A few observations from the final unofficial countywide data

This is still unofficial, and there will still be some overseas/military ballots to be counted as well as some provisional ballots to be cured, but the count of the votes cast by Election Day is over, and we have the current final totals, broken down by vote type for each race. So let’s have a stroll through the data and see what we come up with.

– While Republican voting strength increased on Election Day compared to mail and early in person voting, Democrats still won Election Day. As far as I can tell, every Democrat who was on the whole county’s ballot beat their Republican opponent on Election Day, except for one: Genesis Draper, the appointed and now elected Judge of County Criminal Court #12, who lost Election Day by about 6,000 votes. She still won her election by 78,000 votes, so no big deal. Te’iva Bell, now the elected Judge of the 339th Criminal District Court, won Election Day by fourteen (yes, 14) votes out of 183,492 ballots cast in that race. She won by just over 100K votes overall.

– Democrats did especially well in mail ballots – in the judicial races, the number was usually around 60% for the Democratic candidate. That staked them to an initial lead of 27-40K, with usually a bit more than 160K mail ballots being cast. It’s amazing to realize how much that has shifted from even the recent past – remember, Republicans generally won the mail ballots in 2018, though they lost them in 2016. I don’t know if they quietly walk back all the hysterical “MAIL BALLOT FRAUD” hyperbole and go back to using this tool as they had before, or if that’s it and they’re all about voting in person now.

– As far as I can tell, no one who was leading at 7 PM on November 3, when the early + mail ballot totals were posted, wound up losing when all the votes were in. No one got Ed Emmett’ed, in other words. Gina Calanni and Akilah Bacy led in mail ballots, but lost early in person votes by enough that they were trailing going into Election Day. Lizzie Fletcher, Ann Johnson, and Jon Rosenthal lost the Election Day vote, but had won both mail and early in person voting, and that lead was sufficient to see them through.

– As noted, a very small percentage of the vote was cast on Election Day – 12.28% of all ballots in Harris County were Election Day ballots. That varied by district, however:

Dist     Total   E-Day   E-Day%
CD18   251,623  33,109    13.2%
CD29   161,673  30,274    18.7%

SD04    89,122   8,385     9.4%
SD06   187,819  34,996    18.6%

HD133   91,137   8,650     9.5%
HD134  111,639   9,389     8.4%
HD137   33,344   5,035    15.1%
HD140   33,614   7,325    21.8%
HD143   39,153   6,693    17.1%
HD144   32,522   6,989    21.5%
HD145   44,514   7,774    17.5%

Definitely some later voting by Latinos. Note that Sarah Davis won Election Day with 66% of the vote. There just weren’t enough of those votes to make a difference – she netted less than 3K votes from that, not nearly enough to overcome the 10K vote lead Ann Johnson had.

– There’s a conversation to be had about turnout in base Democratic districts. Countywide, turnout was 67.84% of registered voters. Of the strong-D districts, only HD148 (68.58%) exceeded that. Every strong-R or swing district was above the countywide mark, while multiple strong-D districts – HDs 137, 140, 141, 143, 144, and 145 – were below 60%. HD140 had 51.36% turnout, with HD144 at 51.81%. Harris County is strong blue now because Democrats have done an outstanding job of expanding out into formerly deep red turf – this is how districts like HDs 132, 135, and 138 became competitive, with HD126 a bit farther behind. As we discussed in 2018, deepest red districts are noticeably less red now, and with so many votes in those locations, that has greatly shifted the partisan weight in Harris County. But it’s clear we are leaving votes on the table – this was true in 2018 as well, and it was one reason why I thought we could gain so much more ground this year, to make the state more competitive. The focus now, for 2022 and 2024 and beyond, needs to be getting more votes out of these base Democratic districts and precincts. For one thing, at the most basic level, these are our most loyal voters, and we need to pay them a lot more attention. At a practical level, we need more out of these neighborhoods and communities to really put the state in play. We’ve figured out a big part of the equation, but we’re still missing some key pieces. That needs to change.

