8 October 2020

Will Biden win? US Election 2020

Four years on from 2016, and we're back looking at polls, swings, gains, losses, etc. Well, in a way, it's only one year on from the last time we did this for the 2019 UK General Election.

First, let's get the disclaimer out of the way: I am not a predictor; I am merely attempting to interpret the polls in terms of seats electoral college votes. In 2016, this is something that I had great success with in a way. The takeaway point in 2016 was that the swing in the marginal states was higher than the national swing, which was enough to push Donald Trump into the White House. https://rhysbenjamin.blogspot.com/2016/11/how-did-trump-win.html

However, Trump has a very difficult task if he wants to stay there.

Theoretically, the swing that Biden needs to win is just 0.4%. That's 4 out of every 1,000 Trump 2016 voters going to Biden in 2020. You can also see this in his "easiest" path to the White House: gain Michigan, Pennsylvania, and Wisconsin. 



What we actually find is that, as with 2016, the current swing in marginals appears to be bigger than the national swing. If we use the swingometer we see that if we place the arrow on what the polls say is the national swing, 2.8% (using the average of the different "poll of polls" models, so basically the poll of poll of polls). On a uniform national swing Joe Biden will get a landslide, looking at 350 electoral college votes to 188. 



But what are the "marginal" states in this election? It's time to have a look at...

THE BATTLEGROUND.

This is where the election will be won and lost. These are the states in their 2016 colours, with the white line representing, to an extent, Joe Biden's "winning" post. Starting on 232 electoral votes, he needs to gain Michigan (+16), Pennsylvania (+20), and Wisconsin (+10), if he does it in order of difficulty, to get over the 270 line, because, of course, 232 + 16 + 20 + 10 = 278. 

If Biden gains Florida, that means, however, he can afford to lose any two of those three states and still win the election. Even Florida + Wisconsin gives Biden 271. If Biden loses all three of Michigan, Pennsylvania, and Wisconsin, however, or indeed if Trump makes gains (not currently predicted, it has to be said), in any of these seven targets below, then Biden's task becomes harder.

Looking at the polls across these states, however, the average swing on the current statewide polls is 3.5%, and if you weight these polls by electoral college vote, then the swing in marginals is 3.0%. This indicates that Trump is doing slightly better in the safe seats, but they're no good to him as they're unlikely to be unseated. Sorry, safe "states".

So on election night, here is your cut-out-and-keep guide to the battleground. Mark these off as they are declared, and you will have an idea of what is happening. Trump's easier states to win are at the bottom, and Biden's are at the top, with the white line representing the winning point on either side. As these come in, you should be able to tell who's winning.



Here's one more pic dump, Trump's target list:


Not paid for by either campaign. Just some independent research.