Tuesday, September 15, 2020

Mail-in voting: what to expect as you watch

I'm writing this for people I know that might experience an election with widespread mail-in voting for the first time in a couple months. I've been living in Washington, where most of our voting is done by mail, for 10 years. Local officials and local news in your area will have the most relevant information specific to your situation, especially on how to vote yourself. But I'm worried that newsrooms that are used to covering normal elections won't prepare you for how a largely mail-in election will go. And if you aren't ready for it you could be succeptible to harmful propaganda. So here are some things to expect as you watch the election.

Results will roll in for weeks

We're used to watching election results come in over the course of a few hours. With mail-in voting they'll come in over the course of a few weeks. Some places count early-arriving mail ballots before election day, so by election night a large fraction of votes will already be counted. Other places won't start counting until the day. In either case many ballots that were mailed and postmarked on time will be working their way through the mail for days after election day, and the full count might not be ready for a few weeks. That's normal, it's what you should expect. The count might be slowed down by COVID-related workplace issues, both within USPS and the election system. It might be slowed down by legal challenges or even political wrangling over which ballots should be counted (remember 2000?). It doesn't mean your local election officials or workers are malicious or incompetent.

I mean, your local election officials might be incompetent or even malicious! We have too many election officials trying to suppress votes in this country! But vote counts trickling up through mid-November isn't evidence of a problem, it's just how this kind of election works.

Oh, yeah, I said, “political wrangling over which ballots are counted.” That's going to happen in lots of places, especially places that aren't used to widespread mail-in voting. It won't be over on election night. Be ready to fight for your ballot, and your neighbors' ballots, to be counted, for weeks.

Results may swing dramatically after election day

Alright, so you know all about voter turnout. Every person decides whether and how to vote for their own reasons but in aggregate there are fairly predictable patterns for the voting and turnout patterns for different groups of people. Groups of voters whose turnout varies most from election to election tend to correlate with groups that swing left, so low-turnout elections tend to swing right and high-turnout elections tend to swing left (at least for recent US elections).

Similarly, in aggregate, different groups of people tend to vote at different times. Mail-in voting takes place over a very large time window, so these tendencies are exaggerated. So it's pretty common here in Washington for one candidate to hold a substantial lead on election night, only for their opponent to win resoundingly when all the ballots are counted. This kind of movement is normal. It is not evidence of a conspiracy. Anyone that says so is either ignorant or a liar, and in either case isn't worth listening to.

In Washington late-voters tend to swing left, and left-wing candidates often see large post-election-day increases. If I had to guess, I'd expect that the same pattern would occur nationally. However, that expectation comes with some caveats:

  • Donald Trump and his supporters broke a lot of pollsters' turnout models in 2016 (as did Brexit). Washington, and the Seattle area in particular, has some of the lowest levels of Trump support in the US. It wouldn't shock me to see more late-arriving right-wing votes elsewhere because Trump supporters break the timing model, too.
  • Washington voters, and Seattle voters in particular, tend to be pretty tuned-in and ideological, and our top-2 “jungle primary” system forces voters to decide on their overall top choice in most elections months in advance of the general. So I think we have fewer true “swing voters”, by the time of a general election, than a lot of other places, even in close elections. And I suspect true swing voters would tend to vote late. So there's another late-voting group that might be bigger nationally than in Washington.
  • If you have widespread mail-in and widespread in-person voting all bets are off. I don't know which groups of people will tend to vote in-person, nor whether in-person or mail-in ballots will suffer bigger delays. In fact I'd bet both those things will vary a lot place to place. Washington is almost all mail-in so I don't even have a baseline to go from here.

Exit polling is out the window

Up here we're used to pollsters not bothering with us because we're not a swing state. At the local level we're used to being in the dark until the results come in. The whole thing with staying up and watching exit polls to try to get a jump on official results... it's not going to be a thing this year. Go to bed.

No comments: