Days away from the third Democratic debate in Houston and over a year from the 2020 presidential election, an exclusive poll by Univision News found that 40% of registered voters in the state say they will vote for the Democrat who prevails in the party’s primary, while 33% say they will support president Trump. If undecided voters leaning one way or the other are included, the advantage would be 47% to 42% in favor of the Democrats.
The Latino vote could be decisive. A large majority of Hispanics (69%) surveyed in the state said they intend to vote for the Democratic candidate, compared with 19% who plan to support Trump (also including undecided voters who are leaning one way or another).
According to Census data, Hispanics represented 40% of the population of Texas last year and during the 2018 elections Hispanics constituted more than 24% of registered voters. The Univision national poll, conducted by Latino Decisions and North Star Opinion, measured the presidential preferences of Hispanics in this election cycle and included a specific module for Texas with the support of the Center for Mexican-American Studies at the University of Houston in which included a sample of all state voters.
Senator Bernie Sanders would win the state (48% against 42%). Similarly, former Vice President Joe Biden would win the state (47% to 43%), and the same would happen with Senator Warren (44% to 42%).
According to the national poll, President Trump’s management style continues to generate a strong rejection among Hispanics, with four out of five registered voters disapproving of the president’s performance. At the same time, according to the survey, Democrats have gained support from Latinos who were undecided.
In this latest poll, which comes on the eve of Thursday’s debate in Houston in which all the leading Democratic candidates will share the same stage for the first time, 62% of Latinos reported that they will support any Democrat who opposes the president.
This is the most favorable poll for Dems versus Trump in Texas so far – compare to the Quinnipiac June poll, the UT/Trib June poll, the UT-Tyler July poll, and the August Emerson poll. In part, this is because Trump’s approval numbers are terrible – 44% approve of the job he’s doing, 56% disapprove. There does seem to be a correlation between the two in the polls we’ve seen so far, which makes sense but may not be conclusive. Trump did outperform his approval numbers in 2016, but he wasn’t President then, he didn’t have a record to defend, and he had the good luck to run against someone whose own approval numbers were lousy. At least the first two of those will not be true this time.
The poll has three results for each question – there’s a national Latino result (1,043 Latino RVs), a Texas Latino result (641 Texas Latino RVs), and an overall Texas result (1,004 Texas RVs). You can see how the questions break out, in many different combinations of candidate and query, and you can see it in tabular form with the wording of the questions here. To save you some clicking and scrolling, here are the headline numbers:
Matchup LatinTX All TX =========================== Trump 19% 42% Democrat 69% 47% Trump 19% 43% Biden 60% 47% Trump 19% 42% Sanders 68% 48% Trump 20% 42% Warren 64% 44% Trump 19% 44% Harris 60% 45% Trump 20% 41% Castro 62% 44% Trump 21% 41% Booker 63% 43% Cornyn 22% 41% Democrat 58% 40%
All numbers are for Texas, with the first number being from the Latino subsample and the second number being overall. Not a whole lot of difference, and where there are differences it’s usually in the Undecideds. Note they also threw in a Senate question, though just a generic one. It probably wouldn’t make much difference if they asked about individual candidates, as the polls we’ve seen so far, one from August and one from this week suggest the Dem candidates aren’t sufficiently well known for there to be much difference between them. On that note, here’s the more recent poll of the Texas Senate primary:
Mary Jennings “MJ” Hegar, of Round Rock, an Afghanistan War veteran who nearly unseated a veteran GOP congressman last year, came out on top – barely.
Hegar polled at 12 percent, followed by three candidates with 10 percent: State Sen. Royce West; activist Cristina Tzintzún Ramirez; and Houston City Councilwoman Amanda Edwards. Chris Bell, a former congressman from Houston, had 9 percent in the poll.
The survey of 600 likely Democratic primary voters was conducted from Sept. 3-5 by Ragnar Research Partners of Austin and Washington. The margin of error is 3.9 percent.
I wouldn’t worry too much about Dem Senate primary polls for now. In the Univision poll, John Cornyn does about as well as Trump, with the slight differences on both sides again coming from an increase in the undecideds. Cornyn did basically as well as John McCain in the 2008 election, the main difference in the two races being the larger share of the vote going to the third-party candidate. His 12-point margin over Rick Noriega was identical to McCain’s 12-point margin over Barack Obama. This suggests that there won’t be that much difference between Cornyn and Trump when the 2020 votes are counted. That in turn suggests to me that the not-Trump voters from 2016 who voted more or less Republican otherwise but went much more Democratic in 2018 will likely repeat their more recent behavior in 2020. I don’t want to go too far out on the limb for this – polling data is still preliminary and scarce – but it’s something to keep an eye on. It’s long been my belief that the not-Trump voters will stay with the Dems in 2020, and after that who knows, and if so that makes the path for Dems easier to navigate. But as they say, there’s still a lot of time. Now we wait for the next poll.