If you start the clock the day after the 2016 elections, then were are now over 99 percent of the way through the 2018 election cycle. Less than a week remains before “the only poll that matters.”
Which means, of course, that we have precious limited time remaining to analyze the polls that “won’t matter” come November 6th. In this week’s edition of the DKE House forecast, we will look at the fourth and final critical element of wave elections, which is the ability to persuade non-traditional constituencies to lend their votes to the cause.
In the 2017 Virginia elections, this was critical to the monster night the Democrats enjoyed—as they were able to create what I would call an “asymmetric recovery”. While Democrats were able to recover some in the blue-collar parts of the state that had snapped hard to Donald Trump in 2016, they managed to not just hold onto the well-educated and more political independent suburban precincts in the state, they improved on Hillary Clinton’s 2016 performance.
As we noted at the time, if Democrats could replicate that “asymmetric recovery” in 2018, it could very easily be lights out for the GOP’s House majority. So...how is that working out for the Democrats? Better than we might think—we’ll check out the data in a bit.
THE HOUSE FORECAST: LESS THAN ONE WEEK TO GO
After two weeks where the Democratic standing had grown a bit weaker, there was a bit of a bounce back this week for the blue team. You obviously would like to peak at the right time, and as you will read, that appears to be what the Democrats are doing in the final days of the 2018 campaign:
WEEK # |
DATE |
DEMOCRATIC NET GAIN |
1 |
10/3 |
Democrats +27 |
2 |
10/10 |
Democrats +32 |
3 |
10/17 |
Democrats +27 |
4 |
10/24 |
Democrats +25 |
5 |
10/31 |
Democrats +28 |
From the most likely to flip to the least likely to flip, here are the 32 races currently favored to change parties according to the polling averages on our House race pages. If these held according to form, the Democrats would net a gain of 28 seats, which would put them at a 223-212 majority (last week’s forecast was a Democratic gain of 25 seats):
POTENTIAL DEMOCRATIC PICKUPS (32 SEATS)
PA-05 (--); PA-06 (--); NJ-02 (current average: D+18); CA-49 (D+13); PA-17 (D+13); AZ-02 (D+11); KS-03 (D+10); NJ-11 (D+10); PA-07 (D+10); CO-06 (D+8); MN-02 (D+8); VA-10 (D+8); FL-27 (D+7); MN-03 (D+7); MI-11 (D+4); UT-04 (D+4); IA-01 (D+3); NY-19 (D+3); CA-10 (D+2); IL-06 (D+2); CA-39 (D+1); IA-03 (D+1); KY-06 (D+1); ME-02 (D+1); VA-05 (D+1); VA-07 (D+1); WI-01 (D+1); FL-06 (D+0); FL-26 (D+0); NJ-07 (D+0); PA-01 (D+0); TX-32 (D+0)
POTENTIAL REPUBLICAN PICKUPS (4 SEATS)
PA-14 (--); MN-08 (current average: R+11); AZ-01 (R+3); NV-04 (R+2)
So, this week’s forecast, taken as a whole, gives Democrats cause for what I would call “tempered optimism.”
On a basic level, the simple “needle” of pickups moved in the Democratic direction, as the number of net seats gained by the Democrats in our polling averages moved from 25 seats to 28 seats. This, of course, is a good thing if you are a Democrat.
But if we return to the “headwind/tailwind” model we explored last week, we do get a bit of a cautionary note. To remind folks about this particular model, it is a simple “Algebra I” way of gaming the outcome if there was a uniform swing of just a few points (we used three points as the metric).
Based on this weeks numbers, if the Democrats experience a slight headwind (which we will measure as a uniform swing of 3 points in the Republicans’ favor), that would cost the Democrats the majority, as it would move a net of 18 seats into the GOP column (the sixteen seats listed above, plus the Democratic-held open seats in MN-01 and NV-03). That would leave the GOP with a narrowed majority of 230-205, or a net gain of just ten seats for the Democrats. At this point, this appears to be the “best case scenario” for the Republicans.
But if the Democrats benefit from even a slight tailwind (which we will likewise measure as a uniform swing of 3 points in the Democrats’ favor), not only do the Democrats reclaim two of their “losses” above (AZ-01 and NV-04), but they’d also snag an additional 20 seats for a net gain of an additional 22 GOP-held seats. That would put the Democratic majority at 245 seats, an enormous 50-seat gain.
