The 2026 House Map is No More Favorable to Republicans than the 2018 Map was

KEY POINTS FROM THIS ARTICLE

— Republicans made the House map better for themselves compared to the 2024 map by winning this cycle’s expansive redistricting fight.

— However, comparing this year’s map to the district lines in place in 2018—the last midterm held under President Trump—puts the 2026 House map in a different light.

— By a number of metrics, the 2026 House map is no more favorable to Republicans than the 2018 map, and it may actually be a little less favorable.

— The median House district by presidential performance is less Republican, the number of districts that voted to the right of the nation for president is smaller, and the number of plausible Democratic House targets based on history is comparable.

The House map: 2018 vs. 2026

There’s no question that 2025-2026 mid-cycle redistricting made the overall U.S. House map more Republican-leaning than it was in 2024. The median House seat based on presidential performance, a shorthand way to measure the bias of the overall House map, became roughly 2 points more Republican, and we project Republicans to net a handful of seats from redistricting alone.

But even after these redistricting changes, the 2026 map is not more favorable to Republicans than the 2018 map was. It may even be less favorable, although the differences between the two years are small. Comparing this year’s map to 2018 is warranted given that it’s the most recent election similar to this one, a midterm election held under President Donald Trump.

Compared to the 2018 map, the 2026 map:

1. Has a median House district by presidential performance that is less Republican-leaning compared to the nation.

2. Has fewer districts that voted more Republican than the nation in the most recent presidential election.

3. Has about the same number of districts that profile as plausible Democratic targets based on recent midterm history.

Let’s take these points one by one. The first two don’t require much explanation, but the third one does. Our analysis is based on the Downballot’s invaluable collection of presidential results by congressional district, which the site just updated to account for 2026 redistricting.

The first point involves the median House district. In 2018, the median House district—the one that is right in the middle when sorting the 435 districts from most Republican to most Democratic in a presidential election—voted for President Trump by 2.2 points while he lost the national popular vote by 2.1 points, positioning it 4.3 points to the right of the nation. On the 2026 map, the median district voted for Trump by 4.9 points while he won nationally by 1.5 points, so the median seat is only 3.4 points to the right of the nation—roughly a point less than 2018.

Relatedly, there are fewer districts on the 2026 map that voted to the right of the nation for president than on the 2018 map. On that map, 236 districts voted more Republican than Hillary Clinton’s roughly 2-point national popular vote margin. On this year’s map, 229 districts—7 fewer—voted more Republican than Trump’s 1.5-point popular vote margin.

Again, there is a clear Republican lean on both the 2018 and 2026 House maps, but both of these findings suggest the current GOP lean is a bit more muted.

The final finding requires more explanation, but it also suggests a House map that is no redder than 2018.

This whole analysis is based on how a district votes relative to the nation. This is a way to make comparisons across cycles even as the actual presidential results vary, and we have used it to analyze the kinds of seats that have been vulnerable in recent midterm years (here’s a piece from 2022 looking back at the 2006-2018 midterms, and another from 2025 that incorporates analysis of the 2022 midterm). What we do is set the president’s national share of the presidential vote as a midpoint, and then we categorize districts by how similar they are to that midpoint.

In 2024, Trump won 50% (rounded up from 49.7%) of the national popular vote for president. The districts most similar to the nation, therefore, are ones where he won 49%-51% of the vote, where each district’s Trump share is rounded to the nearest whole percentage point and then slotted accordingly. The districts in the middle of the range encompass D+1 (49% Trump is 1 point more Democratic than 50%) to R+1 (51% Trump is 1 point more Republican than 50%). The R+2-4 range are districts where Trump got 52%-54% of the vote, the D+2-4 range are districts where he got 46%-48% of the vote, and so on and so forth in each direction.

We can also do this for the 2018 districts based on the 2016 presidential vote. In that election, Trump won 46% (rounded up from 45.9%) of the vote, so the midpoint D+1 to R+1 range would be districts where he got 45%-47% of the vote. The findings for both years are shown in Table 1 below, but there’s a bit more to explain before we get there.

