Precinct analysis: 2019 HD148 special election

I started this post while doing other precinct analysis stuff. Didn’t finish it with the others, but now that the legislative special election runoffs are next up on the calendar, I thought I’d finish it off. First, here’s how the main Mayoral candidates did in HD148:


Turner    9,631
Turner%  44.65%

Buzbee    6,280
Buzbee%  29.11%

King      2,947
King%    13.66%

Boykins   1,253
Boykins%  5.81%

Lovell      467
Lovell%   2.16%

Others      993
Others%   4.60%

Not actually all that different than how they did overall in Harris County. Mayor Turner was about 1.7 percentage points lower, while Sue Lovell gained 0.86 points. Oddly, it was the “Other” candidates who collectively gained the most, going from 3.72% overall to 4.60% in HD148, for a gain of 0.88 points. Keeping it weird, y’all.

Since I started this before the runoff, and even before the date for the HD148 runoff was set, I wondered what the effect might be of having Anna Eastman and Luis LaRotta slug it out at the same time as Mayor Turner and that other guy. I decided to zoom in on the best precincts for Eastman and LaRotta and see how the Mayorals did in them:


Eastman top 4

Eastman 1,557
LaRotta   557
Dem     1,508
GOP       547
Others  2,055

Turner  2,389
Buzbee    974
King      592
Others    370

LaRotta top 4

Eastman   242
LaRotta   600
Dem     1,006
GOP       515
Others  1,521

Turner    835
Buzbee  1,001
King      412
Others    245

Putting it another way, Anna Eastman’s best precincts were more Democratic, and more favorable to Turner, than LaRotta’s precincts were Republican and favorable to That Guy. Didn’t much matter in the end, but I was curious, and that’s what I learned.

Finally, there’s always the question of how much turnout efforts from one race can affect another. For sure, the Mayoral race was the big turnout driver in Houston in November, but as overall turnout was below thirty percent, there would still be plenty of people in HD148 who would normally vote in an even-year election, when this race is supposed to be on the ballot, but who may not vote in odd-year races. To try to get a handle on this, I looked at the undervote rate in the Mayor’s race in HD148, and compared it to the overall undervote rate for the Mayorals. In Harris County, 1.59% of the people who showed up to vote in November did not cast a ballot in the Mayor’s race. The undervote rate in the HD148 special was 5.87%, which is another way of saying it was the Mayor’s race that drove the majority of the action.

In the HD148 precincts, all of which are in the city of Houston, there were 22,001 total votes cast, according to the draft canvass sent to me by the County Clerk. That’s a smidge less than what you’ll see on the official election report, which is almost certainly a combination of cured provisional ballots (my canvass does not include provisional votes), split precincts (many voting precincts are partly in and partly not in the city of Houston, which makes all of the calculations I do that also involve non-city entities a little fuzzy), and whatever stupid errors I made with Excel. Be that as it may, of those 22,001 cast ballots, there were 387 non-votes in the Mayor’s race, for an undervote rate in the HD148 precincts of 1.76%, a hair higher than the overall undervote rate. If the voters in HD148 had skipped the Mayor’s race at the same rate as voters everywhere else in Harris County skipped it, there would have been only 350 Mayoral undervotes.

So, I’d say that the turnout effect of the HD148 special election was pretty small, since the voters in that race behaved very much like voters elsewhere. Perhaps if this had been a higher-profile race, with more money and a longer time on the ballot and a clearer partisan split – in other words, a race more like the HD28 special election – we might have seen more people who came out to vote for it and who had less interest in the other races, and thus a higher undervote rate in the Mayoral election. Sadly, we won’t know what that might look like at this time. I should note that I have no idea how many of the 1,288 non-voters in the HD148 special were also non-voters in the Mayoral race; there’s just no way to tell that from the data I have. Maybe some of those people were just there to vote for the Constitutional amendments, or the Metro referendum, or District H, or who knows what. I feel on reasonably firm ground saying that the turnout effect of the Mayor’s race was considerably higher than the turnout effect of the HD148 special election. Anything beyond that needs more study. You’re welcome.

Related Posts:

This entry was posted in Election 2019 and tagged , , , , , , , , , , , , , , , , . Bookmark the permalink.