Off the Kuff Rotating Header Image

Precinct analysis: Congressional districts


All right, let’s get this party started. In the past I’ve generally done the top races by themselves, but any race involving Trump provides challenges, because his level of support just varies in comparison to other Republicans depending on where you look. So this year it felt right to include the other statewide non-judicial results in my Presidential analyses, and the only way to do that without completely overwhelming you with a wall of numbers was to break it out by district types. That seemed to also pair well with a closer look at the competitive districts of interest, of which there were more than usual this year. So let’s begin with a look at the Congressional districts in Harris County. Only CDs 02, 07, 18, and 29 are fully in Harris County – we won’t have the complete data on all Congressional districts until later – so just keep that in mind.

Dist    Trump    Biden    Lib    Grn  Trump%  Biden%   Lib%   Grn%
CD02  174,980  170,428  4,067    969  49.93%  48.63%  1.16%  0.28%
CD07  143,176  170,060  3,416    903  45.09%  53.55%  1.08%  0.28%
CD08   25,484   16,629    520     87  59.65%  38.93%  1.22%  0.20%
CD09   39,372  125,237  1,066    589  23.68%  75.32%  0.64%  0.35%
CD10  101,390   65,714  2,023    431  59.80%  38.76%  1.19%  0.25%
CD18   57,669  189,823  2,382    962  22.99%  75.68%  0.95%  0.38%
CD22   21,912   21,720    522    137  49.47%  49.04%  1.18%  0.31%
CD29   52,937  106,229  1,265    649  32.86%  65.95%  0.79%  0.40%
CD36   83,710   52,350  1,558    402  60.65%  37.93%  1.13%  0.29%

Dist   Cornyn    Hegar    Lib    Grn Cornyn%  Hegar%   Lib%   Grn%
CD02  180,504  157,923  6,215  2,164  52.37%  45.82%  1.80%  0.63%
CD07  152,741  154,670  4,939  2,161  48.90%  49.52%  1.58%  0.69%
CD08   25,916   15,259    846    221  61.67%  36.31%  2.01%  0.53%
CD09   39,404  118,424  2,725  1,677  24.54%  73.76%  1.70%  1.04%
CD10  102,919   60,687  3,168    939  61.71%  36.39%  1.90%  0.56%
CD18   60,111  178,680  4,806  2,468  24.68%  73.35%  1.97%  1.01%
CD22   21,975   20,283    898    377  50.92%  47.00%  2.08%  0.87%
CD29   51,044   99,415  3,022  1,969  33.26%  64.77%  1.97%  1.28%
CD36   83,614   48,814  2,598    913  61.92%  36.15%  1.92%  0.68%

Dist   Wright    Casta    Lib    Grn Wright%  Casta%   Lib%   Grn%
CD02  176,484  153,628  7,631  4,122  51.62%  44.94%  2.23%  1.21%
CD07  149,114  149,853  6,276  3,974  48.22%  48.46%  2.03%  1.29%
CD08   25,558   14,796    992    394  61.23%  35.45%  2.38%  0.94%
CD09   37,090  117,982  2,764  2,570  23.12%  73.55%  1.72%  1.60%
CD10  101,414   58,873  3,758  1,793  61.15%  35.50%  2.27%  1.08%
CD18   57,783  177,020  5,021  3,846  23.71%  72.65%  2.06%  1.58%
CD22   21,026   20,231  1,007    675  48.97%  47.12%  2.35%  1.57%
CD29   46,954  102,354  2,802  2,334  30.40%  66.27%  1.81%  1.51%
CD36   81,424   48,619  2,880  1,300  60.66%  36.22%  2.15%  0.97%

Dist      GOP      Dem    Lib    Grn    GOP%    Dem%   Lib%   Grn%
CD02  192,828  148,374  5,524         55.61%  42.79%  1.59%
CD07  149,054  159,529  5,542         47.75%  50.79%  1.76%
CD08   25,906   15,212    926         61.62%  36.18%  2.20%
CD09   35,634  121,576  4,799         22.00%  75.04%  2.96%
CD10  103,180   60,388  3,496         61.76%  36.15%  2.09%
CD18   58,033  180,952  4,514  3,396  23.51%  73.29%  1.83%  1.38%
CD22   20,953   19,743  2,291         48.74%  45.93%  5.33%
CD29   42,840  111,305  2,328         27.38%  71.13%  1.49%
CD36   84,721   46,545  2,579    985  62.84%  34.52%  1.91%  0.73%

The first three tables are the Presidential, Senate, and Railroad Commissioner results, in that order. Subsequent presentations with State Rep and JP/Constable precincts will be done in the same fashion. For this post, I have also included the actual Congressional results – each Congressional race had both a Dem and a Republican, which doesn’t always happen, so they provide a good point of comparison. The candidate labeled as “Green” in CD18 was actually an independent – only CD36 had an actual Green Party candidate. In the other Congressional races, there were only three candidates.

