I feel I need to say something about this, which ran just before the CD18 runoff.
Early-voting turnout in the runoff election for Texas’ 18th Congressional District is under 4% so far, highlighting the challenges in voter participation as the district prepares to elect a representative after more than a year of instability.
According to unofficial figures from the Harris County Clerk’s Office, 13,675 voters cast ballots during the early voting period, including in-person and mail ballots. The district is home to more than 700,000 residents, according to the 2020 census, with about 381,000 registered voters for the 2025 special election. Votes from Wednesday and Thursday hadn’t been counted yet.
Renee Cross, senior executive director and a researcher at the Hobby School of Public Affairs at the University of Houston, said the low turnout reflects a joining of structural and political factors rather than voter apathy.
“I knew that this would be a low-turnout election,” Cross said, “but I was surprised at how low it has been.”
Cross pointed to the unusual nature of the election itself: a special runoff held in late January on a Saturday for a partial term with another election for a full term scheduled just weeks later in March.
“This one is only going to be for the rest of the year, so I don’t know if that calculation is involved in whether people are deciding whether to vote or not,” Cross said.
University of Houston political scientist Brandon Rottinghaus echoed that assessment, saying the depressed turnout reflects structural challenges unique to the race.
“This is an off-cycle election, and a runoff to boot,” Rottinghaus said. “It hasn’t been subject to aggressive campaigning in the way a highly competitive race would be, and there’s also confusion about what exactly this election is.”
Redistricting – a lot of people who could have voted in this election will not be voting in CD18 going forward, myself included – and the political similarities between the candidates were also cited as factors. All of which is sensible, and I appreciate that no one wagged a finger at the voters. But as is my wont, when I see a quantitative assessment made of something, my first question is always “compared to what?” And with that in mind, I wanted to see what I could come up with.
There haven’t been many special elections to fill vacant Congressional seats in recent years. Since 2006, which is a galaxy or two away in terms of what political universe we now inhabit, the CD18 election is the fourth to occur. There was the CD34 election in 2022, the CD06 election in 2021, and the CD27 election in 2018. Only the CD06 race included a runoff, which is the real point of interest for us. Turnout for CD34 in 2022, which was in June for a term that would expire six months later and would be for a seat that was going to be completely different in November thanks to decennial redistricting, was 7.36%, with 29,069 total votes cast. There are no turnout figures for CD27 in 2018, but there were 36,291 and that election was also in June for a six-month term; at least in that race, the winner was able to continue serving in the same district afterwards.
So the real comparison is with CD06 in 2021, which occurred in April following the death of then-Rep. Ron Wright, who passed on February 8. And yes, the election to succeed him took place three months later, in May, because Greg Abbott actually cared about filling a Republican vacancy. The subsequent runoff was in August. Here’s how the numbers stack up:
Election Votes Turnout
=============================
CD06 special 78,471 16.05%
CD18 special 76,189 18.07%
CD06 runoff 39,166 7.94%
CD18 runoff 27,080 6.41%
I wanted to wait until the provisional ballots were included in the live results before reporting on this. Turns out I was on the nose in guessing there would be about 3500 of those, which brought the early-plus-mail total to about 17K, and I way overestimated how many people would show up on Runoff Day; there were about 10K, when I thought the Saturday turnout would be about the same as the early turnout. Womp womp.
Obviously, 6.41% turnout isn’t great, especially compared to the 15.41% turnout in the SD09 runoff, where there was even worse weather to deal with. What might be the reason for the difference? I don’t think there’s a definitive answer. As Renee Cross and Brandon Rottinghaus noted, the candidates in CD18 were very similar – I’m sure Amanda Edwards’ supporters are disappointed, but I can’t imagine too many of them are mad that Christian Menefee will be their member of Congress instead of her – and as such the stakes were a lot lower than in SD09, even though Taylor Rehmet will likely only ever be in session if he wins again in November. I may, in my copious spare time, look at a bunch of special election runoffs to see where 6.41% and 15.41% fit in the spectrum, both as standalone figures and as a share of the main election turnout. That might at least give some context.
And that’s really my point here. A number by itself can only mean so much. Where it stands in comparison to other numbers of its kind, that might tell me something. All I ask is that we keep that in mind.