Who will control the House of Representatives?

On November 6th, American voters will elect a majority of either Republicans or Democrats to Congress, after nearly 8 years of Republicans controlling the House. In our statistical model, we forecast which party will control the House. We factor in national generic ballot polling, special election results, the partisan lean of all of the nation's 435 districts, and whether each seat's incumbent is running to form a comprehensive and accurate 2018 House forecast. Our forecast fluctuates on a day-to-day basis as we update it for newly-vacated seats, special elections, and polls. We estimate that Democrats will win a 11-seat majority, 207 R - 228 D, and gain 34 seats.

Last updated: Monday 05/27/2019

Probability of winning House majority

DemocratsRepublicans

64.1%▾9.5% over past month
35.9%

Projected seat count207 R - 228 D (▴34)

218 seats are needed to win a House majority.

The current partisan breakdown of seats is 241-194.

Democrats are projected to make a net gain of 34 seats. They need to flip 24 seats for a House majority.

The current partisan breakdown of seats is 241-194.

Democrats are projected to make a net gain of 34 seats. They need to flip 24 seats for a House majority.

435

seats

207-228

seats

207-228

●Republicans207 (▾34)

●Democrats228 (▴34)

●Other0 (▴0)

75% Prediction Interval (Margin of Error)± 17 seats

Accounting for every congressional district, its incumbent's strength, and whether the seat is open or not, we found that Democrats have to win the popular vote by 5.8 points in order to be favored to capture the House majority and overcome significant Republican geographic advantages in the structuring of districts and gerrymandering.

We take into account the win probabilities associated with each seat in order to generate a nationwide estimate of the most likely seat distribution for a given popular vote margin. In contrast to this probabilistic approach, we also count seat-by-seat which party our model predicts to win, creating a more favorable estimate for the Republicans, available on our House ratings page.

We take into account the win probabilities associated with each seat in order to generate a nationwide estimate of the most likely seat distribution for a given popular vote margin. In contrast to this probabilistic approach, we also count seat-by-seat which party our model predicts to win, creating a more favorable estimate for the Republicans, available on our House ratings page.

Projected popular voteD+8.3 pts.

●Republicans44.6%

●Democrats52.8%

●Other2.6%

Popular vote probabilities89.1% D win

Win probabilities over time

Based on our popular vote estimate of D+8.3 points (the average of the latest generic ballot polling and the shift in special election results) and the margin that Democrats have to win to garner a majority of seats, we currently calculate that Democrats have a 64.1% chance of winning the majority of House of Representatives seats (≥218) and Republicans have a 35.9% chance.

75%

Democratic odds 50%

majority win chance 25%

Democratic odds 50%

majority win chance 25%

Current probability of Democratic majority

64.1%

64.1%

Election day

Nov. 6 →

Nov. 6 →

3/27/18

Mar

Apr

May

June

July

Aug

Sept

Oct

Nov

Generic ballot polling average over time

As of today, our latest moving average of polls that ask people which party they would vote for in Congress shows 49.2% indicating a preference for Democrats and 42% for Republicans. For our polling average, we rely on Huffington Pollster's LOESS computations of the trendline, as well as their aggregation of polls, where they only include polls that follow AAPOR standards.

We use the average of the current generic ballot polling (D+7.2) and the partisan lean of special election results (Special Elections Index metric calculated by Daily Kos Elections; D+9.4) to predict the national popular vote (D+8.3) as it will be on November 7th, which we base all of our seat-by-seat predictions on.

Democrats currently lead the generic ballot by 7.2 points, and we estimate that they will win the final popular vote by 8.3 points. Simply put, our popular vote estimate is the average of the latest generic ballot polling and the partisan lean in special election results, which has proved to be very predictive in past elections - moreso than any one of those two measures taken on their own. Probabilistic data, including the prediction error of this popular vote estimate, is presented above. We generate the uncertainty of the popular vote estimate based on past polling error at this stage of the election cycle. As election day approaches, polls tend to become more accurate, and so the level of uncertainty surrounding our predictions decreases.

We use the average of the current generic ballot polling (D+7.2) and the partisan lean of special election results (Special Elections Index metric calculated by Daily Kos Elections; D+9.4) to predict the national popular vote (D+8.3) as it will be on November 7th, which we base all of our seat-by-seat predictions on.

