It’s time for The Forecast!

Over the weekend, CM Election Predictions quietly launched Version 3.0 of The Model. This version boasts several updates, and you’ll see how those updates have changes ratings below.

The Model shifted control of the Senate from Likely R → Solid R over the weekend. This was the result of two things: the upgraded Model version played a role, but also some preliminary “hypothetical” polling came in from Ohio (R-Husted), showing key Democratic recruit, former senator Sherrod Brown, trailing by a decent margin behind incumbent senator Jon Husted. As of today, the Ohio senate race is ranked Lean R.

The other Lean R race is currently in Montana (R-Daines) - though I think this is a highly unlikely scenario. The Model relies on historical data AND polling. In the case of Montana, recent history has shown Democrats winning Senate seats (or at least performing well), and with no polling in for the 2026 race, the Model is relying HEAVILY on historical data. I expect this rating will shift towards Republicans as polls start to come in.

For the Democrats, Georgia (D-Ossoff) is the only “Lean D” seat at the moment. This is a good sign for them - but also is likely impacted by limited polling data, similarly to Montana. This will be a key race to watch. They are also seeing a positive rating in Maine (R-Collins), currently rated at Solid D. I’m less confident, but without any polling, I can’t say for sure.

Finally, the sole Toss-Up seat at the moment is North Carolina (R-OPEN). This, along with Maine, are Democrats’ key pick-up opportunities. However, they need to gain a net +4 seats to gain control of the chamber - something that, at least right now, looks extremely difficult.

The Model remained at Likely D for House control, even with the upgrade to Version 3.0 - Democrats sit at an expect 213 seats, Republicans at 210. However, the number of paths for each party to get to a majority increased pretty aggressively.

The number of toss-up seats increased from three to 12 with the new Model version - those are listed below:

AZ-01 (R-Schweikert)
CO-08 (R-Evans)
IA-03 (R-Nunn)
MI-07 (R-Barrett)
MI-10 (R-OPEN)
NE-02 (R-OPEN)
NJ-07 (R-Kean)
PA-08 (R-Bresnahan)
TX-34 (NEW)
VA-02 (R-Kiggans)
WA-03 (D-Gluesenkamp-Perez)
WI-01 (R-Steil)

The good news for Democrats is that the bulk of these Toss-Up seats are Republican seats - meaning they aren’t playing a ton of defense. The positive for Republicans is that, at least right now, it looks like they’ll have incumbents running in many of these seats.

Below are the seats that currently are ranked Lean D, but belong to Republicans:

AZ-06 (R-Ciscomani)
CA-22 (R-Valadao)
NY-17 (R-Lawler)
PA-01 (R-Fitzpatrick)
PA-07 (R-Mackenzie)

And the seats that currently are ranked Lean R, but belong to Democrats:

ME-02 (D-Golden)
OH-09 (D-Kaptur)

Republicans will certainly be on defense in 2026 - but that’s not unexpected. The question is whether or not they can hold on to a majority during their party’s midterm. The last time an incumbent president’s party maintained their majority in the House of Representatives was in 2002, when Republicans maintained control of the House during George W. Bush’s first midterm.

Following the upgrade to the Model, both elections for governor this year are rated Lean D. In Virginia, this is a FLIP from Republicans. However, predicted margins are currently tight. Here’s where The Model estimates the margins will fall:

New Jersey: D+4.53%
Virginia: D+2.57%

And that’s it for this week’s Forecast! Check out the website for more detailed information, and keep an eye out for the next Forecast!