In last week’s installment of the DKE House forecast, we began to look at the necessary ingredients to generate a wave election in the House. The first ingredient we cited was the ability to put a lot of races on the table, in order to broaden the possible roster of partisan “flips” on Election Day. On that score, when comparing the 2018 landscape to the most recent major wave election (the quite lamentable 2010 election cycle), we can conclude that the mission was largely accomplished: the number of GOP-held seats that Democrats have managed to put into a competitive range (within 5 points in at least one poll) is only slightly shy of what the GOP managed back in 2010.
The second ingredient is having the resources to flip seats. And, on that front, the answer to whether the Democrats met the threshold there has clearly been a resounding “yes.”
In fact, a more appropriate answer might be “holy crap, they didn’t clear the bar, they left the freaking atmosphere.” It is rare in politics, even for those of us who have been following the electoral cycles for a quarter century, to say that we have not seen something before. But it can be said with confidence—we have not seen a fundraising windfall anywhere close to what the Democrats accomplished over the last three months. And that, indeed, may have put a number of races currently on the radar onto the map for November.
We will delve into that all-important money chase a little bit later. Before we do that, however, let’s update the forecast, where there has been some slight shifts, and one important change in methodology.
THE HOUSE FORECAST: THREE WEEKS TO GO
Before we get to the forecast, a methodological shift that was overdue, and benefited neither party. Now that we are just three weeks to Election Day, it is high time to tweak the forecast in order to not let a random and silly poll skew our perception of a race.
Exhibit A would be this little oddity. Now, Kurt Schrader, the Democratic incumbent in Oregon’s 5th district, has had some close shaves over the years. In the 2010 GOP tsunami election, he held off Scott Bruun by a modest five-point margin (51-46). Most recently, in 2016, he took a ten-point win off of Colm Willis (53-43).
That said, in what is universally considered to be a Democratic year, it strains credulity a little to believe a poll that says the Democratic incumbent that survived both the 2010 and 2014 GOP waves would suddenly be trailing by over 20 points.
With that in mind, here is the shift—to be considered for our “balance of power” forecast, a race (with the exception of the three races left for dead by the “incumbent party”—PA-05/PA-06/PA-14) has to have either been polled twice, or polled once by an independent outlet. This took a couple of Democratic “pickups” off the board (like NC-07, which has seen only one Democratic poll), and a couple of Republican “pickups” off the board (the aforementioned OR-05).
This is what remains, and the data this week is incrementally better for the GOP this week. Week one of our forecast say the Democrats projected to gain 27 seats. Week two saw the number creep up to 32 seats. Week three? We see a bounce back to the origin, as the Democrats, according to our polling averages, would be favored to pick up 27 seats. Two things changed—a couple of incrementally Democratic seats fell of the board for the blue team, although just BARELY in most cases. Check this out: the GOP “lead” in our polling averages is one point or less in 11 different races. Also, a couple of Democratic-held seats have slipped to the bad side because of new data (specifically, MN-08 and NV-04).
From the most likely to flip to the least likely to flip, here are the 35 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 27 seats, which would put them at a 222-213 majority (last week’s forecast was a Democratic gain of 32 seats):
POTENTIAL DEMOCRATIC PICKUPS (31 SEATS)
PA-05 (--); PA-06 (--); NJ-02 (current average: D+23); AZ-02 (D+19); PA-07 (D+14); PA-17 (D+13); CA-49 (D+11); CO-06 (D+10); MN-03 (D+6); VA-10 (D+6); CA-10 (D+5); IL-06 (D+5); KS-03 (D+5); MN-02 (D+5); CA-45 (D+4); MI-11 (D+4); NJ-11 (D+4); CA-25 (D+3); IA-01 (D+3); PA-01 (D+3); FL-27 (D+2); KS-02 (D+2); NC-02 (D+2); WA-03 (D+2); KY-06 (D+1); NJ-07 (D+1); NY-19 (D+1); TX-32 (D+1); VA-07 (D+1); CA-39 (D+0); IA-03 (D+0)
POTENTIAL REPUBLICAN PICKUPS (4 SEATS)
PA-14 (--); MN-08 (current average: R+12); AZ-01 (R+5); NV-04 (R+1)
Admittedly, there is a flaw here, but the flaw is a net wash. There is one Democratic seat not listed here that is in legitimate peril (the open seat in MN-01), because only one partisan poll has dropped here. However, that is certainly offset by AZ-01, which only makes the cut because it has two lousy GOP-sponsored polls (IVR polls where the polling memo was so unprofessional that any respectable pollster would laugh it out of the room.
