Off the Kuff Rotating Header Image

John Kerry

The Harris County GOP thinks it can come back in 2020

They’re so adorable.

Never forget

Once a rock of Republican politics in Texas, Harris County has become nothing short of a nightmare for the GOP over the last four years as Hillary Clinton and Beto O’Rourke carried the county and Democrats dominated further down ballot in local races.

But as bad as it has been of late, party leaders say it’s foolish to consider Harris County blue, based on just two election cycles. They insist the party has learned key lessons over the last four years and made changes that will not just stop the Democratic trends, but lead to GOP victories in 2020 and beyond.

“We are still a strong force here,” Harris County Republican Party chairman Paul Simpson said.

He sees 2016 and 2018 as more of temporary Democratic run than a change of the guard. There have already been big changes that will affect 2020, he said, pointing to the end of straight-ticket voting, better minority community outreach and a renewed commitment to registering new voters as three things that will lift GOP candidates in Harris County.

That’s not to discount the pain of the last two election cycles. Shifting demographics and an emboldened Democratic Party that has registered new voters at record speed allowed Clinton in 2016 to win the biggest share of the vote for a Democratic presidential candidate in Harris since Texas icon Lyndon Johnson was on the ballot in 1964.

And in the governor’s race in 2018, Democrat Lupe Valdez — who ran a campaign that was mediocre at best — won Harris County over incumbent Republican Gov. Greg Abbott, becoming only the second Democratic gubernatorial candidate to carry Harris County in 25 years.

“There was shell shock,” Republican media consultant Vlad Davidiuk said.

[…]

Months before the 2018 election, Abbott’s political team was warning allies about what was happening in Harris County. That summer at a training session in San Antonio, Abbott campaign advisers told workers that Democratic-leaning voter registration groups such as Battleground Texas were making big gains registering new voters in Harris County.

Davidiuk, who was working with the Harris County Republican Party then, said others saw it coming, too.

“We didn’t have a response to that,” he said. “If there was a response, it was too fractured.”

That voter registration push has only grown the Democratic advantage at the polls the last two years.

“Our historic voter base is shrinking in both real and absolute terms,” the 2016 post election analysis says. “As a consequence, we are at risk of becoming a minority party within Harris County.”

Later it makes clear that “Donald Trump’s loss in Harris County and its down-ballot impact in 2016 could foreshadow future elections if we do not broaden our voter base.”

I’ve already said most of what there is to say about this. The rationales they give – it was Beto! straight ticket voting! Trump! why don’t those minorities like us? – are as predictable as they are pathetic and self-unaware. The straight ticket thing I’ve beaten to death (but feel free to reread this for one of my responses to that trope), but I think what we need here is to throw some numbers at these claims.


Year    R Pres   D Pres   R Judges   D Judges
=============================================
2004   584,723  475,865    535,877    469,037
2008   571,883  590,982    541,938    559,048
2012   586,073  587,044    563,654    568,739
2016   545,955  707,914    605,112    661,404

Republicans have basically not done any better at the Presidential level in Harris County since George W. Bush in 2004. They have grown some at the judicial level (the numbers you see above are the average totals from the District Court races, my go-to for measuring partisan vote totals), which highlights Trump’s extreme underperformance, but their growth (plus 70K from 2004 to 2016) is dwarfed by Democratic vote growth (plus 192K) over the same period. This is my thesis, which I’ve repeated over and over again and which has clearly not sunk in. This is the problem Republicans need to solve.


Year  R Judges   D Judges    R Str    D Str  R Str%  D Str%
===========================================================
2004   535,877    469,037  370,455  325,097   69.1%   69.3%
2008   541,938    559,048  343,919  391,488   63.5%   70.1%
2012   563,654    568,739  404,165  406,991   71.7%   71.6%
2016   605,112    661,404  401,663  472,030   66.4%   71.4%

These are the countywide straight ticket voting totals, and the percentage of each side’s average judicial total that came from straight ticket votes. Looked at this way, Democratic straight ticket vote total growth is proportionate to their overall vote total growth. In other words, the increase in Democratic straight ticket voters wasn’t inflating their overall strength, it was merely reflecting it. Meanwhile, fewer people voted straight ticket Republican in 2016 than they did in 2012. Sure, some of that is a reaction to Trump, but that’s still a big problem for them, and it’s not something that the elimination of straight ticket voting will help them with in 2020. Note also that Republicans have been pretty heavily dependent on straight ticket voting as well. I do not understand the assumption that its removal will help them.


Year  Voter Reg   R Pres%  R Judge%  D Pres%  D Judge%
======================================================
2004  1,876,296     31.2%     28.6%    25.4%     25.0%
2008  1,892,656     30.2%     28.6%    31.2%     29.5%
2012  1,942,566     30.2%     29.0%    30.2%     29.3%
2016  2,182,980     25.0%     27.7%    32.4%     30.3%

The first column is the total number of registered voters in Harris County in the given year, and the percentages are the percentage of each of the total registered voter population. As a share of all registered voters, Donald Trump did worse than John Kerry, while Hillary Clinton did better than Dubya. The share of all voters choosing Democratic judicial candidates increased twenty percent from 2004 to 2016, while the share of all voters choosing Republican judicial candidates declined by three percent. This is what I mean when I say that the Republicans first and foremost have a “not enough voters” problem in Harris County. Their second problem is that they have no clue what to do about it.

For what it’s worth, here’s a similar comparison for the off years:


Year  R Judges   D Judges    R Str    D Str  R Str%  D Str%
===========================================================
2002   333,009    270,564  185,606  171,594   55.7%   63.4%
2014   359,842    297,812  254,006  210,018   70.6%   70.5%
2018   531,013    651,975  410,654  515,812   77.3%   79.1%

Year  Voter Reg  R Judge%  D Judge%
===================================
2002  1,875,777     17.8%     14.4%
2014  2,044,361     17.6%     14.6%
2018  2,307,654     23.0%     28.3%

Couple things to note here. One is that there wasn’t much in the way of growth for either party from 2002 to 2014, though as we know there were some ups and downs in between. The 2018 election was a lot like a Presidential election in terms of turnout – you’ve seen me use 2012 as a point of comparison for it before – but one in which the Dems did a much better job. No Republican, not even Ed Emmett, came close to getting 600,000 votes. Here, I’ll agree that having unpopular politicians at the top of the ballot, like Ted Cruz and Dan Patrick and Ken Paxton, as well as having to fly under the Trump banner, helped propel Dems, in part because of former Republicans crossing over. But they were starting from a lower point to begin with.

Note, by the way, the jump in voter registrations from 2012 to 2014. Mike Sullivan deserves some credit for that, as he was the first Tax Assessor in a long time to not be hostile to voter registration, but this was also the point at which Dems started really focusing on registering voters. For sure, that has helped, and I’ve no doubt that Abbott’s people had reason to be alarmed going into 2018. I find it kind of amusing that Republicans are turning to voter registration themselves as a way forward. I have to wonder if that will lead to any bills getting advanced that would make voter reg easier and more convenient. My guess is still No, on the grounds that they probably figure they can throw money at the problem and would still rather have it be hard for Dems, but we’ll see.

I could go on, but you get the point. And as a reminder, the numbers themselves aren’t the whole story about why Republicans are struggling and will continue to do so in Harris County:

Simpson, for one, is glad to see the parade of Democratic presidential contenders coming to Harris County because it puts their ideas — particularly on climate change — front and center. Let them bring their calls for banning fossil fuels, he said.

“They don’t want us to eat beef, drill for oil or even use straws.”

Because it there’s one thing younger voters really hate, it’s trying to solve climate change. Way to be on top of the trends there, dude.

As goes Tarrant

The Trib ponders the one big urban county that is not like the others.

Among the state’s five biggest counties, Tarrant is the only one that hasn’t backed a Democratic presidential candidate in the past decade. The 2016 presidential election heightened Tarrant’s status as an outlier. Even as the rest of the state’s big-city territories moved deeper into the Democratic column, Tarrant steadfastly emerged as America’s most conservative large urban county.

