It’s widely assumed that the 2020 voting in Wisconsin will go a long way toward determining whether President Trump serves a second term. And a new poll finds that every single leading Democratic candidate is roughly tied with him in the state.

This leads to a number of conclusions, some of which are promising for Democrats — but others that are heartburn-inducing, and at least one that is truly alarming.

There’s a decent chance — though it’s hardly certain — that any of the leading Democrats would be able to take back Pennsylvania and Michigan, two “blue wall” states Trump cracked. But if Trump holds Wisconsin, he can still prevail in the electoral college, if pretty much all else goes as it did in 2016.

By contrast, if Trump loses in Wisconsin, it’s all but certain he’ll lose Pennsylvania and Michigan as well, and with them the election.

The new poll from Marquette University Law School — the gold standard of polling in Wisconsin — finds that former vice president Joe Biden is slightly ahead of Trump among registered voters in the state by 47 percent to 46 percent.

Meanwhile, Trump just edges the other candidates: Against Sen. Elizabeth Warren (D-Mass.), it’s 45 percent to 44 percent. Against Sen. Bernie Sanders (I-Vt.), it’s 47 percent to 45 percent. And against South Bend, Ind., Mayor Pete Buttigieg, it’s 44 percent to 43 percent.

That means all the leading Democrats are basically locked in statistical ties with Trump in the state he almost certainly must win.

It’s still very early, of course. But this is a reminder that it’s very possible that the 2020 election could come down to Wisconsin — and that it could be incredibly close and hard-fought.

This leads to several takeaways.

First, this is a reminder of why Democrats are so worried about the state. Because Wisconsin has an outsize percentage of blue collar whites, Democrats there feel extremely uncertain about how deep Trump’s pool of support among those voters really runs.

Which means that in Wisconsin, a dynamic we’ve seen again and again could save Trump’s reelection chances. While Democrats have made big gains in the suburbs and among educated whites, driving wins in very red places, those gains have almost been matched by boosts in turnout and vote share for Trump among rural, exurban and non-college white voters.

Trump is further polarizing the electorate. So in Wisconsin, the Democratic nominee very well may make similar gains among suburban and more educated white voters, while enjoying high turnout among young and nonwhite voters — which means running up the score in places like Milwaukee and its suburbs, and in Madison.

But Wisconsin Democrats worry that Trump could still prevail by electrifying turnout in rural and exurban Wisconsin — tapping pools of voters whose depths are unknown. That’s why Democrats there are already organizing feverishly in urban areas to find every voter they can, nearly a year before Election Day.

The closeness of these new numbers is another reminder that the state could be tipped either way by how those efforts go.

The second and related takeaway is that Democrats will have to develop numerous paths to an electoral college win. If the Democratic nominee can win Arizona it could offset a loss in Wisconsin, as could a win in North Carolina or, best of all Florida (which would mean an all-but-certain Trump loss).

This doesn’t lead to a particularly satisfying conclusion, since it’s easy to say that Democrats need to fight everywhere, and it’s hard to figure out exactly how to get that balance right. But it does mean Democrats have to try to build a coalition that can win in the Sun Belt while also fighting as hard as possible to win back all three Rust Belt states — because, after all, if Democrats don’t win an additional Sun Belt state, which is very possible, they’ll need Wisconsin.

In some ways, of course, the fact that Trump is struggling so hard to hang on in archetypal Trump country amid a good economy underscores his weakness and unpopularity, and the unpopularity of policies like his trade wars and his massive corporate tax cuts. Given the economy, Trump surely expected to be seen in such places as a smashing success, but now he’ll likely have to pull out another miracle to win.

Yet Trump could end up doing just that. Which leads to another takeaway.

The continuing disparity we’re seeing between the national polls showing Democrats with comfortable leads, and polls like this new one showing it razor-close in places like Wisconsin, underscores the possibility that Trump could lose the popular vote by even more than he did in 2016, yet still win via the electoral college.

Such an outcome, coming after Trump gets impeached and (presumably) acquitted via a sham trial in the Senate, is an unsettling prospect indeed.

Read more: