The House forecast is the widest page in the model: all 435 districts, every one of them rated, simulated, and placed on the same left-to-right spectrum. That scale is the point. A House majority is not usually won by one clean argument. It is built out of 435 local arguments, most of which are already settled, a few dozen of which are not, and a handful of which end up carrying the whole chamber on election night.
In the locked edition we are preparing to launch with, Democrats are favored. But the interesting part is not the word favored. It is the shape of the advantage. Democrats have more ways to get to 218 than Republicans do, but many of those routes run through seats where the rating is intentionally modest. The model is not saying the House is over. It is saying Republicans need the center of the map to move together, while Democrats can still lose a fair amount of the board and keep the gavel.
| Measure | Current House forecast |
|---|---|
| Seats rated | 435 |
| Majority line | 218 |
| Democratic control probability | 79.1% |
| Republican control probability | 20.9% |
| Median Democratic seats | 241 |
| 80% interval | 201–279 D seats |
| Non-safe races | 84 |
| Tilt races | 38 |
| Tipping-point district | WI-03 |
The map is broad; the door is narrow
There are 351 Safe seats in this edition: 203 Democratic and 148 Republican. That leaves 84 seats outside the safe columns. Those 84 are the live chamber. They are also why a national House probability can look comfortable while the page itself still feels tense.
The Democratic edge comes from the fact that the party does not need to win the entire competitive map. It needs to win enough of it. That sounds obvious until you look at the spectrum. Democrats can afford some misses in Tilt Democratic seats, some Republican overperformance in the Southeast, even some reversion in districts where the local signal is doing more work than the presidential baseline. Republicans need more than that. They need a connected move: Tilt seats, then Lean seats, and then some of the Likely seats that sit just far enough from the center to be painful.
That is why WI-03 matters as the tipping point. It is not a prophecy that Wisconsin will decide the House. It is a way of naming the kind of district that decides the House if the election lands near the current middle of the distribution. The chamber does not pass through the most dramatic race. It passes through the district nearest 218.
The cyan outline is the story we are telling
The map fill tells you who is ahead. The cyan outline tells you where we are calling a seat to change hands. That is where the page stops being a census of districts and becomes an argument about the election.
In this edition, 41 seats are projected flips. The list is asymmetrical, and intentionally so: it is mostly Republican-held territory moving toward Democrats, with a smaller number of Democratic-held seats moving back toward Republicans. Some of the Democratic opportunities are the familiar battlegrounds — AZ-01, CA-22, NE-02, PA-07, WI-03. Others are sharper claims, especially the Colorado seats where the model is reading candidate weakness and primary energy as enough to keep redder districts alive.
That is the bet. Not that every cyan outline is going to be right. A forecast with 41 flips should expect to be embarrassed somewhere. The bet is that the misses will not all point in the same direction. If they do — if the aggressive Democratic calls turn out to be soft, and the Republican holds harden at once — the chamber gets close quickly. If they do not, Democrats have room to spare.
What to watch
| Watch point | Why it matters |
|---|---|
| The Tilt Democratic seats | This is where the Democratic majority gets fat or thin. |
| The Likely Democratic flips | If these start looking merely Lean, the map is moving right. |
| The Republican-held Biden/near-Biden districts | These are the cleanest pickup opportunities and the first place a favorable Democratic environment should show up. |
| The redder reach seats | They tell us whether local candidate quality is really strong enough to overcome the baseline. |
| Market disagreements | A market dot against the rating is not a veto, but it is an invitation to inspect the race page. |
We built the House page to make that inspection easy. Start with the control number, then ignore it for a minute. Read the map for geography, the spectrum for density, the cyan outlines for the actual claim, and the race pages for the places where the claim feels most exposed. The forecast's current answer is that Democrats are favored for the House. The more useful answer is sharper: Democrats have the longer list of plausible paths to 218, but many of those paths run through districts where a small national correction would be enough to close the door.