Democrats currently lead the generic ballot by 7.2 points, and we estimate that they will win the final popular vote by 8.3 points. Simply put, our popular vote estimate is the average of the latest generic ballot polling and the partisan lean in special election results, which has proved to be very predictive in past elections - moreso than any one of those two measures taken on their own. Probabilistic data, including the prediction error of this popular vote estimate, is presented above. We generate the uncertainty of the popular vote estimate based on past polling error at this stage of the election cycle. As election day approaches, polls tend to become more accurate, and so the level of uncertainty surrounding our predictions decreases.

80%
40%
0%

Moving average over 14-day period

D+7.2

D+7.2

Election day

Nov. 6 →

Nov. 6 →

3/27/18

Mar

Apr

May

June

July

Aug

Sept

Oct

Nov

Show generic ballot polls +Expand +

Distribution of polls since May

Below is visualized the relative frequency of polling margins in the generic ballot since May, where taller bars represent more common polling margins. With a sample size of N = 76, the value for the mode, or the most common margin, is {5}.

The simple average of all polls conducted from May on is D+6.4 points.

Of polls conducted from May on, 63.2% have had the Democrats with a lead of 5.8 points or above, which is the margin we estimate that they have to win the popular vote by to have a majority probability of winning the House.

The simple average of all polls conducted from May on is D+6.4 points.

Of polls conducted from May on, 63.2% have had the Democrats with a lead of 5.8 points or above, which is the margin we estimate that they have to win the popular vote by to have a majority probability of winning the House.

Republican win margins →
← Democratic win margins
25% relative

frequency

frequency

R+15

R+14

R+13

R+12

R+11

R+10

R+9

R+8

R+7

R+6

R+5

R+4

R+3

R+2

R+1

0

D+1

D+2

D+3

D+4

D+5

D+6

D+7

D+8

D+9

D+10

D+11

D+12

D+13

D+14

D+15

Polling margins

Probability distribution of seats in the House of Representatives

Below is visualized how likely each distribution of seats Democrats and Republicans will could end up with in the 2018 House of Representatives election are. We ran simulations to determine where the House election will end up in terms of seats, and the probability of each outcome. In this distribution of each party's number of seats won, taller bars occur more often.

4.4% chance
2.0%
Republicans win majority of House seats →
← Democrats win majority of House seats

50
218 seats (majority)
330

Republican party's share of House seats

Seats expected to flip parties (relative to 2016 outcome)

Here we feature every district which we are forecasting to have an outcome different than its 2016 result. These tend to be some of the most hotly contested and closely watched elections. "2016 House margin" refers to the seat's 2016 congressional election margin, most often the latest election in the district and typically from the seat's incumbent. A seat status of "Open" means that the incumbent in the district is not running for re-election.

View our complete House ratings, which we keep update-to-date every day.

View our complete House ratings, which we keep update-to-date every day.

AZ-2

D+3.6

D+3.6

CA-10

D+7.5

D+7.5

CA-21

D+10.8

D+10.8

CA-25

D+6.6

D+6.6

CA-39

D+5

D+5

CA-49

D+6.9

D+6.9

CO-6

D+9.5

D+9.5

FL-26

D+12.7

D+12.7

FL-27

D+24.9

D+24.9

IA-1

D+7.4

D+7.4

IA-3

D+2.7

D+2.7

ME-2

D+2.8

D+2.8

MN-2

D+5.4

D+5.4

MN-3

D+5.4

D+5.4

NE-2

D+2.9

D+2.9

NJ-2

D+1.6

D+1.6

NJ-7

D+1

D+1

NY-19

D+3.2

D+3.2

NY-24

D+5.6

D+5.6

PA-1

D+6.3

D+6.3

PA-5

D+34.3

D+34.3

PA-6

D+12.3

D+12.3

PA-7

D+10.3

D+10.3

PA-17

D+2.3

D+2.3

TX-23

D+6.9

D+6.9

VA-10

D+8

D+8

WA-8

D+3.5

D+3.5

MN-7

R+5.7

R+5.7

27 seats won by Republicans in 2016 are expected to vote Democrat

96.4% of seats (27 out of 28) expected to switch parties from 2016 are forecasted to be won by Democrats.

