As a teacher of US Politics, people often* ask me, “Sam, Biden’s way ahead in the polls, but no-one's prepared to call it for him. So can Trump really win?”

*never
Yes, he can. According to @fivethirtyeight, the go-to place for polling averages, Trump has a c12% chance. That’s around the same likelihood as his face coming up on an eight-sided die. (I’ll be testing that every day from tomorrow – eight days out, eight rolls of the die.)
But there are many more reasons to think that Biden will win. These include:
1. Biden’s lead has been large, and stable, over a long period of time.
2. Historically, races do not change much in the last two weeks.
3. In 2016, Trump was the least popular candidate on record. Unfortunately for Democrats, Hillary was the second least popular. Biden may not be everyone’s choice, but he’s way off that level of antipathy.
4. Last time, Trump was the great usurper, the “let’s throw it all up in the air” candidate. Not any more. He’s the president.
5. Coronavirus.
6. There are far fewer undecided voters this time, so there's even less scope than usual to move the needle at this late stage.
7. In 2016 Trump was tipped over the line by a total of under 100,000 votes across in three key states: Pennsylvania, Wisconsin and Michigan. The Democrats do not need much of a swing in those states – certainly far less than the polls are suggesting they hold at the moment.
8. In those states, the polls in favour of Biden would need to be really quite wrong for Trump to win. He’s up on average 5.5% in Pennsylvania, 6.3% in Wisconsin and 7.9% in Michigan.
It ain’t over til it’s over. There's still time for a final round haymaker, and I haven’t even got into supposed electoral fraud, legal challenges and flat-out refusal to leave the White House. But to paraphrase Ian Dury: reasons to be cheerful, one to eight.
You can follow @MrSamPullan.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: