Donald Trump’s approval rating has fallen to 31 percent in new polling — the lowest recorded for any sitting American president in the modern era, according to Al Jazeera — as the 2026 midterm election cycle has simultaneously become the most expensive in American history, with more than $15 billion raised across all federal races. The twin data points arrive three weeks before election day, with both the House and Senate in genuine contention. The approval number matches the structural drag that has haunted Trump’s second term: inflation that has not fully retreated, an increasingly vocal Republican caucus objecting to the Russia diesel deal, and a string of judicial setbacks on immigration enforcement. For a president whose political identity rests on winning, a 31-percent approval number is not simply a poll — it is a narrative that his opponents will weaponise and his allies will scramble to contextualise.
The received wisdom
The mainstream reading treats Trump’s poll collapse as the predictable endpoint of second-term overreach. A president who came into office with a narrow mandate in 2024 has, the argument runs, governed as if he had a sweeping mandate for transformation: restructuring the federal bureaucracy, suspending Russia sanctions, pressuring Ukraine, and deploying executive power against political opponents. The normal midterm penalty — opposition parties typically gain seats in the first off-year after a presidential election — is compounded, on this reading, by the specific unpopularity of Trump’s policy choices. The $15 billion fundraising figure suggests Democrats have successfully converted public dissatisfaction into donor energy. The received wisdom concludes that November will likely deliver a corrective verdict, restoring some institutional balance, and that the republic’s self-correcting mechanisms are functioning more or less as designed.
A different read
The received wisdom is not wrong, but it is incomplete in at least two important respects. The first is the question of what a “corrective verdict” actually delivers in the current configuration of American politics. Democrats won the House in 2018, producing two years of oversight theatre and impeachment proceedings that did not noticeably constrain Trump’s first term. If the same party wins one or both chambers in 2026, the likeliest outcome is a period of legislative gridlock that benefits no one — not Republican governance, and not Democratic policy ambitions, which would remain hostage to the Senate filibuster and a presidential veto. The self-correcting mechanism, in other words, corrects toward stasis.
The second complexity is about what the 31-percent number actually measures. Approval ratings are structurally depressed in second-term presidencies — Eisenhower fell to the low 40s, Nixon to 24 percent at the Watergate nadir, George W. Bush to 25 percent by 2008. Historical precedent suggests that a 31-percent number for Trump, in a polarised electorate where his base has proven unusually sticky, translates into electoral performance that consistently outperforms the polling average. The Republican base is smaller than it was in 2024 but it is also more consolidated — the voters who have defected are the college-educated suburban moderates who were already unreliable Trump supporters. In Senate races in states like Montana, Ohio, and Wisconsin, the question is not Trump’s national approval but whether the specific Republican candidate has sufficiently differentiated themselves from the White House’s most unpopular positions.
There is also a structural point about the $15 billion fundraising figure that deserves scrutiny. NPR’s reporting notes that the midterms “could be the most expensive on record” — a fact that should prompt reflection about what that money is actually purchasing. Since the Citizens United decision in 2010, the correlation between campaign spending and electoral outcomes has weakened considerably at the Senate level. The party with more money loses Senate races with some regularity. What the money does purchase is infrastructure: field offices, data operations, get-out-the-vote programmes, and the sustained ground-level organisation that democracies actually run on. Democrats have consistently outspent Republicans on this infrastructure; the question is whether that advantage persists in states where cultural alignment runs against them.
The historical comparison worth making is not 2018 but 2010 — the Tea Party wave that gave Republicans 63 House seats under Obama. That wave came from a genuine populist energy on the right, channelled through a narrative about fiscal overreach. What the anti-Trump energy of 2026 most resembles, structurally, is that 2010 moment: a genuine mobilisation of the opposition in response to a presidency perceived as ideologically extreme. But 2010 also showed that wave elections produce their own problems — a class of legislators so committed to opposition that governing becomes impossible. The Tea Party class delivered a Republican majority that could not pass coherent legislation for six years, ultimately catalysing the frustration that produced Trump in the first place. If the 2026 Democratic wave delivers a majority of progressives who define success as maximal confrontation with the executive, the long-term structural damage to American governance may exceed the short-term satisfaction of the corrective verdict.
For conservatives watching this moment, the honest assessment is uncomfortable: Trump’s approval collapse is partly the result of decisions that conservatives have long argued were mistakes — the debt trajectory, the personalisation of foreign policy, the use of the Justice Department against political opponents. The midterm penalty, if it comes, will partly vindicate those critics. But the alternative — a Democratic majority defined by Alexandria Ocasio-Cortez’s wing of the party — is not the reform that those critics were asking for. The American political system continues to operate in a space where the choices presented are consistently worse than the population deserves.
What to watch
- Senate race polling in Montana, Wisconsin, and Pennsylvania: if Republican incumbents or candidates maintain leads despite Trump’s national numbers, it confirms the structural resilience argument.
- Whether Trump endorses a specific midterm narrative — “stolen again” positioning — before election day; this would signal preparations for a contested result regardless of outcome.
- Early voting data: high Democratic early-vote totals historically suggest organisational advantages, but 2024 showed that early-vote leads can be overcome by same-day Republican turnout.
- The California counting timeline: NPR notes that California’s extended counting window could delay clear results in closely contested House races for weeks, creating a media narrative vacuum that partisans on both sides will fill with premature victory claims.
— J