The more I think about the US election, the more I think we’re underweighting the edge scenarios. And of the edge scenarios, I think the one we’re underweighting most is the one where the Democrats have a really good night.

Why?

Well, let’s start with the obvious. President Trump won by the narrowest of margins in 2016. To demonstrate this, let’s play “spin the wheel”. What we’re going to do is run a little simulation with every state in 2016. We’re going to end up with the same final vote shares – 48.2% vs 46.1% – but we’re going to shake things up very slightly in every state. We’re going to apply a random number between -2.5% and +2.5% to the Democrat, and then do the inverse to the Republicans. Our end vote total – for the country as a whole – will remain the same, but we’re just going to randomly change the votes (just a little) in each state. And we’ll run that, say, 10,000 times.

What happens?

Well, the chart shows the frequency of various outcomes in terms of Republican electoral votes (remember kids, 269 to win!).

What’s amazing is that Trump’s 2016 result (304 electoral votes) is on the far right hand side of this probability distribution. Do a little random shaking of the tin, and he loses EVs.

Now, you might think that President Trump’s victory was the result of electoral genius and Brad Parscale. Yeah, that played a role. But so did dumb luck. The votes could hardly have been any more optimally distributed.

The point I’m making is that in 2020, President Trump doesn’t have a lot to play with. Obama could go backwards a bit in 2012, and still be President. Trump doesn’t have that luxury.

His path was narrow before, and is probably narrower now.

So, here’s my three pointer on why I think President Trump might get hammered in 2020.

1. Trump is (significantly) less popular now than in 2016. And one group in particular has really deserted him – white women. According to Pew, Trump won this group by two points in 2016. The opinion polls now show him trailing here by ten points.

Now, some will say “hey, Trump’s unpopular, but so’s Biden”. Well, that’s partly true. But on forced choice between those people who say they dislike both candidates, 49% to 17% say they’ll chose Biden. Ouch.

2. Many Democrats didn’t come out to vote in 2016. There was a general sense of inevitability about Trump losing, and a lot of people didn’t particularly like Ms Clinton. That depressed Democratic turnout. And this feeds through into the 2020 polls: there is strict turnout weighting in the US, and this means lots of 2012 Obama backers, who didn’t vote in 2016, are not being counted.

3. President Trump is now suffering from a bit of an enthusiasm gap himself. Evangelicals used to give Mr Trump 81% favourability ratings, that’s now 61%, a 20 point drop. His drop among religious Catholics is even larger: a 27 point fall between March and May of this year. Now, I’m not suggesting that the deeply conservative and religious suddenly come out and vote Biden. I’m suggesting some of them (and it only takes a few) stay at home on election day.

Put these all together, and what do you have? I think you have the potential for Biden to win by eight to ten points this time around. Yes, yes, I know the last five or six Presidential contests have been really close. But I’m wondering if this time we could see a blow out result.

Nothing is certain, of course. Trump could pull this out the bag. But my gut says he had a winning coalition in 2016 because he managed to combine economic nationalism, a terrible opponent, and a bit of good fortune (the FBI discvovering a bunch of emails – which turned out to be nothing – a week before voting). The ultimate issue here is that Trump – to win – cannot allow his voter base to shrink. And he has done nothing to appeal to people beyond his base to win. The same people, all of them, need to come out in 2020 for him to win – and even that may not be enough, if the Democrats are more motivated.

Let me leave you with one statistic. Right track / wrong track questions tend to have pretty good predictive power. When people think the country is on the right track, they tend to re-elect incumbents. When they think it is on the wrong track, they are more likely to roll the dice.

Right track / wrong track is now at -38%. That is the lowest number of Trump’s Presidency. Now, it may be that he is able to feed off that. He’s the man who can put the country on the right track… But I think it more likely that the voters choose to say “Adieu Mr President”.

Robert Smithson