We’ve been hearing for the last few months that the House of Representatives going Democrat is, for all practical purposes, a done deal. Pundit after pundit, and actual analysts, have sounded off on this note again and again. So, it must be true, right?
*Caveat: I am writing this at a moment in time. I am not predicting. There will be a lot of shifting in the polls over the next few months, and that’s expected. What I’m doing here is essentially a “If the Election Were Held Today” sort of analysis/observation. Nothing more.
Well, RCP started publishing their pre-election maps. It is early, but it doesn’t say what other people are saying. But, before we go to the maps, let’s set the table a bit.
Because of quirks in the polling models in off year elections, it is generally assumed that a 3% Dem lead in the generic congressional poll means that the makeup of the House will stay roughly the same. I truly wish I could find a formal citation for this (maybe there’s one on 538, if I find it, I’ll append it) but this has been my understanding for at least the last 20 years or so.
At any rate, late last year, the Dems were whomping in the generic congressional poll. That’s tightened markedly this year, with the trend curve looking like this:
See that huge gap at Christmastime? That was a 13.8% differential in favor of the Dems. At that figure, the Dems were a sure bet to not just take the House, but do a 2006-style dance on the GOP’s head.
But, times changed. Trump’s tax bill passed at Christmas time, and slowly , through January and February, as the electorate figured out how much shit the Democrats were shoveling regarding the tax plan (never tell a working person that $100 a month is a “crumb”), the lead narrowed, and has been hovering around the 6–8% level since, recently dropping down to the current 5.8%.
So, that’s a pretty clear indicator that the whomping that the Dems were fixin’ to put on the GOP has dropped off quite a bit. But is that still enough to win the House? Fortunately, RCP provides some guidance there, as well:
What we see from the above is that the Dems have 201 safe or leaning seats based on the current polls, while the GOP have 204. That’s pretty close. 30 seats are, at this point, too close to call; they’re in the tossup category. Some observations on that:
- Obviously, if the statistics hold up, and the two parties split those seats right down the middle, the GOP holds the House, 219–217. The Dems have to win 60% of the tossup seats to take the House, assuming all other seats stay in their categories (they won’t, of course).
- However, you quickly note something else about those tossup seats: 26 of them are currently held by Republicans. That tells you a little bit more about the hill the Dems have to climb. Not only do they need to take 60% of the tossup seats, but they have to overcome (a) the advantage of incumbency, in most cases (some of those seats are retirements) but also (b) the tendency of the disgruntled GOP partisan voter to “come home” in the last few days before an election.
- There’s also another wildcard in play, that being that this early prior to an election, all the pollsters are using a “Registered Voter” model for their numbers. The only pollster that uses the “Likely Voter” model (where the pollster doesn’t just ask if the citizen is registered to vote, but asks them questions to determine how LIKELY it is that they’ll vote) this early is Rasmussen; however, there is no Rasmussen data included in the above map. When the pollsters move to “Likely Voter”, there is a notable tendency for the polls to shift GOP, especially in an mid-term year, when the GOP voters are more dependable than the Dems.
All told, and in my opinion, the Dems don’t take the House under the above conditions. They need to shift……probably about 10 seats from “toss-up” to “leans Dem” by election day (and shift a few from “Likely GOP” to “toss-up” as well) to have a reasonable shot at actually winning. (And, it doesn’t help them that the Pennsylvania district they just won is part of the PA Court’s redraw…..and that district, which Conor Lamb won by a nose hair, is about to get a lot redder.)
So, considering all that’s happened….this map doesn’t look so bad for the GOP. They’re going to lose seats; but if the generic poll stays at about the 6% level, they’re still likely to hold the House, it appears.
How about the Senate?
Well, the Senate was never really up for grabs in 2018; it’s a shitty map for the Democrats, who have to defend far more seats than the GOP:
My first thought, looking at this map, is that its the SENATE is catching just as much, and perhaps more, of the guilt-by-association that’s tied to Mr. Trump. With a normal president, the GOP could be looking at 56, even 58 Senate seats.
But, apparently not in this case; they have less than 50 safe and leaning seats. The Dems would have to run the table on the toss-ups to win the Senate, and ….. that’s not gonna happen, folks. Every toss-up Senate seat, save Nevada, is in a red state; West Virginia is one of the two or three reddest states in the country; and all those factors that I listed above still apply. With this map, the GOP are (again) going to underperform in the Senate, picking up perhaps 2–3 seats over their current 51.
One of the most interesting thing about this map, when it comes to narratives, is the position of Sen. Cruz in “Likely GOP”. If you read the left-leaning sites’ discussions on the match between Sen. Cruz and Beto O’Rourke, they are so positive towards Beto that you’d assume that the polls were pointing to him being the likely winner in a close race.
Well, that’s not going to happen. Beto moved Cruz out of “Safe GOP” to “Likely GOP”. That’s something, I suppose.