Tuesday, November 03, 2020

U.S. election watching guide

1. Florida

This is painful for me to write as I have rolled my eyes as pundits said "it all comes down to Florida" in every election cycle since Florida was "decisive" in 2000. (And, in reality, Florida wasn't that decisive; if Gore had won Tennessee or New Hampshire, just as close and easier lifts, he would have won without Florida).

But this year, it really does come down to Florida; mostly.  Particularly if you are looking for an early resolution to the election.

According to Nate Silver's 538 projection model, Trump has a less than 1% chance of winning the election overall if he loses in Florida.

Biden could still win if he loses in Florida, but if he loses in Florida, it means it is a really tight race that won't likely be resolved for several days.

Florida is also one of the few critical states that will complete most of its counting of all ballots on election night, meaning it is likely to be called unless the state is very close.

So watch for the call in Florida.  If Biden wins, he has almost certainly won the election.  Otherwise, we will have to look elsewhere for signs.

2. South Carolina

South Carolina's polls close early (7 p.m. Eastern) and count all of their ballots on election day.  Biden is highly unlikely to win here, but if he does, he has almost certainly won the election.  If it is really close here, he has also almost certainly won.  Similarly if Trump does about as well here or only slightly worse than he did here in 2016 (he won by 14 percentage points), then that bodes well for Trump to hold the neighbouring states of North Carolina and Georgia.

So here is your Carolina voting guide:

  • Biden win to 5 point loss: Biden has likely carried North Carolina and Georgia and won the election;
  • Trump win of 5 to 10 points: Biden is significantly over-performing Clinton in the south and likely wins unless Trump majorly over-performs elsewhere (holding his mid-west victories from 2016 and picking up something like Minnesota or Nevada);
  • Trump win of 10+ points: not a lot to read-in here, move on.

3. Trump?

Other states that count most of their ballots on election day that are worth watching include New Hampshire (8 p.m. Eastern), Missouri (8 p.m. Eastern), and Nebraska (9 p.m. Eastern).

If Trump wins in New Hampshire, a state he has targeted but lost in 2016, that is a very good sign for him.  Though a small state, 538 says Trump's odds jump to 40% from 10% if he wins here.

Not a lot has been written about Missouri which is not expected to be a battleground.  However, traditionally it was and when Obama lost here in 2008 (by only 4,000 votes!) it was only the second time in over a century that Missouri had voted against the overall winner.  Democrats remain competitive here in other statewide contests and the state has been identified on lists of Biden stretch targets were he to be headed for a blowout.  On the other hand, if Trump can hold his margin here (he won by 19 percentage points in 2016!), that is a very good sign for him and perhaps also a good sign for him in neighbouring Iowa--a must win state for Trump.

In Nebraska, it all comes down to the second congressional district.  Nebraska, like Maine, awards electoral votes by district.  Biden is heavily favoured in this district which voted for Obama in 2008.  So much so that if Trump wins it, even though it is worth only 1 electoral vote, it boosts Trump's national chances to 33% in the 538 simulator.  If Trump wins the 5 electoral votes of NH and NE-02, 538 shifts him to the favourite with a 58% chance of winning the national electoral vote!

4. Other States

No other even quasi-competitive state will complete counting their ballots on election night and the networks are likely to be very hesitant to call these states unless there is a blow out.  For example, Ohio and Iowa: they might be called if someone is ahead by 5+ points in the count of what has been counted.  If Trump or Biden were to be declared the winner in these states on election night, they would probably have won the election.  I doubt we will see that happen however.

5. Pennsylvania

The real bad news comes if it all comes down to Pennsylvania.

Imagine a plausible map like this where Biden and Trump each have 259 electoral votes and Pennsylvania is too close to call.

The election result is unlikely to be finalized until late Friday at the earliest, and perhaps not until next week.  And that is before lawsuits begin for recounts or disqualification of certain ballots.  So if it all comes down to Pennsylvania, we will be prognosticating on whether or not 2020 beats the record of 2000 when it took until Dec. 13 to resolve the election.  Not to mention the other unfortunate things that might occur in this circumstance.