State |
District # |
Forecast margin |
2016 House margin |
Incumbent |
Seat status |
---|---|---|---|---|---|

Arizona | 2 | D+3.6 | R+14 | Martha McSally | Open |

California | 10 | D+7.5 | R+3.4 | Jeff Denham | |

California | 21 | D+10.8 | R+13.4 | David Valadao | |

California | 25 | D+6.6 | R+6.2 | Steve Knight | |

California | 39 | D+5 | R+14.4 | Ed Royce | Open |

California | 49 | D+6.9 | R+0.6 | Darrell Issa | Open |

Colorado | 6 | D+9.5 | R+8.3 | Mike Coffman | Open |

Florida | 26 | D+12.7 | R+11.8 | Carlos Curbelo | |

Florida | 27 | D+24.9 | R+9.8 | Ileana Ros-Lehtinen | Open |

Iowa | 1 | D+7.4 | R+7.6 | Rod Blum | |

Iowa | 3 | D+2.7 | R+13.7 | David Young | |

Maine | 2 | D+2.8 | R+9.6 | Bruce Poliquin | |

Minnesota | 2 | D+5.4 | R+1.8 | Jason Lewis | |

Minnesota | 3 | D+5.4 | R+13.8 | Erik Paulsen | |

Nebraska | 2 | D+2.9 | R+1.2 | Don Bacon | |

New Jersey | 2 | D+1.6 | R+22 | Frank LoBiondo | Open |

New Jersey | 7 | D+1 | R+11 | Leonard Lance | |

New York | 19 | D+3.2 | R+8.2 | John Faso | |

New York | 24 | D+5.6 | R+21.2 | John Katko | |

Pennsylvania | 1 | D+6.3 | R+8.8 (Old equivalent) | Open | |

Pennsylvania | 5 | D+34.3 | R+19 (Old equivalent) | Open | |

Pennsylvania | 6 | D+12.3 | R+14.4 (Old equivalent) | Open | |

Pennsylvania | 7 | D+10.3 | R+20.4 (Old equivalent) | Open | |

Pennsylvania | 17 | D+2.3 | R+23.6 (Old equivalent) | Open | |

Texas | 23 | D+6.9 | R+1.3 | Will Hurd | |

Virginia | 10 | D+8 | R+5.8 | Barbara Comstock | |

Washington | 8 | D+3.5 | R+20.4 | Dave Reichert | Open |

1 seats won by Democrats in 2016 are expected to vote Republican

3.6% of seats (1 out of 28) expected to switch parties from 2016 are forecasted to be won by Republicans.

State |
District # |
Forecast margin |
2016 House margin |
Incumbent |
Seat status |
---|---|---|---|---|---|

Minnesota | 7 | R+5.7 | D+5 | Collin Peterson |

Frequency of forecasted House margins

Below is shown the win margins of seats forecasted in the 2018 House of Representatives election for every district with a predicted outcome between 30 points.

● Seat that was won by a Republican in 2016 that is expected to be competititive this year (within 12 points).

● Seat that was won by a Republican in 2016 that is expected to be solidly partisan this year (12-30 points).

● Seat that was won by a Democrat in 2016 that is expected to be competititive this year (within 12 points).

● Seat that was won by a Democrat in 2016 that is expected to be solidly partisan this year (12-30 points).

● Seat that was won by a Republican in 2016 that is expected to be competititive this year (within 12 points).

● Seat that was won by a Republican in 2016 that is expected to be solidly partisan this year (12-30 points).

● Seat that was won by a Democrat in 2016 that is expected to be competititive this year (within 12 points).

● Seat that was won by a Democrat in 2016 that is expected to be solidly partisan this year (12-30 points).

Frequency
Seats won by Republicans in 2016 →
← Democrats win majority of House seats

D+30
0
R+30

Forecasted margin of House seats

Most competitive races (seats rated as "Tossups")

Here we feature every district which we are forecasting to have an outcome within a 4-point margin - the 28 closest races, and the ones most likely to decide the control of the House. The probability of the forecasted winner losing each of the seats listed here is always at least 33%. "Last House margin" refers to last 2016 congressional election margin from the seat's incumbent. A seat status of "Open" means that the incumbent in the district is not running for re-election.

