2018 U.S. House Midterm Elections Forecast

Aug. 2, 2017

2018 Midterms 2018 House election forecast

How does this work?

Note: I now write for The Economist, which has its own midterms forecasting model that is entirely different than this one. My blog will continue to be a home for all my political writing, as well as host updates to this prediction, so don’t go anywhere!



Chance of Winning a House Majority

Dem. Chance: 81% Rep. Chance: 19%
   


Outlined seats are seats that are favored to flip parties.

Fullscreen map here



Get forecast updates and more sent straight to your inbox — subscribe to my weekly newsletter!



Odds Over Time


Popular Vote Forecast


My forecast of the election day vote works in three stages. First, I average all of the generic ballot polls with an algorithm designed to produce the most predictive average for each week in the cycle. Second, I use that average to predict the most likely election day polling average for Republicans, Democrats, other parties and undecided voters. Finally, I combine the projected Democratic margin in election day polls with Democrats’ average performance in special elections between 2017 and 2018 to predict the outcome of the vote on election day.

Average of
Polls Today
Vote Forecast
on Election Day
Dem. Margin 9 8.8


In my projection of the Election Day vote share, based on polls of the generic ballot and the swing toward Democrats in special elections, The Democratic Party is ahead, winning by 8.8% of the vote share on average. The margin of error is roughly 6% points. Below are the generic congressional ballot polls used to make that projection:

