Maybe we shouldn’t call it early voting anymore, with less than a week to go until the polls officially close. Middle voting? Noon voting? Maybe just plain ol’ voting?
Still, with early votes processed from this election already equalling 65% of Utah’s total 2016 turnout, it’s clear that we have a large part of the electorate already represented. And once again, we’re getting daily files from the state reflecting all the votes that have been processed — but not counted — so far. Votes will be counted and results released on Election Day.
These files contain about 68% of those voters, those who didn’t declare themselves private when registering. In all, we have about 514,000 votes of the 752,000 that the state has looked at so far. We don’t know how they voted, but we know their names, their party affiliations and their mailing addresses. That gives us a few tea leaves we’ll try to interpret.
Voting by party registration
There’s obviously a caveat when examining party registration in conservative Utah as a way to analyze the vote: many independents and even some Democrats register as Republican in order to take part in the party’s closed primary. Nearly 100,000 people did so before this year’s primary, for example.
So this doesn’t really reflect who’s likely to win these races (in Utah, Republicans win a big majority of contests). But what we can do is examine trends, such as are Republicans or Democrats voting more? Has their share of the electorate changed as the votes have come in?
It turns out that both Democrats and Republicans are turning out at higher percentages than the average voter. It makes some sense that unaffiliated voters would be less enthusiastic about voting early. They are likely less enthusiastic about politics.
Dems are certainly winning the turnout battle so far, but there is some evidence Republicans are turning the tide. In the first seven days of early vote data, Democrats made up about 23% of the total Utah vote. In the past week, that’s been trimmed to about 15%. This could be due to various counties finishing their vote processing at different rates, or just that Democrats are more focused on turning their vote in early, as polling has suggested.
Check out this comparison of Salt Lake County’s processed votes by precinct compared to its registration — use the slider in the middle of the picture to compare the two maps. You can see the blues are a little bluer in downtown Salt Lake City and Sugar House, and some areas of the south end of the county get a little less red. Overall, it’s not a huge shift, though.
Comparing turnout to previous elections
You know how on election night, the best TV analysts are breaking down data on a county-by-county basis? Usually, they’ll look at county data, and say something like, “Oh, there’s still a lot of votes yet to be counted in Cuyahoga County, so we expect a surge of Democratic votes to come in soon.”
So I was curious how our turnout so far compares on a county-by-county basis to what we’ve seen in the past two elections. Is one county outvoting the others? Here’s the data:
The numbers are relatively similar across the board, but Salt Lake County has had more turnout than average, while Utah County has had less turnout than average. On one hand, you might say that there’s more Utah County vote left to count, on the other, you might say Salt Lake County is banking votes early.
I was also curious to see how Utah’s early turnout compared to other states. Utah’s current total of 65% of their 2016 vote is pretty middle of the pack. Neighboring Colorado, Arizona, Nevada, and New Mexico all have turnout levels beyond Utah’s percentage. Texas is blowing their turnout numbers out of the water. Fully 94% of 2016 voters have come out there, and of course, there’s still days left.
U.S. Elections Project. (https://electproject.github.io/Early-Vote-2020G/index.html)
Nationally, voters have already turned in 57.3% of 2016′s total vote, so Utah’s a little ahead of national averages.
Voting by gender and age
Gender isn’t one of the categories listed in Utah’s scroll of people who have voted so far, but there’s a sneaky way of estimating gender turnout anyway: by using people’s first names. Generally, someone’s name is a pretty good indicator of what gender they are, and in the gender-neutral cases, you can estimate, say, 59% of all of the Caseys in the state are male.
The numbers here reflect an interesting characteristic of Utah voting in recent years: more women are voting than men. According to the Census Bureau, 60.5% of Utah voting-age women actually voted in 2018, whereas only 54.6% of voting-age men did. The 2016 election had only a half-percentage point difference between the genders, but in the 2014 election, the gender split was about two percentage points.
So far, we estimate that about 23,000 more women have voted than men, or about a 52/48 percentage split overall. About 13,000 of that split comes from more female Democratic voters than male ones, and there are about 13,000 more Republican women voters so far than male voters. (Male voters have about a 2,500 vote advantage in the third parties, however.)
While our files don’t give us access to voter age, some that the candidates themselves have access to do. A company called TargetSmart uses these files to estimate turnout among different age groups on a state-by-state and even county-by-county basis.
Here’s how Utah’s age turnout in 2020 so far compares to previous elections.
Age turnout in Utah. (https://targetearly.targetsmart.com/)
As you can see, so far the electorate has been younger than is typical, while those between 50 and 64 have been more reluctant to send their ballots in early in Utah.
Overall, it’s pretty consistent with what was expected: Democrats have a small early voting advantage thanks to their trust of mail-in voting, though the gap is less stark than some other places around the country because of Utah’s familiarity with the process. But nothing here looks likely to shake up the outcome of anything but the closest races.
I agree with something The New York Times' Nate Cohn wrote this week: most elections hinge on the voters preferences on the candidates, rather than big changes in the makeup of who votes. If change is coming in these races, it’s likely to largely come down to voters changing their mind.
And that, we won’t know until election night, when the results are released.
Andy Larsen is a data columnist. He is also one of The Salt Lake Tribune’s Utah Jazz beat writers. You can reach him at alarsen@sltrib.com.
