The House: Democrats Favored on What Starts as a Small Battlefield

KEY POINTS FROM THIS ARTICLE

— History suggested Democrats would be favored to flip the House in 2026 as soon as Donald Trump clinched the presidency in 2024.

— Little that’s happened over the last several months calls into question that initial, gut-level assessment. Democrats should flip the House next year, and we favor them to do so. Republicans do have a couple of factors working in their favor, though, that could help them as they seek to defy history.

— Our initial House ratings reflect a small House map, with Democrats narrowly ahead 209-207 in the seats that at least lean to one party or the other, with 19 Toss-ups.

Our initial House ratings

It would have been reasonable to believe that Democrats were well-positioned to flip the House in 2026 as soon as Donald Trump won the presidency last November. There is a longstanding tradition of the non-presidential party winning control of the House in a midterm—in fact, other than 2002, when Republicans held the House in George W. Bush’s first midterm, the non-presidential party has either flipped or maintained control of the House in every other midterm since Ronald Reagan’s presidency. Setting aside 2002, the last time the presidential party won the House majority in a midterm was 1978, when Jimmy Carter and the Democrats maintained their majority at a time when Democrats functionally held a permanent majority in the House (they held the House majority uninterrupted from the 1954 election all the way to the 1994 election).

Typically in midterms, the non-presidential party improves both their share of seats and share of the popular House vote compared to the previous election. The Democrats’ path became clearer once all of the 2024 races were called and they captured 215 seats to the Republicans’ 220, meaning they only needed to do 3 seats better in 2026 to flip the House.

Since then, Trump has taken office and governed much more aggressively than he did at the start of his first term. His overall approval rating, though better than it was in 2017, has turned net negative (meaning his approval number is lower than his disapproval number) in polling averages, and ratings of his approval on the economy have generally been worse than his overall approval ratings (his overall approval may be buoyed by maintaining positive marks on immigration). That dynamic was taking shape even before his recent unilateral decision to impose massive tariffs followed by his subsequent decision on Wednesday afternoon to delay most of the tariffs. So much about the economic situation is uncertain—let alone the practical implications of the haphazard Elon Musk-led cuts to the federal government or how Republicans craft an extension of Trump’s first-term tax cuts. There is and will be plenty for voters to react to in the midterm: typically that reaction is negative, although it does not have to be.

Electorally, Democrats have been punching above their weight in special elections much like they did in 2017, and they just decisively won the first big statewide election of Trump’s second term, maintaining liberal control of the Wisconsin state Supreme Court last week. Democrats are grappling with some internal strife, as their own voters are starting to express wider dissatisfaction with the party. But fortunately for Democrats, a party’s voters don’t have to love their own party in a midterm context in order to deliver wins—they just have to be motivated enough to turn out against the other guys. For instance, Republicans’ internal issues didn’t prevent them from strong showings in the 2010 and 2014 midterms, although fractious primaries did produce some candidates who kicked away key races, most notably in 2010 Senate contests (the quality of nominees coming out of primaries will be something to monitor during the primary season). Back then, the conventional wisdom was that the smaller midterm electorate had a generic Republican lean. Now that conventional wisdom has been turned on its head, given the migration of higher-turnout voters into the Democratic camp. In a recent episode of the Ezra Klein Show, Democratic consultant David Shor found that Kamala Harris would have beaten Trump in an electorate made up of only 2022 midterm voters, while Trump would have won the popular vote by close to 5 points if all registered voters had cast a ballot. Democrats clearly have some longer-term problems to work through, but that to us is more of a 2028 issue than a 2026 one, at least when it comes to the House.

To sum up this preamble, we would just say the following: Democrats became favorites to flip the House as soon as Trump won, and what has happened since then has not really changed that assessment.

Republicans do have a couple of sources of comfort that may matter if the overall political environment for them is not as bad as it seems like it could be.

Despite holding the majority, Republicans are not overextended into clearly blue-leaning turf. Only 3 House Republicans—skilled incumbent Reps. Don Bacon (R, NE-2, Omaha), Mike Lawler (R, NY-17, exurban New York City), and Brian Fitzpatrick (R, PA-1, Bucks County)—hold districts that Harris carried in 2024. Democrats, meanwhile, hold 13 Trump-won districts. Republicans also may get a slightly better overall map as the result of mid-cycle redistricting: Ohio will have a new map, and the state’s system effectively enables Republicans to draw a map that could allow them to better compete for up to 3 of the 5 seats the Democrats hold in that state, although there’s a lot of uncertainty. Other potential redistricting changes are hazier. But if 2026 is not a clear wave environment against Republicans and instead is a narrower, district-by-district slugfest, like 2022 ended up being, the map is tight enough where Republicans could eke out the majority, as Democrats came close to doing that year.