Polls of the Generic Congressional Ballot
Democrat Republican Date Pollster
54 43 2018-10-28 Gallup
55 42 2018-11-03 CNN/SSRS
56 41 2018-11-03 USC Dornsfire/LA Time
43 40 2018-11-02 Morning Consult
48 40 2018-10-25 CBS/YouGov
50 43 2018-11-03 NBC/WSJ
50 41 2018-11-03 IBD/TIPP
52 44 2018-11-01 ABC/Washington Post
49 42 2018-10-30 Ipsos/Reuters
47 38 2018-10-28 Harvard/Harris
52 43 2018-10-29 NPR/Marist
47 42 2018-10-30 Economist/Yougov
57 40 2018-10-27 USC Dornsfire/LA Time
48 39 2018-10-21 PRRI
50 40 2018-10-23 NPR/Marist
51 43 2018-10-22 Suffolk/USA Today
47 41 2018-10-23 Economist/Yougov
49 42 2018-10-23 Ipsos/Reuters
51 43 2018-10-22 USA Today/Suffolk
55 42 2018-10-20 USC Dornsfire/LA Time
50 41 2018-10-17 NBC/WSJ
49 41 2018-10-19 Harvard/Harris
51 42 2018-10-17 CBS News
51 40 2018-10-16 Ipsos/Reuters
50 38 2018-10-02 Kaiser Family Foundation
46 41 2018-10-16 YouGov/Economist
49 42 2018-10-16 Fox News
44 37 2018-10-14 Morning Consult
47 38 2018-10-07 Pew Research Center
49 40 2018-10-12 Harvard/Harris
55 42 2018-10-11 ABC/Washington Post
47 41 2018-10-09 YouGov/Economist
50 38 2018-10-09 Ipsos/Reuters
54 41 2018-10-07 CNN/SSRS
50 43 2018-10-04 Emerson College
45 40 2018-10-02 YouGov/Economist
48 42 2018-10-01 Marist
45 43 2018-10-03 IBD/TIPP
50 38 2018-10-02 Ipsos/Reuters
45 37 2018-09-30 Harvard/Harris
49 42 2018-09-30 Quinnipiac
45 37 2018-09-30 Harvard/Harris
46 38 2018-09-25 YouGov/Economist
48 41 2018-09-24 Marist
48 41 2018-09-25 Ipsos/Reuters
51 42 2018-09-23 Gallup
52 42 2018-09-19 Pew Research Center
46 38 2018-09-25 YouGov/Economist
48 41 2018-09-24 Marist
55 41 2018-09-24 USC Dornsfire/LA Time
52 44 2018-08-16 USC Dornsfire/LA Time
49 42 2018-09-19 Fox News
51 43 2018-09-19 NBC/WSJ
48 45 2018-09-18 YouGov/Economist
48 41 2018-09-18 Ipsos/Reuters
49 41 2018-09-14 Harris Interactive
49 41 2018-09-11 Ipsos/Reuters
49 41 2018-09-14 Harvard/Harris
52 42 2018-09-09 CNN/SSRS
44 41 2018-09-11 YouGov/Economist
45 35 2018-09-09 Morning Consult
50 38 2018-09-09 NPR/Marist
52 38 2018-09-09 Quinnipiac
49 39 2018-08-07 Harvard/Harris
45 43 2018-09-04 Selzer & Co.
45 40 2018-09-04 YouGov/Economist
44 35 2018-09-04 Ipsos/Reuters
52 39 2018-08-31 Emerson College
42 35 2018-08-31 Harvard/Harris
50 39 2018-08-30 IBD/TIPP
52 38 2018-08-29 ABC/Post
45 39 2018-08-28 YouGov/Economist
50 39 2018-08-28 Suffolk/USA Today
47 34 2018-08-28 Ipsos/Reuters
42 36 2018-08-24 Harvard/Harris
45 37 2018-08-23 Harvard/Harris
50 42 2018-08-22 NBC/WSJ
41 37 2018-08-21 Ipsos/Reuters
49 38 2018-08-21 Fox News
44 38 2018-08-21 YouGov/Economist
48 43 2018-08-19 Monmouth University
45 36 2018-08-18 rv
44 40 2018-08-14 YouGov/Economist
45 36 2018-08-14 Ipsos/Reuters
43 32 2018-08-13 Harvard/Harris
51 42 2018-08-13 Quinnipiac
42 38 2018-08-12 Morning Consult
52 41 2018-08-12 CNN/SSRS
41 39 2018-08-07 Ipsos/Reuters
44 41 2018-08-07 YouGov/Economist
41 39 2018-08-07 Ipsos/Reuters
42 36 2018-08-06 Morning Consult
45 45 2018-08-02 Live Phone
44 40 2018-07-31 YouGov/Economist
42 36 2018-07-31 Ipsos/Reuters
44 37 2018-07-30 Morning Consult
40 36 2018-07-27 YouGov/CBS
52 42 2018-07-27 SurveyMonkey
43 36 2018-07-25 Harvard/Harris
44 38 2018-07-24 YouGov/Economist
42 37 2018-07-24 Ipsos/Reuters
41 38 2018-07-23 Morning Consult
51 39 2018-07-23 Quinnipiac
47 40 2018-07-22 NPR/Marist
49 37 2018-07-22 Kaiser Family Foundation
49 43 2018-07-18 NBC/WSJ
45 37 2018-07-17 YouGov/Economist
44 35 2018-07-17 Ipsos/Reuters
45 37 2018-07-14 Morning Consult
48 40 2018-07-11 Fox News
42 38 2018-07-10 YouGov/Economist
42 37 2018-07-10 Morning Consult
47 34 2018-07-10 Ipsos/Reuters
49 42 2018-07-07 Emerson College
42 38 2018-07-03 YouGov/Economist
47 37 2018-07-02 Washington Post/Schar School
50 41 2018-07-01 Quinnipiac
44 34 2018-07-02 Ipsos/Reuters
43 35 2018-06-29 Morning Consult
48 40 2018-06-29 IBD/TIPP
43 38 2018-06-26 YouGov/Economist
45 36 2018-06-25 Harvard/Harris
41 38 2018-06-26 Ipsos/Reuters
44 37 2018-06-24 Morning Consult
44 37 2018-06-19 YouGov/Economist
42 37 2018-06-18 Morning Consult
42 36 2018-06-19 Ipsos/Reuters
45 39 2018-06-18 Suffolk/USA Today
50 42 2018-06-17 CNN/SSRS
49 43 2018-06-17 Quinnipiac
48 43 2018-06-17 Gallup
48 43 2018-06-12 Pew Research Center
48 41 2018-06-13 Monmouth University
43 37 2018-06-12 YouGov/Economist
45 35 2018-06-12 Ipsos/Reuters
51 42 2018-06-05 Democracy Corps
46 40 2018-06-10 PPP
43 38 2018-06-10 Morning Consult
48 39 2018-06-06 Fox News
44 38 2018-06-05 YouGov/Economist
43 32 2018-06-05 Ipsos/Reuters
50 40 2018-06-04 NBC/WSJ
48 40 2018-06-05 Quinnipiac
44 37 2018-06-04 Morning Consult
47 40 2018-06-05 IBD/TIPP
42 37 2018-05-29 Morning Consult
42 39 2018-05-29 YouGov/Economist
39 37 2018-05-29 Ipsos/Reuters
46 39 2018-05-25 Echelon Insights
44 37 2018-05-22 Harvard/Harris
43 38 2018-05-22 YouGov/Economist
40 37 2018-05-22 Ipsos/Reuters
42 37 2018-05-15 YouGov/Economist
38 37 2018-05-15 Ipsos/Reuters
42 35 2018-05-07 Morning Consult
44 35 2018-05-08 YouGov/Economist
47 44 2018-05-05 CNN/SSRS
42 39 2018-05-01 YouGov/Economist
43 37 2018-05-01 Ipsos/Reuters
48 43 2018-05-01 Pew Research Center
49 41 2018-04-30 Monmouth University
43 34 2018-04-24 Harvard/Harris
44 35 2018-04-23 Politico/Morning Consult
48 40 2018-04-24 Quinnipiac
45 34 2017-04-24 Ipsos/Reuters
43 38 2018-04-24 YouGov/Economist
43 38 2018-04-17 YouGov/Economist
43 38 2018-04-17 Politico/Morning Consult
44 39 2018-04-14 NPR/Marist
44 34 2018-04-17 Ipsos/Reuters
47 43 2018-04-11 ABC/Post
47 40 2018-04-11 NBC/WSJ
46 43 2018-04-09 Quinnipiac
42 38 2018-04-07 Politico/Morning Consult
44 36 2018-04-10 YouGov/Economist
43 34 2018-04-03 Ipsos/Reuters
43 36 2018-04-03 YouGov/Economist
43 34 2018-04-03 Ipsos/Reuters
45 34 2018-03-29 Harvard/Harris
42 35 2018-03-27 YouGov/Economist
42 35 2018-04-03 Ipsos/Reuters
50 39 2018-03-25 PPP
50 44 2018-03-25 CNN/SSRS
46 41 2018-03-21 Fox News
44 39 2018-03-21 Marist
40 37 2018-03-20 Ipsos/Reuters
49 43 2018-03-20 Quinnipiac
44 38 2018-03-20 YouGov/Economist
50 40 2018-03-14 NBC/WSJ
43 38 2018-03-13 YouGov/Economist
38 30 2018-03-13 Ipsos/Reuters
49 40 2018-03-08 GWU/Battleground
43 37 2018-03-06 YouGov/Economist
48 38 2018-03-05 Quinnipiac
50 41 2018-03-05 Monmouth University
38 31 2018-03-06 Ipsos/Reuters
37 31 2018-02-27 Ipsos/Reuters
40 38 2018-02-27 YouGov/Economist
41 39 2018-02-26 Politico/Morning Consult
47 32 2018-02-24 Suffolk/USA Today
54 38 2018-02-23 CNN/SSRS
46 39 2018-02-21 Marist
43 35 2018-02-20 YouGov/Economist
38 30 2018-02-20 Ipsos/Reuters
53 38 2018-02-19 Quinnipiac
41 36 2018-02-19 Harvard/Harris
42 38 2018-02-13 YouGov/Economist
39 30 2018-02-13 Ipsos/Reuters
38 39 2018-02-12 Politico/Morning Consult
49 41 2018-02-11 PPP
42 38 2018-02-09 Ipsos/Reuters
49 38 2018-02-07 Marist
43 37 2018-02-06 YouGov/Economist
38 31 2018-02-06 Ipsos/Reuters
49 40 2018-02-05 Quinnipiac
42 38 2018-02-04 Politico/Morning Consult
46 41 2018-02-02 IBD/TIPP
45 39 2018-02-01 NBC/SurveyMonkey
47 45 2018-01-30 Monmouth University
42 37 2018-01-30 YouGov/Economist
37 31 2018-01-30 Ipsos/Reuters
41 37 2018-01-23 YouGov/Economist
44 38 2018-01-23 Fox News News
51 38 2018-01-23 Quinnipiac
38 30 2018-01-23 Ipsos/Reuters
43 37 2018-01-21 Politico/Morning Consult
42 37 2018-01-20 Politico/Morning Consult
51 39 2018-01-18 ABC/Post
51 41 2018-01-18 CNN/SSRS
49 43 2018-01-17 NBC/WSJ
42 36 2018-01-16 YouGov/Economist
37 32 2018-01-16 Ipsos/Reuters
50 39 2018-01-16 Quinnipiac
41 38 2018-01-16 Politico/Morning Consult
52 35 2018-01-16 Quinnipiac
53 39 2018-01-15 Pew
46 40 2018-01-10 NPR/Marist
44 37 2018-01-09 YouGov/Economist
40 31 2018-01-09 Ipsos/Reuters
44 36 2018-01-05 Politico/Morning Consult
42 36 2018-01-02 YouGov/Economist
38 31 2018-01-02 Ipsos/Reuters
44 36 2017-12-26 YouGov/Economist
44 35 2017-12-19 YouGov/Economist
39 27 2017-12-19 Ipsos/Reuters
44 34 2017-12-18 Politico/Morning Consult
56 38 2017-12-17 CNN/SSRS
51 40 2017-12-12 PPP
44 36 2017-12-12 YouGov/Economist
51 36 2017-12-12 Monmouth University
37 29 2017-12-12 Ipsos/Reuters
41 36 2017-12-11 Politico/Morning Consult
50 37 2017-12-07 Marist
42 36 2017-12-05 YouGov/Economist
37 31 2017-12-05 Ipsos/Reuters
40 38 2017-12-03 Politico/Morning Consult
41 35 2017-11-28 YouGov/Economist
36 30 2017-11-28 Ipsos/Reuters
42 36 2017-11-25 Politico/Morning Consult
37 28 2017-11-22 Ipsos/Reuters
42 33 2017-11-21 YouGov/Economist
43 36 2017-11-19 Politico/Morning Consult
43 40 2017-11-15 NPR/Marist
42 34 2017-11-14 YouGov/Economist
38 31 2017-11-14 Ipsos/Reuters
44 36 2017-11-11 Politico/Morning Consult
51 36 2017-11-09 Marist
40 33 2017-11-07 YouGov/Economist
38 30 2017-11-07 Ipsos/Reuters
38 39 2017-11-06 Politico/Morning Consult
51 40 2017-11-05 CNN/SSRS
51 40 2017-11-01 ABC/Post
39 36 2017-10-31 YouGov/Economist
40 31 2017-10-31 Ipsos/Reuters
43 38 2017-10-30 Politico/Morning Consult
44 37 2017-10-30 PRRI
50 35 2017-10-24 Fox News
40 33 2017-10-24 YouGov/Economist
36 28 2017-10-24 Ipsos/Reuters
42 36 2017-10-23 Politico/Morning Consult
41 33 2017-10-16 YouGov/Economist
42 37 2017-10-16 Politico/Morning Consult
54 38 2017-10-15 CNN/SSRS
40 33 2017-10-10 YouGov/Economist
43 37 2017-10-09 Politico/Morning Consult
39 32 2017-10-03 YouGov/Economist
41 39 2017-10-01 Politico/Morning Consult
40 34 2017-09-26 YouGov/Economist
48 37 2017-09-25 PPP
42 38 2017-09-24 Politico/Morning Consult
49 43 2017-09-20 CNN/SSRS
39 33 2017-09-19 YouGov/Economist
43 37 2017-09-17 Politico/Morning Consult
40 31 2017-09-12 YouGov/Economist
40 37 2017-09-11 Politico/Morning Consult
39 33 2017-09-05 YouGov/Economist
43 38 2017-09-03 Politico/Morning Consult
40 32 2017-08-29 YouGov/Economist
40 36 2017-08-28 Politico/Morning Consult
39 33 2017-08-22 YouGov/Economist
49 35 2017-08-21 PPP
41 36 2017-08-19 Politico/Morning Consult
46 40 2017-08-17 GWU/Battleground
41 33 2017-08-15 YouGov/Economist
40 38 2017-08-14 Politico/Morning Consult
47 40 2017-08-12 Marist
39 34 2017-08-08 YouGov/Economist
51 42 2017-08-06 CNN/SSRS
43 36 2017-08-06 Politico/Morning Consult
39 34 2017-08-01 YouGov/Economist
44 37 2017-07-29 Politico/Morning Consult
40 34 2017-07-25 YouGov/Economist
40 40 2017-07-24 Politico/Morning Consult
40 34 2017-07-18 YouGov/Economist
50 40 2017-07-17 PPP
42 40 2017-07-15 Politico/Morning Consult
39 32 2017-07-11 YouGov/Economist
43 40 2017-07-09 Politico/Morning Consult
37 35 2017-07-04 YouGov/Economist
43 39 2017-06-30 Politico/Morning Consult
47 41 2017-06-27 Fox News
41 35 2017-06-27 YouGov/Economist
48 38 2017-06-25 NPR/Marist
42 40 2017-06-24 Politico/Morning Consult
38 35 2017-06-20 YouGov/Economist
43 37 2017-06-19 Politico/Morning Consult
38 36 2017-06-13 YouGov/Economist
42 39 2017-06-12 Politico/Morning Consult
50 40 2017-06-11 PPP
41 37 2017-06-06 YouGov/Economist
42 39 2017-06-02 Politico/Morning Consult
39 33 2017-05-30 YouGov/Economist
43 39 2017-05-30 Politico/Morning Consult
38 36 2017-05-23 YouGov/Economist
41 37 2017-05-22 Politico/Morning Consult
40 33 2017-05-16 YouGov/Economist
42 35 2017-05-14 Politico/Morning Consult
49 38 2017-05-14 PPP
42 37 2017-05-11 Politico/Morning Consult
40 35 2017-05-09 YouGov/Economist
42 36 2017-05-06 Politico/Morning Consult
38 35 2017-05-02 YouGov/Economist
41 41 2017-04-30 Politico/Morning Consult
47 42 2017-04-25 Fox News
39 34 2017-04-25 YouGov/Economist
50 41 2017-04-25 CNN/SSRS
40 44 2017-04-24 Politico/Morning Consult
47 42 2017-04-20 Gravis Marketing/Orlando Political Observer
47 41 2017-04-18 PPP
38 32 2017-04-18 YouGov/Economist
40 40 2017-04-15 Politico/Morning Consult
45 38 2017-04-12 McClatchy/Marist
45 38 2017-04-12 Marist
40 37 2017-04-11 YouGov/Economist
43 40 2017-04-09 Politico/Morning Consult
41 39 2017-04-01 Politico/Morning Consult
48 43 2017-03-28 PPP
47 38 2017-03-27 McClatchy/Marist
46 41 2017-03-12 PPP
46 43 2017-02-22 PPP
49 41 2017-02-08 PPP
45 42 2017-01-31 PPP
48 40 2017-01-24 PPP


