Here’s why you should be very cautious about proclaiming a winner on Election Day

Well, yes — for a number of reasons, not all of which are related to the coronavirus pandemic.

Political junkies are used to being able to tune in to cable news on election night or to log on to their favorite news websites to track returns as they come in. This often leads to frustrating misinterpretations of what’s happening, as people seize on how candidates are faring with, say, 20 percent of precincts reporting. Those results could change! Which, of course, is half the fun for those watching, tracking how each candidate is faring over time, as though what’s being measured is an actual shift in voter sentiment. But it isn’t. It’s like describing the final score of the football game by telling people how each team scored in random order. The result is set; tracking returns by precinct just adds artificial drama.

This expectation that the evening will build to a result is normally safe because most races aren’t that close. But it again sets a false expectation — one that politicians can abuse.

When then-Gov. Rick Scott (R-Fla.) ran for the Senate in 2018, he seemed to be winning once the Election Day vote was totaled. There were a lot of absentee votes to count, though, many in heavily Democratic counties. Scott and President Trump tried to claim that those votes were riddled with fraud to boost the sense that the results on election night were the “real” results. At one point, Trump even insisted that those results should for some reason stand. There wasn’t any fraud (as a review by the state later determined), and Scott won anyway.

This year, we can expect a similar delay in counting in many or most states. The pandemic means that far more voters plan to vote by mail, meaning that far more votes may need to be counted after Election Day than in years past. Which, by extension, means that the results may be uncertain for longer.

Again, this doesn’t mean that an accurate result is being modified any more than the 2018 Florida Senate results were somehow more accurate on election night. It means that the results on election night should be viewed in many cases as though only 20 percent of precincts are reporting.

It’s a bit more complicated than that, of course. For one thing, Democrats are more likely to vote by mail this year, meaning that the vote which comes in after Election Day will probably be more heavily Democratic — and that the votes tallied on the day of the vote will probably be more heavily Republican.

Axios earlier this week reported on what it called the “red mirage,” an Election Day which could end with an apparent lead for Trump that wouldn’t hold when all the votes were counted. The story was unfortunately framed in a way that suggested Trump would have a “landslide” on that day which would then be taken away from him. Which, again, is like saying that a Republican was elected governor of New York in a landslide after all the non-New York City vote was counted — only to see that massive victory evaporate.

The story led to an inevitable tweet from the president.

It will be a rigged election, in the sense that “counting valid ballots” is “rigging” things. Trump, again, wants to prompt suspicion among voters of any votes counted after election night, for obviously self-serving reasons. As in Florida two years ago, we should not assume his complaints are offered in good faith.

So how could this play out? There are a lot of factors here. One is that absentee ballots are more likely to be rejected as counties run checks to protect against the sort of fraud Trump claims is rampant and which isn’t. Another is that split in how likely Democrats and Republicans are to vote by mail. In a new Selzer and Co. poll conducted by Grinnell College, 31 percent of Democrats said they planned to vote by mail, compared with 11 percent of Republicans. Many of those votes will arrive before Election Day, but varying rules around when they can arrive and still be counted mean that some could come in after Election Day has passed.

In an effort to show how the results could change over time, we created this interactive, allowing you to adjust the frequency of mail balloting, the extent to which mail ballots are rejected, the duration of the vote and, to gauge the actual result, how the margin this year will differ from 2016. (The shift in the vote is compared with each state’s 2016 margin, and the vote totals use totals from those states four years ago.)

Give it a shot.

Using the default settings, you can see how the apparent results shift over time. Again, the actual result is the one shown after all of the votes are in. The result on election night is only a fraction of the actual votes cast, a percentage which we’ve included in the interactive. You can think of that as the “percent of precincts reporting” for the election, if you’d like to continue the analogy.

Again: It’s no secret why Trump wants people to focus on the results as they stand on Election Day. You can tweak the controls and come up with a scenario where the results don’t change or where things improve for Trump over time, but that’s probably not how things will play out.

Instead, we’ll go to bed on Nov. 3 with some understandable and natural uncertainty about the outcome. Be patient. If you need something to do while you wait, watch “Floor is Lava.”