4 trends in the early voting that could determine who wins Tuesday’s primaries

Some things should not be lumpy.

Mattresses and mashed potatoes, for instance.

Some things often turn out lumpy, anyway.

Mattresses and mashed potatoes — and elections.

It looks as if we’ll have some lumpy results tonight in primary elections across the state.

By “lumpy,” I mean some places will vote at higher or lower rates than others. That’s the case in every election. Case in point: In the 2021 governor’s race, 71% of registered voters in Goochland County cast ballots, while only 38% of those in Petersburg did. Every other locality in the state fell somewhere in between. 

Many of those disparities, though, are quite expected. Petersburg’s 38% turnout was low, but it was also exactly the same as four years prior. Higher-income localities always have higher turnout than lower-income localities, which means suburban localities have higher turnout than rural areas or central cities. Those differences are just factored into what’s “normal.” 

Sometimes, though, we get some changes that aren’t so normal. Four years ago, we saw turnout spike in Republican-voting areas but generally stay flat in Democratic-voting areas. That meant those 2021 results were a little lumpier in a Republican direction, which helps explain the Republican sweep that year.  

With that explanation out of the way, let’s look at how the primary results might be lumpy — because the early voting trends were. That’s more significant for Democrats than Republicans because Democrats are far more likely to vote early than Republicans. Based on recent elections, I’d expect about half the Democratic vote in this year’s primary has already been cast but only about one-quarter of the Republican vote has. That said, early voting is still a relatively new option in Virginia, so we don’t have that many data points to go by. 

Thanks to the data folks at the Virginia Public Access Project, who have poured all the early voting data into some easy-to-understand charts, we have a good picture of just how lumpy this primary may be.

Here are some of the things we see so far that might — a very important conditional word — make a difference when we start to see some actual results tonight.

1. The early vote in the Democratic primary is weighted toward the Richmond metro, especially Henrico County.

One way to look at the early vote is through the lens of House of Delegate districts since they’re all roughly the same size, although some are obviously more Democratic or more Republican than others. Still, we have this potentially noteworthy development: Three of the six heaviest-voting districts are in the Richmond area, and two of those are mostly or entirely in Henrico County; the other is entirely in the city of Richmond. That’s potentially significant because one of the two Democratic candidates for attorney general is Henrico County Commonwealth’s Attorney Shannon Taylor; also two of the six Democratic candidates for lieutenant governor are from the Richmond area: Levar Stoney, the city’s former mayor, and state Sen. Ghazala Hashmi, whose district takes in part of Chesterfield County and part of Richmond but who would also be well-known to Henrico voters from local news coverage. The shape of the turnout would seem to be beneficial to all of them, assuming, of course, they have support from hometown Democrats.

There’s a famous quote attributed to Woody Allen that “80% of success is showing up.” (Sometimes the percentage gets bumped up but the original quote appears to be 80.) This is certainly true in low-turnout primary elections. The winner here will probably be whoever does the best job getting their supporters to turn out, and these figures suggest that Taylor is doing a good job of ensuring that her home county comes in strong for her.

Before we jump to too many conclusions, let’s remember this: In the Democratic primaries four years ago, Henrico County accounted for 5.5% of the total vote. So far this year, it’s accounting for 6.2%. That may not seem like much of a difference, but close elections are often won around the margins. Still, there’s a whole lot of Virginia that is not Henrico County, so let’s keep looking. 

2. The Northern Virginia share of the vote in the early voting is down.

The high-voting House districts in Henrico are strongly Democratic ones, anyway, but by no means are they the brightest blue districts in the state. House District 81 (the third-highest voting House district in the primary) voted 66.5% Democratic in last year’s election, House District 80 (the fifth-highest) voted 74.8%.

Let’s see how the early vote there compares to other House districts that are equally Democratic.

House District 2 in Arlington County is even more blue; it voted 76.12% Democratic last year.

House District 3, which covers parts of Arlington and Alexandria, is bluer still; it voted 76.90% Democratic last year.

Both of those districts are more Democratic than those two Henrico districts that are coming in third and fifth statewide. So how many early votes did they produce? Let’s compare.

If they all voted at the same rate, those two Northern Virginia districts should be producing more Democratic votes than the two Henrico districts. They’re not. House District 3 in Arlington/Alexandria is producing less than two-thirds of the vote that House District 81 in Henrico is.

What’s going on?

