Democrats and Republicans both have reasons to feel optimistic about 2026

DDHQ is moving to Substack — our first newsletter looks at the trends that give each party hope in the midterms

Welcome to The Bellwether, Decision Desk HQ’s revamped weekly newsletter devoted to the ins and outs of electoral politics. Below you will find a special edition of The Bellwether in celebration of our launch of the new Decision Desk HQ Substack! Moving forward, subscribers will receive the same leading analysis covering election results and race calls, but with plenty of expanded content offerings. Please join us!

---

The 2026 midterm elections are more than a year away, but with all the focus they receive, you might think they were happening tomorrow. President Donald Trump is pushing for Republican-controlled states to redraw congressional maps to help the GOP’s chances, while Democrats are celebrating the entry of high-profile Democratic contenders in the North Carolina and Ohio U.S. Senate contests. In the contest for political power, it’s never not Election Day.

Yet this far out, you can choose your own adventure regarding how the midterms might play out. As the president’s party often does, the GOP could lose significant ground in the House due to Trump’s middling approval rating and energized Democrats winning the turnout battle. At the same time, the anti-presidential party effect might be muted by the circumstances surrounding this election. Broadly, our highly-polarized political era could limit GOP losses, while Republicans stand to benefit from the ongoing redistricting fight and the small number of obviously vulnerable seats the GOP must defend in both the Senate and House.

With all this in mind, let’s take a look at why each party can have hope — and concern — about their outlook for the 2026 midterm elections.

The Democrats have history on their side

Democrats can hope that 2026 features the typical midterm gravity that drags down the president’s party. Midterms almost always give that party some trouble — particularly U.S. House elections, which are most telling about midterm conditions because they constitute the only national vote in a midterm.¹ Tellingly, the president’s party has lost ground in the House in 38 of 42 midterms since the two-party system became Democrats versus Republicans in 1856.²

In the age of public opinion polling, we can see how a president’s job approval rating influences the extent of such losses, too. In midterms dating back to 1946, there is a clear relationship between how low a president’s approval is and how many seats the president’s party loses in the House. Over that period, the president’s party has lost an average of 25 House seats, but the presidents with especially high approval have tended to suffer the fewest losses and those with the lowest have tended to experience the most.

Ahead of 2026, the big question is where Trump could land within this historical picture. His approval rating stands in the mid-40s — about 45%, based on an average of five approval polling trackers, including Decision Desk HQ’s.³ Historically, that would be in line with about a 30-seat loss for the president’s party, though the range of likeliest outcomes around that estimate spans from 22 to 39 seats. If we take the highest and lowest approval figures in those trackers — about 47% and 42% — and look at the range of likeliest outcomes, past results would point to a loss of in the range of 19 to 44 seats.⁴

Only parties with presidents enjoying approval ratings north of 60% have avoided midterm House losses. In 1998, Bill Clinton had a 65% approval rating amid Republican efforts to impeach him, which proved to be political overreach as Democrats gained five seats. Four years later, George W. Bush’s approval rating had spiked to nearly 90% after the terrorist attacks on Sept. 11, 2001, and views of his job performance remained quite positive in November 2002. Republicans wound up gaining six seats in the midterms. Still, high approval hasn’t guaranteed loss avoidance: In 1954 and 1962, Dwight Eisenhower and John F. Kennedy, respectively, still saw their parties drop a small number of House seats while sporting approval above 60%.

Another challenge for Republicans is that Democrats have only a small hill to climb to claim a House majority. At present, Democrats only need a net gain of three seats, which the out party has accomplished in 18 of 20 midterms since World War II. Republicans currently hold a 220-215 majority,⁵ effectively the smallest advantage a party has enjoyed in the House since 1931. Even if Democrats underperform historical trends — say, by gaining only 10 seats — they would still need to clear a very low bar to retake a majority. Understandably, betting markets give Democrats about a 7 in 10 chance of capturing the House.

Moreover, the president’s party nearly always loses vote share in the House from the previous presidential election. In the aggregate House vote, the president’s party’s margin has worsened from two years earlier in 19 of 20 midterms going back to 1946 — with Bush and Republicans in 2002 as the only exception.

And because Democrats already have a small lead in generic ballot polling, they could continue this trend in 2026. A pollster’s generic ballot question asks respondents which party they plan to vote for in the next election. While American voters technically cast votes for a candidate rather than a party, most voters who identify with or lean toward a party will vote for the candidate from that party in November. The generic ballot has a strong track record of forecasting the midterm vote, when there is no presidential race at the top of the ticket to influence the outcome.