When applied to the 2014 and 2016 House elections, our model correctly predicted the winners of 68.75% of seats (32 out of 48) rated as "Tossups". For the rest of our House predictions, the accuracy rate is 98.42%.

View our complete House ratings, which we keep update-to-date every day.

When applied to the 2014 and 2016 House elections, our model correctly predicted the winners of 68.75% of seats (32 out of 48) rated as "Tossups". For the rest of our House predictions, the accuracy rate is 98.42%.

View our complete House ratings, which we keep update-to-date every day.

Frequency
Seats forecasted to be won by Republicans →
← Seats forecasted to be won by Democrats

AZ-2

IA-3

ME-2

MN-8

NE-2

NY-19

WA-8

NJ-2

PA-17

MN-1

NJ-7

IL-6

KS-3

CA-45

MI-11

NJ-3

NJ-11

NY-22

CA-48

FL-18

IL-13

MI-8

TX-32

IL-12

NY-1

NY-2

NC-13

VA-2

CO-3

FL-15

IL-14

MI-6

NY-11

NY-23

OH-1

PA-10

VA-7

D+4
0
R+4

Forecasted margin of House seats

State |
District # |
Forecast margin |
Last House margin |
Incumbent |
Seat status |
---|---|---|---|---|---|

Arizona | 2 | D+3.6D | R+14 | Martha McSally | Open |

California | 45 | R+1.1R | R+17.2 | Mimi Walters | |

California | 48 | R+2.2R | R+16.6 | Dana Rohrabacher | |

Florida | 18 | R+1.5R | R+10.5 | Brian Mast | |

Illinois | 6 | R+0.2R | R+18.4 | Peter Roskam | |

Illinois | 12 | R+2.9R | R+14.6 | Michael Bost | |

Illinois | 13 | R+1.8R | R+19.4 | Rodney L. Davis | |

Iowa | 3 | D+2.7D | R+13.7 | David Young | |

Kansas | 3 | R+0.3R | R+10.7 | Kevin Yoder | |

Maine | 2 | D+2.8D | R+9.6 | Bruce Poliquin | |

Michigan | 8 | R+2.3R | R+16.8 | Mike Bishop | |

Michigan | 11 | R+0.6R | R+12.7 | Dave Trott | Open |

Minnesota | 1 | D+1.3D | D+0.8 | Tim Walz | Open |

Minnesota | 8 | D+2.7D | D+0.6 | Rick Nolan | Open |

Nebraska | 2 | D+2.9D | R+1.2 | Don Bacon | |

New Jersey | 2 | D+1.6D | R+22 | Frank LoBiondo | Open |

New Jersey | 3 | R+0.8R | R+20.4 | Tom MacArthur | |

New Jersey | 7 | D+1D | R+11 | Leonard Lance | |

New Jersey | 11 | R+0.7R | R+19.1 | Rodney Frelinghuysen | Open |

New York | 1 | R+3.5R | R+16.4 | Lee M. Zeldin | |

New York | 2 | R+3.4R | R+24.2 | Peter T. King | |

New York | 11 | R+3.7R | R+24.9 | Daniel M. Donovan Jr. | |

New York | 19 | D+3.2D | R+8.2 | John Faso | |

New York | 22 | R+1.2R | R+5.4 | Claudia Tenney | |

North Carolina | 13 | R+3.4R | R+12.2 | Ted Budd | |

Pennsylvania | 10 | R+3.7R | (New court-ordered district) | Open | |

Pennsylvania | 17 | D+2.3D | (New court-ordered district) | Open | |

Texas | 32 | R+1.7R | R+100 | Pete Sessions | |

Virginia | 2 | R+3R | R+23 | Scott Taylor | |

Washington | 8 | D+3.5D | R+20.4 | Dave Reichert | Open |

Probability of various likely scenarios

Republican majority of seats35.9%

Democratic majority of seats64.1%

Republican popular vote win10.9%

Democratic popular vote win89.1%

About this forecast and attribution

Learn how we created this forecast, and our methodology. Credit goes out to the Daily Kos for the hexagonal congressional map SVG we used. Follow @plural_vote on Twitter for daily updates and subscribe to our newsletter for ocassional ones.

Email us to report any problem you notice in our forecast.

Email us to report any problem you notice in our forecast.