President-elect Donald Trump, who takes office this week, won the county by an 8.6-point margin. It was the narrowest win for a GOP presidential nominee in decades in Tarrant. But among the country’s 20 largest counties, Tarrant was only one of two that swung Trump’s way in November — and it had the wider margin.

Across Tarrant County, Democratic pockets are fewer and less powerful than their Republican counterparts. All four of the state senate districts that fall in Tarrant County are represented by Republicans. The GOP also holds eight of the county’s 11 state House seats. Four of the five county commissioner court seats are held by Republicans.

Residents, elected officials and experts here point to a nuanced union of demographic, cultural and political forces to explain why.

“There’s just all kinds of interesting numbers out there that make Tarrant County a lot different,” said U.S. Rep. Marc Veasey of Fort Worth, the only Democrat holding one of the county’s five congressional seats.

Tarrant’s minority population, which tends to lean Democratic, hasn’t caught up to the state’s other big urban counties. At the same time, many Tarrant voters have a storied history of preferring practical governance to partisanship, according to officials and political observers. They say that helps support the moderate faction of the GOP, especially in Fort Worth, the nation’s 16th-largest city.

Then there’s the county’s development pattern. A lot of Tarrant remains rural. And, unlike Harris, Dallas and Travis counties, many of Tarrant’s affluent suburbs and conservative bedroom communities lie within its borders, not outside them. That’s helped give rise to the NE Tarrant Tea Party, a passionate and organized group that simultaneously supports far-right local candidates and serves as a powerful base for statewide Republicans.

[…]

Part of what has helped Tarrant become the state’s lone Republican urban county is that its minority populations, which largely and traditionally tend to lean Democratic, haven’t caught up to the state’s other big urban counties.

White residents’ share of the Tarrant population is falling, but it hasn’t declined as quickly as it has in Harris, Dallas, Travis and Bexar, said state demographer Lloyd Potter. The county’s Hispanic population is growing quickly, but it still lags behind the other big counties in terms of raw numbers, Potter added.

But that’s likely to change.

While Tarrant remains more white than Texas as a whole, it’s experienced a more significant drop in its share of white residents in the past 10 years compared to the state. In 2015, the county’s white population dropped to 48.5 percent — down from 56.4 percent in 2005.

Whites’ falling numbers in the county aren’t limited to its urban core in Fort Worth. In fact, the white population experienced a bigger drop in its share of the population in the suburbs from 2005 to 2015.

Here’s a fun fact, which I believe I have mentioned before: Tarrant County is a really good predictor of the overall Presidential race result in Texas. Witness the past four elections:

2004

Statewide – Bush 61.09%, Kerry 38.22%
Tarrant – Bush 62.39%, Kerry 37.01%

2008

Statewide – McCain 55.45%, Obama 43.68%
Tarrant – McCain 55.43%, Obama 43.43%

2012

Statewide – Romney 57.17%, Obama 41.38%
Tarrant – Romney 57.12%, Obama 41.43%

2016

Statewide – Trump 52.23%, Clinton 43.24%
Tarrant – Trump 51.74%, Clinton 43.14%

Almost spooky, isn’t it? One perfectly rational answer to the question “when will Texas turn blue?” is “when Tarrant County also turns blue”.

Anyway. The article is correct that Tarrant differs from the other big urban counties in that it’s actually a lot less urban than they are. Much of Tarrant is suburban, even rural, and that’s just not the case in Harris, Dallas, Bexar, and Travis. Tarrant’s demographics are changing, as the story notes, but I have no idea if there’s anything to suggest its demographics are changing any faster than the state’s are. The statewide judicial races and the one contested district court race were all in the 13-16 point range, which is consistent with the statewide results. I wish I could say I saw something to suggest change was coming faster, but at least in the numbers, I can’t. Maybe someone who is more familiar with the county can chime in.

Having said all this, one big opportunity in 2018 is in Tarrant, and that’s SD10, the Senate seat formerly held by Wendy Davis. Even in the dumpster fire of 2014, freshman Sen. Konni Burton only won by nine points, with 52.83% of the vote. If 2018 is a less hostile year, this is a winnable race, and as I’ve said before, any competitive Senate race is a big deal. Whatever we can do to hasten change in Tarrant County, 2018 would be a good time to do it.

Early voting, one week in

We have completed one full week of early voting, and through Sunday a total of 362,827 people had voted in person, with an additional 53,131 ballot being cast by mail, for a grand total of 415,958. The updated spreadsheet is here for your perusal. For comparison, there were 314,252 in person ballots cast through Sunday, 2008, so this year represents a 15.5% increase in non-absentee early votes. Another way to look at it is that there were 1,892,65registered voters in 2008, and this year there are 2,003,436 registered voters, which is 5.8% more. If the increase in early voting turnout were driven entirely by the increase in voter registration, we would have had 332,479 early votes by now. The actual total of 362,827 is therefore an increase of 9.1% over what might have been expected.

Again, all this suggest what we are seeing is the new normal. The totals are high-water marks, but they’re not a quantum leap like what we saw in 2008. It’s not out of the question to me that we could see the pace of early voting slack off a bit next week, with 2012 numbers losing some of their lead over 2008 numbers. I’m confident that 2012 will have more early voting, even accounting for registration growth, but the percentage margin may be less at the end of this week than it is right now. Just a feeling, I have no objective evidence for this. We’ll see.

One more thing to talk about is not just how many people are voting early, but which people are voting early. In 2004, 45.5% of all straight-ticket Republican votes and 43.8% of George W. Bush’s votes were cast early, while 40.7% of both straight-ticket Democratic votes and John Kerry’s votes were cast early. In 2008, those numbers were 61.6% of straight-ticket Republican votes and 59.4% of John McCain’s votes, and 66.6% of straight-ticket Democratic votes and 66.4% of Barack Obama’s votes were early. I suspect in the end that the share of each party’s early votes will be about the same, and at least as high as the Democratic share was in 2008. And, as foolish as it is to make predictions this far out, I suspect we’ll see the same sort of behavior in 2016.

Mail ballots

Campos has been tracking mail ballot requests to the Harris County Clerk.

Here is what the County put out yesterday evening:

As of this evening we have approved 71,101 applications and sent out 67,376 ballots. We have received 19,468 voted ballots returned.

25,848 have been generated by the GOP and 20,866 by Dems.

In an earlier entry, Campos noted that “In 2008 in Harris County, 76,187 requested mail ballots and 67,612 (88.7%) were returned and counted. In 2010 in Harris County, 69,991 requested mail ballots and 55,560 (79.4%) were returned and counted.”

So 55.3% of the ballots that have been requested by people with an identifiable primary voting history are going to Republicans. Out of curiosity, I looked at the two most recent Presidential-year elections in Harris County for a point of comparison. In 2004, 29,926 absentee ballot voters went for George W. Bush, and 17,010 went for John Kerry. That’s 63.8% for Bush. In 2008, the numbers were 41,986 absentee voters for John McCain, and 24,503 for Barack Obama; that’s 63.1% of absentee votes for McCain. Already we have more absentee ballots requested with more than a week to go before early voting starts than were cast in 2008, and about as many that can be identified by party primary voting as were cast in 2004. We don’t know how many of those requesters from each group will actually return their ballots, and we don’t know how many of those 19,000 or so non-primary voters really belong to each party, but early on at least it looks like Democrats may have closed the absentee gap a bit. Whether that means anything for the final totals I couldn’t say – it’s entirely possible that most if not all of the new absentee ballot requesters are folks who would have voted in person anyway, in which case this is just shifting things around a bit. I make note of this because I’m a numbers guy and these are some interesting numbers. Are you voting by mail this year, and if so have you done it before? Leave a comment and let us know.

How about those judicial races?

The Chron takes a look downballot.