The competitive map, at least as it stands at the starting point, is not that large, and our initial ratings show a competitive race for control. We have 209 seats rated Safe, Likely, or Leans Democratic, 207 seats rated Safe, Likely, or Leans Republican, and 19 Toss-ups. Democrats would need 9 of the Toss-ups to win the majority, Republicans would need 11. So the ratings show a very evenly divided race for the House even though our overall assessment is that the Democrats are more significant favorites than the ratings show. But the ratings also show a path for Republicans, if they can defend the bulk of their most vulnerable members, get a better map in Ohio and perhaps elsewhere, and play offense in at least a handful of Democratic seats. Notably, if either side sweeps the Toss-ups, that party would still win fewer than 230 seats. So in order to break into the 230s—which neither side has done the last 3 elections—the winning side will have to play offense beyond the current Toss-up column.

Table 1: Crystal Ball House ratings

Let’s go through some of the ratings. The big-picture factors discussed above definitely have a bearing on the ratings: Democrats are generally getting more of a benefit of the doubt here than Republicans to start.

— In Ohio, the specter of redistricting has a bearing on the ratings. It’s possible that we would have listed either Reps. Marcy Kaptur (D, OH-9, Toledo) or Emilia Sykes (D, OH-13, Akron-Canton) as Leans Democratic were it not for the possibility of redistricting making their districts more hostile. As it stands now, Kaptur barely won in 2024 in a district Trump carried by a little under 7 points, while Sykes won by 2 points in a district Harris carried by less than a tenth of a point. Sykes’s credible 2024 opponent, former state Sen. Kevin Coughlin (R), just announced he is running again. Rep. Greg Landsman (D, OH-1, Cincinnati) would be rated as Likely Democratic if his Harris +6 district was guaranteed to exist for another term, although his district may be harder for Republicans to dismantle without creating ripple effects elsewhere, so we’re starting him as Leans Democratic pre-redistricting. There’s been no public movement on redistricting, and as Republicans figure out what they want to do, they may also not want to hurt some of their other members too much in order to damage the Democrats. For instance, Reps. Mike Carey (R, OH-15, Columbus/western Ohio), Mike Turner (R, OH-10, Dayton), and Max Miller (R, OH-7, Northeast Ohio), all hold Safe Republican districts that probably are not Democratic targets as currently constituted, but perhaps could be if they were made less red and if the environment is really good for Democrats next year. History suggests that only a few, if any, Democratic House incumbents will lose next year; in past years, redistricting has helped explain some of those losses and could again if these Democrats run and lose in more difficult districts. Sykes or Landsman could also potentially run statewide, maybe for Senate, if their districts become unfavorable.

— The explosive growth in GOP performance in South Texas is the main reason we are listing Reps. Henry Cuellar (D, TX-28, Laredo) and Vicente Gonzalez (D, TX-34, Brownsville/McAllen) as Toss-ups. Cuellar is a really interesting case: He is a strong electoral performer and probably the least progressive member of the Democratic House caucus, a good fit for his culturally conservative, red-trending district. He is also under indictment on corruption charges, although his potential legal problems have been known since the 2022 cycle (and he was indicted in advance of the 2024 election). Gonzalez, meanwhile, is more of a national Democrat, although he has taken some steps to craft a more moderate image as his district has changed and is a member of the moderate Blue Dog Coalition.

— Two very narrow California Democratic winners, first-term Reps. Adam Gray (D, CA-13, Central Valley) and Derek Tran (D, CA-45, western Orange County), also start in Toss-up contests. Aside from holding a district where Harris performed better, Tran may have the better case for Leans Democratic, as he generated some crossover support in the Harris +1.5 district’s Vietnamese pockets in his defeat of former Rep. Michelle Steel (R, CA-45) that he may be able to retain (he is Vietnamese American himself). Gray, meanwhile, won by less than 200 votes even as his district swung wildly from backing Joe Biden by 11 points to Trump by 5.5, so he clearly ran well ahead of the presidential ticket but his district is moving against his party. Rep. George Whitesides (D, CA-27, northern Los Angeles County), the other California Democrat who flipped a Republican seat in 2024, is less of a Republican target in a Harris +4 seat and starts as Leans Democratic.

— Speaking of California, Rep. Young Kim (R, CA-40, eastern Orange County) had easy races in 2022 and 2024 in what is on paper a competitive Trump +2 seat. She is the only 1 of the 8 Republicans holding presidential bellwether seats—districts that voted for Trump in 2016 and 2024 but Biden in 2020—who starts outside of Toss-up (my colleague J. Miles Coleman wrote about these seats for the Crystal Ball earlier this year). The others are swing seats that Republicans won in competitive 2024 races and should be prime Democratic targets in 2026: Reps. Juan Ciscomani (R, AZ-6, Tucson), David Schweikert (R, AZ-1, Scottsdale), Gabe Evans (R, CO-8, north of Denver), Tom Barrett (R, MI-7, Lansing), Tom Kean Jr. (R, NJ-7, suburban northern New Jersey), Ryan Mackenzie (R, PA-7, Lehigh Valley), and Jen Kiggans (R, VA-2, Virginia Beach). Rep. Zach Nunn (R, IA-3, Des Moines) holds a district similar to the others mentioned here, although Trump carried his district by less than half a point in 2020, which means this district is not a bellwether like the others (Trump carried it by 3 points in 2016 and 4 points in 2024). If Democrats do win the House next year, they likely will have made significant gains from among this group, but there also are not any easy layups for Democrats among them at this early point (and Republicans surely would argue that a few of these Toss-ups should be Leans Republican).