Currently, Democrats hold about a 3-to-4 point lead over the GOP in generic ballot polling trackers. It’s still early, but the president’s party usually loses more ground in the polls as we get closer to the election. A small early deficit for Republicans could become much more substantial by the time November 2026 rolls around.

Lastly, midterm turnout trends could help Democrats. Overall, turnout in presidential elections is always higher than in midterms, so the 2026 electorate may differ a fair bit from the 2024 electorate. As a result, surges and declines in participation among certain groups of voters can disproportionately affect the outcome.

One example is the effect of “differential turnout,” whereby voters from the party not in the White House are, all else being equal, more likely to show up and vote in a midterm than voters from the president’s party. This concept connects to the unhappiness with the status quo felt by voters from the out party, which can be a superior driver to contentment among voters who identify with the president’s party. Correspondingly, early polls suggest Democrats are more motivated to vote in 2026 than Republicans.

The “diploma divide” may also be working to Democrats’ benefit in midterms. College-educated voters have shifted toward Democrats, while voters without a four-year college degree have swung toward the GOP. More highly-educated Americans vote at higher rates, so in a lower-turnout midterm, such voters will make up a larger share of the electorate. And because those voters have trended blue, Democrats stand to do better when fewer votes are cast. This won’t necessarily outweigh differential turnout or persuading independent voters in importance, but it helps explain Democrats’ stronger performances in recent non-presidential races — including in the 2022 midterms, when they lost only nine House seats despite Joe Biden’s 41% approval rating.

The GOP has fewer vulnerable seats and redistricting

Yet Republicans have their own positives to extoll when it comes to the 2026 midterms. For one thing, the broad polarization of the electorate into mostly pro- and anti-Trump camps has made it so that Trump’s approval rating experiences relatively small fluctuations. This could mean that his standing is unlikely to significantly worsen between now and next November (although no one knows what the future will bring). And with his high approval rating among Republicans, Trump won’t have to worry about keeping his party unified, despite (for instance) all the coverage of GOP disagreements over the Trump administration’s handling of the Jeffrey Epstein files. These factors could help insulate the GOP from suffering deep House losses.

Another way polarization could benefit Republicans is the sorts of seats the GOP must defend in 2026. Now, this is going to take a couple of charts to explain, so bear with me. Nate Cohn astutely pointed out earlier this month that Republicans have just about the smallest edge in the partisan lean of congressional districts that they have held in the past three decades. That’s good for Democrats: They do not have to win as many GOP-leaning seats to claim a House majority as in years past.

Under the current district lines (more on redistricting shortly), 224 House seats voted more Republican than the country as a whole in the 2024 presidential election, while 211 voted comparably more Democratic. This relatively close split resulted from a more evenly-balanced redistricting process (in partisan outcomes) after the 2020 census.

This marked a big change from the 2010s, when Republicans dominated redistricting after the 2010 census thanks to their huge victories in the 2010 midterms; or in the 1990s and 2000s, when Republicans benefited from factors such as the increased nationalization of politics, the South’s realignment toward the GOP, and reapportionment shifting House seats from the Northeast and Midwest to the South and West.

Yet the more closely-balanced House seats have also aligned in a way that has left the GOP with few immediately vulnerable seats to defend. The 2024 election produced just 16 “crossover districts” represented by a member from the opposite party that carried the seat at the presidential level — tying 2020 for the modern low. Of those, fewer Republicans ended up in left-leaning seats than Democrats landed in right-leaning seats. The GOP finds itself defending only seven seats that leaned more Democratic than the nation under the current lines. That’s the smallest number a presidential party has had to defend in any midterm dating back to 1994 (excluding the 2002 and 2022 post-redistricting midterms, when new seats lacked a clear incumbent party).

The above chart helps explain why some midterms have gone particularly badly for the president’s party. In 1994 and 2010, Democrats had to defend far more seats that leaned to the right of the country (86 and 66, respectively). When those elections went sharply against the Democrats, Republicans had a huge number of seats to readily target. In both midterms, more than 2 in 3 Republican seat flips came in GOP-leaning seats that Democrats were defending.