Democratic judges who surprised Harris County in a 2008 rout because of strong turnout for Barack Obama are bracing for a tough fight in November after seeing the GOP, which had a clean sweep in 2010, continue to bolster its position statewide.

In the county’s 23 contested state district court races, 18 Democrats will have to overcome strong Republican momentum to keep their benches.

“It doesn’t look great,” said Brandon Rottinghaus, a political science professor at the University of Houston. “The state is trending conservative, so it will be difficult for Democrats to retain a lot of those seats.”

Although the judges are countywide races, they are too far down the ballot for most voters to study and make choices outside of party affiliation.

“A lot of these races are consumed and swept up in the general partisan trends,” said Rottinghaus. “Harris County certainly has flecks of blue, but there are elements that will drive the state to be more red this year.”

He said anger with Obama, as evidenced by the tea party, and the popularity of critics of the president, like Senate candidate Ted Cruz, will influence the election.

“The people at the top of the ticket are driving not only the turnout, but also much of the debate we’re having nationally, statewide and locally.”

Other experts said they do not expect a sweep, while well-known candidates on both sides will rise above the fray.

“I do not see a partisan sweep either way,” said political analyst Robert Miller. “Strong Democrats such as (Sheriff) Adrian Garcia will win, as will strong Republicans such as (district attorney candidate) Mike Anderson.”

It’s hard to argue with Miller’s prediction, but that doesn’t answer the question about the judicial races, which are primarily a function of base turnout. I’m not sure what Prof. Rottinghaus is basing his opinion on. Here’s a look at Presidential turnout levels in Harris County since 1992:

Year Republican Democrat =========================== 1992 406,778 360,771 1996 421,462 386,726 2000 528,965 418,143 2004 584,723 475,865 2008 571,883 590,982

I know from past study of 2004 races that George W. Bush received a number of Democratic crossover votes, so his total is a bit inflated. Still, the average Republican judicial candidate in a contested race received about 536,000 votes in 2004, and about 540,000 votes in 2008, while Democratic judicial candidates got 470,000 and 562,000, respectively. Was overall Republican turnout depressed in 2008? Maybe. Is it likely to be better this time around? Again, maybe. The Tea Party was clearly a factor for them in 2010, but that was largely due to bringing out Presidential year voters in a non-Presidential election. Those people are already factored into the equation for this year. How many Republicans who didn’t vote in 2008 are likely to come out this year, that’s the question. I suppose, as Prof. Rottinghaus suggests, that Ted Cruz could be amping up their excitement levels – Lord knows, Mitt Romney ain’t doing it – but if so he’s doing it while executing a Dewhurst-style avoidance campaign. (And I don’t know about you, but the only campaign ads I’ve seen lately are Obama ads, which run in fairly high frequency on cable. I swear, I never saw an Obama ad at this time in 2008, nor a Kerry ad ever.) The bigger question is where are new Republican voters coming from? Turnout levels in the Republican parts of Harris County were already very high in 2008, while turnout levels in the Democratic strongholds didn’t change much from 2004. As we know from the polls, the GOP’s base of support comes from Anglo voters, yet Harris County’s Anglo population is on the decline, at least relative to other populations. So again, where are new Republicans coming from?

On the flip side, it is certainly plausible that Democrats hit a peak in 2008 and that a fair number of new and irregular voters who showed up that year won’t bother this time around. Democratic enthusiasm and engagement seems pretty good from where I sit, certainly better than it was earlier this year, but 2008 was a historic year, and 2012 is a defensive one. This I think is the biggest factor, and it’s one I have a hard time quantifying. Demography and the current national atmosphere favor the Dems, but the Democratic base is more prone to enthusiasm deficits, and the effects of voter registration restrictions and voter intimidation efforts are unknown. Overall, I think the Democrats are in the better position, but I really don’t know how to feel about this election locally.

A look at HD136

The Statesman takes a look at the new State Rep. district in Williamson County.

Matt Stillwell

All county and state elected officeholders from Williamson County are Republicans. The party has long dominated the area. But Democrats are eyeing the new district as a potential weak spot in the Republican stronghold, counting it among a handful of districts they hope to take in November.

The race is a high priority, said Bill Brannon, executive director of the state Democratic Party.

“I would say it’s either top tier or very, very close to top tier — it’s a high target,” Brannon said. “It’s a race that presents a lot of opportunities.”

The Democrats hope that [Matt] Stillwell, a father of three who owns an insurance agency and has never held elected office, can beat out Republican candidate Tony Dale and Libertarian Matthew Whittington for control of House District 136. The new district covers portions of Northwest Austin, Cedar Park, Leander, Round Rock and the Brushy Creek Municipal Utility District.

[…]

Democrats say demographics in the newly drawn district provide an opportunity for the party.

According to data from the Williamson County Elections Department, more than a third of registered voters in the district live in Austin and nearly 20 percent of them are 30 or younger — a group that generally leans left.

The party sees the district as a possible repeat of Democratic state Rep. Mark Strama’s 2004 grab of a northern Travis County seat.

Strama, who still holds the seat immediately south of House District 136, said that like the new district, his was drawn for an easy Republican win.

“Frankly it was hard to convince anybody it was a winnable race — which I think is the challenge Matt has now,” Strama said.

Growth in the area — especially in Pflugerville — from Central Austin, as well as California and other states, made his district more Democratic than anyone realized at the time, Strama said.

HD136 is a race that isn’t quite on the radar. It’s a second tier Back To Blue race, outside observers like Robert Miller have not taken it into account. The numbers are daunting but not overwhelming, and there is certainly some hope that the landscape has changed. A comparison of the 2004 and 2008 numbers is instructive:

2004 Bush Kerry Keasler Molina ================================== 63.8% 36.2% 63.2% 36.8% 2008 McCain Obama Wainwright Houston ================================== 51.8% 45.9% 51.2% 42.9%

Data can be found here and here; both links are XLS files. The dropoff from Obama to Sam Houston is mostly accounted by the 5.9% received by the Libertarian candidate in that race. The difference between the two years is striking, and it’s magnified by the raw vote totals. John McCain barely beat George Bush’s number – McCain received 32,977 votes, Bush got 32,413 – but Obama’s total was more than fifty percent greater than John Kerry’s – Obama got 29,227, Kerry just 18,403. I’m sure some of that was “surge”, and maybe that will be hard to repeat, but still. That’s a huge difference. Part of Stillwell’s challenge is identifying and reaching out to the new voters in the district, and part is making sure that those who vote for Obama stick around for him as well – Sam Houston’s vote total, by contrast, was only 25,734, a much bigger decrease from Obama than Dale Wainwright’s 30,696 was from McCain. The comparison to Rep. Mark Strama, who won a rapidly-changing district in 2004 that had been thought to be solidly red in 2002, is instructive, but there is one key difference here: Stillwell has a lot less money than Strama did at the time. Maybe that’s why this race isn’t as high profile. Keep an eye on this one, though, it could easily be a surprise on November 6.

Sadler’s challenge

Democratic Senate hopeful Paul Sadler is a strong candidate with limited resources. Where have I heard that before?

Paul Sadler

In Victoria on a recent Saturday afternoon, the candidate for the U.S. Senate had the crowd on its feet, the shouts and applause washing over the meeting room like waves on the nearby Gulf. As he wrapped up his 15-minute jeremiad warning of the havoc his opponent would wreak on the Lone Star State and, as he began making his way to the back of the room, shaking hands and posing for photos along the way, an older woman in a red pantsuit sought to recapture the crowd’s attention.

“This campaign costs money,” she shouted into the microphone several times, but only those within a few feet of her were listening. One of them eventually doffed his straw hat, which became a makeshift collection basket for a statewide campaign tossing nickels and dimes at an opponent awash in money and nationwide ardor.

The Victoria experience represents the Paul Sadler campaign in miniature. Little-known statewide and underfunded, the lawyer and former state representative from Henderson is a capable campaigner, an experienced lawmaker and a credible candidate for a party desperately in need of new faces and arresting ideas.

