President Trump’s closing argument ahead of midterms has been imploring voters to pretend that he is on the ballot.
That being said, midterms are referendums on the president whether or not he asks it to be, and Trump’s approval has fallen so far that Republican candidates may be best served if voters pretended the opposite – that Trump’s name is nowhere near theirs on ballots.
Comparing Trump’s approval now – overall and on the issues – versus where it was in 2018, when Democrats won a 41-seat landslide highlights the problems facing Republicans.
Indeed, Trump’s current average approval rating sits slightly above 36%, per Nate Silver’s aggregator.
Individual polls show it lower. Last week’s Washington Post-ABC-Ipsos survey put Trump’s approval at 33%.
Similarly, a recent Reuters/Ipsos poll revealed just 31% of Americans approve of Trump’s job performance.
And in the final New York Times/Siena survey before midterms, Trump’s approval sat at 34%, down three points since September.
Notably, in all three, Trump’s approval was the lowest either of his political career or since the survey’s creation.
Critically, history provides important context, particularly looking at 2018.
At this point in the 2018 cycle, Trump’s average approval sat at 44%, while 52% disapproved, per RealClearPolitics.
To be sure, the overall approval numbers may have declined since Trump’s first term, but the biggest difference is how Americans feel about specific issues.
In October 2018, seven-in-10 (70%) Americans said the economy was in good shape, prompting CBS News to write at the time, “Americans are feeling the most positive about the state of the nation’s economy heading into a midterm…since 1998.”
On the economy, 52% of Americans approved of Trump’s handling of the economy in that 2018 poll, while 42% were optimistic that the economy would continue improving.
Today, Trump’s average approval on the economy (33%) is 19-points lower than it was in the 2018 CBS News poll.
And, contrary to how Americans felt in 2018, just 20% now say the economy is in good shape, with a similarly dismal 18% optimistic about the economy’s future, according to Thursday’s CNBC All-America Economic survey.
Yet, despite Trump’s relatively high grades in 2018, Democrats still routed Republicans that year, winning a massive 41 House seats. According to polling from the time, the overwhelming majority (69%) of voters cited Trump’s behavior as a key factor in their vote.
Put another way, in 2018, voters were fed up with “Trump the man” but generally approved of his policies. Eight years later, voters are angry at the policies in addition to disapproval with some of the personal acts.
In that same vein, frustration with Trump’s policies is now showing up among his past supporters. According to Pew Research, Trump’s approval among his 2024 voters has fallen from 95% in February 2025 to 67% now.
Within the larger GOP voter base, whereas 91% of Republicans approved of Trump’s handling of the economy in 2018, that has fallen to 69% according to Economist/YouGov trackers.
Midterms are often decided by turnout, so a demoralized base could be fatal to Republican candidates.
Zooming out, there is one more historical reason for Republicans to be concerned: the statistical connection between a president’s approval and his party’s midterm performance.
In eleven of the 20 midterms since 1946, the president has entered November with an approval rating below 50%. In those eleven, the average loss for the president’s party was 35 House seats, according to the American Presidency Project.
Even popular presidents struggle with midterms. In the nine midterms where the president’s approval was above 50%, they still lost an average of 15 House seats.
To that end, the relationship is real, but it is not a perfect indicator, accounting for roughly one-third of the variation in House outcomes since 1946.
Even so, Trump is towards the bottom of the chart. His approval is slightly above former President Truman in 1946 (27%), and essentially tied with former President George W. Bush in 2006 (37%), though they lost 56 and 30 seats, respectively.
Does that mean Republicans will suffer a 1946 or 2018 style defeat this time?
While anything is possible, there are reasons to think that even if Republicans do lose, it likely won’t be a landslide.
Of the 220 House seats Republicans are defending, only 16 are considered toss-ups by Cook Political Report, a result of the battlefield narrowing significantly due to gerrymandering and polarization.
Mid-decade, Republican-friendly redistricting also shifted a theoretical nine seats in their favor.
Nevertheless, with the House often seeing considerable midterm changes, the Senate then becomes the sharper test of the impact of Trump’s ratings.
Since 1946, unpopular presidents have lost an average of 4.5 Senate seats. With Democrats needing a net-gain of four seats, that should put the Senate squarely in play, and Trump’s unpopularity is tilting a number of races.
Six Republican-held Senate seats are now considered toss-ups, including Kansas, Alaska, Iowa, and Texas, all typically safe for Republicans.
Democratic candidates are also ahead in battleground states North Carolina, Ohio, Maine, Michigan, and Minnesota – although outside of North Carolina, all are within the margin of error, thus essentially tied.
However, the impact of presidential approval in the 30’s is that the range of outcomes for Democrats to retake the Senate has widened considerably.
To be clear, these races are very much winnable for Republicans, even if the field is shifting towards Democrats. It is possible that polls could be undercounting Republicans, particularly in Kansas, Iowa, Ohio, and Michigan.
Ultimately, the full impact of Trump’s approval numbers on the midterms will only be known after extensive post-election analysis.
However, the current and historical data is clear: unpopular presidents often drag down their entire party, and with Republicans’ House and Senate margins relatively thin, that could be enough to shift both chambers back to Democrats.
Douglas Schoen is a longtime Democratic political consultant