We now know how many early votes were cast in Harris County for the November 2023 election. I’m now going to run through a few scenarios to show what the range of possibility is for overall final turnout. None of this is intended to be a prediction, just what I think the ballpark is.
First things first, I suggested what I thought the turnout rate was for the city of Houston as of the end of early voting, but I didn’t say what the total number of voters that implied. Based on my calculations, which I remind you are based on several assumptions and thus should be considered to be for entertainment value only, a total of 150,775 Houston voters cast ballots in Harris County through Friday.
This is also where I remind you that there are Houston voters in Fort Bend and Montgomery Counties, much more so in the former, and as a general rule my calculations are based only on the Harris County voters who are in Houston. That’s mostly because I only have current EV data for Harris County, and partly because it’s just easier this way. Whatever the Harris County total ends up being, assume there will be another 5-7K total for Houston, nearly all of which will come from Fort Bend. In general, unless I specify otherwise, assume I mean “Houston voters in Harris County” when I refer to Houston voters.
I will also remind you that the number of mail ballots that get reported on Tuesday will be slightly higher than the number that was reported on Friday, because mail ballots are still arriving at the County Clerk’s office.
All right. In 2015, which is the election we will be basing comparisons on, there were 268,872 votes cast. That’s out of 979,401 registered voters, for 27.45% turnout. If all we get is an identical number of voters, that would mean turnout this year was 23.24%, quite a bit lower than in 2015. I think that is unlikely (though more on that in a minute). If instead, we have the identical level of turnout as a percentage of RVs, that will mean a final tally of 317,640, based on my estimate of 1,154,157 RVs. We’ll know when the results start getting posted what the actual number of RVs is, and we can adjust that accordingly, but it’s unlikely to be more than a couple of thousand one way or the other.
If in fact we end up with the 2015 total of 268,872 votes cast this year, that would mean that 115,107 people show up tomorrow. That would also mean that 56.08% of the vote was cast as of Friday, which would easily be a new high water mark for a city of Houston election. As noted before, past history suggests that somewhat less than half of the vote has been cast already. Going by 2015, 49.88% of the vote was early; if we get that this year, then the final total will be 302,275. In 2019, only 44.55% of the vote was cast early. At that rate, we’d see 338,440 voters this year.
I’ve been talking about the possibility of seeing the highest vote total for a city of Houston election. I will confess that as I’ve written and thought about this, I was primarily focusing on more recent elections, in particular since 2003. I finally got the bright idea to doublecheck those assumptions, by looking at the older election data from the City Secretary page. I stand corrected:
Year Turnout ============== 1991 316,278 1997 313,323 2001 288,283 2001r 321,030 2003 300,768 2009 177,940 2015 274,002
“2001r” refers to the 2001 runoff election between incumbent Mayor Lee Brown and then-Council Member Orlando Sanchez. If you’re looking at this and thinking “Huh, we sure used to have higher turnout in city elections”, then yeah, I agree, we did. Why it was higher then and why it took a downturn, I don’t know. We’re not in the 90s anymore. I don’t have voter registration numbers for those 90s elections so I can’t say what turnout as a percentage of RVs was for those years. It was certainly less than it is now – voter registration for the Harris County part of Houston was 1,006,301 in 2001, 955,205 in 2003, and 935,073 in 2009 – so even if we break the numeric record we’re still going to be short of those percentages. And yes, our VR numbers were declining in the city in the aughts even as population increased. Some of that may be attributable to changes in the city population – i.e., a greater share of non-citizens – but for sure some of it had to do with the policies of then-Tax Assessor Paul Bettencourt, who cared a lot more about removing voters from the rolls than he ever did about adding them.
Note that those totals above are for the whole city, including Fort Bend and Montgomery. I think we are very likely to surpass 2015 and November 2001, we are likely to surpass 2003, and we have a decent chance of surpassing the rest. It definitely helps that we have more registered voters now, as the rate of turnout doesn’t have to be as high. We’ll have answers to some of these questions tomorrow night.