By contrast, Democrats have had to win more right-leaning seats to make major gains in their recent wave elections. In 2006 and 2018, the GOP defended comparably fewer left-leaning seats (28 and 18, respectively). Democrats' majorities expanded by pushing into redder turf — more than 60% of Democratic seat flips in those years came in GOP-leaning seats.

On the one hand, Democrats would grab a majority if they flipped all seven of the left-leaning seats Republicans hold — remember, Democrats only need three seats to win a one-seat majority. However, a clean sweep would not be guaranteed. In 2010, the GOP flipped 50 of 66 right-leaning seats that Democrats defended, while Democrats did better in 2018 by flipping 16 of 18 left-leaning seats defended by the GOP. Yet to gain a larger majority — by winning, say, 20 seats in 2026 — Democrats will have to capture most of the seats with a small GOP lean. Namely, the 14 Republican-held seats that sit less than 5 points to the right of the country.

Now, for the GOP’s other advantage: The specter of mid-decade redistricting that is haunting the 2026 midterms. In two of Decision Desk HQ’s final Election Spotlight newsletters (succeeded by The Bellwether on Substack), I delved into the partisan makeup of the Texas (Republican-drawn) and California (Democratic-drawn) redistricting proposals. Each plan aims to flip five seats held by the opposition party, so they seemingly could cancel each other out.

The thing is, Trump has called for more Republican-run states to redistrict to the party’s benefit — and there is little question that Republicans stand to gain more seats than Democrats from further redraws. That’s because the GOP has unified control over state government in far more states in which the state legislatures hold redistricting power than Democrats do. The New York Times recently examined the playing field and found that four Democratic-run states could redistrict,⁶ while a whopping 15 Republican-controlled states could potentially redraw.

Although not every state might redraw, even a few more GOP seats from redistricting could complicate the midterm math tremendously. If Democrats could be hard-pressed to flip 20 GOP-held seats, they would have an even harder time if potential GOP-run redistricting states like Missouri and Indiana pick off at least one Democratic-held seat, and Florida targets three or more. Moves such as these would shift the number of total GOP-leaning seats closer to 230 heading into the 2026 midterms. And this doesn’t include the possible political earthquakes of 1) California voters rejecting the Democrats’ plan to create a temporary new map, or 2) the Supreme Court further weakening the Voting Rights Act, which could allow Southern states to eliminate some majority-Black seats that vote heavily Democratic.

Lastly, there’s the Republicans’ advantage in the race for the U.S. Senate. I previously discussed the Senate picture in Maine and North Carolina — the Democrats’ two best targets on the 2026 map — but the Democrats' path to additional seats is steep even if the midterm environment proves highly favorable. After all, Ohio is Democrats’ next-best target, but that’s a state Trump carried by 11 points in 2024. Beyond those, Florida, Alaska, Iowa, and Texas all went for Trump by 13-14 points last year.

Democrats could win one of those races, but could they win at least two while also flipping Maine and North Carolina? That is what they will have to do to win the four seats necessary to overcome the GOP’s 53-47 edge in the Senate (three would be insufficient because Vice President JD Vance can break 50-50 ties for Republicans). Betting markets give the GOP about a 7 in 10 shot of winning the Senate in 2026 — the reverse of the House picture.

***

When it comes to winning the House, you would probably rather be Democrats at this point, given Trump’s approval rating and the tendency for the president’s party to lose ground. But Republicans do not have that many vulnerable seats to defend in the House (or Senate), and the GOP could give itself more seats to work with via redistricting. As a result, it’s easy for Democrats to dream of breaking the GOP’s grip on power in Washington, but also understandable why Republicans believe they could be the first presidential party to retain or gain full control of the federal government in a midterm since the GOP did so in 2002.

1. All 435 U.S. House seats are on the ballot in a midterm, whereas only around one-third of U.S. Senate seats are up every two years.

2. This includes President Andrew Johnson as a Democrat. “Lost ground” signifies a party losing seat share in the House even if it technically increased its seat total in an election after the number of seats in the House increased.

3. The five trackers are Decision Desk HQ, The New York Times, RealClearPolitics, Silver Bulletin, and Strength in Numbers.

4. Data as of Aug. 19, range of outcomes based on a 95% confidence interval.

5. This includes vacant seats with the party that previously held them (currently three Democratic seats and one Republican seat).

6. This includes California, where the state legislature is working to pass a constitutional amendment (that will then need to be approved by the voters in November) to allow it to pass temporary maps in response to Republicans’ mid-decade redistricting efforts.