Decision Desk provides an early view of the 2018 Congressional election.
The 2018 House Midterm Election is bound to be one of the more interesting in recent memory. With Donald Trump in the White House, infighting on both sides of Congress, and an American public that is bursting at the seams we have a recipe for a perfect political storm. Keep your eye on this page, which houses our forecasts for all 435 congressional districts, and stick with us as we attempt to answer the ultimate questions: who will win majority control of the US House of Representatives?
[…]
The Democratic Party is ahead in generic ballot polls up 7.1% in our average. They hold an 8.2 percentage point lead in our projection of the election day two-party vote.. We get all of our polling data from Huffington Post Pollster, which you can investigate here.
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?
Democrats earn a median of 218 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.
See here for ratings of individual races, and here for an explanation of the methodology. Note that latter entry is from August, when Dems had about a four percent lead in the generic Congressional ballot, and the model predicted a gain of nine seats, well below the amount needed to retake the majority. Things have improved considerably for them since then, and it shows up in the probabilistic model for each district. Farther down in the original link above is a table highlighting the relevant data and odds of a D victory in each district. I’ve cut out the relevant info for Texas. Feast your eyes:
| District | Dem 2016/14 (%) | Clinton (%) | Forecast Dem 2018 (%) | Dem Win Prob. |
| TX-02 | 37.3 | 45.1 | 49.4 | 45.8 |
| TX-03 | 36.1 | 42.6 | 46.9 | 29.6 |
| TX-06 | 40.1 | 43.6 | 44.9 | 15.0 |
| TX-07 | 43.8 | 50.7 | 49.6 | 46.3 |
| TX-10 | 40.1 | 45.2 | 45.6 | 18.6 |
| TX-14 | 38.1 | 39.8 | 42.4 | 6.1 |
| TX-17 | 36.7 | 40.8 | 42.2 | 5.7 |
| TX-21 | 39 | 44.7 | 49.0 | 43.4 |
| TX-22 | 40.5 | 45.9 | 46.0 | 20.9 |
| TX-23 | 49.3 | 51.8 | 52.4 | 69.2 |
| TX-24 | 41.2 | 46.7 | 46.7 | 24.9 |
| TX-25 | 39.3 | 42.2 | 43.9 | 11.0 |
| TX-27 | 38.3 | 37.8 | 41.6 | 4.5 |
| TX-31 | 38.5 | 43.3 | 44.0 | 11.3 |
| TX-32 | 36.4 | 51.0 | 46.4 | 23.1 |
| TX-36 | 22.5 | 25.9 | 29.6 | 1.0 |
Kind of amazing, isn’t it? One Dem takeover favored, three tossups, and four more seats for which the odds are around one in four. That was before the Joe Barton nude photos scandal, and who knows what effect that could have. CD02 is rated much more highly as a pickup opportunity than CD32, likely due to Ted Poe’s retirement. As the authors take pains to note, this kind of forecast provides a range of outcomes, and some amount of error is to be expected. Such errors are likely to go exclusively in one direction, and things can change quickly. We’ll need to keep an eye on this going forward – I expect there will be updates about once a quarter – but if there’s a main takeaway, it’s that we really need good candidates in every race. We have them in most districts, but there are a few that could still use an upgrade. There’s a ton of opportunity here, we need to be in a position to grab it.
I would be surprised if CD 2 flips to the Democrats. The Clinton vote is a misleading indicator of Democrat support. Many Republicans voted for Clinton and pretty much a straight R ticket in sectors of the Heights, Montrose, Rice University areas which form part of the district.