Who’s going to win the American election? With a week to go, the polling and the data all point in one direction, and yet the lessons of 2016, alongside the enthusiasm for President Trump’s campaign, all point in another.

Losing campaigns, of course, always point to polling errors and massive enthusiasm for their side right up until the final moments. What else are they going to say? “Yeah, actually, the polls are right and we’re probably going to lose badly”?

Even if you’re losing, it’s very important for any campaign to make sure that its supporters believe that victory is just within grasp, otherwise they might not vote at all.

With that said, the enthusiasm for President Trump’s campaign is, in fact, quite remarkable. Look at his supporters here, lining the streets for him in Circleville, Ohio:

And this piece, from Byron York in the Washington Examiner provides a glimpse of the remarkable grassroots effort going in to re-elect President Trump:

“It was the biggest political rally no one saw. And gatherings like it have been happening for months in some of the places President Trump needs most to win if he is to be reelected. And, remarkably, the rallies are not the work of the Trump campaign. The road rally in Washington, Pennsylvania, was organized and staged by local Trump supporters, linked together largely by Facebook, who want to show that enthusiasm for the president in western Pennsylvania and surrounding areas is not just strong but stronger than it was when Trump eked out a victory in Pennsylvania in 2016. If Trump wins this critical state, it will owe in significant part to this organic movement and the energetic organizers who have nothing to do with his campaign.”

To the naked eye, President Trump’s campaign sure doesn’t look like a losing campaign, does it? And yet, and yet, and yet… every available bit of data says that that is just what it is.

There are eight full days of campaigning left, including today, until polls open on Tuesday week. Because of the pandemic, though, polls have actually been open in many states for weeks now – and Americans are voting in incredible numbers. In Texas, for example, the total number of votes cast until yesterday was 80% of the total votes cast in 2016. Unless turnout falls dramatically over the last week of the campaign, several states are going to see massive increases in voter turnout compared to last time:

And what of those polls, then, the ones that, for some reason, nobody on either side seems really willing to believe? Well, the truth is that they’re just not pointing to an especially close election. As of this morning, the realclearpolitics average has Biden winning the national popular vote by 8 per cent – 50.8% to 42.8% for the President.

But the national popular vote, as we learned in 2016, isn’t everything. Biden is going to win massive population centres like California and New York at a canter. Even if he loses Texas, he’s going to keep Trump’s margin so small there that realistically, the President’s chances of winning the popular vote are close to zero.

It is worth noting how unusual this is: There have only been five Presidential elections in American history where the winner lost the popular vote and won the electoral college: 1824, 1876, 1888, 2000, and 2016. Nobody has ever done it twice. Until 2016, only once had someone lost the popular vote by more than 1% and still won. Rutherford Hayes lost it by 3% in 1876, and Donald Trump by 2% in 2016.

For Trump to lose by 8% and still win the election would be scarcely believable. It would also, in all likelihood, cause a constitutional crisis.  Whether Trump supporters like it or not, the re-election of a President who had been not just defeated, but roundly defeated, in the popular vote, would lead to many people questioning whether the system itself is just.

In the battleground states, then, which will decide the election, what’s the situation?

Here are the averages for some of the key states:

What you’ll notice there is that there’s no real divide between the national and the state polls, at all. Alaska, a safe Republican state, is closer than Pennsylvania, usually a swing state. South Carolina, a deep red state, is closer than Michigan. And the most competitive states – the ones where the election is genuinely on a knife edge? Here they are:

Texas, Ohio, Georgia, and Iowa are all states that in a normal election, a Democrat should have no business winning. If they’re that close, then that generally points to a landslide victory for the Democrat.

So that leaves one of two possibilities: Either the polling in general is wildly, comically, wrong, from one end of the country to another, and the President’s enthusiastic support will see him home, or the polling is correct, and Biden is cruising to victory.

Go back up and take a look at the graphic above about the massive early voting turnout, by the way. Do you see where those votes are being cast? Texas. Georgia. North Carolina. Montana. Does that tell you that the polls are wrong, or does it tell you that those states are massively competitive, and people are coming out to vote in large numbers in them?

Of course, we’re all spooked by 2016, when the evidence seemed fairly clear that Mrs. Clinton would win, until it was wrong. But the evidence is even stronger this time that Mr. Biden is in the box seat. President Trump pulled a once in a generation rabbit from a hat four years ago. He’s going to have to do something even more incredible over the last week. The one thing he can count on, though, is the boundless enthusiasm of his supporters.