Sadler’s problem, of course, is that his GOP opponent, tea party darling Ted Cruz, has been all but anointed the successor to retiring U.S. Sen. Kay Bailey Hutchison, R-Texas. Cruz has money, star power and the overwhelming advantage of being a Republican in the most fervid of red states. In last month’s Senate runoffs, 1.1 million Texas Republicans cast a ballot, compared to 235,000 Democrats.

I’m going to begin by going off on a tangent here. I don’t know exactly how one defines fervidness in this context, but at least by 2008 results, Texas isn’t the reddest of red states. It’s not even in the top half, if one uses margin of victory as the metric. Here are the 2008 results by state. I’ve helpfully plucked out the states carried by John McCain and sorted them by margin of victory below:

State Obama McCain Margin ================================= Wyoming 32.54 64.78 32.24 Oklahoma 34.35 65.65 31.29 Utah 34.22 62.24 28.02 Idaho 35.91 61.21 25.30 Alabama 38.74 60.32 21.58 Alaska 37.89 59.42 21.54 Arkansas 38.86 58.72 19.85 Louisiana 39.93 58.86 18.63 Kentuky 41.15 57.37 16.22 Tennessee 41.79 56.85 15.06 Nebraska 41.60 56.53 14.93 Kansas 41.55 57.37 14.92 Mississippi 43.00 56.17 13.17 W Virginia 42.51 55.60 13.09 Texas 43.63 55.39 11.76 S Carolina 44.90 53.87 8.98 N Dakota 44.50 53.15 8.65 Arizona 44.91 53.39 8.48 S Dakota 44.75 53.16 8.41 Georgia 46.90 52.10 5.20 Montana 47.11 49.49 2.38 Missouri 49.23 49.36 0.13

Fourteen states were redder than Texas in 2008. Even in 2004, when George W. Bush was running for re-election and beat John Kerry by 22 points here, Texas was only the tenth-reddest state. Now I admit that even an 11.76 point margin is still daunting, and if you go by vote margin instead of percentage margin Texas was indeed the reddest state in 2008 – McCain got 950,000 more votes than Obama; only Oklahoma and Alabama had margins greater than 450,000 – but that’s a function of population, not popularity. I mean, Alabama and Oklahoma had barely more total votes for both candidates combined than Texas had for just Obama. If fervidness is a synonym for intensity, then Texas was at best #15 for the GOP in the last Presidential election.

But numbers are one thing, perception is another, and the perception that Texas is as red as it gets is a big factor working against candidates like Sadler and other Democratic statewides. Fundraising is obviously affected by this – it’s one thing to give to an underdog, another to a hopeless cause. I believe Sadler is the former, and I’m putting my money where my mouth is by cohosting a fundraiser for him on Monday, September 24 at the Continental Club. There obviously isn’t much time for fundraising at this point, and I don’t even know what a realistic target that can make a meaningful difference might be, but I do believe a difference can be made. If you think so as well, come out and help the cause and meet the candidate on the 24th at the Continental Club. Thanks very much.

PPP: Perry takes lead on Obama in Texas

Not by that much, however.

Rick Perry has an under water approval rating in Texas and he’s leading Barack Obama by a smaller margin than John McCain won the state by in 2008…but at least he is leading Obama, which is more than he could say the last time we polled the state.

45% of Texas voters approve of the job Perry is doing to 48% who disapprove. Those aren’t good numbers but they do represent improvement from a June PPP poll of the state when Perry was at 43/52. The better numbers are attributable to Republicans really rallying around him. He was at 73/21 with them before but now it’s 78/14. He continues to be very unpopular with independents though (32/61) and even in a state that still has a lot of conservative Democrats his crossover support is virtually nonexistent with just 13% of voters approving of him across party lines. The numbers with independents are particularly troublesome for Perry- if that’s where he is with swing voters where they know him best, can he expect to do well with those folks in key swing states like Ohio, Colorado, and Virginia?

Perry leads Obama in a head to head 51-44. Those aren’t terribly impressive numbers given that John McCain defeated Obama by 12 points in the state, but they do at least represent an improvement for Perry since June when he actually trailed the President 47-45. Perry polls the best of any of the Republicans in Texas- Mitt Romney leads Obama by 6 points at 47-41, Ron Paul’s up by a single point at 43-42, and Obama actually leads Newt Gingrich and Michele Bachmann by 46-45 and 45-43 margins respectively.

See here for the earlier poll details. As I’ve been saying, there’s no evidence so far in Texas that the electorate looks all that different than it did in 2008. Standard disclaimers apply, but so far there have been no poll results that make you sit up and say “Whoa!”

Couple things to ponder here. One is that in recent Presidential elections, the Republican Presidential nominee has done better than Republican downballot candidates. The reason for this has been a greater dropoff in Republican voting downballot than there has been for Democrats – basically, Democratic downballot candidates get almost as many votes as John Kerry and Barack Obama – in some cases, more votes than them – while Republican candidates lose between five and ten percent of the vote total that George Bush and John McCain received. If that pattern continues, it’s not hard to imagine downballot races being quite close, possibly being won with less than 50% of the vote given the three or four percent that Libertarians generally take. On the other hand, as we saw in 2010, there’s a not-insignificant number of Republicans who dislike Perry enough to cross over against him. That’s not quite the case in this poll, as the crosstabs make clear – Perry does best among Republicans by far, but Obama gets a clear majority of independents. I suspect, however, that a significant number of those “independents” are otherwise fairly reliable Republican voters, so it’s hard to say exactly how different this is from 2010. Point being, I don’t have a good feel yet for whether Obama would generally lead or trail other Dems in a matchup with Perry.

Also, as noted by Stace, Obama does well with Hispanics in Texas, but could do even better:

There are a couple things keeping him from getting completely crused in the state though. One is the Hispanic vote- he’s up 28 points on Perry, 35 on Romney and Paul, 43 on Bachmann, and 45 on Gingrich with those voters. In the case of Perry that margin is equal to what Obama won Hispanics by in Texas in 2008 and with the others it’s a wider spread. This is one state anyway where he is not slipping with Latino voters.

NewsTaco has previously noted that Obama polls quite well among Latinos against all of the GOP hopefuls, Perry included. Perry for his part is hoping to do better among Latinos, assuming he doesn’t get teabagged on “sanctuary cities” and his prior support of the DREAM Act. I’ve looked through the 2010 results to see how Perry did in Latino districts compared to other Republicans, and it’s kind of a mixed bag; the numbers get skewed by the races that feature Latinos, and by differing dropoff rates. Having said that, he did better in South Texas than I would have predicted, less well in the urban areas. Put Marco Rubio on the ticket with him and I’d certainly be concerned.

Anyway. Just another data point, which comes just as the real campaign is getting under way. I’ll keep track of these things to see if any trends develop. Greg has more.

Yeah, it is too early to be polling for 2012

But that won’t stop anyone from doing them.

2012 could be the year Democrats are finally competitive for President in Texas…but only if the Republicans nominate Sarah Palin.

There are vast differences in how the various different potential GOP contenders fare against Barack Obama in Texas. Mike Huckabee is very popular in the state and would defeat Obama by 16 points, a more lopsided victory than John McCain had there in 2008. Mitt Romney is also pretty well liked and has a 7 point advantage over the President in an early hypothetical contest, a closer margin than the state had last time around but still a pretty healthy lead. A plurality of voters have an unfavorable opinion of Newt Gingrich but he would lead Obama by a 5 point margin nonetheless. It’s a whole different story with Palin though. A majority of Texas voters have an unfavorable opinion of her and she leads the President by just a single point in a hypothetical contest.

Part of the reason Obama looks like he could be competitive against the right Republican opponent is that his position in the state has improved. 42% of voters approve of the job he’s doing to 55% who disapprove. His average approval rating in 4 surveys conducted in PPP over the course of 2010 was 38% so he’s seeing the same sort of uptick in his numbers there that he’s seeing nationally right now.

