On the menu today: As the midterm elections get closer, it’s worth keeping some key lessons in mind while perusing the latest polls. First, a recent ABC News/Washington Post poll buried the lead, which shows Republicans well ahead in the key swing districts — and this tells us a lot more than the usual generic-ballot numbers. Second, it’s getting a little late in the cycle for polls of registered voters, but they keep showing up, and there’s good reason to think that the polls are using a way-too-generous definition of a “likely voter.” Third, even if the midterm outlook isn’t as good for Republicans as it was in the early summer, that doesn’t mean the outlook is bad. And finally, NBC News commentator Jen Psaki, Biden’s former White House press secretary, offers some surprisingly blunt truth for her party and former boss.
Checking the Numbers . . .
There are a few things to keep in mind about polling as the days grow shorter and September turns to October:
One: The generic-ballot question isn’t focused on the districts that matter most.
The generic-ballot question is an imperfect measurement because it’s just asking people across the country whether they’re voting for Republicans or Democrats. We don’t know if they live in a swing district, a heavily Democratic district, or a heavily Republican district. When we want to know which party is going to control the House, we care a lot about those swing districts and districts that lean just a little toward one party or the other. We don’t care that much about New York’s 14th congressional district because we know Alexandria Ocasio-Cortez is getting reelected by a wide margin. We don’t care that much about Wyoming’s lone House seat because we know Harriet Hageman is going to win it by a wide margin. And we don’t really care about the margins in those safe districts. If a Democrat wins a heavily Democratic-leaning district with 90 percent instead of 60 percent, that doesn’t change anything.
RealClearPolitics rates 32 House races as “toss ups.” It rate s22 districts as “lean Republican,” 17 as “likely Republican,” and 179 as safe Republican. It classifies 17 districts as “lean Democratic,” 20 seats as “likely Democratic,” and 185 seats as safe Democratic. One of the reasons analysts feel fairly comfortable projecting that the GOP will win the House is that even if you put all of the “toss up” districts in the Democratic pile, Republicans still have a bare majority of 218 seats.
It’s a similar story at FiveThirtyEight, although it puts only 13 seats in the “toss up” category. The site rates 195 seats as solid Republican, 14 as likely Republican, and five as lean Republican — showing Republicans knocking on the door of a majority at 214 seats. FiveThirtyEight rates 167 seats as solid Democrat, 30 as likely Democrat, and twelve as lean Democrat.
When you see someone talk up Democrats’ chances of keeping the House, that’s mostly wish-casting. Democrats need to win just about every competitive seat and not have any lean-Democrat seats slip through their fingers.
If anything, the numbers point to Republicans’ winning a slew of those competitive seats. The most recent ABC News/Washington Post poll this weekend buried the lead*:
Among those living in congressional districts that are rated as at least somewhat competitive by ABC’s FiveThirtyEight (neither solid Republican nor solid Democratic), registered voters favor Republican candidates by a wide 55-34 percent — nearly as big as the Republican lead in solid GOP districts (+24 points). Democrats lead by 35 points in solid Democratic districts, pointing to a potential overvote where they’re most prevalent.
If Republicans really are leading in competitive districts by 21 percentage points, then this is indeed a red-wave election, and we should see some unexpected Democratic losses.
My back-of-the-envelope math suggests that this would come out to a GOP majority of 235 seats — a pickup of 21 districts.
Two: It’s getting a little late in the cycle for polls of registered voters.
Right now, in the RealClearPolitics average of the generic ballot, Democrats barely lead — by three-tenths of a percentage point. That number comes from averaging out most recent results from eleven polling firms: CBS News, Emerson, Rasmussen Reports, ABC News/Washington Post, Economist/YouGov, Politico/Morning Consult, NBC News, Fox News, New York Times/Siena, Trafalgar Group, and Harvard-Harris.
But some of those firms are polling registered voters on the generic-ballot question, and others are polling likely voters on the generic-ballot question.
Once you sort out the pollsters using registered voters from the pollsters using likely voters, a clear pattern emerges. CBS, Emerson, Rasmussen, ABC, and Trafalgar are using likely voters, and their results average out to Republicans leading by 2.8 percentage points.