— Reps. Mariannette Miller-Meeks (R, IA-1, Davenport/Iowa City), Scott Perry (R, PA-10, Harrisburg-York), and Derrick Van Orden (R, WI-3, La Crosse/Eau Claire) all hold districts that voted for Trump by at least 5 points apiece, but each ran clearly behind the top of the ticket in 2024. They all could face rematches with their 2024 challengers. So while the districts themselves suggest Leans Republican, the members themselves turned in Toss-up-style performances in 2024, and that’s the rating we’ll start with.

—  A pair of strong performers on either side, Reps. Jared Golden (D, ME-2, non-coastal Maine) and Mike Lawler (R, NY-17, exurban New York City), start with a little benefit of the doubt in their respective Leans columns. Both are also mentioned as possible gubernatorial candidates in their respective states. Either district would instantly be a Toss-up if open, and maybe even Leans Republican in the case of ME-2, which at Trump +9 is the reddest seat held by any Democrat. And these seats may very well be Toss-ups even if the incumbents run for reelection.

— Speaking of statewide runs, Rep. Chris Pappas (D, NH-1, southeastern New Hampshire) just announced a Senate bid to replace the retiring Jeanne Shaheen (D-NH), while Rep. John James (R, MI-10, northern Detroit suburbs) just jumped into the state’s open-seat gubernatorial race. Both seats are rated more competitively than they would be if the incumbents were running for reelection: The Harris +2 NH-1 is Leans Democratic, and the Trump +6.5 MI-10 is a Toss-up. Again, the political situation and usual midterm dynamic helps explain this disparity—if Kamala Harris was president, NH-1 would be a Toss-up and MI-10 would be Leans Republican as open seats. As it is, MI-10 is probably a little more competitive than the presidential topline would indicate: 2024 Republican Senate candidate Mike Rogers only carried it by 3 points, and James himself barely won the seat in 2022 (he won by a better 6-point spread in 2024, similar to Trump). As of now, these are the only two open seats listed in our ratings, but of course that total will grow. As an aside on the Senate, popular former Gov. Chris Sununu (R-NH) just announced that he will not be a candidate in 2026, depriving Republicans of the one candidate who would have instantly made that open-seat race a Toss-up.

— All 13 of the Trump-district Democrats are rated as Toss-up or Leans Democratic. Four are Toss-ups, the aforementioned Adam Gray, Marcy Kaptur, Henry Cuellar, and Vicente Gonzalez. Of the others, Golden and Rep. Marie Gluesenkamp Perez (D, WA-3, Vancouver) were the others we most considered for the Toss-up category. Gluesenkamp Perez could be a Toss-up if she draws a credible challenger—she dispatched a weaker Republican, National Counterterrorism Center director nominee Joe Kent, in 2022 and 2024. Though statewide Democrats typically do not carry WA-3, in the context of the 2024 election, it was one of the nation’s most left-trending districts: Despite Trump’s improvements across the board, his margin there dropped from 4.2 points in 2020 to 3.3 points last year.

— Of the 3 Harris-district Republicans, Don Bacon—a surprising 2024 winner—starts in Toss-up. Mike Lawler is discussed above; Brian Fitzpatrick starts in Leans Republican, perhaps a little aggressive for a proven performer (he could be in Likely Republican), although he did have a somewhat close call in 2018 against a well-funded but flawed Democratic challenger. We’ll see how actually vulnerable he is; he did draw a notable challenger late last week in Bucks County Commissioner Bob Harvie (D). But the main reason we have Fitzpatrick rated as Leans and not Likely is the potential midterm environment (the aforementioned Young Kim is in the same boat).

— Several House Democrats in the Leans Democratic column occupy seats that were not top Republican targets in 2024 but have moved toward Republicans at the presidential level: Reps. Jim Costa (D, CA-21), Jared Moskowitz (D, FL-23), Frank Mrvan (D, IN-1, Gary/Hammond), Nellie Pou (D, NJ-9, Paterson/Meadowlands), and Tom Suozzi (D, NY-3, Nassau County) fit this basic description, among others. We suspect these are closer to being in the Likely column than the Toss-up column, but they’re the kinds of seats that Republicans may be able to win at some point in the near future, maybe even 2026 under the right set of circumstances.

Conclusion

Our bottom line assessment of the battle for Congress is that Democrats should win the House, and Republicans should win the Senate (as we explained in our initial ratings back in February). One way to track 2025 and 2026 is whether either of these assessments look like they are in danger of being upended—does the race for the House seem like more of a true Toss-up in fall 2026, or does the Senate majority seem to be in doubt?

Kyle Kondik is a Political Analyst at the Center for Politics at the University of Virginia and the Managing Editor of Sabato's Crystal Ball.

See Other Political Commentary by Kyle Kondik.

See Other Political Commentary.

Views expressed in this column are those of the author, not those of Rasmussen Reports. Comments about this content should be directed to the author or syndicate.