March 09, 2006
Primary notes

Just a few random observations from the Tuesday primaries...

The thing that really strikes me about Tom DeLay's win is that his poorest showing was in his home county of Fort Bend:

County DeLay vote Total vote DeLay Pct
Harris 6,532 8,892 73.46
Galveston 798 1,161 68.73
Brazoria 2,341 3,508 66.73
Fort Bend 10,887 19,599 55.55

The turnout disparity between Harris and Fort Bend is presumably due to the numerous contested countywide primaries in FBC, including one for the county GOP chair. You can look at DeLay's win as vindication for him, but I look at that 62% figure and I see weakness. He'll certainly need more than 62% of the Republican vote to win in November, and as long as Steve Stockman is on the ballot, the anti-DeLay Republicans who can't bring themselves to vote for Nick Lampson have a place to go.

I also view DeLay's even weaker performance in FBC as a good sign. As I see it, there's no path to victory for Lampson that doesn't include carrying Fort Bend. DeLay got only 53% of the FBC vote in 2004, so that's very much an attainable goal, especially after nearly half of the GOP primary voters there picked someone else. I thought DeLay needed a big win on Tuesday to help dispel doubts about his electoral viability. This was not a big win.

Moving over to CD28, there's a myth that needs debunking.

There is no way to prove this, but I still bet that if this was a closed primary and Republicans and Independents had been unable to vote, we would have at least forced a run-off, if not won outright. The margin was so close, I can easily imagine a few thousand registered Republicans and Republican-leaning indepdents wanting to vote for Cuellar because of his connections to Bush. The Texas primary system would have allowed them to do so.

As a commenter noted there, you are a Democrat by definition if you vote in the Democratic primary. Bowers is talking about registering with a party as a prerequisite for voting in that party's primary, which is not how it works in Texas.

So yes, it is theoretically possible that people who might normally vote Republican chose to take part in this primary race since it would be their only chance to express a preference. If such people exist, they probably went with Cuellar. Let's stipulate that for the sake of argument and go from there.

The first question is how do you distinguish such voters from the "real" Democrats? Perhaps if they only voted in the CD28 race, since (in theory) that's the only one that interests them. I'd need to see precinct-level data before I could judge if turnout in CD28 was significantly higher than in the other contested primaries. Even if I found such a pattern, it's suggestive but hardly conclusive - I mean, this was the highest-profile race on the ballot. Who's to say that some number of "real" Democrats didn't just care about that race? The best comparison here would be to the HD42 race, featuring incumbent State Rep (and erstwhile CD28 challenger) Richard Raymond, his former staffer former Webb County Judge Mercurio Martinez whom Raymond tried to force off the ballot, and two others. If CD28 drew a lot more votes than HD42 in the same precincts, you might be able to convince me that it was stuffed with ringers. I don't see precinct data on the Webb County Clerk site, and it won't be available through the Secretary of State for some time, so I'll leave that task to someone more motivated than I.

It's my guess that you won't find any evidence of such, at least in Webb County where one expects the crossover motivation would have been highest. For one thing, turnout in 2006 was about what it was in 2002, prior to the redistricting. Indeed, it was a little higher in 2002, when hometown boy Tony Sanchez was on the ballot for Governor. That race drew over 31,000 votes in Webb; the 2006 topper, for Lt Gov, got just over 26,000, while Proposition 1, on raising the minimum wage, got 28,000 votes. The election day results (very slow loading PDF) on the Webb County Clerk site also includes the uncontested CD23, where Rick Bolanos pulled in 11,583 votes; add that to the 14,543 cast in CD28 and you're right at 26,000. Had CD23 also been contested, these two races might have combined for something like 28,000 or even 29,000 votes, so that's a little high, but again - CD28 was a high profile race. So was HD42, which largely overlaps CD28. In my opinion, there's nothing remarkable here.

What about turnout in the Republican primaries? Well, in Webb County, a grand total of 847 votes were cast in the GOP gubernatorial primary. In 2004, there were 846 votes cast in the Presidential primary, and in 2002 there were 604 ballots in the Lite Guv primary, all of which being the high scorers for their cycles. In other words, there is no Republican primary to speak of in Webb, so there's no basis for comparison here. If there are Republicans voting in the Webb County Democratic primary, they've always been there. Having Henry Cuellar on the ballot was only a factor as far as his Laredo roots were concerned. That's why Cuellar won - his home county supported him more than Bexar supported Ciro Rodriguez. I said before that Ciro needed a boost in Bexar turnout to win, and he didn't get it. That was and is the story of this race.

Rep. Carter Casteel is weighing a recount request in her 45-vote loss to Leininger puppet Nathan Macias.

"We'll probably take a look at that very seriously," Casteel said. "I probably owe it to the people who voted for me and contributed to me. I have until March 29 to decide."

