As I type this on a Saturday morning, somewhere between 33 and 35 million Americans will have already cast ballots in the 2018 midterm elections. In the 2014 midterms – the last time we did this – about 83 million total votes were cast. To say that interest in voting in 2018 is high almost feels like underselling it. Here’s another way to frame it. According to a report prepared by NonProfit Vote, in 2010, 25% of all votes cast were “early vote” or in-person absentee ballots. In 2014, the early vote percentage was 29%. Even if 2018 follows a trendline to 33%, the chances that 110-120 million votes will be cast for Tuesday’s election is incredible!
But that’s nerdy stuff. What you’re here for is the red meat. Will Democrats wrest control of at least one house of Congress? Will a presidential closing argument geared around nativism and exclusion carry the day? Well, we’ll get to that shortly. But here are the numbers that matter.
As of now, Republicans basically control 240 seats, with 5 of those sitting vacant due to retirements or resignations.
Democrats control 195, with 2 of that number sitting vacant due to resignation or death.
The number here is simple. Republicans maintain control by keeping the Democrats from flipping control of seats. Democrats win control if they can flip 23 or more.
23. That’s the first number. It’s an important one, and not one to take for granted on either side heading to Tuesday.
On the Senate side, The Republican Party currently controls 51 seats. 49 senators (including two independents) caucus with the Democrats. To keep control of the Senate, for Republicans they simply need to prevent Democrats from gaining more than a single seat on them. Democrats have to get a net gain of 2 seats. 2 seats….and that’s not a number to take for granted on either side of the aisle, either.
So. Who’s gonna win? How’s this all going to shake out?
As we sit here 72 hours before people are voting on Election Day, here’s my best answer: I haven’t a clue. No one does.
What we can do is look at history, polling, and actual hard data to offer some clues, perhaps.
For one thing, we know that it can be brutally hard for the party holding the White House to not start shedding seats in the midterm elections. That does happen from time to time (2002 and 1998, for instance) but it doesn’t happen too often.
And we can look at polling, too. As Nate Silver noted on Friday, for the last two weeks the generic congressional ballot (in which respondents are simply asked which party they would prefer as the majority) has held at a steady 8-10 point lead for Democrats. That’s a good indicator that Democrats could gain a significant number of seats. And in other polling news, the Siena/Upshot polls that have run in hotly contested districts have been fascinating. They started out blue…ish. Then for a week the blue trend really seemed to level off. In the past 10 days, the blue trend has been growing, and heading into the final weekend, it’s looking like one of those things at the beach, where the water comes in and breaks onshore? Forget what they call that.
BUT…the other thing that the live polling by Siena/Upshot illustrated dramatically is just how tough it is to poll congressional races. In one Florida district, they had to make 50,000 calls to get to the sample they were trying for of 500 voters. That’s insane!
And that also means that when you see a margin of error of 5 points or 7 points either way on a congressional poll, take that seriously. They’re not kidding. These district-level polls are incredibly noisy! And a bunch of polls of districts will be pretty close after the smoke clears and the votes are counted. But…there are also going to be a number of surprises. There just are. Democrats are going to lose districts they didn’t expect to. So are Republicans. Polling by phone may not be a viable thing for much longer.
There’s one other data point worth discussing here, though, that may provide some evidence as to what might happen on Tuesday. Fundraising numbers are not perfectly correlated to election-day performance by any means. But fundraising totals, especially by small donors and especially especially by small donors within the constituency of a candidate (be that precinct, district, or state) might be a decent indicator of voter enthusiasm.
This year, that indicator points to Tuesday being pretty good for Democrats. Q3 2018 fundraising numbers for Democrats in the House and Senate were absurd. Record-breaking. And yes, I’m as disturbed by the money in politics as anyone. But this is the system we’ve got (for now), so play it as it lays.
It’s not just federal elections for the House and some Senate seats going on Tuesday, either. A bunch of Governorships are up for grabs. Polling and fundraising seems to indicate that Democrats might have a good night on that front, too. And control of statehouses that are up for voting will also be worth watching…
…because we’ve got a 2020 census coming up, and redistricting may be on the menu – either by court order or census indications – in a number of states. How important is that? Let’s put it this way: this would be a 50/50 fight for Democrats at best for House control, were it not for court-ordered redistricting in Pennsylvania, Florida, Virginia, and North Carolina.
So. That’s my left-leaning commentary to lead off this thread. Full disclosure, in case you’ve been living under a QT3 rock: I grew up in a house that had framed portraits of Franklin Roosevelt and Jack Kennedy on the walls…and I cast my first vote ever in the 1984 presidential election (a month after my 18th birthday!) for Walter Mondale. So yeah, you know where my viewpoint is skewed.
But hey. Let’s fill this thread up! Let’s talk about election specific stuff here now. If you voted early, and especially early, in-person, tell us about it here! And when you vote Tuesday (because you haven’t voted yet, and OBVIOUSLY if you’re eligible to vote in this country YOU ARE GOING TO VOTE…) tell us about that, too!
And then finally on Election night, let’s all hang out here together and wring our hands and see what happens!
