Trump’s Poll Numbers Look Brutal, But Midterm History Is More Complicated

Another week, another batch of ugly poll numbers for President Trump. But with the midterm elections now just weeks away, the more interesting question isn’t whether Trump’s approval rating is bad. It is what numbers like these have historically meant for the party occupying the White House when voters head to the polls in November.

On that front, there is a lot of history to work with, and much of it should make Republicans uncomfortable.

A new American Research Group poll released Monday found just 29 percent of Americans approve of Trump’s job performance, compared with 68 percent who disapprove. His approval on the economy is even lower at 26 percent, while 71 percent disapprove.

Those are brutal numbers, but there is an important caveat: ARG is currently on the low end of national polling of Trump. Other recent surveys have put his approval higher, including 35 percent in a Reuters/Ipsos poll and 39 percent in a Fox News poll.

That spread matters. No single poll should be treated as a prediction of what will happen in November. But even the better numbers for Trump put him in territory that has historically been difficult for the president’s party in a midterm election.

The Midterm History

Gallup has tracked the relationship between presidential approval and midterm elections going all the way back to 1946. The basic pattern is hard to miss: presidents with approval ratings below 50 percent have seen their parties lose an average of 37 House seats.

Presidents above 50 percent have fared considerably better, with their parties losing an average of 14 seats. There are exceptions in both directions, but presidential approval has historically been closely associated with the political environment facing the president’s party.

The table also shows why approval ratings should not be used as a seat-loss calculator. Obama was at 45 percent approval in 2010 and Democrats lost 63 House seats. Four years later, Obama was at a nearly identical 44 percent and Democrats lost only 13.

Trump was at about 40 percent before the 2018 midterms and Republicans lost 40 House seats. Biden was also at 40 percent in Gallup’s final pre-election reading in 2022, yet Democrats lost only nine.

There are plenty of reasons for the enormous differences. The number of seats a party already holds matters. So do the economy, candidate quality, congressional approval, district boundaries, turnout and the political issues dominating a particular election.

Still, the larger pattern is difficult to ignore. Low presidential approval has repeatedly accompanied House losses for the president’s party, even if it cannot tell us how large those losses will be.

Trump Is Running Below 2018

Perhaps the most useful comparison for 2026 is Trump himself.

Gallup measured Trump’s approval at 40 percent in mid-September 2018 and 40 percent again immediately before that year’s election. Republicans went on to lose 40 House seats as Democrats captured the majority.

We don’t have a current Gallup number to make a direct 2026 comparison, but other pollsters provide some context. Pew Research Center had Trump at 34 percent in July and noted that his rating was lower than it had been at a comparable point in his first term.

ARG offers an even cleaner Trump-to-Trump comparison because it polled both elections. The organization had Trump at 37 percent approval in September 2018. Its latest survey has him at 29 percent, an eight-point drop from the same point eight years ago.

Polls use different methodologies and survey populations and should not be treated as directly interchangeable. Pew’s survey was conducted July 6–12; the other results listed are from September.

The methodologies and survey populations differ, so those numbers shouldn’t be averaged together casually. What they do show is that Trump’s standing varies considerably by pollster while remaining underwater across these surveys.

The Economy May Be the Bigger Problem

The ARG numbers become even more striking when the questions turn to the economy.

Only 26 percent approve of Trump’s handling of the economy, compared with 71 percent who disapprove. Among independents, Trump’s economic approval falls to 19 percent, while 75 percent disapprove.

ARG also found just 5 percent of Americans saying the national economy is getting better, while 72 percent said it is getting worse. Only 6 percent expect the economy to be better a year from now, and 70 percent expect it to be worse.

Even people who approve of Trump are pessimistic in the survey. Among Trump’s approvers, 56 percent told ARG they expect the national economy to be worse a year from now. ARG surveyed 1,100 adults from Sept. 16-20 and reports a theoretical margin of error of plus or minus three percentage points.

Other polling paints a less extreme but still difficult economic picture. The latest Fox News poll found roughly three-quarters of voters rating economic conditions negatively, while Reuters/Ipsos found Trump’s overall approval at 35 percent and Democrats leading Republicans 44 percent to 37 percent when respondents were asked which party they would support for Congress.

History Is a Warning, Not a Forecast

There is no formula that takes a president’s September approval rating and spits out the number of House seats his party will lose in November. The 2010, 2014 and 2022 elections alone should put that idea to rest.

There is, however, a long historical relationship between presidential popularity and midterm performance. Gallup’s data going back to 1946 shows that unpopular presidents have generally created a much more difficult environment for their party’s House candidates.

That matters even more when the House majority is narrow. Republicans began the current Congress with a 220-215 advantage, meaning control of the chamber does not require anything resembling the 40-seat swing of 2018 or the 63-seat wave of 2010.

There are still weeks left before Election Day, and presidential approval can move. Individual House races will ultimately be decided by individual electorates rather than a national approval number.

But the question hanging over Republicans is becoming harder to dismiss: If Trump’s approval remains somewhere in the 30s as voters begin casting ballots, can GOP candidates outrun a political environment that has historically cost the president’s party seats?

We will find out in November.