Redistricting aside, Dems' glass is half full
If Trump’s approval rating doesn’t improve, Republicans will find it tough to overcome.
The eyes of the political world are turning to three elections slated for November: gubernatorial races in New Jersey and Virginia, and a retaliatory Democratic gerrymandering initiative in California, in response to Republican redistricting efforts in Texas and likely several other red states.
It would be surprising if Democrats don’t win all three, even if it’s not by landslide margins. New Jersey’s score in The Cook Political Report’s Partisan Voting Index is D+4, which means that in presidential races the Garden State votes more Democratic than the country as a whole by 4 percentage points. Virginia is D+3, while California checks in at D+12, so I’m not exactly walking out on a limb. In this era of hyper-partisanship, unless a blue state votes red or a red state votes blue, it’s not really news.
Yes, Republican Glenn Youngkin did win the Virginia governorship four years ago, and another Republican lost by less than expected in New Jersey that year. However, that was at a time when President Biden’s approval ratings had tanked, dropping 14 points in the Gallup poll in just four months. By Nov. 2, 2021, it was becoming increasingly clear that the Biden presidency was hopelessly mired in a political Bermuda Triangle comprising the border, the cost of living, and a botched withdrawal from Afghanistan.
If Democrats do indeed hold serve in those three campaigns, they will inevitably claim it is a great omen for 2026, while Republicans will argue it is meaningless.
My hunch is that Democrats are probably in better shape in terms of winning back a majority in the House than most think. Let’s start by examining the macro side, looking down from 30,000 feet.
National elections are always about something. Just as the 2020 election was about Donald Trump and his four years in office, not about Biden, the 2024 election was about Biden and his four years, not about Trump. Biden did not win the former; Trump lost it. Trump did not win the latter; Biden lost it, even if he had technically dropped off the ballot 107 days before the formal vote occurred. In such cases, the victorious challengers flatter themselves by thinking it was about them. Kamala Harris was as much collateral damage as anything else.
Midterm elections are almost always about the sitting president, although exceptions sometimes occur. In the long run, Democrats would have been better off if the 2022 midterms had been a referendum on Biden; a shellacking might have sent them in a different direction for 2024. However, while the national Republican congressional vote in 2022 increased and the Democratic vote decreased from four years earlier, Republican primary voters in about two dozen critical races for the House and Senate inexplicably nominated candidates who might have been found in the bar scene from Star Wars. Thanks to this rogue’s gallery of red-state characters running in purple states, Democrats did not have as bad an election as one might expect for a party whose president had a Gallup approval rating of just 40 percent (37 percent among independents). The 2022 election was not about the Dobbs Supreme Court decision.
Next year’s election will be about Trump, whose current Gallup approval rating is 40 percent (56 percent disapproving), exactly the same as it was in the final poll before the 2018 midterm election, when Republicans lost 42 seats. Among those who identify as Republicans, Trump’s approval rating is 93 percent; among independents, it is 33 percent; and among Democrats, 1 percent. Keep in mind that most partisans are packed, either naturally or cartographically, in districts of the same ilk, giving those independents a lot of sway in the few competitive districts in the middle.
A second macro factor is turnout, another reason midterms don’t usually go so well for the “in party.” The makeup of voters in midterm elections tends to differ from that in presidential elections. The drop-off is particularly large when the president has a personality-driven appeal. President Obama’s base turned out in sufficient numbers to elect him in 2008 and reelect him in 2012, but it did not bother to show up in the 2010 and 2014 midterms, when Democrats lost 64 and 13 net House seats, respectively. As a Democratic pollster dryly noted to me at the time, “They don’t call them Obama voters for nothing.” There is little reason to believe that it will be any easier for Republicans to get the Trump vote out in 2026 than it was in 2018.
On the micro, seat-by-seat level, Republicans are defending three seats in districts with Democrat-tilting scores for their districts in the Cook PVI: D+3 in Rep. Don Bacon’s open seat in Nebraska, D+1 in Rep. Mike Lawler’s New York district, and D+1 in Rep. Brian Fitzpatrick’s Pennsylvania district. Five GOP seats have a dead-even PVI: Virginia-02, New Jersey-07, Colorado-08, Arizona-06, and Michigan-07. Eight more are R+1 or R+2, and a dozen more are either R+3 or R+4. When the political climate gets tough in a midterm, these are the kinds of districts that are most likely to go over the side.
It is doubtful that Republicans can redraw enough favorable districts in Texas, Indiana, Missouri, and perhaps Florida to offset the macro-environmental factors, not even allowing for a handful of seats that Democrats will likely secure in California. At most, the gerrymandering efforts from both sides will introduce uncertainty into a situation that would otherwise be highly predictable, i.e., the House flipping.
Just as Biden didn’t have a wide mandate in 2020, Trump lacks one today. That didn’t stop each man from governing as if he had one. The tendency to be delusional clearly knows no partisan bounds. Impersonating a mandate may not be a criminal offense, but it is a political one.