There are 36 Seats that will decide control of the the House.
The Math: if there is no polling a projection is made based on the 2016 results using the 2016 incumbent’s margin or the Clinton margin if the seat is open.
Polling Leads: 5 points or more | 22 | 28 |
Seats Projected as Safe based on 2016 results | 185 | 142 |
| Subtotal-Total | 207 | 170 |
| Seats Leaning based on 2016, no polling | 2 | 20 |
| Total Safe and Leaning | 209 | 190 |
| Needed for majority | 9 | 28 |
| Seats with leads less than 5 (1 Tie) | 15 | 20 |
Right now the Democrats lead in enough seats to take the House. They need to win 9 of the 15 seats where they lead, or win some of the 20 seats where the GOP lead is less than 5.
The numbers here are of ONLY non-partisan polls since September 1st. My list of House polls is here. The 2016 margin is the Incumbent’s margin, of the Clinton margin if the seat is open.
| CA-10 | -3.4 | Incumbent | Tossup | 5 | |
| NJ-03 | -20.4 | -11 | incumbent | Tossup | 4 |
| MI-11 | -4.4 | -10 | open | Leans D | 3.5 |
| NJ-07 | -11 | -12.5 | incumbent | Tossup | 3.5 |
| FL-27 | 19.7 | -9 | open | Lean D | 2.5 |
| MN-01 | -14.9 | -3 | open | Tossup | 2 |
| PA-01 | -6.1 | -10 | incumbent | Tossup | 1.5 |
| CA-25 | -6.3 | -9 | incumbent | Tossup | 1 |
| IA-03 | -13.7 | -12 | incumbent | Tossup | 1 |
| KS-02 | -18.4 | 3 | open | Tossup | 1 |
| NC-02 | -13.4 | -14 | incumbent | Lean R | 1 |
| NY-22 | -5.4 | 10 | incumbent | Toss up | 1 |
| TX-32 | -71.1 | -8 | incumbent | Tossup | 1 |
| VA-05 | -11.1 | 4 | open | Lean R | 1 |
| NY-19 | -8.6 | -1 | incumbent | Tossup | 0.5 |
| FL-15 | -10 | 3.5 | open | Tossup | 0 |
| NC-09 | -11.6 | 1 | open | Tossup | -0.5 |
| CA-39 | 8.6 | -3 | open | Tossup | -0.67 |
| CA-48 | -16.6 | -3.5 | incumbent | Tossup | -0.67 |
| FL-26 | -11.8 | -7.5 | incumbent | Tossup | -0.67 |
| IL-06 | -18.4 | -21 | incumbent | Leans D | -1 |
| KY-06 | -22.2 | -6 | incumbent | Tossup | -1 |
| PA-10 | -15.2 | open | Lean R | -1 | |
| ME-02 | -9.6 | 0 | incumbent | Tossup | -1.25 |
| MI-08 | -16.9 | 0.5 | incumbent | Tossup | -1.33 |
| UT-04 | -12.5 | 0 | incumbent | Tossup | -1.5 |
| AK-AL | -14.3 | 8 | incumbent | Lean R | -1.67 |
| VA-07 | -15.3 | 1 | incumbent | Tossup | -2 |
| NM-02 | -10.2 | 2 | open | Tossup | -2.33 |
| FL-18 | -10.5 | -3 | incumbent | Likely R | -3 |
| NY-27 | -34.4 | 18 | incumbent | Lean R | -3 |
| TX-07 | -12.3 | -8 | incumbent | Tossup | -3 |
| GA-06 | -3.7 | -6 | incumbent | Lean R | -4 |
| NC-13 | -12.2 | 3 | incumbent | Tossup | -4.5 |
| CA-50 | -26.9 | 10 | incumbent | Lean R | -5 |
| IL-12 | -14.6 | 4.5 | incumbent | Lean R | -5 |
The following is worth noting with respect to this list:
- The Democrats hold leads in 6 more seats than they need to take the House as of this moment.
- A minor swing could have enormous consequences. A 3 point swing to the Republicans, and the Democrats win only 214 seats. A 3 point swing to the Democrats and they win 238 seat
In the past I have compared state polling to national polling. In 2012, for example on the day before the election I noted that the State polling was pointing to a larger Obama margin than the National Polling was suggesting.
Right now the House Polling suggests a larger Democratic victory than generic ballot polls are finding.
The generic ballot average had hovered between 7 and 8 points over the last month (or indeed even since the Spring). But the story the polling form individual House districts is very different. It is suggesting a bigger, perhaps even a significantly bigger win..
I have not studied the extent to which House polling swings can predict the National House vote.
But it is hard to reconcile the polling over the last 31 days, and particularly over the last 10 days (where the Democratic margin is growing) with the leads found in national polling.
The swing is calculated by comparing the predicted result in each House District (based on the incumbent’s margin in 2016 or Clinton’s margin if the seat is open). There are 2 seats which are such outliers that they by themselves effect the averages (W VA-3 and TX-31)