Is a wave coming? What North Carolina’s early-vote numbers actually show

Aqua-themed metaphors abound this time of year. Is it a wave? A tsunami? A king tide?!

The answer is that nobody really knows yet.

Fortunately, North Carolina has an extended early-voting period, giving us fresh data each morning to examine who’s already casting ballots. I’ve been tracking these numbers during the 5 a.m. hour, nearly every day of every election cycle, for almost 20 years. And if there’s one thing I’ve learned, it’s that early voting can tell us quite a bit about the makeup of the electorate, emerging trends, and the direction races may be heading. That’s especially true when you drill down into the more granular data.

We’re now nine days into early voting, with eight days remaining. That’s enough data to begin identifying some meaningful statewide trends, even from a 30,000-foot view.

And this year, I’m particularly interested in what the numbers tell us when compared with 2018.

***

The 2018 yardstick

The last time North Carolina went through a midterm election with major Republican legislative losses was 2018. Republicans lost their supermajorities in both chambers. Their Senate delegation dropped from 35 seats to 29, and their House delegation from 75 to 65.

That is why I have been watching the 2018 early-voting numbers since Day 1. We now have nine reporting days behind us, with eight days left in early voting, including today. There is enough data to see some meaningful trends, even if there is not enough to call an election.

Start with the size of the vote. At time of this writing (in the wee small hours of Saturday morning) North Carolina has accepted 1,453,517 ballots. That’s about 386,000 more than at this point in 2022 and 493,000 more than in 2018. That is a 36.1% increase over 2022 and a 51.3% increase over 2018. North Carolina is a growing state, so larger raw totals are not automatically proof that voters are more enthusiastic.

The party-registration numbers are where the 2018 comparison becomes more interesting. Democrats account for 41.5% of accepted ballots so far, compared with 43.1% at this stage in 2018. Republicans account for 24.0%, compared with 30.3% in 2018. Unaffiliated voters have climbed from 26.4% to 34.3%.


In other words, the Democratic share is now 1.7 percentage points below its 2018 level, but the Republican share is still 6.3 points below. On these measures alone, Republicans face a bigger registration-share shortfall than they did at the same stage eight years ago.

There is an important catch: unaffiliated voters now make up a much larger slice of the electorate. A ballot from an unaffiliated voter tells us little about which candidate that person supported. Comparing party registration across elections is useful, but it is not quite the same thing as comparing actual candidate votes.

The raw ballot gap is even harder to ignore

Friday’s cumulative count shows 602,460 accepted ballots from Democrats and 348,445 from Republicans. That is a Democratic lead of 254,015 ballots. At the same reporting point, the Democratic advantage was 123,040 in 2018 and 83,054 in 2022.

The current lead is even larger than the roughly 249,000-ballot Democratic advantage recorded by the final day of in-person early voting in 2018. Those are raw totals, and today’s electorate is bigger, but the comparison shows how large the gap has become.

Republicans have, however, made real gains this week. The Democratic registration-share advantage shrank from 26.1 points on Sunday to 17.5 points Friday. Republican share increased for five straight weekdays, while the Democratic share declined.

The daily raw-ballot numbers tell a similar—but not identical—story. Democrats added about 39,000 more ballots than Republicans Monday. By Thursday that daily margin was down to about 13,500. Friday it edged back up to 13,821. So the daily gap has shrunk dramatically, but the cumulative Democratic advantage is still growing.

That brings us to the question I keep asking: are Democrats bringing voters to the polls earlier this time, essentially using up turnout that would otherwise come later? Maybe. But a slowdown in daily additions does not prove that. We will have to compare the remaining early-vote days and Election Day participation before drawing that conclusion.

The age of the electorate 

The age numbers are worth a look, too. For the first time this year, voters 65 and older have fallen below half of accepted ballots, at 49.8%. That remains well above the 43.9% senior share in 2018. Meanwhile, voters ages 45–64 are 30.4% of this year’s electorate, compared with 37.2% eight years ago.

Here is the twist: combine everyone 45 and older, and this year looks remarkably similar to 2018—80.2% now versus 81.1% then. The difference is that far more of those older ballots are coming from people 65 and up. And despite the attention around younger voters, ages 18–30 account for just 8.4% of accepted ballots, albeit more than at the same point in 2018 or 2022.

Eight days remain

So, is this a wave? It seems to be headed towards a good year for Democratic candidates. The numbers tell us the Democratic raw-ballot lead is much larger than it was at this stage in 2018, while Republicans are recovering registration share faster this week than they did in either previous midterm. Both things can be true.

Now we reach another weekend, when voting schedules and the mix of participating counties may shift the numbers again. Republicans have eight early-voting days to keep changing that picture. Democrats have to show whether their enormous head start reflects more overall participation or simply ballots cast earlier than usual.

A look back at historical Election Day performance is useful as well. Republicans have traditionally performed well on Election Day, and I have a feeling they’ll be more focused on the first Tuesday in November than at any point in the past seven election cycles as they look to make up lost ground from early voting.

The data tell us where the electorate stands today. They do not yet tell us how North Carolinians will vote on Election Day, or exactly how unaffiliated voters are filling out their ballots. That is why I will keep watching the daily numbers rather than declaring a wave before it reaches shore.

Thanks for reading this morning, and every other morning. If you have questions or comments, please feel free to reach out. mrusher@theresultscompany.com.