Here are some new poll numbers for Texas from UT and the Tribune.
Republican Mitt Romney has a commanding lead over Democrat Barack Obama in the presidential race in Texas, according to the latest University of Texas/Texas Tribune Poll. The survey of likely voters found that 55 percent support Romney while 39 percent support the incumbent. The remaining 6 percent said they support someone else.
The survey results illustrate the continuing dominance of the GOP in Texas — Republican John McCain got 55.5 percent of the Texas vote in 2008, to Obama’s 43.7 percent — and illuminate a significant gap in Texans’ feelings about national and state officeholders and government.
“At the top of the ticket, in the big marquee races, there are no surprises,” said Jim Henson, who teaches government at the University of Texas at Austin, heads the Texas Politics Project there and co-directs the poll. “We see the basic structure of the state, in terms of partisanship, pretty stable.”
Numbers in the U.S. Senate race were similar to those in the top contest, with Republican Ted Cruz holding 54 percent of the support to Democrat Paul Sadler’s 39 percent, according to the poll. John Jay Myers, the Libertarian candidate, had 3 percent, and Green Party candidate David Collins had 2 percent.
[…]
Republican Christi Craddick held the lead in the contested race for Texas Railroad Commission, with 50 percent of the support to Democrat Dale Henry’s 36 percent.
“What you have in these results is a pretty decent idea of what a Democrat with warm blood and a pulse can get in Texas,” said Daron Shaw, co-director of the poll and a professor of political science at UT-Austin.
Actually, I suspect that what these results give is about five to eight points below what a Democrat with warm blood and a pulse can get in a non-2010 year, but as they say, the only poll that matters is taking place right now. The poll’s summary is here and the description of its methodology is here. I don’t think the connection to YouGov had clicked with me until I looked at that. This result is similar to but not the same as the recent YouGov result we saw. For comparison, here’s the May UT/TT poll that had it at Romney 46 – Obama 38 among registered voters but 55-35 when their ridiculously restrictive “likely voter” screen was applied, and their February poll back when the nominee was not yet decided. Of somewhat peculiar interest is that the February poll asked respondents how they voted in 2008, and the result was McCain 46 – Obama 39. The May poll had no such question, but this one did, and the result there was McCain 43 – Obama 40. That would translate to a 52-48 McCain win if you filter out the “other” and “didn’t vote” respondents – and by the way, if you simply used these people as your “likely voter” screen for this poll, it would be a sample size of about 665, considerably larger than the 540 actually used. That suggests two possibilities to me: One, Obama has lost a number of supporters from 2008. The Wilson Perkins poll suggests that possibility as well. And two, the “likely voter” screen they used screened out a disproportionate number of Obama supporters. The overall sample is 65% white, which is perfectly reasonable, but we don’t know what the screened sample looks like. If it’s anything like that crappy Lyceum poll, with it’s 5% African-American share, you can see how things might get wacky. For what it’s worth, Greg says that Harris County’s early vote pattern is suggestive of 2008 so far. You can make up your own mind. As I said, we’ll have a fact check on this soon enough.