📈 Our latest midterm forecast
Updated October 30, 2026
As Nate discussed in today’s Still Counting post: “We’ve patched FLIPR’s assumptions in the Montana Senate race, which is a weird one because there’s both an actual Democrat (Alani Bankhead) and a de facto Democrat (independent Seth Bodnar) on the ballot.” As a result, Kurt Alme (the Republican candidate) is up to a 93 percent chance of winning that race.
We’ve also made some updates to the list of political scandals that feed into the model. None of those additions moved our topline numbers dramatically. Democrats remain overwhelming favorites to retake the House and are also clearly favored in the Senate, especially given their excellent state-level polls over the past couple days. -NS & EMD, 10/30/26
ELSEWHERE IN FLIPR:
See also: Generic congressional ballot dashboard and Donald Trump approval rating dashboard.
This is the landing page for FLIPR1, Silver Bulletin’s 2026 election forecast.2This page is designed to be viewed on the web rather than in our email client or in the Substack app. If you’re curious about how the model works, and what’s new for 2026 (quite a lot, actually), we also have a very detailed methodology article.
Your paid subscription gives you full access to our Election Night livestream with The Argument, GD Politics, and Slow Boring. silverbulletin.com/liveblog
The polls
The Silver Bulletin polling averages account for whether polls are conducted among registered or likely voters, and for house effects. They weight more reliable polls more heavily. They also include a “timeline adjustment”, where outdated polls are revised to reflect trends in the generic ballot. Speaking of the generic ballot, you can find a version of it with our likely-voter adjustment below; this is the version that FLIPR uses.
We draw most of the election polls used in our model from The New York Times poll database3, but collect generic ballot and congressional approval polls ourselves.
The topline
Here are today’s topline numbers for Senate and House control. They are typically based on 40,000 simulations and include a small amount of sampling error. FLIPR models House, Senate and gubernatorial elections jointly, meaning that we can calculate the probability of things like Republicans holding the Senate but losing the House.
Our default Deluxe model incorporates polls, fundamentals such as fundraising and incumbency, and expert race ratings from the Cook Political Report, Inside Elections and Sabato’s Crystal Ball. We also run Classic (polls and fundamentals) and Lite (just polls) versions of FLIPR; you can switch between these using the buttons in each chart.
House, Senate and governor
We’ve moved our full House, Senate, and governor forecasts to separate pages. You’ll find links to those below.
Topline forecast · House · Senate · Governors
Bug reports? Feature requests? Thoughts about what you like about FLIPR — or how it could be improved? Don’t hesitate to drop us a line.
Forecast with Leading Indicators, Polls and (Expert) Ratings.
Please email us at silver.bulletin.customer.service@gmail.com if you notice any missing polls.
We also include polls from a few pollsters that The New York Times excludes from their database, and exclude surveys from firms we ban or that don’t meet our standards for being scientific surveys.