But, because Democrats are clustered in cities and face harsh gerrymanders, they aren’t expected to win an equivalent share of the seats in Congress. What does electoral geography tell us about the actual outcome?


Seat Forecast


Seats

Median number of Democratic seats in our simulations (NOTE this may be more/fewer seats than the strict prediction of seats in the Individual Seat Projections section below, for reasons explained here):

Dem. Seats: 230 Rep. Seats: 205
   


Outlined seats are the top 20 “tipping point” districts.

Simulated Seats Over Time

Democrats earn a median of 230 seats in our simulations of the 2018 midterms. This may differ from the strict predictions below because of the larger number of Lean Republican seats than Lean Democratic seats in the current Congress. Effectively we are saying that the below number is an ideal estimate, meant to give you context as to which seats are competitive, but that we expect Democrats to overperform expectations based on the assessment of our error in past predictions.


Individual Seat Projections

Using the average vote share for each district over all of our simulations, we can identify both the districts that have the best chance of flipping parties and the chance that that happens.

Democrats: 221
Republicans: 214

Democrats are likely to pick up 29 seats on November 6, 2018. Republicans are favored to gain 2 seat, for a net gain of 27 seats for the Democratic Party.

Seats Likely to Flip Parties in 2018
District Dem 2016/14 (Margin, %) Partisan Lean (Margin, %) Forecast Dem 2018 (Margin, %) Dem Win Prob.
AZ-02 -14 1 10 96
CA-10 -3 1 4 82
CA-25 -6 2 3 73
CA-39 -14 3 1 56
CA-49 -1 1 9 98
CO-06 -8 5 7 91
FL-26 -12 12 4 74
FL-27 -10 14 10 94
IA-01 -8 -2 7 90
IA-03 -14 -4 0 53
IL-06 -18 1 1 54
KS-03 -11 -4 6 86
ME-02 -10 -8 2 64
MI-11 -13 -7 4 77
MN-01 1 -13 -1 43
MN-02 -2 -3 7 92
MN-03 -14 5 6 87
MN-08 1 -13 -3 32
NE-02 -1 -6 1 56
NJ-02 -22 -4 12 98
NJ-07 -11 -3 4 76
NJ-11 -19 -5 3 70
NY-19 -8 -6 3 74
NY-22 -6 -14 1 61
PA-01 -12 -1 2 61
PA-06 -6 5 12 93
PA-07 -10 0 10 96
PA-17 -11 -6 9 94
TX-32 -16 -5 0 53
VA-10 -6 5 9 96
WA-08 -20 0 3 72

Forecasts for every House race on the bottom of the page.


Who’s Vulnerable?

The graph below stacks each House seat on top of each other above the percentage share of the vote I forecast for the Democratic candidate in the district. The gray shaded area represents a 5% margin of error — roughly what we expect given past error in the national generic ballot polls — identifying vulnerable seats that could be won by either Democrats or Republicans.

Tipping Points and the Majority Power Indicator (MPI)

Districts that usually fall in the middle of the pack are “tipping point” districts. They tell us that, in a tied election, these districts most likely land the 218th seat for the winning party. For Democrats, the “tipping point” district is the one that gives them the 24th seat they need to win the election given that they’ve already won 23 other seats (most likely the ones detailed above).

The Majority Power Indicator (MPI) is simply a measure of the increase in the probability that a given party wins the House majority given that they win a given seat. Mathematically, MPI is equal to (1) difference between (A) the number of trials a party wins a given seat and wins the House majority minus the number of trials they win that seat but lose the majority and (B) the number of trials that a party loses a given seat and wins the House majority minus number of trials they lose that seat but lose the majority, (2) all divided by the number of trials/simulations in our forecast model.

Together, the tipping point index and MPI tell us which House districts are most instrumental in producing control of the House majority. More information can be found here.

