Trump-vs.-Biden Polls: A Close Race Gets Even Closer

Photo-Illustration: Intelligencer; Photos: Getty Images

Three weeks after Donald Trump’s criminal conviction in Manhattan, his close presidential contest with Joe Biden has gotten closer — but the fundamental contours of the race really haven’t changed. In national polls, Trump’s 0.8 percent lead on May 30 in the RealClearPolitics averages of head-to-head matchups has marginally declined to 0.5 percent. In five-candidate matchups, Trump’s lead has dropped from 2.2 percent on May 30 to 1.3 percent now. In FiveThirtyEight’s unique averages that include Kennedy but not other rivals, Trump’s 1.7 percent lead on May 30 has actually turned into a 0.1 percent Biden lead.

There isn’t a lot of publicly available battleground-state polling since Trump’s conviction, but what we have reinforces the impression of a close and stable race with Trump perhaps losing a few inches of ground. In Arizona, Trump’s head-to-head lead in the RCP averages actually rose from 4 percent on May 30 to 4.6 percent now, though it declined slightly from 5.8 percent to 5.7 percent in the five-candidate averages. In Nevada, similarly, RCP shows Trump’s head-to-head lead rising from 5.4 percent to 5.7 percent since the jury verdict in New York was announced but dropping from 7.2 percent to 6.5 percent in the five-way contest. In Michigan, Trump’s tiny 0.5 percent lead over Biden shrank to an even tinier 0.2 percent lead since the trial verdict; in the five-way contest, Biden and Trump were tied, but now Trump is up by 0.1 percent. In North Carolina, Trump has again gained ground marginally in the head-to-head race since May 30, from 4.8 percent to 5.3 percent. There’s only been one post-conviction five-way poll in the Tar Heel State, and it’s a possible outlier, but his lead at RCP has risen from seven points to eight points. In Georgia as well, Trump has gained in the RCP averages in both the head-to-head contest (4.8 percent to 5 percent) and in the five-way race (6 percent to 6.4 percent).

There are two battleground states, Pennylvania and Wisconsin, where an Emerson poll provides the only post-conviction data, which means comparing the numbers to earlier surveys from the same outlet makes sense. Trump’s two-point lead in a late-April head-to-head Emerson poll in Pennsylvania is identical to his lead on June 18; in the five-way race, Trump has lost a point from his 4 percent lead in April. In Wisconsin, Trump has gained a point against Biden since the April Emerson survey (from 2 percent to 3 percent), but has lost three points from his lead in the five-way contest (from 5 percent to 2 percent).

It’s unclear why Trump might be (nationally and in some states as well) losing a bit of ground in the five-way contest, though Politico (from the Morning Consult tracking poll) reports a post-conviction slippage among independents that could reflect small defections to minor-party candidates or to an undecided position.

All in all, it’s hard to claim evidence for a significant change in the race now that the 45th president is a convicted criminal. Though perhaps the small signs of erosion in his support will quell some of the ever-present panic among Democrats who don’t understand why Biden isn’t rolling toward an easy victory. Trump does, however, maintain a much broader path to 270 electoral votes given his sizable leads in Sun Belt battleground states (Arizona, Nevada, and particularly Georgia and North Carolina). At this point, Biden still needs to sweep the Rust Belt states of Michigan, Pennsylvania, and Wisconsin and avoid upsets elsewhere. Many Trump backers appear to believe that the former president’s overall national lead in most polls guarantees victory given his Electoral College advantage in both 2016 (when he won) and 2020 (when he narrowly lost). But there’s a lot of evidence that changes in both candidates’ bases of support since 2020 could shrink or even reverse Trump’s ability to outperform his popular vote in the Electoral College.

The residual effect of Trump’s criminal conviction may not be fully known until he’s sentenced on July 11. And another criminal trial before November (still possible in the federal case involving the January 6 insurrection) could affect perceptions of the scofflaw former president as well.