Last week, the “headwind/tailwind” model disproportionately favored the Democrats, as there were only 13 seats in peril for the Democrats if they were to experience even a small uniform loss of support. For the GOP, their number of seats in immediate peril from a small uniform swing changed very little (from 23 seats down to 22).
To put it another way—while the Democratic seat gain, based on a raw look at who is leading the polling averages, has improved, it is instructive to remember that roughly half of these leads are really, really close. But this tightness cuts both ways: if the undecided remainder of the electorate disproportionately favors the party out of power in the last week of the election (as happened in 2010), the Republican losses could border on catastophic.
ANATOMY OF A WAVE, PART FOUR: BRINGING NEW VOTERS TO THE PARTY
Democrats in the 2017 Virginia gubernatorial and legislative elections, as I wrote about shortly after the elections, enjoyed an “asymmetric recovery”. In short, what that means is this:
The playing field also seems to have shifted, in that a lot of historically red territory might be more amenable than ever to Democratic overtures. What’s more—the early evidence is that some of those ancestrally blue working class patches of turf that flirted hard with Trump don’t seem terribly interested in a long-term relationship. If both of those tendencies that we saw in the (admittedly smallish) sample size of 2017 elections hold up, it could be a LONG 2018 for the Republican Party.
2016 was not a realigning election in the classic sense, but much ink has been spilled in describing the palpable movement (in differing directions) during the Trump-Clinton election. While blue-collar and predominantly white enclaves moved into the Republican column, college-educated and more diverse suburban districts that usually trended Republican in the past turned their backs on Trump and the GOP.
What made 2017 such a nightmare for Ed Gillespie and his Republican legislative colleagues is that these trends did not hold. While the suburban and ancestrally Republican counties of northern Virginia continued their blue trend, the territory that drifted more red in 2016 did not keep apace in 2017. The net result? Big Democratic gains and a win for Democratic Gov. Ralph Northam that was several points wider than Hillary Clinton’s triumph in 2016.
Where do we stand on this metric in 2018? The answers have to be heartening for the Democrats.
There are a total of 21 House districts where Barack Obama carried the district in 2012, but the districts turned red in 2016 by supporting Donald Trump. Prior to the 2018 elections, Republicans held a total of 12 of these districts, while the Democrats held nine of these districts. Right now, if the polling averages we see right now were to hold next week, the Republicans would actually LOSE ground in these “Trump surge” districts. While they would likely pick up one of the Democratic seats (MN-08), they are also currently projected to lose six of these seats currently held by Republicans (IA-01, IA-03, ME-02, MN-02, NJ-02, and NY-19).
To put it another way, if the polling averages hold, Democrats would hold two-thirds of the districts that were Obama-Trump districts. This would hint that these districts are, to an extent, returning to form.
Meanwhile, there are also a total of 13 House seats which shifted from Republican presidential support in 2012 to Democratic presidential support in 2016. As it happens, the Republicans control all thirteen of these “Romney-Hillary” districts.
For now, at least.
The current polling averages show the Democrats on track to pick up a total of eight of these seats (AZ-02, CA-39, CA-49, IL-06, KS-03, NJ-07, TX-32, and VA-10). What’s more: of the five seats the Republicans are currently projected to hang onto, here are the requisite margins in our current polling averages (incumbent in parentheses):
- CA-25 (Steve Knight): Knight 48, Katie Hill (D) 45
- CA-45 (Mimi Walters): Walters 49, Katie Porter (D) 44
- CA-48 (Dana Rohrabacher): Rohrabacher 48, Harley Rouda (D) 46
- TX-07 (John Culberson): Culberson 47, Lizzie Panill Fletcher (D) 45
- TX-23 (Will Hurd): Hurd 53, Gina Ortiz Jones (D) 37
So, to put it another way, while the GOP is on pace to win just 33% of the “Trump surge” districts, the Democrats are on pace to win 61% of the “Clinton surge” districts, despite the fact that they presently hold none of them. And they are within shouting distance of winning 92% of them.
If this pattern holds, especially when we consider how many races are underpolled (or not polled at all!), Tuesday night could get very interesting.