Tipping Point Districts and the MPI
District Tipping Point (% of trials) D. Tipping Point R. Tipping Point MPI (% of trials)
WA-08 2.4 2.3 2.5 51
CA-25 2.2 2.2 2.2 51.1
CA-39 2.2 2.2 2.4 33
NJ-07 2.1 2 2.2 54.8
NY-22 2.1 2.1 2.1 38.1
CA-45 2 2 2.3 23.6
ME-02 2 2 2.4 40.3
NE-02 2 1.9 2.2 32.8
NJ-11 2 1.9 2.2 47.9
NY-19 2 2 2 52.3
TX-32 1.9 1.9 2 29.5
CA-10 1.8 1.8 1.5 52
IA-03 1.8 1.8 1.8 27.9
IL-06 1.8 1.8 1.9 28.3
MI-11 1.7 1.7 1.7 53.4
PA-01 1.7 1.7 1.9 35
TX-07 1.7 1.7 1.8 22
CA-48 1.6 1.6 1.7 14.5
VA-07 1.6 1.6 1.6 18.1
KS-03 1.5 1.5 1.8 61.7
NJ-03 1.5 1.6 1.4 17.6
FL-26 1.4 1.4 1.5 41.8
IL-14 1.4 1.4 1.4 6.6
MN-01 1.4 1.4 1.4 13
MN-03 1.4 1.4 1.5 59.1
UT-04 1.4 1.4 1.6 10.3
IA-01 1.3 1.3 1.3 61
MI-08 1.3 1.4 1.2 11
NC-09 1.2 1.3 1.2 3.7
NC-13 1.2 1.2 1.4 2.4
TX-23 1.2 1.1 1.3 -4
CO-06 1.1 1.1 1 64
FL-15 1.1 1.1 1.1 1.2
MN-08 1.1 1.1 1.2 -4.6
NV-03 1.1 1.1 1.2 63.6
AZ-01 1 1 1.1 62.7
MN-02 1 1 1.1 62.6
IL-12 0.9 0.9 0.9 -12.6
NM-02 0.9 0.8 1.1 -17.3
NV-04 0.9 0.9 0.9 61.8
NY-23 0.9 0.8 0.9 -9.7
VA-05 0.9 1 0.9 -9.9
FL-06 0.8 0.8 0.7 -18.3
GA-06 0.8 0.8 0.7 -11.9
NC-02 0.8 0.8 0.7 -18.1
NH-01 0.8 0.8 0.8 62.5
NJ-05 0.8 0.8 0.9 63.8
PA-16 0.8 0.8 1 -15.2
PA-17 0.8 0.8 0.7 62
AK-AL 0.7 0.7 0.7 -19.5

Race Ratings

Each seat is given a discrete race rating based on the following scale for either party:

  • Tossup: neither party has a win probability greater than 60%
  • Lean Dem./Rep.: win probability between 60% and 80% for either party
  • Likely Dem./Rep.: win probability between 80 and 95% for either party
  • Safe Dem./Rep.: win probability greater than 95% for either party


Possible Outcomes


Of course, there is error in our forecast — measurable error. We can account for that error by simulating the election thousands of times, asking the computer each time to pick random error for national polling that is based on the error from past predictions. Then, for the national error in each “election,” we also generate an error for each congressional district, basing each of those 435 individual errors on what the error is in districts that have performed the same in the past. (So, for example, if a simulated error adds 4% for the Democratic candidate in a district that voted for Romney and Trump by 5%, the error would around 4% for other districts that voted similarly. This is called correlated random error.)

Here’s the distribution of total number of Democratic House seats after engaging in that exercise:

And here’s a graph of the possible seat outcomes that includes a breakdown of the possible national popular vote shares for Democrats:

We can take all of those simulations of Democratic seat shares and further ask “what is the chance that Democrats win the election, given the chance that Democrats or Republicans beat expectations?”

Democrats have a 81.2 percent chance of winning a House majority on November 6th, 2018. You can imagine that the rectangle below is a dart board, and if you randomly threw a dart anywhere at the rectangle whichever square you land on corresponds to the party that wins the House majority.


Acknowledgements, Notes, and Data


Acknowledgements

Notes

  • This page updates four times daily: at 8:30 AM CST, 12:30PM CST, 4:30PM CST, and 8:30PM CST. Come back regularly for a different look at the 2018 midterm race. If you notice something missing let me know.



