For the second midterm in a row, we’re talking about how the out-party isn’t dominating polls quite to the degree that one would expect given how badly the president is doing.
In the summer of 2022, I wrote in my first column for National Journal, about the changing relationship between presidential approval and midterm performance in this hyper-polarized era. The structural discussions largely fell under the radar, though, since the lack of a Republican wave was largely attributed to—or blamed on—the Supreme Court's overturning of Roe v. Wade that summer.
The structural conversation is back this summer, with roles reversed. Republicans have the trifecta, holding the White House and both chambers of Congress, and Democrats are the insurgents trying to take the House and the Senate. President Trump is wildly unpopular except among his base. Under similar conditions in 2018, the midterm was a Democratic romp.
Yet Democrats are currently running behind where they were in 2018 on the generic congressional ballot. This poll question asks respondents whether they plan to vote for the Republican or Democratic candidate for Congress in their district rather than naming the candidates—it’s used because it’s impractical to ask about 435 sets of candidates in a national poll.
Right now, the Democratic advantage has slipped to 6 points in the Silver Bulletin average, down from 7 points in early June, and well below the 7.6-point advantage at this point in 2018. Because of population patterns and gerrymandering, Democrats typically need to be ahead by 2-4 points in the generic ballot to take the majority. So a 6-point lead isn’t as solid as it might sound.
On the other hand, Trump’s approval rating is lower than it was at this point in 2018, when it hovered near 41 percent. Silver Bulletin has measured it at under 40 consistently since April. And Democrats are much more enthusiastic than Republicans about the midterms—even more than they were in 2018, according to the Washington Post-Ipsos poll.
There are several reasons the enthusiasm gap and low presidential approval are not showing up more clearly in the generic ballot: some structural, others attitudinal.
The structural reasons are not new. Gerrymandering has utterly destroyed competition in much of the country. The Cook Political Report with Amy Walter lists 37 districts in competitive territory for 2026, which is down from 55 in 2022 and 75 in 2018. At this rate, there will barely be any competitive districts left in 2034, especially after the 2030 Census and subsequent redistricting. In the meantime, a great year for Democrats might mean winning 12-15 seats, compared to flipping 41 seats in 2018.
That means there is little reason for most people to seriously consider who they are voting for when asked a generic poll question. They simply respond with their own long-held partisan preference or a general sense of the political vibes, because in most of the country it is a foregone conclusion which party will win the seat.
The attitudinal problem for Democrats right now is a damaged brand. Approval for Democrats in Congress is in the 20s or lower, with even self-identified Democratic voters sour on their representatives. Independents certainly aren’t big fans. Republicans in Congress aren’t viewed much more favorably, but they score better than Democrats because their own partisans are happier with their representatives’ performance.
Most pollsters are currently reporting generic ballot numbers among all registered voters, but public pollsters usually shift to reporting likely voters in late summer. The gap between registered voters and likely voters will be informative: Does the higher enthusiasm among Democrats matter?
From a polling perspective, that depends on how pollsters define a likely voter. Some use enthusiasm as a part of the equation; others do not. The former will show a larger advantage for Democrats. Of course, what ultimately matters is whether Democrats do turn out in higher numbers than Republicans, or whether the enthusiasm gap either closes or doesn’t matter much for actual turnout.
By any metric, you’d rather be a Democrat in this political environment, even if the picture isn’t as rosy as we might expect. But if Democratic gains are limited, the question becomes whether our historical guideposts are still useful. We may be in an era of such dramatic gerrymandering and polarization that presidential approval has far less relationship to midterm outcomes than it used to.