What we’re seeing here is a well-known phenomenon in Virginia politics: While Northern Virginia is the state’s population center, it’s one so focused on the nation’s capital in Washington that it feels disconnected from its own state capital in Richmond. That makes it hard for Democrats to mobilize that vote in a summer primary, in particular. The upshot here is that, overall, Northern Virginia is likely to produce a smaller share of the Democratic primary vote than it could. There are some exceptions: House District 1 n Arlington, which has both a three-way House primary and some local primaries, is now the heaviest voting in the state. House District 5, in Alexandria, has shot up to fifth heaviest.

Who does that help or hurt? That’s harder to say. In the attorney general’s race, there’s no candidate from Northern Virginia, so there’s no home field advantage. In the lieutenant governor’s race, there are three candidates from Northern Virginia: the lightly funded Alex Bastani and Victor Salgado and the better-funded Babur Lateef. The latter has focused much of his campaign in Northern Virginia on his opposition to a proposed casino at Tysons Corner. If these or any other Democratic candidates had counted on a big vote out of Northern Virginia, that candidate might be disappointed.

Here’s how much the Democratic vote in Northern Virginia is down so far. In the primary four years ago, 17% of the Democratic vote was from Fairfax County. So far this year, it’s 14.8%. That’s not a total collapse by any means, but some elections are won around the margins. Maybe Fairfax County Democrats will make that up with a stronger turnout at the polls today, but so far they are muting their very strong voice in one of the most important party decisions: who to nominate. 

3. Southwest Virginia will have almost no say in who Democrats nominate.

Last week, I wrote that only three of the 12 statewide candidates in both parties had ever been to Virginia’s westernmost county (and one of those, Jones, had been to Lee County long before this campaign). Tut-tut. The other two were both Republicans who have run statewide before: Republican gubernatorial nominee Lt. Gov. Winsome Earle-Sears and Attorney General Jason Miyares, who is seeking reelection. 

So why do so many candidates, especially Democrats, seem to ignore Southwest Virginia? Here’s the math: Four years ago, Lee County accounted for 0.3% of the statewide Democratic primary vote. So far this year, it’s 0.02%. 

If Lee County voters don’t like Democrats, that’s their choice, but the consequence of being so all-in for the other party is that Democrats have no practical reason to pay attention. I can make the argument that all candidates should be familiar with Southwest Virginia because they’ll have to deal with policy issues there whether they get votes from the region or not — in fact, I did just that in another column last week. However, for our purposes today, this means Southwest Virginia Democrats — what few of them there are — will have very little say in who their nominees are.

Still, there will be some Democratic votes cast in Southwest Virginia. Who will they be for? None of the candidates have a natural connection to the region, unless you count lieutenant governor hopeful Aaron Rouse, who played football at Virginia Tech. I’m surprised he hasn’t done more to play that up and try to run up the score among a small number of voters who might yet prove important in a razor-close race.

4. The lumpiest early voting is in a Republican House primary in Southwest Virginia. 

If you subscribe to our weekly political newsletter, West of the Capital, you know that I’ve been calling attention to the curious early voting trends in House District 46 for about a month now. If you’re not a subscriber, you can fix that right now — it’s free!

The lay of the land: This is a deep red House of Delegates district where the Republican nomination is tantamount to election. Del. Jed Arnold, R-Smyth County, is retiring. Adam Tolbert, a community college administrator from Smyth County and the chair of the 9th District Republican Party, was the first candidate to enter, and he entered with the endorsement of almost every Republican leader — from Gov. Glenn Youngkin on down.

Then Mitchell Cornett — a farmer and Grayson County supervisor — jumped in. Tolbert has more money, but Cornett has run a spirited campaign, devoting much of it to criticizing Appalachian Power. 

The early voting trends appear to be in Cornett’s favor: 18% of the voters in that district are from Grayson County but so far 39.9% of the early voters are. That Grayson County vote is also likely pumped up because of a Republican primary for county supervisor in one magisterial district. 

Republicans still overwhelmingly prefer day-of voting, but it would seem that Tolbert needs a big day-of turnout from his home base in Smyth (and elsewhere) to smooth out what otherwise looks like a lumpy turnout that favors Cornett’s home county.

Those are some of things I’ll be watching tonight as the returns come in. The polls close at 7 p.m. I’ll be posting commentary through the evening, so check back and let’s see who wins.

If you’re still undecided and need some last-minute inspiration, you can see how the candidates answered the questionnaire on our Voter Guide.