closes the episode with a personal tribute to his mentor, Dave Sitton, who passed away 12 years ago.