I guess I'm not entirely sure about that but you can certainly make a case for that proposition. Now that the early spate of Trump-will-win-because-of-the-economy-and incumbency stories have died down, we're now seeing a number of assessments that look across all the evidence and see him as being in pretty bad shape.
Of course, most Democrats are reluctant even to breathe the sentiment, given how Trump seemed to beat the odds in 2016. But that doesn't mean it isn't true. As Josh Kraushaar of National Journal--not at all a Democratic;leaning guy--puts it:
"Democrats still have so much post-traumatic stress from the last presidential campaign that they’re unable to recognize the obvious: President Trump is a serious underdog for reelection.
It’s remarkable to see the contortions that otherwise-savvy politicians, operatives, and analysts take in order to avoid this reality. President Obama’s former press secretary, Ben LaBolt, fretted in The Atlantic that Trump’s campaign is out-strategizing its Democratic opposition. Obama auto czar Steve Rattner warned in The New York Times that leading economic models predict a Trump landslide....
The reality? Trump is in the weakest political shape of any sitting president since George H.W. Bush. Despite a historically strong economy, his job approval ratings are still badly underwater. He’s never hit 50 percent job approval in any reputable national poll throughout his presidency. At least 40 percent of voters are fired up to vote against him, no matter what happens in the next year. He’s already lost ground with the working-class voters who defected from the Democrats to support him in 2016, with his favorability rating dropping 19 points among that critical Obama-Trump constituency in the last two years.
The latest wave of polling is even more alarming for Trump. His campaign’s own internal polling reportedly shows him trailing in many of the must-win battleground states. A new Quinnipiac survey shows Trump trailing all six Democrats tested against him; what's more, he couldn’t win more than 42 percent of the vote against anyone....In the latest Morning Consult tracking survey, Trump hits 50 percent disapproval ratings in North Carolina, Arizona, Wisconsin, Ohio, Pennsylvania, Michigan, and Iowa—all states he carried in 2016."
Stuart Rothenberg--another not particularly pro-Democratic analyst--considers some of the same data and renders a somewhat more reserved, but consistent, judgement.
"No matter how many economists, political scientists or investment bankers are involved, predictive models based solely on economic data miss the point because they look at only one aspect of a presidency and only one facet of a presidential election. My column from Sept. 18 last year, “Why it’s NOT the Economy, Stupid,” sought to explain why the economy would not be decisive in the midterms and why it might well be less important than usual next year.
Models predicting a Trump wave strike me as more about clicks and being contrarian than about taking a dispassionate look at the 2020 election...
Given Trump’s inability to broaden his appeal and the likelihood that Democrats will be more united and energized than they were in 2016, the Democratic ticket deserves to be given a narrow but clear advantage.
“Tilting Democratic” still seems a reasonable rating to me at this early stage of the race."
And Harry Enten of CNN assesses the situation this way:
"The 2020 election is a long way off. We don't know what Trump will do over the next 17 months. We don't know who the Democrats will nominate.
But Trump likely needs something to change if he wants to win reelection."
Of course, "underdog" doesn't mean "for sure going to lose". The other side of the equation is a strong Democratic campaign that doesn't make too many unforced errors. Gulp.