Economist/YouGov, Politico/Morning Consult, NBC News, Fox News, New York Times/Siena, and Harvard-Harris are using registered voters, and their results average out to Democrats leading by 2.83 percentage points.
In other words, there’s probably a bunch of registered-but-not-likely voters out there telling pollsters they’re likely to vote for the Democrat in their district’s House election . . . but it’s anybody’s guess as to whether they’ll actually show up and cast a ballot.
We know that not all registered voters will cast ballots in the midterms. In 2018, voter turnout was extremely high for a midterm election . . . and that was 50.3 percent of the voting-eligible population. The “voting-eligible population” is not the same as “registered voters.” In Virginia that year, 59.5 percent of registered voters cast a ballot. In Ohio, 55.7 percent of registered voters cast a ballot. In North Carolina, 53 percent of registered voters cast a ballot.
In the CBS News Battleground Tracker poll, which has Republicans ahead by one percentage point, 74 percent of registered-voter respondents said they would definitely vote, and another 13 percent said they probably would vote. Based upon history, it seems exceptionally improbable that 87 percent of registered voters will actually turn out and vote this year.
It’s the end of September. Pollsters probably should have put a likely voter screen on their samples around Labor Day, and those likely voter screens probably aren’t tight enough.
(Considering that four states have started early voting, maybe it’s time to ask, “Have you voted already?”)
Three: Republicans may need just a modestly good outlook for November.
Nate Silver contends that the polls do not show much of a GOP “bounce back” in recent weeks. No doubt, this is not the near-ideal environment for Republicans like it was in early summer, when gas prices were at record highs, and before the repeal of Roe v. Wade fired up the Democratic grassroots. Silver warns that:
We’ve reached that part of the election cycle where you ought to be wary of media narratives about which party has ‘momentum.’ There are a lot of competitive elections this cycle, and since Labor Day, we’ve started to see a lot more polling. It’s very, very easy to cherry-pick your way into a story that fits your preconceived notions about the race.
Okay, but how many races do you need to examine before you’re no longer “cherry-picking”?
- I’m not sure that GOP Pennsylvania Senate nominee Mehmet Oz is going to win, but he’s now down by increasingly modest single-digit margins — a far cry from those eleven-point and 13-point John Fetterman leads we saw earlier in the summer.
- Out in Nevada, Republican Adam Laxalt has quietly been building a small but consistent lead over incumbent Democrat Catherine Cortez Masto.
- In Wisconsin, incumbent Republican senator Ron Johnson, who was supposed to be a dead man walking — just like in 2016! — keeps enjoying one small lead after another.
- In Ohio, J. D. Vance enjoys a modest lead in most, but not quite all, of the recent polls. (The Siena poll that has Tim Ryan ahead of Vance by three percentage points also has GOP incumbent governor Mike DeWine leading by 23 points. Are we really going to see a 20-point split between the Senate race and the governor’s race?)
- In North Carolina, Republican Ted Budd isn’t out of the woods yet, but he’s looking a little safer.
- In Georgia, maybe you can argue that incumbent Democrat Raphael Warnock is ahead, but it’s not by much.
- The much-touted Val Demings challenge to Marco Rubio in Florida’s Senate race isn’t amounting to much so far.
It’s not all coming up roses for Republicans, however. Blake Masters is turning in a terrible performance in Arizona, unless Trafalgar is seeing something in that state that no one else is. It’s a similar story in Colorado and Washington’s Senate races. In New Hampshire, Don Bolduc does not appear to be giving Maggie Hassan any serious stress.
It is still fairly easy to envision a scenario that gets Republicans to 51 Senate seats. Beyond that, you need something unexpected to fall in their laps — perhaps a win in Pennsylvania or in one of the western races.
*I know most journalists spell it “lede” in this context, but no one knows what the heck we’re talking about when we do that.
ADDENDUM: Biden’s former press secretary, Jen Psaki, dished out a bit of unexpectedly blunt honesty on Meet the Press this weekend:
Yeah, look, I think that Democrats, if the election is about who is the most extreme, as we saw Kevin McCarthy touch on there with Marjorie Taylor Greene — I’ll say her name, sitting over his left side — then they’re going to win. If it is a referendum on the president, they will lose. And they know that.