Forecasts For Every House Seat


Forecasts for Every House Seat
District Party Dem 2016/14 (Margin, %) Partisan Lean (Margin, %) Forecast Dem 2018 (Margin, %) Dem Win Prob. Inc.
AK-AL R -14.3 -17.5 -3.8 22.9 R
AL-01 R -34.3 -30.7 -23.1 0.2 R
AL-02 R -8.3 -33.1 -6.4 20.9 R
AL-03 R -34 -33.7 -24 0.1 R
AL-04 R -68.5 -62.5 -58.3 0 R
AL-05 R -33.5 -34.9 -24.1 0.1 R
AL-06 R -49.1 -48.5 -39.9 0 R
AL-07 D 45.7 39.7 100 100 D
AR-01 R -33.5 -34.8 -28.8 0 R
AR-02 R -21.5 -13.5 -8.7 6 R
AR-03 R -57.7 -34.6 -31.8 0 R
AR-04 R -23.1 -33.7 -22.4 0 R
AZ-01 D 7.3 -4 7.3 92 D
AZ-02 R -14 0.7 10.1 95.9 OPEN
AZ-03 D 24.1 26 38 100 D
AZ-04 R -43 -41.8 -32.5 0 R
AZ-05 R -28.2 -25.7 -17 1 R
AZ-06 R -24.2 -15.2 -12 1.3 R
AZ-07 D 50.5 45.7 100 100 D
AZ-08 R -50.9 -24.6 -22.2 0 R
AZ-09 D 21.9 10.8 25.1 99.8 OPEN
CA-01 R -18.2 -21.4 -9.2 10.3 R
CA-02 D 53.8 42.2 56.2 100 D
CA-03 D 18.8 9.7 22.8 99.8 D
CA-04 R -25.4 -18.2 -9.2 2.1 R
CA-05 D 53.8 41.7 100 100 D
CA-06 D 50.8 41.3 100 100 D
CA-07 D 2.4 7 13 99.8 D
CA-08 R -24.6 -17.4 -100 0 R
CA-09 D 14.8 15.8 21.8 99.3 D
CA-10 R -3.4 0.6 4.4 81.7 R
CA-11 D 44.2 43.5 52.1 100 D
CA-12 D 75.3 73.5 80 100 D
CA-13 D 81.6 77.5 100 100 D
CA-14 D 61.8 54.1 74.3 100 D
CA-15 D 47.6 41.3 49.3 100 D
CA-16 D 16 18.5 21.2 99.8 D
CA-17 D 55.1 49.1 56.4 100 D
CA-18 D 42.2 47.2 51.5 100 D
CA-19 D 47.8 47.2 62.2 100 D
CA-20 D 41.6 44 100 100 D
CA-21 R -13.4 11.9 -7.2 9.6 R
CA-22 R -35.2 -13.4 -11.1 1 R
CA-23 R -38.4 -25.4 -28.5 0 R
CA-24 D 6.8 15.4 14 99.7 D
CA-25 R -6.2 2 2.8 72.9 R
CA-26 D 20.8 16.5 24.2 99.8 D
CA-27 D 34.8 32.6 100 100 D
CA-28 D 56 45.8 58.2 100 D
CA-29 D 58.5 57.3 63.4 100 D
CA-30 D 45.2 38.3 47.9 100 D
CA-31 D 12.2 17.4 19 98.2 D
CA-32 D 32.4 34.8 47.1 100 D
CA-33 D 32.8 34.4 40.9 100 D
CA-34 D 75.4 69.4 100 100 D
CA-35 D 44.8 37.3 41.5 100 D
CA-36 D 24.2 4.9 20.6 99.3 D
CA-37 D 75.9 72.6 90.9 100 D
CA-38 D 41 35.1 53 100 D
CA-39 R -14.4 3 0.6 55.7 OPEN
CA-40 D 71.8 65.8 100 100 D
CA-41 D 30 24.7 41.6 100 D
CA-42 R -17.6 -15.3 -13 5.3 R
CA-43 D 52.2 58.2 54.8 100 D
CA-44 D 79.8 68.3 100 100 D
CA-45 R -17.2 -1.5 -0.2 48.5 R
CA-46 D 28.3 32.6 34.3 100 D
CA-47 D 27.4 26.8 29.8 100 D
CA-48 R -16.6 -4.2 -0.8 42.3 R
CA-49 R -0.6 1.4 9.2 98.3 OPEN
CA-50 R -27 -19.5 -12.6 0.2 R
CA-51 D 45.6 44.3 44.9 100 D
CA-52 D 13 15.9 22.7 99.9 D
CA-53 D 34 29.9 37.7 100 D
CO-01 D 40.2 41.9 55 100 D
CO-02 D 19.7 18 29.8 99.9 OPEN
CO-03 R -14.3 -13.1 -7.9 7.2 R
CO-04 R -31.8 -24.7 -19.1 0.5 R
CO-05 R -31.4 -25.8 -19.2 0.5 R
CO-06 R -8.3 5.4 7.2 91.5 R
CO-07 D 15.4 10.2 26.2 100 D
CT-01 D 30.3 21.7 40.6 100 D
CT-02 D 29.5 3 32.9 100 D
CT-03 D 38 15.7 43.5 100 D
CT-04 D 19.8 17.5 31.9 100 D
CT-05 D 16 2.6 17.2 98.1 OPEN
DE-AL D 14.5 10.7 27.9 100 D
FL-01 R -38.2 -41.6 -29.9 0 R
FL-02 R -37.4 -36.9 -27.5 0 R
FL-03 R -16.8 -18.2 -6.5 19.5 R
FL-04 R -42.6 -31.9 -29.1 0 R
FL-05 D 28.4 23.6 40.1 100 D
FL-06 R -17.2 -16.7 -4.1 23.8 OPEN
FL-07 D 3 2.9 9.3 95.5 D
FL-08 R -30.6 -21.7 -17.1 1.1 R
FL-09 D 15 10.2 17.4 99.8 D
FL-10 D 29.8 23.2 100 100 D
FL-11 R -33.8 -31.6 -23.1 0.1 R
FL-12 R -37.2 -18.4 -19.1 0.1 R
FL-13 D 3.8 2.5 15.4 98.4 D
FL-14 D 23.6 15.2 100 100 D
FL-15 R -15 -11.5 -2.2 35.1 OPEN
FL-16 R -19.6 -12.7 -6.8 10.8 R
FL-17 R -27.6 -27 -18.3 1.9 OPEN
FL-18 R -10.5 -10.4 -4 22.6 R
FL-19 R -31.8 -24.6 -10.5 3.1 R
FL-20 D 60.6 60.4 100 100 D
FL-21 D 27.6 17.5 100 100 D
FL-22 D 17.8 12.6 28.7 100 D
FL-23 D 16.2 22.9 31.6 100 D
FL-24 D 76.4 66.2 100 100 D
FL-25 R -24.8 -6.2 -6.3 18.6 R
FL-26 R -11.8 12.4 4.2 74.3 R
FL-27 R -9.8 13.9 10.1 93.7 OPEN
GA-01 R -20.3 -17.4 -8.5 14.4 R
GA-02 D 22.4 10.7 31.1 99.9 D
GA-03 R -36.6 -34.4 -26 0 R
GA-04 D 51.4 49.3 65.4 100 D
GA-05 D 68.8 69.1 100 100 D
GA-06 R -23.4 -9.5 -3.2 27.1 R
GA-07 R -20.8 -12.8 -4.7 21.3 R
GA-08 R -35.2 -30.3 -100 0 R
GA-09 R -61.8 -60.8 -53.1 0 R
GA-10 R -31.3 -28.2 -20.1 0.5 R
GA-11 R -34.8 -30.2 -23.2 0.1 R
GA-12 R -23.2 -17.7 -10.6 10.1 R
GA-13 D 46.6 40.6 58.9 100 D
GA-14 R -60.2 -54.2 -49.5 0 R
HI-01 D 49.2 32.1 49.5 100 OPEN
HI-02 D 62.4 32.5 66.4 100 D
IA-01 R -7.6 -1.8 6.6 90.1 R
IA-02 D 7.5 -2.4 16.8 99.9 D
IA-03 R -13.7 -4.1 0.4 52.7 R
IA-04 R -22.6 -25.1 -11.7 1.3 R
ID-01 R -36.4 -39.5 -30.3 0 OPEN
ID-02 R -33.5 -28.8 -21.8 0.2 R
IL-01 D 48.2 52.7 64.6 100 D
IL-02 D 59.6 57.2 74 100 D
IL-03 D 26.1 12.3 34.2 100 D
IL-04 D 66 65.1 79.2 100 OPEN
IL-05 D 40.3 41 54.8 100 D
IL-06 R -18.4 0.7 0.5 54.2 R
IL-07 D 68.4 75 86.7 100 D