How competitive CD02 looks depends very much on how you’re looking at it. On the one hand, Joe Biden came within 1.3 points, with Trump failing to reach fifty percent. On the other hand, Dan Crenshaw won by almost thirteen points, easily exceeding his marks from 2018 while clearly getting some crossover support. In between was everything else – MJ Hegar and Chrysta Castaneda trailed by about six and a half points each, with third-party candidates taking an increasing share of the vote. As we’ll see, most of the time the spread was between seven and nine points. That doesn’t tell us too much about what CD02 will look like going forward, but it does tell us that it doesn’t have a large reserve of Republican votes in it that can be used to bolster other Republicans. One possible outcome is that the map-drawers decide that Crenshaw will punch above his weight – he certainly fundraises at a very high level – which will allow them to leave him in a seemingly-narrow district while tending to more urgent matters elsewhere. The downside there is that if and when Crenshaw decides he’s made for bigger things, this district would be that much harder to hold with a different Republican running in it.

Another possibility is that Republicans will decide that they’re better off turning CD07 into a more Dem-friendly district, and using the space Republican capacity from CD07 to bolster CDs 02 and maybe 10. Lizzie Fletcher didn’t win by much, though I will note that Wesley Hunt’s 47.75% is a mere 0.28 points better than John Culberson in 2018. (There was no Libertarian candidate in 2018; do we think that hurt Hunt or Fletcher more in this context?) But other than Biden, no Dem came close to matching Fletcher’s performance – Hegar and Castaneda were among the top finishers in CD07, as we will see going forward. Like Crenshaw, Fletcher got some crossovers as well. It’s a big question how the Republicans will approach CD07 in the redistricting process. In years past, before the big blue shift in the western parts of Harris County, my assumption had been that the weight of CD07 would continue to move west, probably poking into Fort Bend and Waller counties. I’m less sure of that now – hell, I have no idea what they will do. I have suggested that they make CD07 more Democratic, which would enable them to shore up CD02, CD10, maybe CD22. They could try to add enough Republicans to tilt CD07 red, and at least make Fletcher work that much harder if not endanger her. Or who knows, they could throw everything out and do a radical redesign, in which case who knows what happens to CD07. Harris is going to get a certain number of full and partial Congressional districts in it no matter what, and there are Republican incumbents who will want to keep various areas for themselves, and the Voting Rights Act is still in effect, so there are some constraints. But there’s nothing to say that CD07 will exist in some form as we now know it. Expect the unexpected, is what I’m saying.

None of the other districts had as large a variance in the Trump vote. He trailed Cornyn and Wright in total votes in every district except CDs 29 and 36 (he also led Wright in 22). He trailed the Republican Congressional candidate in every district except 09, 18, and 29, the three strong D districts. Conversely, Joe Biden led every Democratic candidate in every district except for Sylvia Garcia in CD29; Garcia likely got about as many crossover votes as Lizzie Fletcher did. I’m amused to see Trump beat the designated sacrificial lamb candidate in CD18, partly because he was one of the co-plaintiffs on the state lawsuit to throw out all of the drive-through votes, and partly because I saw far more yard signs for Wendell Champion in my mostly-white heavily Democratic neighborhood (*) than I did for Trump. Maybe this is what was meant by “shy Trump voters”.

One more point about redistricting. Mike McCaul won the Harris County portion of CD10 by 43K votes; he won it by 46K in 2012 and 47K in 2016. He won overall by 30K, after squeaking through in 2018 by 13K votes. He had won in 2012 by 64K votes, and in 2016 by 59K votes. Now, a big driver of that is the ginormous growth in the Travis County Dem vote – he went from a 14K deficit in Travis in 2012 to a 57K deficit in 2020. The point I’m making is that there’s not a well of spare Republican votes in CD10 that could be used to redden CD07, not without putting CD10 at risk. Again, the Republicans could throw the current map out and start over from scratch – there will be new districts to include, so to some extent that will happen anyway – it’s just that Harris County is going to be of limited, and decreasing, use to them. They have to work around Harris, not with it. It’s going to make for some interesting decisions on their part.

I’ll have a look at the State Rep districts next. Let me know what you think.

(*) The two main precincts for my neighborhood went for Biden over Trump by a combined 68-28.

Related Posts:


  1. blank says:

    I have suggested that they make CD07 more Democratic, which would enable them to shore up CD02, CD10, maybe CD22.

    That’s my bet. Or at least they will create a 4th Democratic voter sink in the area. How much the 4 sinks look like the current 4 Democratic districts is still an open question.

    The State House data will be interesting, because I still don’t know how well they can draw a safe 10th Republican State House seat.

  2. ken roberts says:

    I suppose it isn’t a legal barrier for Congress, but they could make a subtle change to redraw CD07 so that Fletcher no longer lived in the district. She lives in River Oaks, and about a mile away from CD02. If they were willing to chop off CD02’s lower appendage, they could even draw Fletcher’s home into CD18.

    That still doesn’t solve what to do with the district as a whole. I agree with you that Republicans may be best off punting on CD07 and use it to soak up Democratic votes. It would certainly save Republicans a lot of money because that district is expensive.

  3. […] Introduction Congressional districts […]

  4. Thanks for running these numbers. Much appreciated. I hope you don’t mind me linking to them (with attribution) on my page.