The dilemma for Republicans here is that they already, by this metric, would lose the majority, and it would take only a tiny tailwind to be in more peril. Also, for the first time, we are starting to see a little bit of a gap developing in those “coin flip” races. There are 11 Democratic “flips” that have margins at two points or less in our averages, just as there were two weeks ago. But there are now 15 GOP-held seats that are within the same margin of flipping to the Democrats. That’s a marked increase from two weeks ago, when just nine GOP-held seats were in similar peril.
What’s more—the Republicans have been flooding the zone lately with internal polling, probably in an effort to convince donors that all is not yet lost. A prime example of that is in IA-01, where every public poll has had Democrat Abby Finkenauer up handily on GOP Rep. Rod Blum, but Blum has released two polls in the past two weeks that showed him trailing, but by a not-insurmountable margin. Since our averages do not take the partisan sponsor into account, it sits in our averages as Democrat +3. But this is a prime case of the GOP candidate benefitting from the asymmetry of polling releases. For all we know, team Finkenauer could be sitting on really good polling, but why foster complacency? Meanwhile, team Blum needs to release its polling, because they have to convince the press and donors that he is not being tossed on the scrap heap. It’s just one of those little quirks in polling that means that, while averages are the best tool we have, they are not perfect.
ANATOMY OF A WAVE, PART TWO: THE MONEY CHASE
For those who were in a self-directed news or information blackout this week, the Democrats got a not-unexpected, but very welcome, boost this week as the third quarter FEC reports were submitted to the government. The outcome was an utterly massive Democratic advantage over the GOP, even in Republican-held seats, as we head into the final month of the election cycle.
In the most competitive races on the docket for November, the advantage was substantial:
The average Democratic House candidate across these districts raised $1.48 million last quarter, from July through September, and has $959,000 banked away for the rest of the election.
By comparison, the average Republican candidate in these districts raised $559,000 over that same period and has $817,000 left in the bank.
Now, money alone does not portend an electoral wave. But it certainly can be a leading indicator. Especially when campaign cash is largely driven by a wide swath of small-dollar donations, which was the case in this cycle, as our own Kerry Eleveld noted earlier in the week. ActBlue, the best-known Democratic campaign donation clearinghouse, noted that they have three times as many donors this cycle as they did in 2014.
Furthermore, even net fundraising per party can be an indicator. Check this out this graphic by Nate Silver:
An analyst responding to Silver via Twitter noted that their data showed that this was the widest disparity dating back to 1980. Either way, Democrats have to be heartened that the two elections that showed substantial shifts (a shift of 10 points or more) on Silver’s chart happened to be 2008 (a great Democratic year, which saw them expand their majority to over 250 seats), and 2010 (a great Republican year, which saw the red wave bring in a 63-seat pickup and a GOP majority).
Aside from the potential predictive volume, the Democrats also reap an additional benefit from this fundraising windfall, one that refers to last week’s forecast. If a key to building an electoral tsunami is broadening the political battlefield, it is imperative that challengers in second-tier and third-tier Republican targets raise enough money to get their message out. Mission accomplished, given that over 150 Democrats raised at least $250,000 for the quarter, and well over a hundred Democrats (!) raised over a half-million dollars.
The reason why this matters is because waves are built on a steady dose of late-breaking races that nobody saw coming. In the 2010 GOP wave election, Republicans would eventually pick up a total of 16 seats where there was not a single poll showing the GOP challenger in the lead as of October 1st. In fact, five of those seats (AL-02, MN-08, NY-24, TX-27, and VA-09) did not even show a competitive poll until we were within a month of Election Day. Democrats, it is now evident, have the resources to take a couple of races that have been on the fringes of competitiveness for most of the cycle and bring them rapidly to “real race” status.
And the danger for Republicans, one that they may already be staring in the face, is this: when the field widens that markedly, for the party in-power, it becomes an immensely frustrating game of whack-a-mole. If the field expands from 60 to 70...or even 80-100 different GOP held seats (right now, our race ratings have 61 Republican-held seats rated from Safe D to Lean R), the odds of the GOP’s resources simply being overwhelmed becomes quite high.