There is a real chance Donald Trump will be re-elected.
I think a lot of Democrats haven't really processed this, in the same way we didn't process the possibility that he could be elected in 2016.
Between viewing the initial election as a fluke, and being buoyed by 2018's "blue wave", it's easy to think that Trump is destined to be a single-term president.
But I'm not sure. And I think some of last night's election results -- specifically, the Wisconsin Supreme Court race (in recount territory, but with the Republican candidate up by about 6,000 votes) and a Pittsburgh-area Pennsylvania state senate special election (a Democratic red-to-blue flip) -- can give us some hints about the 2020 electoral landscape.
Let's start with the basics: it is very likely that Democrats will do worse in 2020 than they did in 2018. Midterm elections often represent high water marks for the opposition party, and a monster performance in 2018 doesn't make 2020 a gimme for Democrats anymore than huge GOP victories in 2010 signaled a defeat for Obama in 2012. One could argue that Democratic constituencies tend to drop off in the midterms, but I think that effect is more than canceled out by the advantages midterms give to the opposition party.
And while Trump remains relatively unpopular, there hasn't been the widespread, broad-based repudiation of him that one might have hoped for were we an actually functioning liberal state -- indeed, his net approval rating has been remarkably stable for the past year at roughly -10 (currently it's 42/53). That's obviously pretty bad, but it's not disastrously so -- it hasn't reached that tipping point where "unpopular" turns into a rout.
So my baseline assumption is that Republicans are likely to overperform compared to 2018, and the question is how much Democrats can arrest that trend.
Begin by looking at Wisconsin. If 2018 gave us a narrow Democratic victory in the marquee statewide match up (the Governor's race), that doesn't give us a lot of breathing room in 2020. And lo and behold, the statewide race in 2019 -- a pivotal one, since it effectively determines the partisan composition of the court ahead of 2020 redistricting -- ends up yielding what looks like a squeaker GOP victory. And this was an election where the Democratic nominee had a lot of institutional support (including a massive spending edge), whilst the Republican had been hammered for anti-LGBT views.
This, to me, bodes poorly for 2020. In my "trendlines" post assessing the political movement of each (well, most) states, I put a big ol' question mark next to Wisconsin, and this is part of the reason why. Yes, beating Scott Walker felt great. But barely doing it in a year like 2018 actually wasn't that great a sign. Given Minnesota's increasingly-reddish tilt (albeit from a bluer base), it was hard to imagine that its more conservative neighbor would be more favorable, on net, to the Democratic Party in the immediate future.
A similar problem afflicts Florida. It was perhaps the only "swing" state where Republicans didn't see any drop off in support from 2016 to 2018. So it's hard to imagine that it will be friendlier turf for Democrats in 2020, where they're almost certainly going to have less momentum than they did in the midterms.
That's the bad news. Onto the better news: the Pennsylvania result. This was the first red-to-blue special election flip in 2019, and it was a good one to get -- and not just because it puts the Pennsylvania State Senate in play in 2020. Of all the "blue wall" states that weren't in 2016, Pennsylvania to me seems the most likely to return to the Democratic column in 2020 -- and obviously that's a must-win state for the Democratic Party. And results like this show that Democrats are solidifying their appeal in districts like this one (suburban, recently light-red, well-educated, etc.). That bodes well for Democratic performance even outside of Pennsylvania where there is a considerable cache of voters in districts that share this profile.
So that's the good news. But where does that leave us? Suppose Wisconsin and Florida go red, and Pennsylvania and (we'll say) Michigan return to team blue. If everything else holds (including Trump sniping an electoral vote off Maine), that's still a 270-268 GOP victory. If Maine goes blue down the line, then it's a tie.
Where could Democrats pick up another state? The place where they've seen the best recent trendline is Arizona, followed by maybe North Carolina. But as heartening as Arizona's recent blue-ish shift has been, do you really want to hinge our presidential prospects on a Democratic winning Arizona for just the second time since 1952?
I actually think the popular vote -- for what that's worth, which is basically nothing -- is in good shape for Democrats, if only because of the huge Democratic margins we're likely to see in California. But the electoral college is a different story. The fact is, the tea leaves aren't looking good for Democrats winning either Florida or Wisconsin in 2020, and without those two states the map suddenly looks really, really difficult.