The final, unofficial tally put Macias ahead with 10,176 votes to Casteel's 10,131 in the four-county district.

Macias said he's "telling everyone I'm victorious and will represent the district with strong, conservative values."

But he later added that with the number of ballots that still could show up and be counted, "I'm not sure anybody can claim victory."

The 45-vote margin could narrow when mail-in absentee ballots sent to out-of-country military personnel are counted. They must be postmarked by election day but can be received until March 20.

Officials are awaiting 93 ballots that were mailed out, but not yet returned, including 79 from Comal County, Casteel's base of support. However, election officials expect relatively few of those ballots to come in.

Four were delivered to the Comal County courthouse Wednesday, said Comal County Elections Administrator Linell Hinojosa.

There also were nine provisional ballots cast, seven from Comal County. Those will be checked by voter registration officials and counted if the voters registered properly.

"When I found out she lost by 44, I thought, 'Hey, this isn't over yet,'" Hinojosa said.

"It's workable, I guess, but my gut tells me not to bank on it," Casteel said of the final results.

Recounts seldom make a difference, but in any race that close I think you have to request one. You never know.

The primary fallout stories are coming: from the Chron, the Morning News, and an editorial from the Statesman, all of which focus on the effect on the upcoming special session on school finance. Oh yeah, this is gonna be fun.

Finally, if you want to read some tea leaves in certain vote totals for a barometer on November, there's good news and bad news for each party. The good news for Democrats is that in the open HD118, formerly held by soon-to-be-State Senator Carlos Uresti, there were only 1924 votes cast in the GOP primary, while the Democrats totalled 5703. The bad news for the Dems is that they had only 3639 votes cast in the HD47 primary, compared to 6090 for the GOP. I wouldn't read too much into this, especially given that HD118 is also within both SD19 and CD28, but consider it noted for future reference anyway.

UPDATE: As noted, Mercurio Martinez is not a former Raymond staffer - that was Sergio Mora. Thanks to Chito in the comments for the correction.

Posted by Charles Kuffner on March 09, 2006 to Election 2006 | TrackBack

I think HD 47 also suffered from lack of attention, being in the shadows of HD 48. The papers didn't cover it that much here in town or on the news (especially not compared to Howard), so a lot of folks didn't pay attention.

That, and there's lots of D's that don't want to vote for/against Earle, as they're unsure if his name hurts/helps him on the ballot. That prevented much money from going to Bolton and Earle, which in turn kept them from doing much GOTV stuff.

I'd imagine that will all quickly change come April 12.

Posted by: Phillip Martin on March 9, 2006 12:25 PM

Good analysis. From downtown San Antonio south to Laredo, there are a negligible amount of Republicans. There may be a few in San Marcos and New Braunfels that voted in CD 28, but not enough to affect the race. One correction - Mercurio Mrtinez is a former Webb County judge, not a Raymond staffer. The Raymond staffer that was in the HD42 race is Sergio Mora, who finished third. A few other things worth mentioning about the race: 1) most of the internet money for Ciro seems to have gone to attack ads. Maybe this was all that could be done since the flood of cash came so close to the end, but I don't think this is a winning strategy; 2) It seems that Ciro never even tried to campaign in Webb County. You cannot cede 1/3 of the voters to your opponent like that an expect to win. It was telling on election night when the vote tabulation problems came up that Ciro was "sending" a team down to investigate. Shouldn't he have had team in place?; 3) An article from the Laredo Morning Times explains the vote counting problems. Embarrassing? Yes. Fraudulent? No.

Posted by: Chito on March 9, 2006 1:53 PM

So DeLay's team turned out less than 35,000 voters. He got 150,000 votes in the 2004 general election, right?

Sounds like there are a lot of Republican voters who didn't feel strongly enough about DeLay to show up for him when he called.

Posted by: JWilson on March 9, 2006 3:02 PM

I think it's interesting that DeLay did relatively poorly in Ft. Bend, too. You got close to something when you observed that DeLay also fared relatively poorly in the general election of 2004. Don't stop there, DeLay has repeatedly underperformed in Ft. Bend County primaries and general elections. A cynic might say that those who know him best support him least. And I am a cynic.

Regarding 35,000 votes in the TX22 GOP primary. That's an unreasonable expectation. The fact is that GOP turnout went up ~18% in TX22 over 2002 (I know, redistricting and all). Quite outstanding for an off-year primary. DeLay would be hard pressed to argue that increased turnout helped him, though. His vote total declined by ~10%.

Posted by: Greg in TX22 on March 9, 2006 9:46 PM

CD28 and HD42 have barely any precincts in common. HD42 is moslty south Webb County.

Posted by: David on March 10, 2006 9:38 AM