The other reason for Obama’s closeness is the weakness of the Republican candidate field. He’d have no shot against a GOP nominee that voters in the state like. Huckabee’s favorability rating is a 51/30 spread and he blows Obama out of the water. But none of the other GOP hopefuls come close to matching that appeal. Romney’s favorability is narrowly in positive territory at 40/37, but Gingrich’s is negative at 38/44, and Palin’s is even worse at 42/53. Texas voters certainly don’t like Obama but for the most part they don’t see the current Republican front runners as particularly great alternatives.

What’s maybe most striking about Obama’s competitiveness in these numbers is that they’re from the same sample that showed Democrats had virtually no chance of picking up Kay Bailey Hutchison’s Senate seat earlier this week, trailing all 12 match ups we tested by double digit margins.

The previous poll results are here. I’m going to disagree with the analysis in that I think it really is all about name recognition. In the end, Obama may or may not perform better than whoever the Democratic candidate for Senate is – I’ll take the over if it’s Gene Kelly, the under if it’s John Sharp, and would consider it a tossup otherwise – but he isn’t about to perform 10 to 15 points better than any of them. The level of support Obama gets is roughly going to be the base Democratic performance level.

Yeah, sure, candidates and campaigns and fundraising matter, but only so much in a Presidential year. John Cornyn had Senate incumbency, several terms as a statewide officeholder, and something like a 3-1 financial advantage over Rick Noriega, yet he finished behind John McCain in both total votes and vote percentage, and did only one to three points better than downballot Republicans. Barring a Gene Kelly situation, I expect all the Democratic statewide candidates in 2012 to be within a few points of each other.

The question is what is the ceiling for Democrats in 2012. About a million more people voted in Texas in 2004 than in 2000, and at both the Presidential level and downballot, the Republicans got about 70% of those votes. About 900,000 more people voted in 2008 than in 2004, and again at all levels the Democrats got about 90% of those votes. There are a number of reasons for this, but one factor I’d point to is Latino support. Obama did more than ten points better among Latino voters than John Kerry did, and that was a big part of it. Call me crazy, but I don’t think any Republican Presidential candidate is going to appeal to Latinos like George W. Bush did in 2004. Given that our state, and our electorate, isn’t getting any whiter, I like those odds.

I’ll say this much, if Team Obama actually spends some money in Texas, it would make a difference. If they consider the 2010 results in a vacuum, they’ll run screaming in the other direction, but this was a tough year all over, and one presumes they’re smart enough to realize that the 2012 electorate will be very different, here and elsewhere. Even if they (quite reasonably) think our electoral votes are out of reach, there’s still an excellent reason to play here, and that’s for the Congressional races. Two Republicans won in 2010 with less than 50% – Blake Farenthold and Quico Canseco – and of course there will be four new seats to fight over. Winning back CDs 23 and 27, and taking two of the four new seats, would mean a net +2 for Dems in Texas. If Obama hopes to start his second term with a Democratic (or at least a more Democratic) Congress, that sure would help.

All right, I don’t really plan to talk about this much between now and the end of the year, so file this away for later. We’ve seen how quickly and significantly the winds can change over a few months, so we’ll see where things stand once the Republicans begin to coalesce around a single contender.

Cap and trade would cut the federal deficit

Surely this means all those “deficit hawks” I keep hearing about will rush to embrace the American Power Act now. Right?

The CBO analysis of the American Power Act, championed by Sens. John Kerry (D-Mass.) and Joe Lieberman (I-Conn.) found that government revenues would grow by about $751 billion from 2011 to 2020 if the bill became law. By contrast, the legislation would create direct spending of $732 billion over the same 10-year period.

Authors of the proposal called the CBO report a “powerful message” ahead of a floor debate next month. They are still searching for a formulation that will draw 60 votes.

“There is no more room for excuses; this must be our year to pass comprehensive climate and energy legislation and begin to send a price signal on carbon,” Kerry and Lieberman said in a joint statement. “Many of our colleagues have said they flatly oppose anything that adds a penny to the deficit, so we hope they look anew at this initiative, which reduces it.”

The CBO report is here. Of course, no one actually believes the American Power Act can pass, because we can’t afford it or some such, so the talk is about various alternate approaches that may have a chance of surviving the Senate. No, I don’t understand that either. Texas Vox has more.

Population and voting trends: 2004 and 2008 judicial elections

So we’ve seen how county returns changed in the Presidential election between 2004 and 2008. Obviously, there are many factors that can affect a Presidential election, even when there’s not really an active campaign going on in the state. How do things look at the judicial level, which is probably a closer reflection of party ID? To try to answer that, I compared two races for the Supreme Court, and two for the Court of Criminal Appeals: Scott Brister versus David Van Os in 2004 and Dale Wainwright versus Sam Houston in 2008; Mike Keasler versus JR Molina in 2004 and Tom Price versus Susan Strawn in 2008. My observations:

– Houston improved on Van Os’ percentage by six and a half points, going from 40.76% to 47.31%; Strawn did a bit less than five points better than Molina, 42.14% to 46.86%. (Note that both 2008 races included a Libertarian candidate, while neither 2004 race did. All percentages are based strictly on R/D vote totals only.) In doing so, Houston cut the 2004 deficit by 875,000 votes, while Strawn improved by 616,000 votes over 2004.

– One corollary to that is that Houston gained in more counties than Strawn did. There were only 28 counties in which Houston’s deficit was greater than Van Os’, with Montgomery and Parker being the places he moved backwards the most. Strawn did worse in 69 counties, adding Orange and Jefferson to the biggest loser list. Recall that there were 107 counties in which Barack Obama lost ground compared to John Kerry.

– The 20 counties in which Obama lost the most ground from Kerry differed somewhat from the counties in which Houston and Strawn combined did worse than Van Os and Molina. Counties that appeared in the former list but not the latter were:

Bowie: Obama’s deficit increased by 3436 votes; Houston gained 1303 while Strawn lost 867.
Galveston: -3082 for Obama, +2720 for Houston, and -1307 for Strawn.
Jasper: -1488 for Obama, +866 for Houston, and -656 for Strawn.
Liberty: -1416 for Obama, +1185 for Houston, and +155 for Strawn.
Harrison: -1385 for Obama, +530 for Houston, and -11 for Strawn.
Johnson: -1280 for Obama, +2745 for Houston, and +2005 for Strawn.
Henderson: -1239 for Obama, +1076 for Houston, and +427 for Strawn.
Tyler: -1094 for Obama, +501 for Houston, and -260 for Strawn.
Van Zandt: -1075 for Obama, +656 for Houston, and +178 for Strawn.
Lamar: -993 for Obama, +2185 for Houston, and +1208 for Strawn.

Obviously, the worst 20 counties for Houston and Strawn were not identical to those for Obama, but I did not find any examples where Houston and Strawn combined to lose votes while Obama gained them.

