November 17, 2008
Precinct analysis: The Obama effect

Ready for some precinct data analysis? Let's get started. There will be a lot of these entries, because there's a lot of interesting data. Today let's take a look at how Barack Obama's performance in Harris County compared to the county Democratic judicial slate, since it was conventional wisdom that as Obama went, so went the judicial hopefuls.

The first question to consider is the coattail effect. I've written before that in 2004, Republican judicial candidates lost a lot more votes compared to George Bush than Democratic candidates did compared to John Kerry. My reason for this is that if you were voting for John Kerry in 2004, you were a real true Democrat, and you went on down the line. Bush, however, seemed likely to attract voters who were only interested in his race, and that was borne out by the numbers:

2004 avg GOP 536,000, Bush 584,000, dropoff 48,000
2004 avg Dem 470,000, Kerry 475,000, dropoff 5,000

Bush may have lifted the tide for the incumbent judges in 2004, but they didn't really need it. There were plenty of Republican voters for them to win. This year, not so much:

2008 avg GOP 540,170, McCain 569,860, dropoff 29,690
2008 avg Dem 558,324, Obama 587,538, dropoff 29,214

Both judicial slates saw an equivalent dropoff from the Presidential vote. This year there were more Democrats voting overall, and that was the difference maker for most. Not for all, since there was a 44,000+ vote gap between the top Democratic judicial candidate and the bottom one, but for 23 of 27.

That's at the countywide level. Going down to the State Rep district level, you begin to see some patterns. For the most part, you'd expect the average judicial candidate to lose around 1000 votes from the Presidential level in each HD. For many of them, what actually happened was in line with that expectation, adjusted for relative levels of turnout and partisan makeup. In fact, in 17 of the 25 State House districts, the judicial candidates received a percentage of the vote that was within 1.25 points of the Presidentials. One of the eight exceptions was HD142, which is an African-American district. In HD142, where Obama received 79.78% of the vote, Democrats voted down the line at a very high rate, whereas the Republican judicials lost over ten percent of the meager vote total McCain got, from 8401 votes for McCain to 7506 for the judges. The result there was an 81.35% share for the average Democratic judicial candidate.

The remaining seven districts that defied the pattern had some shared characteristics with each other. They are:

1. The Bubba districts

In HDs 128 and 144, which cover places in eastern Harris County like Baytown and Pasadena, Barack Obama did significantly worse than the rest of the Democratic ticket:


Dist McCain Obama M Pct O Pct GOP avg Dem avg GOP % Dem %
==================================================================
128 27,322 13,973 66.16 33.84 24,621 14,542 62.87 37.13
144 23,656 16,167 59.40 40.60 21,172 16,563 56.11 43.89

Here, the GOP suffered its biggest and third-biggest dropoffs (they lost 2,517 votes on average in HD127, but Dems lost 1000 votes as well, which was proportionate to the percentages), while the Dems gained votes in each case. These are working-class districts with a white voting majority, both friendly to the GOP. If Obama had any problems locally with that demographic here, these districts are where you can see the effect.


2. The Hispanic districts

There are four State Rep districts in Harris County that are represented by Hispanic politicians - HDs 140 (Rep.-elect Armando Walle), 143 (Rep. Ana Hernandez), 145 (Rep-elect Carol Alvarado), and 148 (Rep. Jessica Farrar). In all four, Barack Obama received over 60% of the vote. But in three of them, the downballot Democrats did significantly better:


Dist McCain Obama M Pct O Pct GOP avg Dem avg GOP % Dem %
==================================================================
140 6,396 11,450 35.84 64.16 5,111 11,943 29.97 70.03
143 6,729 10,299 39.52 60.48 5,270 10,982 32.43 67.57
145 7,574 12,508 37.72 62.28 5,999 13,082 31.44 68.56
148 13,421 20,137 39.99 60.01 12,198 19,119 38.95 61.05