(Yes, I know, we have just talked about how perhaps some low-propensity Latino voters are much more Republican than their higher-propensity counterparts. We do need a strategy that has some thought and nuance to it, to make sure we’re not committing a self-own. But to put this in crass marketing terms, your strongest customers are the ones who have already bought your product in the past. We need to do better with them, and we start by doing better by them.)

– I’ll have more data going forward, when I get the full canvass. But in the meantime, there was one other group of people who had a propensity for voting on Election Day – people who voted Libertarian. Get a load of this:

Race         E-Day%  Total%
President     1.89%   1.03%
Senate        3.33%   1.81%
CD02          3.18%   1.59%
CD07          3.57%   1.77%
CD09          5.82%   2.97%
CD22          8.23%   5.33%
RRC           3.62%   2.08%
SCOTX Chief   4.50%   2.35%

You can peruse the other races, but the pattern holds everywhere. Seems to be the case for Green candidates as well, there are just far fewer of them. Not sure what that means, but it’s a fun fact. By the way, the Libertarian candidate in CD22 got 3.87% overall. Not sure why he was so much more popular in Harris County.

Related Posts:

This entry was posted in Election 2020 and tagged , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , . Bookmark the permalink.

3 Responses to A few observations from the final unofficial countywide data

  1. E says:

    Is there really any evidence that increasing Latino turnout in Harris County, or any non-border-area urban/suburban counties, would NOT be a net benefit to Democrats? It seems to me that the narrative now developing about Latinos supporting Trump in states other than Florida is misleading. The results in counties like Zapata and Starr don’t necessarily say anything about Texas Hispanics in general.

    The increased turnout for Trump in those counties could be attributed to the fact that they are not near large urban areas, and Trump overperformed in rural/small-town areas. All this means is that Trump had an appeal to some low propensity Hispanic voters in rural areas and small (or mid-sized) cities along the border in South Texas.

    First, remember that these areas are OVERWHELMINGLY Hispanic; something like over 80% of the population of these counties is Hispanic. Trump’s racist rhetoric probably did not repel as many Hispanics in these areas, compared to other areas where Hispanics are in the minority, because there is simply not that much day-to-day racial division in these areas. Low-propensity voters are low-information voters. They don’t follow developments outside their immediate lives very much, so even though Trump has stoked the flames of racism on a national scale, they probably didn’t feel it in their daily experience.

    Second, most Hispanics in the border area are not immigrants and don’t have family ties to immigrants. These Hispanic voters are from families that have been in Texas for generations and even centuries. That means Trump’s attacks on immigrants are less likely to repel at least a subset of these voters, as compared to Hispanics in urban areas who are more likely to be, or have close familial or social ties to, immigrants.

    Taking these two points together, I think you end up with the following conclusion: low-propensity Hispanic voters in South Texas tend to behave more like rural/small-town voters in the rest of Texas, at least when Trump is on the ballot. (Note, though, that they are not Republicans to the same degree as rural Anglos in Texas–look at the drop-off in support from Trump vs. down ballot Republicans in counties like Starr.)

  2. Mainstream says:

    The high level of election day support for the Libertarian running in CD 22 is probably Republican voters concerned about the harsh TV attacks that Troy Nehls had been fired as a law enforcement officer, but unwilling to back a Democrat.

    E–I will leave it to political scientists to sort out, but the strong backing of GOP and Trump by Hispanics is the big news out of this election cycle. Look at how close CD 15 was! A GOP candidate I had not even heard of took 48% of the vote against an incumbent Democrat. Henry Cuellar’s opponent picked up a solid 40% and Tony Gonzalez held Will Hurd’s seat by a much wider margin than Hurd was able to eke out. Adrian Garcia, Mary Ann Perez, maybe others could be in for tougher fights in the future.

  3. Mainstream says:

    from Patrick Svitek:

    (D): 52.3% (56,540)
    (R): 47.7% (51,566)

    Biden: 60.5% (66,889)
    Trump: 38% (42,027)

    Davis finished 9.7% ahead of Trump, per unofficial results. Her margin (-4.6) is ~18 points ahead of Trump’s (-22.5)

Comments are closed.