The ten best counties for Houston and Strawn:

County Brister W'wright Change Van Os Houston Change Dem net ================================================================== FORT BEND 87,872 96,887 9,015 66,748 95,069 28,321 19,306 DENTON 132,244 138,359 6,115 56,112 86,738 30,626 24,511 COLLIN 165,017 167,840 2,823 64,159 100,302 36,143 33,320 HIDALGO 39,076 32,270 -6,806 60,122 87,197 27,075 33,881 EL PASO 62,780 50,627 -12,153 93,239 118,844 25,605 37,758 TRAVIS 142,841 127,796 -15,045 190,168 228,493 38,325 53,370 BEXAR 234,526 222,471 -12,055 212,415 260,152 47,737 59,792 TARRANT 327,136 320,585 -6,551 201,026 266,375 65,349 71,900 DALLAS 328,697 280,688 -48,009 324,165 406,857 82,692 130,701 HARRIS 555,454 523,101 -32,353 464,815 577,134 112,319 144,672 County Keasler Price Change Molina Strawn Change Dem net ================================================================== WILLIAMSON 77,666 80,967 3,301 42,377 61,373 18,996 15,695 DENTON 130,850 139,868 9,018 57,294 83,774 26,480 17,462 EL PASO 58,240 53,893 -4,347 99,152 115,154 16,002 20,349 HIDALGO 35,930 33,109 -2,821 64,087 86,441 22,354 25,175 COLLIN 164,805 169,377 4,572 64,188 96,476 32,288 27,716 TRAVIS 140,473 125,335 -15,138 190,769 228,492 37,723 52,861 TARRANT 321,497 322,531 1,034 206,841 263,585 56,744 55,710 BEXAR 224,983 215,807 -9,176 220,717 267,444 46,727 55,903 DALLAS 319,890 283,343 -36,547 329,484 402,483 72,999 109,546 HARRIS 540,632 521,753 -18,879 474,278 574,945 100,667 119,546

Williamson was Houston’s eleventh-best county, with a net gain of 18,502, while Fort Bend was Strawn’s eleventh-best county, with a net gain of 13,574. Not much variance on this end, in other words.

– Finally, I said in my previous entry that if 2012 is to 2008 as 2008 was to 2004, Texas would be a tossup state at the Presidential level. That’s true, but all else being equal, the Republican candidate would still win Texas by a bit more than 200,000 votes. That same level of improvement would be more than enough to win both of these judicial races, however. Sam Houston would win by more votes in 2012 than he lost by in 2008, while Strawn would win by about 150,000 votes. Given that even Republicans think the political landscape in Texas could be quite favorable to Democratic candidates, we may see as much interest in Supreme Court and CCA nominations as we saw in Harris County this year for district and county benches. All standard disclaimers apply, of course, but keep that in the back of your mind.

Next in the series will be a closer look at the 2002 and 2006 judicial elections, which will be done in two parts. As always, your feedback is appreciated.

Population and voting trends: 2004 and 2008 Presidential election

Taking a look at the voting trends in the fastest growing counties made me want to know more about this, so I broke out the spreadsheets and took a look. I’ll present the results in a three-part series, starting today with a comparison of the 2004 and 2008 Presidential election. Basically, I took the county by county canvass report for the two elections from the Secretary of State webpage, loaded them into a spreadsheet, and went to town on it. Here’s what I learned:

– At a macro level, there were 7,359,621 votes cast in the 2004 Presidential election in Texas, and 8,007,961 votes cast in 2008, for an increase of 648,340. Note that in all cases all I’m considering is the sum of the Republican and Democratic votes – third parties and write-ins are not counted. Bush/Cheney got 4,526,917 votes, while McCain/Palin got 4,479,328, for a decline of 47,589. Kerry/Edwards received 2,832,704 votes and Obama/Biden received 3,528,633, for an increase of 695,929.

– For each county, I compared the total number of votes cast for each party, and the difference between the Democratic and Republican totals. The spreadsheet is sorted by the difference in the Democratic performance from 2004 to 2008, so a negative number means that the Republicans did better in terms of vote total than Democrats did, while a positive number means that Democrats gained ground.

There were a total of 107 counties in which Democrats did worse in 2008 than in 2004. A total of 1,394,368 votes were cast in those counties. They broke down as follows:

– 60 counties in which Republicans gained votes from 2004 to 2008 and Democrats lost them, for a net of 633,754 total 2008 votes.

– 24 counties in which both parties gained votes but the GOP gained more, for 583,941 votes total.

– 21 counties in which both parties lost votes but Dems lost more, for 174,956 votes total.

– Comanche County, which had the same GOP total but 97 fewer Democratic votes. It was 3813 to 1431 in 2004, and 3813 to 1334 in 2008.

– And finally, Loving County, which had the same Dem total, but 2 more GOP votes. It was 65 to 12 in 2004, and 67 to 12 in 2008.

Some highlights from each group, starting with the first. Here are the six counties in which the Republican gains plus the Democratic losses were the greatest:

County Bush McCain Gain Kerry Obama Loss Dem Net =============================================================== Orange 20,292 21,509 1,217 11,476 7,646 -3,830 -5,047 Bowie 21,791 24,162 2,371 11,880 10,815 -1,065 -3,436 Hardin 15,030 16,603 1,573 5,608 3,939 -1,669 -3,242 Galveston 61,290 62,258 968 43,919 41,805 -2,114 -3,082 Cass 7,383 8,279 896 4,630 3,490 -1,140 -2,036 Jasper 8,347 9,022 675 4,471 3,658 -813 -1,488

These are not fast-growing counties. In fact, three of them – Orange, Cass, and Jasper – lost population this decade, according to the Census population estimates. Galveston County has actually grown by more than ten percent for the decade, with no reported drop in population in 2008 or 2009. Much of that growth is at the north end, in Republican territory like Friendswood and League City. And of course, we know what was going on, especially in the more Democratic-friendly south end of the county, in late 2008.

Next, the counties in which everyone lost ground:

County Bush McCain Loss Kerry Obama Loss Dem Net =============================================================== Polk 13,778 13,771 -47 6,964 6,230 -734 -687 Jefferson 44,423 42,905 -1,518 47,066 44,888 -2,178 -660 Milam 5,291 5,217 -74 3,445 3,044 -401 -327 Eastland 5,249 5,165 -84 1,582 1,271 -311 -227

You get into some mighty small counties after that. Jefferson County’s population has declined by about three percent over the decade, though it’s ticked up a bit since a big drop from 2005 to 2006. Milam and Eastland have basically stayed the same, but Polk County actually grew by more than ten percent. I have no idea why its turnout dropped as much as it did given that.

Finally, some of the growers:

County Bush McCain Gain Kerry Obama Gain Dem Net ================================================================== Montgomery 104,654 119,884 15,230 28,628 36,703 8,075 -7,155 Parker 31,795 36,974 5,179 8,966 10,502 1,536 -3,643 Johnson 34,818 36,685 1,867 12,325 12,912 587 -1,280 Chambers 8,618 9,988 1,370 2,953 3,188 235 -1,135 Erath 9,506 10,768 1,262 2,710 3,128 418 -844 Hood 16,280 17,299 1,019 4,865 5,087 222 -797 Angelina 18,932 19,569 637 9,302 9,379 77 -560 Comal 31,574 35,233 3,659 9,153 12,384 3,231 -428 Kaufman 21,304 23,735 2,431 8,947 11,161 2,214 -217

Montgomery and Kaufman, you know about. Comal probably just missed being on that fastest-growing list, as its population grew by about 50% between 2000 and 2009. Angelina and Erath grew modestly, less than ten percent each; Chambers grew by a bit less than 20%, mostly in the last two or three years; the others all grew by 25% or more.

How about the flip side? There were 23 counties in which both parties lost ground, but the Republicans lost more, so the Democrats had a net gain. Most of these were tiny, with the five largest as follows:

County Bush McCain Loss Kerry Obama Loss Dem Net =============================================================== Gray 7,260 6,924 -336 1,289 1,153 -136 200 Hutchinson 7,480 7,029 -451 2,663 2,545 -118 333 Bee 5,428 4,471 -957 4,045 3,645 -400 557 Jim Wells 5,817 4,841 -976 6,824 6,706 -118 858 Atascosa 7,635 5,462 -2,173 4,421 4,415 -6 2,167

Other than Atascosa, which actually grew by about 15% during the decade but apparently replaced a bunch of Republicans with even more non-voters, there not really much to be said about this group. There were 34 counties in which both parties received more votes, but the Democrats increased by more than the GOP. Those 34 counties accounted for 1,615,855 votes, or more than all 107 in which the Dems lost ground. Some highlights:

County Bush McCain Gain Kerry Obama Gain Dem Net ================================================================== Collin 174,435 184,897 10,462 68,935 109,047 40,112 29,650 Denton 140,891 149,935 9,044 59,346 91,160 31,814 22,770 Fort Bend 93,625 103,206 9,581 68,722 98,368 29,646 20,065 Williamson 83,284 88,323 5,039 43,117 67,691 24,574 19,535 Hays 27,021 29,638 2,617 20,110 28,431 8,321 5,704 Brazoria 63,662 67,515 3,853 28,904 36,480 7,576 3,723 Guadalupe 28,208 30,869 2,661 10,290 16,156 5,866 3,205 Smith 53,392 55,187 1,795 19,970 23,726 3,756 1,961 Bastrop 13,290 13,817 527 9,794 11,687 1,893 1,366 Kerr 16,538 16,752 214 4,557 5,570 1,013 799

There’s the rest of the fastest growers, plus a few others that are no slouches – Guadalupe, which abuts Comal, grew by 30%; Brazoria and Bastrop by 25%, Smith by more than 15%, and Kerr by more than 10%. Together, these ten counties by themselves shaved 108,878 votes off the Democrats’ deficit.

