Recent state-by-state polls tracking President Trump's approval rating have Democrats feeling very optimistic about their 2020 chances. This week, polling company Civiqs unveiled a poll showing Trump's approval rating to be negative in swing state after swing state, including North Carolina (-9), Michigan (-11), Pennsylvania (-6), and Iowa (-3). His approval was dead even in Florida and Wisconsin. The only swing state in which his number was in the black was Ohio. Ominously for the president, he is also struggling in map-expanding states for the Democrats like Georgia and Arizona.

If Trump in 2020 lost all the states where his approval is currently net negative, he would get clobbered in the Electoral College in a nearly precise inverse of his 2016 margin, 311-227. This holds true even if you give him Florida and Wisconsin. But before Democrats start prematurely popping those sparkling wine corks and envisioning their triumphant election night parties, they should consider recent history, which suggests that Trump, while deeply vulnerable, may be in better shape for the general election than these grim numbers suggest.

Trump's weakness is real, and has been since even before he took office. The president begins his campaign for re-election with unique vulnerabilities not shared by the last three presidents re-elected for a second term. He lost the popular vote decisively in 2016, and was only elected after winning a series of purple states by achingly narrow margins, a fluke he would have to repeat once again in 2020, since he is unlikely to carry any states Hillary Clinton won in 2016. And unlike former presidents Bill Clinton, George W. Bush, or Barack Obama, whose early days in office were marked by broad and bipartisan approval, Trump has never spent a single day in office with an average approval mark above 50 percent. In fact, only the highly suspect Rasmussen poll has ever shown him with an approval rating at or above 50 percent after inauguration day, a truly remarkable record of consistent and deep unpopularity unmatched in the history of polling.

The president is also hampered by the ongoing unpopularity of his signature policy issues. As of January, 60 percent of Americans said they opposed significant new construction of border fencing. Still, the president chose to declare a national emergency and claim for himself the right to re-appropriate Congressional funds for building a wall. And even though most Americans support keeping immigration levels the same, the administration is doubling down on both its inhumane border policies and its quest to reduce legal immigration.

The administration also decided to back a frivolous lawsuit seeking to declare the Affordable Care Act unconstitutional, despite public backing of the law by an average of 11 points. The GOP's health-care nihilism was a major driver of their rout in the House of Representatives last November, and there is no sign that the fundamental dynamics of that issue have changed: Voters rightly fear that Republicans will eliminate essential health benefits, including protections for pre-existing conditions, and trust Democrats over Republicans to craft solutions to the system's ongoing problems.

With regard to trade, while the president has succeeded in turning many Americans against China, thus far voters do not believe his tariffs have benefited the American economy.

The president's team has been crowing incessantly about his "total exoneration" in Special Counsel Robert Mueller's investigation, and promising to use the Justice Department to conduct a revenge quest against anyone involved in launching the probe. But there are still countless potential scandals looming that could cripple the president at critical moments during the campaign, including of course the Mueller report itself, should it ever see the light of day.

Does all of this mean Trump is doomed? Certainly not. The most important thing to remember is that a certain percentage of people who say they disapprove of the president's job performance are going to vote for him anyway, because an estimation of job performance is not equivalent to voting intentions.

Let us also not forget that in August of 2012, Gallup's state-by-state approval numbers for Obama looked grim. But on election night, he beat those approval numbers by an average of about 4.5 points, including by 8 in Colorado, 9 in Maine, and 5 in Virginia, all blue-leaning states where pre-election polling found that a majority disapproved of his performance. In the end, many of those voters reluctantly supported Obama's bid for a second term.

The same thing could certainly happen with Trump next year. If you add 4.5 points to Trump's Civiqs numbers, he would win Wisconsin, Iowa, Florida, Arizona, Georgia, and Pennsylvania, and thus re-election by a comfortable margin in the Electoral College. And there are a variety of ways for him to get there. His approval with Republicans was 84 percent in the most recent Politico/Morning Consult survey, and he's likely to improve that number on election night, since he won 88 percent of self-identified Republicans in 2016. He could also cut into his deficit with independent voters, especially if the economy remains strong or he succeeds in re-setting the terms of trade with China in ways that obviously benefit American manufacturers.

But just as Democrats shouldn't assume that Trump's lousy approval numbers mean he is condemned to a single term, neither should Republicans count on the president exceeding his job performance ratings. While state-by-state data is hard to come by for 2004, in May of that year, Bush led in that cycle's battleground states 50-47. He ended up getting an average of just under 50 percent of the purple state vote, meaning that his election night numbers barely diverged from approval ratings at all. And even though Bill Clinton's approval rating before the 1996 election was 58 percent, he only captured 49 percent of the vote. Despite the unusual three-candidate dynamics of that particular race, this is proof that even some people who do approve of a president's job performance will nevertheless cast a ballot for someone else.

Trump is a dour and divisive figure whose lawlessness and corruption have disfigured American politics in unprecedented ways. Yet he is likely to be a formidable opponent for any Democrat in 2020, unless there is an economic crisis that cuts into his floor. So while Democratic hopefuls may be pleased the president is unpopular in so many critical states, they should nevertheless prepare themselves for what will be a difficult, expensive, and ugly contest.