The Washington PostDemocracy Dies in Darkness

The Democrats haven’t turned the midterms around yet

Abortion rights activists protest outside the Planned Parenthood Reproductive Health Services Center in St. Louis on June 24 after Roe v. Wade was overturned. (Angela Weiss/AFP/Getty Images)

When the Supreme Court released its anticipated decision in Dobbs v. Jackson Women’s Health Organization overturning Roe v. Wade, Democratic leaders were mocked for their immediate response: Ya gotta vote this November.

“That’s all you’ve got?” an incredulous base replied. Telling Democrats to do what they did in 2020 and 2021 and give the party a majority? That was the plan?

Well, most of it, yes. On Friday, President Biden signed an executive order offering some federal protections for abortion after the ruling, but otherwise, that same mantra: vote. You need to vote.

In short order, though, frustration at the knee-jerk “it’s up to you on Election Day” crumpled a bit since, after all, Election Day is the point at which voters can reshape the direction of the country. In the weeks after the Dobbs verdict, some Democrats started to even get hopeful. Would anger at overturning Roe be enough to turn around the historic pattern of midterm elections going against new presidents? Was that an uptick in generic congressional ballot polling? Was enthusiasm surging?

When I wrote Thursday about the likelihood that Republicans would fare well in November’s elections, I got a lot of pushback: Hadn’t this expectation reversed?

So far, the answer is no.

Sign up for How To Read This Chart, a weekly data newsletter from Philip Bump

Let’s start with the generic ballot. This is a question pollsters use to gauge how the House will look after an election, asking respondents whether they plan to vote for the Democrat or Republican in their local House contest, but not by name. This year, Republicans opened up a healthy lead on this question, according to FiveThirtyEight’s polling average. Since Dobbs, Democrats have inched up a bit — 0.7 points, to be precise.

But the GOP still leads and it isn’t clear what happens next. The graph above shows only a narrow range to accentuate the recent movement; since the beginning of April, the average has only moved within a range of about a point for both parties. It could certainly move more before November.

For the sake of argument, though, let’s assume it doesn’t. What would a 1.8-point margin for the GOP translate to, given past polling averages?

We can plot midterm years for which we have both averages and seat shifts. (Those years are 2002, 2006, 2010, 2014 and 2018.) You can see them all on the graph below, with the gain — or usually loss — of seats for the sitting president’s party shown from right to left and the generic ballot margin for his party from top to bottom. The dotted diagonal line is a rough trend for the available data. The horizontal blue line shows where the value for 2022 is.

If the trendline were perfect and the generic ballot margin didn’t change, we’d expect the Democrats to lose the number of seats shown where the two dotted lines intersect. But neither of those things will end up being true.

Nor is the generic ballot the only indicator at play. Consider another older one: presidential approval.

Usually, less popular presidents have rougher midterms. That’s certainly been true of late. And if that pattern holds, Biden and his party are in trouble. Using Gallup’s most recent data, the intersection of the trend and 2022 lines lands somewhere in the “Democrats lose more than 60 seats” range.

(This graph and those that follow use values closest to the election for every year except 2022.)

This is not preordained, though. Look at 2006, when George W. Bush was reeling from the Iraq War. He was less popular than Biden was in Gallup’s most recent poll, but the GOP still managed only to lose 30 seats. I mean, that still throws the House to the Republicans in 2022, but you get the point: This stuff is imprecise.

Another metric Gallup isolated last month was congressional approval. The trendline here makes its way through a very loose cloud of results. Will 2022 be a 2010 upending for the president’s party? Or a 2014 not-that-bad-considering one?

Gallup’s data on satisfaction with how things are going in the United States offers a more compact trend from bottom left to upper right. Here, we have a different problem: Americans are more dissatisfied than in any previous recorded midterm. The next least-happy (if you will) voters had ever been was in 2010 — the most recent mega-blowout against the Democratic Party.

When Donald Trump faced a similarly difficult-seeming midterm in 2018, I noted a weird divide: Americans thought the economy was doing well but that he was doing terribly. Did that mean they would reward the GOP or punish the president? The answer was the latter: That year was a blowout in the Democrats’ favor.

The good news for Biden, if you will, is that he faces no such perplexing discrepancy. He’s got low approval and Americans are worried about the economy.

Now we come to enthusiasm about voting, that metric that many Democrats have seized on of late as an indicator of fury about Dobbs. As it turns out, Gallup has new data on that. Enthusiasm for voting among Democrats is higher now than in any recent year besides 2006 and 2018 — two elections that went very well for the party. But Republican enthusiasm is 10 points higher.

If we look at the gap between the two parties’ enthusiasm, we again get a murky picture. That said: If the trend line is perfect and the enthusiasm gap stays unchanged (neither of which is the case), 2022 will be a rough year for the left.

Infinite caveats apply, even on top of those sprinkled throughout the article above. Poll numbers can change (though it seems unlikely in this highly polarized era that Biden’s approval is going far). Republican enthusiasm can sink; Democratic enthusiasm can surge. Exceptionally bad or exceptionally good general election candidates can shift a lot of results unexpectedly. It really does come down to turnout.

All of which is to say: If you are a Democrat who wants to shift all of the graphs above to the right, your best bet isn’t to parse individual polls or cross your fingers. It is, in fact, to vote.