You may have noticed that some of the big counties have been absent in this discussion. Well, here the are now:

County Bush McCain Loss Kerry Obama Gain Dem Net =================================================================== Harris 584,723 571,883 -12,840 475,865 590,982 115,117 127,957 Dallas 346,246 310,000 -36,246 336,641 422,989 86,348 122,594 Bexar 260,698 246,275 -14,423 210,976 275,527 64,551 78,974 Tarrant 349,462 348,420 -1,042 207,286 274,880 67,594 68,636 Travis 147,885 139,981 -10,904 197,235 254,017 56,782 67,686 Hidalgo 50,931 39,668 -11,263 62,369 90,261 27,892 39,155 El Paso 73,261 61,783 -11,478 95,142 122,021 26,879 38,357 Cameron 34,801 26,671 -8,130 33,998 48,480 14,482 22,612 Bell 52,135 49,242 -2,893 27,165 40,413 13,248 16,141 Webb 17,753 13,119 -4,634 23,654 33,452 9,798 14,432 Lubbock 70,135 66,304 -3,831 22,472 30,486 8,014 11,845 Nueces 59,359 52,391 -6,968 44,439 47,912 3,473 10,441

Sometimes I think people don’t fully appreciate what happened in Harris County in 2008. Because the Democrats didn’t quite win all of the countywide races, some people consider the effort that year to have failed. All I can say is that I look at the numbers, I see the magnitude of the swing in four years, and I’m just amazed. Dallas is technically more amazing, since their swing was nearly the same size but was done with far fewer voters, but since they had their blue breakthrough in 2006, it too gets a bit lost in the shuffle. Bexar and Cameron, along with Harris and Dallas, flipped from red to blue, while Tarrant, Bell, and Nueces became officially purple. The only deep red county up there is Lubbock, and even it moved in the right direction.

I bring all of this up for two reasons. One is because even though I’ve covered some of this ground before, I feel like it needs to be repeated every now and again, as a reminder. Texas is a very different place than it was as recently as six years ago. That hasn’t shown up in the statewide elections yet, but the shift from one cycle to the next is unmistakable. And two, as a delayed response to Paul Burka, who recently wrote that “National Democrats have done a good job of spinning the myth that Democrats are resurgent in Texas. In fact, the D’s success has been limited to one area, the Texas House of Representatives.” I pointed out in the comments that this completely overlooked the gains that Democrats had made in county elections in places like Dallas and Harris, but it’s more than that. Democrats were in a huge hole after 2004, and it’s hard to overstate how far they came in just four years. If 2012 is to 2008 as 2008 was to 2004, Texas will be a tossup state. Obviously, a lot has to happen between now and then, but the point is that a lot has already happened. We shouldn’t lose sight of that.

Next up, a look at judicial races from 2004 to 2008, and a similar comparison for 2002 to 2006.

Where the votes are going

Matt Stiles looks at Census data and notes a political point.

Seven Texas counties — Rockwall, Williamson, Collin, Hays, Fort Bend, Montgomery and Denton — are listed among the nation’s 30 fastest-growing areas, according to U.S. Census Bureau estimates released [Tuesday].

They are also Republican-voting counties, according to results in the 2008 general election. Sen. John McCain won these counties by a 20-point margin, well over 240,000 votes.

It’s actually a hair shy of 260,000 votes – Stiles had missed Rockwall County in his initial post, and though he added it in for an update, he did not re-do the math. There’s a bit more to this than that, however. Let’s have a look at how these counties voted in 2004:

County Name Party Votes Total Pct ==================================================== Collin Bush R 174,435 243,370 71.67 Collin Kerry D 68,935 243,370 28.33 Denton Bush R 140,891 200,237 70.36 Denton Kerry D 59,346 200,237 29.64 Fort Bend Bush R 93,625 162,347 57.67 Fort Bend Kerry D 68,722 162,347 42.33 Hays Bush R 27,021 47,131 57.33 Hays Kerry D 20,110 47,131 42.67 Montgomery Bush R 104,654 133,282 78.29 Montgomery Kerry D 28,628 133,282 21.71 Rockwall Bush R 20,120 25,440 79.09 Rockwall Kerry D 5,320 25,440 20.91 Williamson Bush R 83,284 126,401 65.89 Williamson Kerry D 43,117 126,401 34.11 Total Bush R 644,030 938,208 68.64 Total Kerry D 294,178 938,208 31.36 Total McCain R 699,183 1,139,175 61.38 Total Obama D 439,892 1,139,175 38.62

Putting it another way, those counties had about 200,000 more voters in 2008 than in 2004. 145,000 of those new voters – 72.5% – voted Democratic, 55,000 voted Republican. That’s change I can believe in, but it shouldn’t come as a surprise. Obama did do about five and a half points better overall in Texas than John Kerry did, going from 38.22% to 43.68% of the absolute vote (38.49% to 44.06% in the two-party matchup). It would be strange indeed if he didn’t markedly improve on 2004 in these counties. Notice, however, that he improved by a point and a half more than he did in the state as a whole. That’s a good trend, too.

To which you may say, “Oh sure, compare a historic election for which Democrats were super-excited to one where a highly popular Texas Republican President was on the ballot. That’s fair.” Well, how about we compare the election of 2002 to the election of 2006? Since there are no Presidential candidates, I’m going to look at a couple of Supreme Court races, because 1) they’re usually more about party identification than anything else, and 2) we have a couple of races with similar R/D performances: Margaret Mirabal versus Steven Smith in 2002, and Bill Moody versus Don Willett in 2006. Here are the numbers:

County Name Party Votes Total Pct ===================================================== Collin Smith R 88,762 122,655 72.37 Collin Mirabal D 33,893 122,655 27.63 Denton Smith R 69,899 100,260 69.72 Denton Mirabal D 30,361 100,260 30.28 Fort Bend Smith R 47,008 84,153 55.86 Fort Bend Mirabal D 37,145 84,153 44.14 Hays Smith R 14,238 26,129 54.49 Hays Mirabal D 11,891 26,129 45.51 Montgomery Smith R 53,977 71,428 75.57 Montgomery Mirabal D 17,451 71,428 24.43 Rockwall Smith R 10,148 13,304 76.28 Rockwall Mirabal D 3,156 13,304 23.72 Williamson Smith R 46,480 71,981 64.57 Williamson Mirabal D 25,501 71,981 35.43 County Name Party Votes Total Pct ===================================================== Collin Willet R 82,834 125,348 66.08 Collin Moody D 42,514 125,348 33.92 Denton Willet R 63,475 99,380 63.87 Denton Moody D 35,905 99,380 36.13 Fort Bend Willet R 49,953 92,843 53.80 Fort Bend Moody D 42,890 92,843 46.20 Hays Willet R 13,644 27,775 49.12 Hays Moody D 14,131 27,775 50.88 Montgomery Willet R 54,018 74,650 72.36 Montgomery Moody D 20,632 74,650 27.64 Rockwall Willet R 10,331 14,233 72.58 Rockwall Moody D 3,902 14,233 27.42 Williamson Willet R 43,193 75,659 57.09 Williamson Moody D 31,466 75,659 42.91 2002 Total R 330,512 489,910 67.46 2002 Total D 159,398 489,910 32.54 2006 Total R 317,448 508,888 62.38 2006 Total D 191,440 508,888 37.62

Once again, improvement by the Democrats across the board. Dems picked up 32,000 voters, while the Rs lost 13,000. It’s not an exact apples to apples comparison because there was a Libertarian candidate in 2006, but even if you assign all of his votes (23,730 in these seven counties) to Willett, the Dems still have a 32,000 to 10,000 advantage in voters gained. All without any of that hopey-changey stuff.