IL-08 D 16.6 17.9 29.9 100 D
IL-09 D 33 39.3 49.2 100 D
IL-10 D 5.2 23.6 24.4 99.9 D
IL-11 D 20.8 19.4 33.3 100 D
IL-12 R -14.6 -13.3 -3.3 27.2 R
IL-13 R -19.4 -6.8 -4.6 19.2 R
IL-14 R -18.6 -8 -1.5 38.4 R
IL-15 R -48 -44.7 -37.6 0 R
IL-16 R -30 -17.4 -14.7 0.3 R
IL-17 D 20.6 1.2 26.2 100 D
IL-18 R -44.2 -28.9 -29 0 R
IN-01 D 24.3 12.9 33.2 100 D
IN-02 R -22.4 -23.5 -12.3 4.9 R
IN-03 R -47.1 -35.5 -29.7 0 R
IN-04 R -34.1 -34.1 -25.6 0.2 OPEN
IN-05 R -27.2 -15.6 -9.3 3.7 R
IN-06 R -42.4 -38.6 -31.5 0 OPEN
IN-07 D 24.3 21.4 36.5 100 D
IN-08 R -32 -32.5 -22.2 0.2 R
IN-09 R -13.6 -26.9 -7.5 7.5 R
KS-01 R -39.1 -46.9 -24.8 0 R
KS-02 R -28.3 -19.8 -6.1 13.1 OPEN
KS-03 R -10.7 -4 5.6 86.1 R
KS-04 R -31.1 -29.3 -22.9 0 R
KY-01 R -45.3 -47.5 -36.9 0 R
KY-02 R -39.6 -39.5 -30 0 R
KY-03 D 27 11.9 34.6 100 D
KY-04 R -42.6 -36.6 -30.9 0 R
KY-05 R -60 -62.1 -52.4 0 R
KY-06 R -22.2 -17.4 -3.9 23 R
LA-01 R -55.1 -45.1 -42.6 0 R
LA-02 D 73.2 50 100 100 D
LA-03 R -63.2 -39.6 -46 0 R
LA-04 R -41.6 -25.4 -26 0.1 R
LA-05 R -36.9 -30.4 -24.7 0.2 R
LA-06 R -48.9 -36.4 -35.1 0 R
MA-01 D 26.4 20.4 100 100 D
MA-02 D 22.9 16.9 33.8 100 D
MA-03 D 37.5 18.4 35.6 100 OPEN
MA-04 D 40.3 19.6 100 100 D
MA-05 D 44.2 38.2 56.3 100 D
MA-06 D 19.1 13.6 29.9 100 D
MA-07 D 74.3 68.3 100 100 OPEN
MA-08 D 44.9 21.2 100 100 D
MA-09 D 22.1 8.6 30.1 100 D
MD-01 R -38.4 -29.5 -25.4 0 R
MD-02 D 29 22.9 40.2 100 D
MD-03 D 29.3 26.4 41.7 100 D
MD-04 D 52.7 55.7 68.8 100 D
MD-05 D 38 29.2 48.7 100 D
MD-06 D 15.9 12.7 24.7 99.8 OPEN
MD-07 D 53.1 52.1 67.7 100 D
MD-08 D 26.4 29.2 40.9 100 D
ME-01 D 16.1 13.9 26.2 100 D
ME-02 R -9.6 -8.2 1.8 63.9 R
MI-01 R -14.8 -20.6 -8.4 5.1 R
MI-02 R -30.1 -19 -15.7 1.9 R
MI-03 R -22 -11.4 -7.4 15.9 R
MI-04 R -29.5 -23.2 -16.9 1.1 R
MI-05 D 26.1 6.2 31.9 100 D
MI-06 R -22.2 -9.2 -4.3 20.3 R
MI-07 R -15.1 -16.1 -4.4 24.6 R
MI-08 R -16.8 -8.4 -1.2 41.3 R
MI-09 D 20.5 7.1 22 99.3 OPEN
MI-10 R -30.8 -29.6 -20.3 0.3 R
MI-11 R -12.7 -7.2 4.2 77.1 OPEN
MI-12 D 35 25.5 45.3 100 D
MI-13 D 61.4 60.7 100 100 OPEN
MI-14 D 59.8 58.7 74.7 100 D
MN-01 D 0.7 -13.4 -1.1 42.7 OPEN
MN-02 R -1.8 -3.4 7.5 92.3 R
MN-03 R -13.7 4.7 6 86.5 R
MN-04 D 23.4 27.4 38.1 100 D
MN-05 D 46.8 51.2 62.9 100 OPEN
MN-06 R -31.3 -25.6 -19.1 0.5 R
MN-07 D 5.1 -28.1 9.4 94.2 D
MN-08 D 0.6 -12.9 -2.6 31.9 OPEN
MO-01 D 55.5 56.4 70.9 100 D
MO-02 R -20.8 -14.2 -5.7 15.6 R
MO-03 R -39.9 -38.3 -29.7 0 R
MO-04 R -40 -35.8 -28.9 0 R
MO-05 D 20.6 12.5 30.6 100 D
MO-06 R -39.6 -31.6 -27 0 R
MO-07 R -40.1 -46.2 -32.9 0 R
MO-08 R -51.7 -51.8 -42.9 0 R
MS-01 R -40.8 -33.5 -28.1 0 R
MS-02 D 38 27.2 100 100 D
MS-03 R -35.8 -26.2 -22.9 0.1 OPEN
MS-04 R -37.2 -42.6 -25.2 0 R
MT-AL R -15.7 -21.4 -4 21.3 R
NC-01 D 39.6 34.5 51.8 100 D
NC-02 R -13.4 -13 -3.9 23.7 R
NC-03 R -34.4 -24.8 -100 0 R
NC-04 D 36.4 35.2 49.9 100 D
NC-05 R -16.8 -19.3 -6.9 17.9 R
NC-06 R -18.4 -16.8 -7 18.3 R
NC-07 R -21.8 -19 -5 22.1 R
NC-08 R -17.6 -16.4 -6.3 20.8 R
NC-09 R -16.4 -14.2 -1.9 37 OPEN
NC-10 R -26.2 -25.5 -15.6 1.9 R
NC-11 R -28.2 -29.8 -18.6 0.6 R
NC-12 D 34 36.7 48.9 100 D
NC-13 R -12.2 -11.3 -1.8 36.7 R
ND-AL R -45.4 -34.8 -22.4 0 OPEN
NE-01 R -39 -22.7 -20.4 0 R
NE-02 R -1.2 -6 0.7 55.6 R
NE-03 R -53.2 -54.3 -44.9 0 R
NH-01 D 1.4 -3.4 8.4 93.9 OPEN
NH-02 D 4.4 1.7 17.8 99.9 D
NJ-01 D 23.2 23.7 36.6 100 D
NJ-02 R -22 -4 11.7 98.3 OPEN
NJ-03 R -20.4 -6.1 -0.6 45.2 R
NJ-04 R -30.2 -16 -14.7 2.4 R
NJ-05 D 4.4 -4.2 8.6 95 D
NJ-06 D 28.8 15.2 37.1 100 D
NJ-07 R -11 -3.3 3.7 76.4 R
NJ-08 D 58.5 52.5 71.5 100 D
NJ-09 D 41.7 30.2 51.6 100 D
NJ-10 D 73.8 70.9 88.8 100 D
NJ-11 R -19.1 -4.7 3 70.4 OPEN
NJ-12 D 30.9 30.9 44.5 100 D
NM-01 D 30.2 13.8 18 99.8 OPEN
NM-02 R -25.5 -11.9 -4 26.8 OPEN
NM-03 D 24.8 13.5 26.9 100 D
NV-01 D 33.1 27.5 37 100 D
NV-02 R -21.4 -13.8 -11.5 1.7 R
NV-03 D 1.2 -3.1 7.2 91 OPEN
NV-04 D 4 3.8 7.8 90.4 OPEN
NY-01 R -17.9 -11.7 -7.6 7.6 R
NY-02 R -24.9 -8.3 -8.1 14.2 R
NY-03 D 5.6 2.7 16.7 99 D
NY-04 D 19.1 8 27.8 100 D
NY-05 D 72.4 72.6 100 100 D
NY-06 D 45.4 31.4 100 100 D
NY-07 D 81.5 74.4 100 100 D
NY-08 D 76.5 70.5 100 100 D
NY-09 D 83.5 67.1 93.9 100 D
NY-10 D 56.1 54.2 70.5 100 D
NY-11 R -24.8 -8.8 -6.2 12.1 R
NY-12 D 66.3 63.7 81 100 D
NY-13 D 81.7 85.1 99.6 100 D
NY-14 D 65.7 56.5 72.7 100 OPEN
NY-15 D 91.7 87.6 100 100 D
NY-16 D 55 49 100 100 D
NY-17 D 19.9 16.4 100 100 D
NY-18 D 11.2 -2.9 18.3 99.6 D
NY-19 R -8.5 -6.5 3.3 74.1 R