In recent years, it’s been hard for either party to win districts with a clear lean toward the other party, which we’ll define as D or R+5 or more. Based on Trump’s 2024 performance, that would be districts where he got 45% or less (the more Democratic districts) or 55% or more (the more Republican ones). In fact, Democrats did not win any R+5 or greater districts in 2024, nor did Republicans win any D+5 or greater districts, although redistricting means that each party is technically defending some of these districts this year. Democrats, aided by the midterm political environment, may be able to flip or hold some R+5 or more districts. In our ratings, one incumbent Democrat impacted by redistricting is actually favored despite holding a district in this group—Rep. Henry Cuellar (D, TX-28)—and several others are rated in Toss-up or Leans Republican. Democrats also have some offensive targets in R+5 or greater districts, although we favor Republicans in those districts in our current ratings. Republicans very likely won’t win any D+5 or greater seats this year.

Recent history bears out that to the extent the non-presidential party gains seats in a midterm, those gains tend to peter out above the D or R+5 range (depending on which party is on offense). Here’s the brief history:

2006: Even at a time when Democrats still held or were able to flip many more clearly Republican-leaning districts than they are able to now, about two-thirds of their gains in 2006 came in districts that were R+4 or bluer.

2010: Despite a gigantic overall net gain that pushed them from just 178 seats in 2008 to 242 in 2010, Republicans did not flip any seats that were D+5 or greater, although they did successfully defend a couple.

2014: Republicans netted 13 seats while winning 247 districts overall, but they only netted a pair of additional seats in the D+5 or more group.

2018: Democrats did cut a little deeper into this group, although only about a fifth of their 41-seat net gain came in R+5 or greater districts, while the rest came in districts that were less red than that.

2022: Republicans netted a modest 9 seats, although they, like the Democrats this year, were starting from a high floor, as they had won 213 seats in 2020. But Republicans only flipped a single seat in a D+5 or greater district.

We note this history just to say that while Democrats may be able to target or even win some R+5 or greater districts this year, the R+4 line (or D+4 line in a Republican-leaning midterm) is a reasonable point to separate the more plausible targets for the non-presidential party from the less plausible ones.

So let’s compare 2018 to 2026 with that marker in mind. Table 1 compares the two maps, categorizing all 435 districts based on the rubric laid out above.

Table 1: House districts by presidential performance, 2018 vs. 2026

Note: Districts are categorized by rounding President Trump’s 2016 or 2024 share of the vote in each district to the nearest whole percentage. The 2018 districts account for redistricting in Pennsylvania, which was the only state that redrew that cycle, while the 2026 districts account for the 10 states that redrew this cycle.

Source: Compiled by author from the Downballot.

There are a lot of numbers here, but the key one for our purposes is summed up at the bottom of the table: On the 2026 map, there are 234 districts that are R+4 or bluer. On the 2018 map, there were 233 such districts.

That’s basically the same, but on this metric too, the 2026 map is very slightly better for Democrats than the 2018 map was.

To get a sense as to how the 2026 map might have performed in 2018, we looked at the percentage of races Democrats won in each of these categories in the actual 2018 elections, and then we applied those percentages to the 2026 map.

That produces 239 Democratic seats, or 4 more than the 235 Democrats actually won in 2018.

Now, we wouldn’t take this too literally: beyond map changes, the country has also shifted politically since 2018, with changes that have altered both party coalitions.

But here’s the point: If the 2026 House map was worse for Democrats than the 2018 map, we’d expect that if we applied the 2018 win percentage for Democrats in each of these categories to the 2026 map, they would win fewer seats, not more.

None of this should be taken to suggest that we think Democrats are going to do better in 2026 than 2018 (at this point we don’t think they will, even though we continue to favor Democrats to win the majority). There’s much more than just the map that goes into producing the actual House results. And however Democrats perform in this election, it’s very likely that they would have done better had there been no redistricting this cycle.

But if Democrats could win the House on the 2018 map, they can win it on this one, too.

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.