If you want to see the effect in pictures, I’ve got you covered there as well:

Democratic vote share - click to enlarge

Democratic vote share - click to enlarge

GOP vote share - click to enlarge

GOP vote share - click to enlarge

The GOP vote share ticked up a bit from 2006 to 2008 in Montgomery, and to a lesser extent in Hays, but overall the trends are pretty clear. It’s especially clear when you simply compare 2002 to 2006, and 2004 to 2008.

Does any of this mean anything for 2010? Well, elections are all about the candidates, and every election is different, and blah blah blah. What I’ll say is simply that these counties start out with a higher floor for Democrats than they had eight years ago – I’ll be surprised if Bill White doesn’t carry Fort Bend and Hays, and he has a decent shot at Williamson, too – and I expect that this year there will be a lot more organizing done in them as well; in some cases, that may be the first time there’s been a real, funded, organizing effort. All things being equal, that should certainly have a positive effect. The whole point of this exercise was to show that while these counties are still challenging territory for Democrats, they’re a lot friendlier overall than they once were, and the prospect of them being the fastest growing areas in the state is not a daunting one for the Ds.

Challenging Chet

Via Eye on Williamson, I see the national GOP is once again looking to try to beat Rep. Chet Edwards in CD17.

There’s little question Republicans are looking to target Rep. Chet Edwards (D-Texas), who could face yet another tough re-election in his solidly conservative Waco-based seat. The question is who his opponent will be.

Both experienced and inexperienced Republicans are preparing their Federal Election Commission forms in Texas’ 17th district, encouraged by a strong showing by poorly funded 2008 nominee Rob Curnock.

Curnock held Edwards to 53 percent of the vote, despite receiving almost no support from the national party. Curnock, a small-business owner from Waco, plans to run again and hopes this time he’ll receive more support from national and local party leaders.

I think the key here is to compare Edwards’ 2004 performance with his 2008 performance, since I believe the non-Presidential year will be more favorable to him as it was in 2006. Here’s a Google spreadsheet that compares Edwards’ performance in each of CD17’s counties to John Kerry in 2004 and to Barack Obama in 2008. What I did in each was compare Edwards’ performance to that of the Democratic presidential candidate, and then compared the ratio from 2004 to that of 2008.

I think the story of these two elections is in the three biggest counties: Brazos, Johnson, and McClennan. In 2004, Edwards barely eked out a plurality in Brazos, got clobbered in Johnson, and won big in McClennan. In 2008, Edwards won a solid majority in Brazos, improved noticeably in Johnson, and won a smaller majority in McClennan.

His improvement in Brazos, I believe, can be largely attributed to an overall improvement in Democratic performance there. John McCain got almost exactly as many votes as George Bush did, while Barack Obama added over 4000 votes to John Kerry’s tally; meanwhile, Curnock did almost as well as Arlene Wohlgemuth while Edwards increased his total by 5000 votes. While there were probably a few Wohlgemuth voters who switched to Edwards in 2008, for the most part there were just a lot more people voting Democratic.

By contrast, Edwards’ improvement in Johnson is all him. McCain gained 1800 votes over Bush, and Obama added 600 to Kerry’s total, leaving their percentage almost identical to 2004, while Curnock lost 1500 votes and Edwards added 4200. Clearly, Curnock was a weaker candidate than Wohlgemuth, who was also from Johnson County and surely benefited from being a hometown girl, but Edwards did more than just take advantage of that difference.

Finally, McClennan presents an interesting case. Edwards won it by 23,000 votes in 2004, and was in net negative territory everywhere else. In 2008, he would have won even if all of McClennan’s votes were thrown out, but he only carried McClennan by 16,000 votes, and that was with Obama getting 37% to Kerry’s 33%. Here, Curnock’s residency in Waco likely helped him. Similarly, a local issue having to do with water rights that Edwards tied around Wohlgemuth’s neck back in 2004 was not on the table this time around. Unlike Johnson County, not being Arlene Wohlgemuth, especially not being Arlene Wohlgemuth in 2004, worked to the GOP’s advantage.

Based on all this, I’d venture that Edwards will likely do fine in 2010, barring any national headwinds against the Dems. If the NRCC dream candidate of State Sen. Steve Ogden jumps in, that would make for a hell of a race, but Ogden is up for re-election himself in 2010, so he’d have to give up his Senate seat and his powerful spot as chair of the Finance Committee to do that. I don’t know that a chance to maybe be in the House minority is worth that, but we’ll see.

Presidential results by Congressional district

Swing State Project has compiled a list of Presidential results by Congressional district, for all 435 DCs around the country. I’ve pulled out the Texas numbers and put them in a Google spreadsheet for ease of viewing. Here are a few notable ones:

CD Incumbent Obama Kerry Gore =================================== 03 Johnson 42 33 30 07 Culberson 41 36 31 10 McCaul 44 38 34 21 Smith 41 34 31 22 Olson 41 36 33 24 Marchant 44 35 32 26 Burgess 42 35 38 31 Carter 42 33 32 32 Sessions 46 40 36

The numbers represent the percentage of the vote the Democratic Presidential nominee got in that district in that year. I believe this is a two-party comparison, so Nader votes were excluded; in other years, the third-party Presidential vote is small enough to not matter much. “Incumbent” refers to the 2008 officeholder.

(By the way, my assumption is that the 2000 results are derived from taking the data from the existing precincts for that year, regardless of which actual CD they were in at that time. That must be the case, because CDs 31 and 32 didn’t exist in 2000.)

You can also now see similar figures from the Cook Political Report, which just released its updated PVIs to reflect the 2008 Presidential cycle. What does this mean?

The Cook Political Report Partisan Voter Index (PVI) Explained

In August of 1997, The Cook Political Report introduced the Partisan Voting Index (PVI) as a means of providing a more accurate picture of the competitiveness of each of the 435 congressional districts. Using the 1992 and 1996 major-party Presidential voting results, the PVI measured how each congressional district performed compared to the nation as a whole.

Using the results of the 2004 and 2008 elections, we have updated these PVI ratings and have even more information to draw upon to understand the congressional level trends and tilts that will help to define upcoming elections.

Developed for The Cook Political Report by Polidata, the index is an attempt to find an objective measurement of each congressional district that allows comparisons between states and districts, thereby making it relevant in both mid-term and presidential election years.

While other data such as the results of senatorial, gubernatorial, congressional and other local races can help fine tune the exact partisan tilt of a particular district, those kinds of results don’t allow a comparison of districts across state lines. Only Presidential results allow for total comparability.

A Partisan Voting Index score of D+2, for example, means that in the 2004 and 2008 presidential elections, that district performed an average of two points more Democratic than the nation did as a whole, while an R+4 means the district performed four points more Republican than the national average. If a district performed within half a point of the national average in either direction, we assign it a score of EVEN.

To determine the national average for these latest ratings, we have taken the average Democratic share of the two-party presidential vote for 2004 and 2008, which is roughly 51.3 percent, and that of Republicans, which is roughly 48.7 percent. So, if John Kerry captured 55 percent of the vote in a district and Barack Obama carried 57 percent in the district four years later, the district would have a PVI score of roughly D+5.

And here are the PVIs for the Texas districts:

(more…)