Special elections are held when a seat suddenly becomes vacant—either because of a death or resignation. So far, the 2025-2026 cycle has seen 105 special elections—with more to come. These run the gamut from congressional and legislative races to gubernatorial and mayoral contests. According to University of Virginia Professor Kyle Kondik, who has studied special elections held since 1957, they more often than not break against the party that holds the presidency. The same pattern persists in midterm elections, where the president’s party typically loses seats in Congress. Special elections tend to be low-turnout affairs; to the extent they preview the November midterms, it is because they serve as a reliable predictor of the enthusiasm within each party.
The pattern continues so far this cycle: Democrats are performing well. This trend is evident when comparing the margins between the parties in previous regular elections against more recent special elections. The shift in performance from one election to the next serves as an indicator of where partisan winds are blowing.
The following table shows congressional special elections for 2025 and 2026. Note that these results exclude the 2025 gubernatorial and statewide races in New Jersey and Virginia, where Democrats performed strongly. In every special election held to date, Republicans lost ground compared to their 2024 results, even in districts they won. Conversely, Democrats improved upon their 2024 vote shares in every race they won. This table also excludes the 2026 special election for Texas’ 18th congressional district, as it was an all-party “jungle” primary that resulted in a runoff election between two Democrats.
A word of caution: We included raw vote totals in this table to illustrate just how low turnout typically is in special elections—far lower than in presidential elections and generally below midterm levels. With one exception, the raw vote totals for the winners in this table are less than half of what the same party received in the same district in 2024. While Republican numbers historically rise when Trump is on the ballot, it remains unclear if that trend will carry over in a nationalized midterm election without him in 2026. Dating back to the Democrats’ disastrous 1946 midterms, presidents in political trouble are rarely welcome on the campaign trail. However, motivating the Trump base without Trump himself may prove difficult. In any event, the record demonstrates that the president tends to get involved in campaigns, even if his presence might boost the opposition.
Beyond tracking elections where districts changed parties, we also analyze races where the incumbent party’s margin of victory narrowed significantly—potentially putting those districts in play for future cycles. This trend surfaced in the special election for Georgia’s 14th congressional district; while Republicans held onto the seat, winner Clayton Fuller ran 8 percentage points behind Marjorie Taylor Greene’s 2024 performance.
Special elections for state legislative seats were also instructive. Most of these vacancies resulted from incumbent resignations, in some cases to seek other offices, or deaths. In Mississippi, three of the districts that flipped had been redistricted.
One of the most notable contests took place in Florida’s House District 87, which includes Mar-a-Lago, President Trump’s home. The district was held by a Trump-endorsed Republican and ended up in the hands of a first-time Democratic candidate. The result surprised many, perhaps including the president, who voted by mail in the contest.
To date, 12 state legislative seats have flipped from Republican to Democratic control in special elections. When accounting for Democratic gains in the 2025 off-year elections in Virginia and New Jersey, the total number of flips from Republican to Democratic is 30. By contrast, no seats have flipped in the opposite direction. Democrats have overperformed by 4.5 percentage points on average.
If these races are viewed alongside the substantial 2025 Democratic wins in New Jersey and Virginia, the midterm outlook appears favorable for Democrats. The party picked up seats in Congress and state legislatures, won a seat on the Wisconsin Supreme Court, and secured a mayoral race in Waukesha, Wisconsin—a city where Republicans had won for years.
The swings toward Democrats have been unusually large, likely reflecting sharp declines in the president’s approval ratings, the unpopularity of his immigration policies, and persistent frustration over high prices. But there are still seven months until Election Day, providing Trump time to convince voters that his policies are working. While Democrats have consistently performed well in recent contests, the party’s brand is often described as “toxic,” a sentiment reflected in its low polling. The combination of Republican fears over a potentially catastrophic election cycle and the vulnerability of the Democrats’ image is likely to lead to an exceptionally negative campaign.
For Democrats, it’s too early to be changing the drapes in the Capitol, but some might be starting to check out the fabric.
Related Content
Authors
The Brookings Institution is committed to quality, independence, and impact.
We are supported by a diverse array of funders. In line with our values and policies, each Brookings publication represents the sole views of its author(s).
