Donald Trump has given Republicans a simple instruction for the midterms: vote as if he were on the ballot.
At the Republican midterm convention in Dallas, he asked supporters to “pretend, please, that I am on the ballot.” The line worked because it captured the problem Republicans are trying to solve. Trump is still the party’s dominant political figure, but November will be decided in an election in which his name does not appear.
That makes the 2026 midterms more interesting than the usual referendum on an incumbent president. The central question is not only whether voters are dissatisfied with Trump. It is whether the voters who changed the Republican coalition in 2024 have become reliable Republican voters, or whether many of them remain voters who are most likely to participate when Trump himself is running.
Pew Research Centre’s validated voter study found that Trump won 48 per cent of Hispanic voters in 2024, up from 36 per cent in 2020. His support among Black voters rose from 8 per cent to 15 per cent. He also improved among Asian voters and remained dominant among voters without college degrees. These gains did not amount to a wholesale partisan conversion. Pew found that turnout changes mattered more than mass switching between the parties.
Among people who voted in the 2020 presidential election, Trump and Kamala Harris were nearly even in the 2022 midterms and the 2024 presidential election, with Trump ahead 50 per cent to 48 per cent. Among eligible voters who participated in 2024 but had skipped both 2020 and 2022, Trump won 55 per cent to 41 per cent. That group represented 12 per cent of the 2024 electorate, according to Pew’s turnout analysis.
Trump also drew irregular voters into the electorate, creating a different problem for Republicans: a coalition built partly on less-frequent voters may be harder to reproduce when the candidate who mobilised them is absent.
But durability has more than one meaning. Trump’s voters could remain Republican in preference without becoming regular Republican voters; they could also support congressional candidates without transferring their loyalty to a presidential successor. The midterms therefore test partisan alignment and turnout. Whether another Republican can inherit Trump’s personal connection with these voters is principally a question for 2028.
Early signs in 2026 suggest the problem is real, although they do not yet prove it will decide the election. In July, Pew found Democrats ahead 43 per cent to 37 per cent on the generic congressional ballot among registered voters. More relevant to the turnout question, 39 per cent of Democratic and Democratic-leaning voters said they had thought a lot about the midterms, compared with 22 per cent of Republican-leaning voters. Seventy per cent of Democrats said it really mattered which party-controlled Congress, while 60 per cent of Republicans said the same.
The gap has persisted as the campaign has intensified. A Reuters/Ipsos poll released this week put Democrats ahead 44 per cent to 37 per cent in congressional preference and measured Trump’s approval at 35 per cent. The latest CBS News Battleground Tracker estimated 228 Democratic House seats, enough for a majority, and found that Democratic voters were more enthusiastic about participating than Republicans.
Those numbers should not be reduced to a story about Trump’s personal appeal. CBS found that prices and inflation remain central to voter sentiment, while opposition to new tariffs and the Iran conflict is also shaping preferences. Those pressures can weaken Republican enthusiasm independently of Trump.
Republican strategists have noticed the broader turnout problem. Americans for Prosperity Action, a conservative political organisation, found that about one in five reliable Republican base voters in six states were at risk of staying home or considering Democratic candidates, according to reporting by The Washington Post. Those voters were primarily worried about inflation and the economy. Their hesitation sits alongside the party’s longer-standing challenge of turning out less frequent voters who support Trump in presidential elections.
This produces a peculiar asymmetry. Trump may be more effective at mobilising marginal Republican voters when he is personally on the ballot than when he is campaigning for others. Yet his political presence appears capable of mobilising opponents even when he is absent from the ticket. Pew found in July that 42 per cent of registered voters viewed their congressional choice as a vote against Trump, compared with 22 per cent who saw it as a vote for him. Republicans therefore face a double challenge: reproducing the turnout benefits of Trumpism without Trump as the candidate, while continuing to bear much of the counter-mobilisation his presidency generates.
That also complicates how to interpret November. Republican losses would not by themselves prove that Trump’s coalition has dissolved. They could reflect the familiar costs of incumbency, dissatisfaction with prices or foreign policy, or unusually motivated opposition. The more revealing evidence will be found underneath the headline result: whether the demographic and less-frequent voters who moved towards Trump in 2024 remain Republican in preference but fail to turn out, or whether they actually move away from the party.
Trump faced a version of this problem in 2018, when Democrats gained 41 House seats and took control of the chamber even as Republicans expanded their Senate majority. Trump was central to the election: 64 per cent of voters said he influenced their decision, with more casting their ballot against him than for him. But 2018 is an imperfect precedent. The coalition Republicans must mobilise today contains substantially more of the less-frequent voters Trump brought into the electorate by 2024.
The difficulty of transferring a president’s personal coalition to his party is not unique to Trump. Barack Obama’s 2008 victory rested partly on exceptional mobilisation among young and minority voters. Two years later, participation among young Hispanic voters fell sharply relative to older and white voters, and Republicans captured 242 House seats. Obama and Trump built very different coalitions, but they illustrate the same institutional problem: distinctive presidential candidates can reach beyond their parties’ habitual electorates, yet winning those voters once is different from turning them into regular partisan voters.
The comparison points to a broader feature of candidate-centred politics. Charismatic or unusually distinctive presidential candidates can assemble electoral coalitions that exceed their parties’ normal organisational reach. But winning those voters once is different from incorporating them into a durable partisan electorate. Parties institutionalise political realignments only when voters begin turning out for less compelling candidates, in lower-salience elections and eventually for successors they have never supported before. Obama’s experience showed how difficult that conversion can be. Trump’s Republican Party is now confronting its own version of the same problem.
If Republicans can reproduce Trump-era turnout patterns when Trump is not a candidate, his transformation of the electorate will look more durable than a personal political phenomenon. It would suggest that voters who entered or returned to politics because of him have developed a partisan attachment that can survive his absence from the ballot.
The question will follow the party into 2028, when a successor will have to persuade Trump voters to participate without being Trump. Vice President JD Vance used the Dallas convention to step more clearly into the role of a possible heir to the movement. He or another Republican contender can inherit much of Trump’s agenda on immigration, trade and culture. What cannot be transferred by endorsement is the relationship Trump built with voters who saw him as something different from a conventional Republican. A successor can inherit a platform. It is much less clear that he can inherit an electorate.
If Democrats win the House in November, the immediate interpretation will be that voters rejected Trump’s second term. There would be good reason for that conclusion, particularly when prices, inflation and opposition to the Iran conflict are already shaping voter preferences. But a midterm electorate is not the same as a presidential electorate. Some voters who backed Trump in 2024 may simply sit out this November and return when the presidency is at stake again in 2028.
A Democratic House victory would naturally be read partly as a rejection of Trump’s second term, particularly amid dissatisfaction over prices and foreign policy. But it would not by itself establish that his 2024 coalition had dissolved. Some Trump voters may simply sit out a midterm and return when the presidency is at stake.
The most important evidence on election night will therefore lie beneath the House seat count. If Trump’s less-frequent 2024 voters remain Republican but disproportionately stay home, the party will have a mobilisation problem rather than a coalition problem. If they turn out but move back towards Democrats, the realignment itself will look less durable. If they both participate and remain Republican, Trump may have accomplished something more consequential: turning a personal electoral breakthrough into a partisan one.
Trump has already shown Republicans how to reach voters whom the party once struggled to mobilise. What he has not shown them is how to keep those voters once he is gone. November will offer the first serious indication of whether Trump has transformed the Republican Party—or merely transformed the electorate when he is on the ballot.