NY-20 D 35.8 12.7 40.9 100 D
NY-21 R -35.2 -11.5 -13.8 1.1 R
NY-22 R -5.5 -14.3 1.5 61.4 R
NY-23 R -15.2 -14 -3.2 28.7 R
NY-24 R -21.1 4.1 -4.5 20.8 R
NY-25 D 12.3 14.6 19.9 100 D
NY-26 D 49.2 19.6 52.6 100 D
NY-27 R -34.4 -24 -8.2 6.8 OPEN
OH-01 R -18.4 -9 -3.8 24 R
OH-02 R -32.2 -17.4 -14.9 0.3 R
OH-03 D 37.2 36.5 50.9 100 D
OH-04 R -36 -31.3 -24.3 0 R
OH-05 R -41.8 -23.8 -25.5 0 R
OH-06 R -41.4 -37.6 -30.5 0 R
OH-07 R -35 -27.2 -15.8 0.1 R
OH-08 R -41.8 -34.8 -29.7 0 R
OH-09 D 37.4 23.3 46 100 D
OH-10 R -31.4 -8.5 -12.6 5.5 R
OH-11 D 60.6 61.7 76.4 100 D
OH-12 R -36.8 -13.7 -11.2 2.6 R
OH-13 D 35.4 9.2 39.3 100 D
OH-14 R -25.2 -12 -9.8 9.2 R
OH-15 R -32.4 -15.5 -16 1.6 R
OH-16 R -30.6 -17.1 -10.5 4.9 OPEN
OK-01 R -38 -32 -23.3 0 OPEN
OK-02 R -47.4 -49 -27.1 0 R
OK-03 R -56.6 -54 -38.9 0 R
OK-04 R -43.5 -39.2 -32.6 0 R
OK-05 R -20.3 -17.2 -10.6 2.5 R
OR-01 D 22.6 18.9 34.4 100 D
OR-02 R -43.7 -21.7 -19.3 0.1 R
OR-03 D 54.4 45.4 100 100 D
OR-04 D 15.8 -0.8 22.2 99.8 D
OR-05 D 10.5 1.5 19.5 99.6 D
PA-01 R -11.8 -0.7 1.7 61.4 R
PA-02 D 56.6 47.4 68.2 100 D
PA-03 D 82.2 81.4 98.5 100 D
PA-04 D 14.7 15.3 26.2 99.7 OPEN
PA-05 D 13.6 25.5 33.5 99.9 OPEN
PA-06 R -6 5.2 12.3 92.5 OPEN
PA-07 R -9.6 0 10.2 96.3 OPEN
PA-08 D 0.2 -6.8 12 99 D
PA-09 R -31.8 -32.1 -22.1 0 OPEN
PA-10 R -19.8 -10.9 -3.9 22.3 R
PA-11 R -31.6 -27.5 -12 1.2 R
PA-12 R -41.6 -36 -30 0 R
PA-13 R -38.2 -45.5 -27 0 OPEN
PA-14 R -45.2 -28.7 -25.1 0.2 OPEN
PA-15 R -37.7 -41.8 -29.6 0 R
PA-16 R -15 -18.8 -3.5 25.4 R
PA-17 R -11 -5.6 9 94.5 OPEN
PA-18 D 44.3 25.2 100 100 D
RI-01 D 29.4 25.2 41.4 100 D
RI-02 D 27.4 8.2 33.5 100 D
SC-01 R -21.8 -16.9 -10.4 4.6 OPEN
SC-02 R -24.3 -20.8 -12.5 5.2 R
SC-03 R -45.7 -38.7 -33.8 0 R
SC-04 R -36.2 -28.3 -22 0.8 OPEN
SC-05 R -20.5 -19.3 -9.4 12.2 R
SC-06 D 42.5 35.5 54.1 100 D
SC-07 R -22.1 -19.3 -10.5 9.9 R
SD-AL R -28.2 -29.4 -17.4 0.1 OPEN
TN-01 R -63 -57.1 -52.5 0 R
TN-02 R -51.2 -38.2 -34 0 OPEN
TN-03 R -37.6 -36 -27.3 0 R
TN-04 R -30 -41.5 -24.3 0.1 R
TN-05 D 25.2 14.6 34.5 100 D
TN-06 R -49.3 -49.1 -41.5 0 OPEN
TN-07 R -48.7 -40.2 -34.7 0 OPEN
TN-08 R -43.7 -37.6 -32 0 R
TN-09 D 59.8 55.1 73.3 100 D
TX-01 R -49.8 -48.8 -40.4 0 R
TX-02 R -24.6 -16.4 -9.2 7.9 OPEN
TX-03 R -26.6 -20.7 -13.3 5.9 OPEN
TX-04 R -61.1 -55.1 -50.5 0 R
TX-05 R -59.1 -31.4 -31.4 0 OPEN
TX-06 R -19.3 -16.1 -7.5 14.8 OPEN
TX-07 R -12.4 -6.8 -0.3 48 R
TX-08 R -71.8 -53 -56.9 0 R
TX-09 D 61.2 57.7 100 100 D
TX-10 R -18.9 -14.5 -4.8 22.2 R
TX-11 R -67.5 -61.5 -57.3 0 R
TX-12 R -42.5 -34 -30.4 0 R
TX-13 R -69 -65.2 -59.7 0 R
TX-14 R -23.8 -22.4 -12.8 5.5 R
TX-15 D 19.6 14 30.5 100 D
TX-16 D 42.1 35.4 49.7 100 OPEN
TX-17 R -25.6 -21.4 -14.1 1.3 R
TX-18 D 49.9 53.2 65.9 100 D
TX-19 R -52.8 -51.5 -43.5 0 R
TX-20 D 54.3 22.3 100 100 D
TX-21 R -20.6 -15.5 -8.6 6.7 OPEN
TX-22 R -19 -14.8 -6.6 18.3 R
TX-23 R -1.3 -0.7 -2.1 37.4 R
TX-24 R -16.9 -12.8 -4.5 27.3 R
TX-25 R -20.6 -19.3 -9.4 10.1 R
TX-26 R -36.8 -31.7 -25.1 0 R
TX-27 R -23.4 -25.8 -13.8 4.6 R
TX-28 D 34.9 17.7 100 100 D
TX-29 D 48.5 40 55.1 100 OPEN
TX-30 D 66.5 58.1 100 100 D
TX-31 R -21.9 -17.4 -10.5 2.6 R
TX-32 R -16.4 -5 0.4 52.7 R
TX-33 D 47.4 45.6 61.3 100 D
TX-34 D 25.4 19.2 36.4 100 D
TX-35 D 31.5 29.8 44.5 100 D
TX-36 R -52.4 -49.5 -42.5 0 R
UT-01 R -39.5 -37.3 -31.2 0 R
UT-02 R -27.7 -22.8 -14.3 0.4 R
UT-03 R -47 -35.2 -36.6 0 R
UT-04 R -12.5 -16.8 -1.2 40.7 R
VA-01 R -23.3 -15 -9.6 9.9 R
VA-02 R -22.8 -5.7 -4.2 21.9 R
VA-03 D 33.6 30.2 100 100 D
VA-04 D 15.7 19.1 29.8 99.9 D
VA-05 R -16.6 -12.8 -3.2 29.1 OPEN
VA-06 R -33.5 -26.2 -19.6 1.4 OPEN
VA-07 R -15.3 -10.2 -0.6 45.9 R
VA-08 D 41.1 45.8 57.1 100 D
VA-09 R -40.3 -40.8 -23.8 0 R
VA-10 R -5.8 4.6 9.2 96.2 R
VA-11 D 30.2 33.4 45.1 100 D
VT-AL D 35.1 26.2 41.5 100 D
WA-01 D 10.8 12.4 24.1 99.9 D
WA-02 D 28 19.3 100 100 D
WA-03 R -23.6 -8.5 -3.5 21.1 R
WA-04 R -31.1 -25.1 -18.7 0.9 R
WA-05 R -19.2 -14.8 -3.6 20.5 R
WA-06 D 23 10.4 32.5 100 D
WA-07 D 65.9 65.2 81.3 100 D
WA-08 R -20.4 0.1 2.5 71.8 OPEN
WA-09 D 45.8 42.5 100 100 D
WA-10 D 17.4 9.8 29.1 100 D
WI-01 R -34.7 -11.3 -4.1 23.2 OPEN
WI-02 D 37.5 34.5 100 100 D
WI-03 D 10.2 -3.2 17.6 99.1 D
WI-04 D 51.7 49.5 65.7 100 D
WI-05 R -37.4 -23.5 -22.4 0.1 R
WI-06 R -19.9 -17.1 -8.9 4.2 R
WI-07 R -23.4 -18.7 -11.1 6.5 R
WI-08 R -25.3 -16.7 -11.6 5.7 R
WV-01 R -38 -40.4 -30.9 0 R
WV-02 R -16.4 -35.4 -11.1 1.7 R
WV-03 R -43.9 -47.5 -16 0.3 OPEN
WY-AL R -32 -48.4 -27.8 0 R

