With just 24 days to the presidential election, Joe Biden has opened an almost ten-point lead in national polls. That’s a wider lead than Hillary Clinton had at the same point in 2016. Not that it matters. Our eighteenth-century process makes election for president not one unified referendum, but 51 separate state contests (including the District of Columbia), and what counts is who receives at least 270 votes from the Electoral College on December 14. To reach 270, Biden needs to hold all the states Clinton won and reclaim three that Trump won by a collective 77,000 votes: Pennsylvania, Michigan, and Wisconsin. How’s that looking?

The California 2020 mail-in ballot.

Good news on job one: Biden is likely to win all the states Clinton won in 2016. According to FiveThirtyEight, the weakest link is New Hampshire, which as of this morning is 83% likely to vote for Biden. FiveThirtyEight puts Trump’s probability of winning any state that went Democratic in 2016 at just 29%.

Good news on job two also: Biden is ahead by much more than the polling margin of error in Pennsylvania (7.1%), Michigan (8.4%), and Wisconsin (7.1%). FiveThirtyEight gives Biden a 91% likelihood of winning Michigan, an 86% likelihood of winning Pennsylvania, and an 84% likelihood of winning Wisconsin.

FiveThirtyEight gives Biden an 85% likelihood of winning the Electoral College.

But what if the polls are getting it completely wrong somewhere, say Nevada or Minnesota?

In the purple states, Trump’s damning tax returns, rancid debate performance, COVID-19 diagnosis, and dexametha-mania seem to be working against him. (The phenomenon has gained a name: Trump Fatigue.) Biden is 70% likely to take Florida, worth 29 electoral votes. Also leaning Biden’s way are Arizona (11 electoral votes) and North Carolina (15 electoral votes). And Ohio, with 18 electoral votes, has become a toss-up. For Biden, a win in any of these states would be insurance against a surprise loss elsewhere; Trump, on the other hand, needs to win them all.

So even after accounting for potentially significant polling errors, Biden looks like the winner — on Election Day.

After Election Day, maybe not. The specifics will vary, but most scenarios boil down to this: Trump will claim voting irregularities in states that swing to Biden via mail-in ballots tabulated after Election Night; those states’ legislatures, secretaries of state, or governors will appoint a rival, pro-Trump delegation to the Electoral College; and the courts (many of which have Republican majorities) will determine which delegations are official. If Trump’s challenges aren’t resolved by December 14, the disputes will go to the House of Representatives, where each state, no matter how many representatives in its delegation, gets just one vote — and most state delegations have a Republican majority.

If the last four years have shown us anything, it’s that Trump is willing to damage or destroy anything — or anyone — if he perceives it’s to his advantage. We have to assume that in the likely event he loses the election, he will resort to underhanded tactics to remain in power. If you want to be part of the non-violent effort to resist Trump’s post-election desecration of democracy, sign up to receive messages from Protect the Results. If you encounter any resistance to your voting, contact the ACLU’s election protection hotline at 1–866-OUR-VOTE.

And if you want to laugh a little about the political vortex we may be entering, try this:

Former Risk Manager at UC Berkeley, author of four books, ectomorphic introvert.