I'm including HD148 for comparison, even though I consider it to be one of the non-exceptional districts. If there were any remaining Hillary diehards who refused to embrace Obama, I'd have to say this is where they'd be found. Clinton dominated these districts in the primary, so perhaps some of her loyalists refused to push the button for Obama. That did not carry down the ballot, however, and it wasn't just the judicials who did better than Obama. Everyone from Adrian Garcia (76.2% in HD140, 73.2 in 143, and 75.6 in 145) to Rick Noriega (71.0, 69.3, 73.0) to the statewide Dems, the appeals court candidates, and the rest of the county slate did better than Obama. Maybe McCain still had some unique appeal to these voters, but my guess is they just never got as on board the Obama Express as others did. I'll refrain from making any remarks about the lack of effort from the Obama campaign to reach these voters like what they did in some other states and move on to the last group. I will, however, make the point that it was only Barack Obama who really underperformed here. Whatever antipathy some voters may have had towards the Presidential nominee was limited to him. Be sure to look at Greg's take on this as well.


3. The Lawyer districts

Whereas those five districts showed an increase in votes for the downballot Dems over what Obama got, the last two had the exact reverse outcome:


Dist McCain Obama M Pct O Pct GOP avg Dem avg GOP % Dem %
==================================================================
134 35,430 34,740 50.49 49.51 35,977 28,357 55.92 44.08
136 37,958 19,490 66.07 33.93 37,472 16,391 69.57 30.43

These were the only districts in which the GOP judicial incumbents gained votes over John McCain. They weren't the only ones to do so - John Cornyn did, though Rick Noriega lost about half as many votes as the Dem judicials, and thus was closer percentagewise; Wallace Jefferson also gained votes, as did a couple of the appeals court judges in HD134 - but the effect is striking. My hypothesis here is that these are two of the most affluent districts around, covering places like River Oaks, West U, Bellaire, the Galleria area, and Memorial, and they probably have more lawyers per capita than the rest of the county. It's not a shock to me that they might have preferred the status quo on the local benches at a higher rate than everyone else, if only to preserve the devils they knew. As foolish as the prediction-making business is, especially this far out, I'll bet that the new judicial incumbents perform much like their ballotmates in four years' time, and for the same reason. Someone please make a note to remind me to check that in 2012.

And finally, since there's been so much written lately about "funny names", I present the following:


Dist Obama Pereira Pierre Mahendru Murray Mincberg Trautman
===================================================================
128 33.84 35.80 36.10 33.73 35.35 33.19 33.78
144 40.60 42.88 42.69 40.20 41.99 38.80 39.15

140 64.16 70.44 69.12 66.51 67.63 64.86 64.35
143 60.48 67.53 66.37 63.60 65.29 62.29 61.03
145 62.28 69.21 67.39 64.32 65.73 62.14 61.10
148 60.01 60.55 59.46 58.74 58.50 53.42 54.23

134 49.51 42.11 41.65 42.50 41.76 39.69 39.59
136 33.93 28.94 28.95 28.40 28.66 28.03 27.37

Other 51.86 50.82 51.12 49.57 50.55 47.96 48.57


You can make of all that what you will. I note with some interest that Ashish Mahendru did better than the three other unsuccessful Democratic judicial candidates in HD134; perhaps that's a statement about incumbent Sharon McCally, who had the lowest Republican performance in Harris County in 2004 when Kathy Stone opposed her.

How's that for a start? Much more to come soon, so stay tuned.

Posted by Charles Kuffner on November 17, 2008 to Election 2008
Comments

I think some of the points on Item #2 warrant comparison to 2004. In short, I don't see any hint of a PUMA problem whatsoever.

Districts 143 and 148 have some noticably Republican-leaning precincts in the district whereas the other two Hispanic seats don't. This is part of the reason why the results are somewhat dampened in their Dem support.

In looking at 2004, the districts had better Democratic support down ballot (just as they did this time around) and they also both showed improved support for the Presidential candidate.

I'm sure there are some minor disparities somewhere in a comparison with 2004, but the larger patterns look awfully familiar. My sense is that the real story in these areas may either be moderate Hispanics who vote GOP at the top of the ticket, but Dem elsewhere ... or Hispanic voters who vote GOP at the top of the ticket and just plain *don't* vote down ballot.

Posted by: Greg Wythe on November 17, 2008 11:00 AM
Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)