Who’s Retiring?


  • Democrats

    District Partisan Lean Dem Win Prob.
    AZ-09 D+ 11% 100
    CO-02 D+ 18% 100
    HI-01 D+ 32% 100
    IL-04 D+ 65% 100
    MA-03 D+ 18% 100
    MA-07 D+ 68% 100
    MD-06 D+ 13% 100
    MI-13 D+ 61% 100
    MN-05 D+ 51% 100
    NM-01 D+ 14% 100
    NY-14 D+ 56% 100
    PA-04 D+ 15% 100
    PA-05 D+ 26% 100
    TX-16 D+ 35% 100
    TX-29 D+ 40% 100
    MI-09 D+ 7% 99
    CT-05 D+ 3% 98
    NH-01 R+ 3% 94
    NV-03 R+ 3% 91
    NV-04 D+ 4% 90
    MN-01 R+ 13% 43
    MN-08 R+ 13% 32
  • Republicans

    District Partisan Lean Dem Win Prob.
    CA-49 D+ 1.4% 98.3
    NJ-02 R+ 4% 98.3
    PA-07 D+ 0% 96.3
    AZ-02 D+ 0.7% 95.9
    PA-17 R+ 5.6% 94.5
    FL-27 D+ 13.9% 93.7
    PA-06 D+ 5.2% 92.5
    MI-11 R+ 7.2% 77.1
    WA-08 D+ 0.1% 71.8
    NJ-11 R+ 4.7% 70.4
    CA-39 D+ 3% 55.7
    NC-09 R+ 14.2% 37
    FL-15 R+ 11.5% 35.1
    VA-05 R+ 12.8% 29.1
    NM-02 R+ 11.9% 26.8
    FL-06 R+ 16.7% 23.8
    WI-01 R+ 11.3% 23.2
    TX-06 R+ 16.1% 14.8
    KS-02 R+ 19.8% 13.1
    TX-02 R+ 16.4% 7.9
    NY-27 R+ 24% 6.8
    TX-21 R+ 15.5% 6.7
    TX-03 R+ 20.7% 5.9
    OH-16 R+ 17.1% 4.9
    SC-01 R+ 16.9% 4.6
    FL-17 R+ 27% 1.9
    VA-06 R+ 26.2% 1.4
    SC-04 R+ 28.3% 0.8
    WV-03 R+ 47.5% 0.3
    IN-04 R+ 34.1% 0.2
    PA-14 R+ 28.7% 0.2
    MS-03 R+ 26.2% 0.1
    SD-AL R+ 29.4% 0.1
    ID-01 R+ 39.5% 0
    IN-06 R+ 38.6% 0
    ND-AL R+ 34.8% 0
    OK-01 R+ 32% 0
    PA-09 R+ 32.1% 0
    PA-13 R+ 45.5% 0
    TN-02 R+ 38.2% 0
    TN-06 R+ 49.1% 0
    TN-07 R+ 40.2% 0
    TX-05 R+ 31.4% 0


Full Spreadsheet


Note: the projections listed in this spreadsheet are old ad-hoc measures of partisanship and possible swing. Ultimately, the following is a “sandbox.” If you’re looking for a spreadsheet of forecasts from the model output, use the download links at the bottom of the page.


Download Model Output


The Latest 2018 House News

  • September 8, 2018: I made a modeling adjustment to ensure that district-level polls fielded by partisan outlets get less weight in the model than nonpartisan, public polls.
  • July 20, 2018: The House model now incorporates district-level polling when predicting the seat level outcomes. This step can help account for weak candidacy, a redder-than-predicted downballot environment in a given state or region, etc. In the coming weeks, I’ll publish a short piece with more details on how and why this makes a difference, but know that today it increased the Democrats’ chance of winning the election from 74 to 78 percent.
  • July 11, 2018: I corrected an error in the forecast that was adding too much uncertainty in the simulation stage of the simulation model. The chance that Democrats win the majority has increased accordingly. Read more about that here.
  • June 30, 2018: The forecast has been updated to assign automatic winners in the uncontested races from New York. (They were all already rated Safe, so the 100% win probability now does not change much)
  • June 9, 2018: The forecast has been updated to take into account uncontested seats from this week’s primary elections. Democrats are certain winners in 17 seats, whereas (as of currently counted votes in California) Republicans are certain to win 2.
  • June 1, 2018: Model v3.0 (the final version?) officially released. Changes: (1) an updated map for the new Pennsylvania congressional district plan; (2) the incorporation of special election polling in the model’s final stage of predicting the House popular vote; (3) a smoother averaging technique that draws the most predictive (based on back-tested historical forecasting) trend line between polls for each week in the cycle.
  • May 27, 2018: My future data journalist colleagues at The Economist “>released their own model predicting the midterm elections. They have Democrats as 2-in-3 favorites for winning control of the chamber.
  • May 8, 2018: The incumbent Republican congressman for North Carolina’s ninth congressional seat lost election in a primary challenge today. The seat moves to open for 2018.
  • April 14, 2018: Democrats came within 4 points of flipping control of Arizona’s eight congressional district in a special election to this Trump +22 seat.
  • April 11, 2018: House Speaker Paul Ryan announced that he will not seek reelection to the House come November. His seat moves to roughly Tossup odds in the upcoming midterm elections.
  • March 23, 2018: Democrat Conor Lamb scored a 20-point upset in the special election to Pennsylvania’s 18th congressional district, adding to the chance that they take back the House majority in the fall.
  • March 7, 2018: Democrats had a strong, but possibly unsatisfactory, showing in the Texas first-of-the-cycle primaries last night. This comes after their massive early voting performance that some interpreted as a blue wave washing over the state. And though Republicans cast 500,000 more ballots in their U.S. Senate primary, a pickup is still very possible for Democrats.
  • March 1, 2018: Democrats are bouncing back in generic ballot polls, and right about the time history suggested it would happen.
  • Feb, 19, 2018: The Pennsylvania Supreme Court redrew congressional district lines today after it decided last month that the current plan, drawn in 2011, violated the state constitution and was a partisan gerrymander. The new map shifts the state’s seats toward Democrats by 2-3 seats and increases their odds of winning a House majority by roughly 4%. My piece on the map.
  • Feb, 13, 2018: Democrats flipped more state legislative districts tonight — and saw losses in another — putting their average overperformance of Hillary Clinton’s 2016 vote margins at 12%. If these data predict November’s outcome this year as well as they have in the past, Democrats will be strongly favored to win back the House majority. The question is, will that last?
  • Feb, 9, 2018: Democratic Representative Rick Nolan from Minnesota’s eighth congressional district announced today that he will not seek reelection in November. MN-08 joins MN-01 as the two Democratic seats most likely to flip red this fall. Both voted for President Trump by double-digit margins in 2016 after voting for President Obama over Mitt Romney in the 2012 presidential election
  • Feb, 7, 2018: Democrats in Missouri last night flipped a state legislative district that voted for Donald Trump in the 2016 presidential election by a 28-point margin, adding to a long list of special elections that have swung heavily towards Democrats over the past one and a half years. Of course, each special election alone is not informative, but in aggregate the double-digit shift towards Democrats could be indicative of a large wave to come this fall. The question is, are these special elections overstating possible Democratic gains? I think they may be, and historically, the generic ballot — which currently forecasts a 10-point, not 16-point in the case of special election results, race — is a better predictor of Election Day outcomes.

    Update: My takeaway from these special elections (with all the usual caveats about their unrepresentativeness) continues to be that backlash to Trump and Republicans is focused in Republican districts. If this pattern holds true in the November midterms, it could push our forecasts of Democratic seats up as much as 15 seats.
  • Feb, 5, 2018: The Supreme Court of the United States rejected an appeal by Pennsylvania Republicans to “stay” orders by the PA Supreme Court to redraw the lines of their U.S. House districts, which last month had been declared a unconstitutional partisan gerrymander. With no other recourse, the PA GOP now turns to drawing fair maps for the 2018 elections — maps which are likely to move 1-2 currently GOP seats into Democrats’ column in November.
  • Feb, 2, 2018: The New York Times Upshot’s Nate Cohn argues today that your evaluation of the national environment will largely depend on how conservative you are when averaging generic ballot polls, especially this far out from November. This is something I have been advocating for some time; this far away from Election Day, a 30-day average is much more predictive than a two-week average.
  • Jan. 31, 2018: A Monmouth University Poll finds Democrats up just 2% in the generic House ballot. That’s a large decrease from their D+15% number last month, telling us that even if the 2-point Democratic lead in this poll is underestimating the opposition party, there has been at least some amount of movement towards the Republicans since 2017.
  • Jan. 31, 2018: Democratic Representative Bob Brady from Pennsylvania’s 1st Congressional District announced his retirement today. His district, situated in the heart of Philadelphia, has a partisan lean of nearly D+60%. It is considered Safe Democratic by most handicappers (at least, those who even bother to give it a rating).
  • Jan. 31, 2018: Republican Congressman Trey Gowdy (SC-04), chair of the powerful House Oversight Committee, announced today that he’s retiring from office. He summarised his statement on twitter, writing “There is a time to come and a time to go. This is the right time, for me, to leave politics and return to the justice system.” Gowdy’s seat is considered Safe for Republicans.
  • Jan. 30, 2018: CNN just released their initial ratings of competitive 2018 House seats. 36 Republicans and 13 Democrats are in the Lean or Toss-up categories.
  • Jan. 29, 2018: GOP Rep. Rodney Frelinghuysen (NJ-11) won’t seek reelection in 2018. His district leans 5% towards Republicans and is now rated as a Toss-up district by most handicappers. The forecast model, updated for his absence, says Democrats have a 65% shot at flipping the district in November.