From the inbox, the promised Day Two results:
Statewide poll numbers released today by the Texas Lyceum, the state’s premier, non-partisan, nonprofit statewide leadership group, show U.S. Republican Senator Ted Cruz of Houston (Lyceum Class of 2004) isn’t guaranteed another term as Texas’ Senator according to early trial ballots pitting the incumbent against his two likely Democratic challengers: U.S. Congressmen Beto O’Rourke of El Paso and Joaquin Castro of San Antonio.
Senator Cruz is tied with Congressman O’Rourke, who entered the contest last month, at 30 percent each. However, 37 percent of registered Texas voters say they haven’t thought about the race yet. Congressman Castro fairs slightly better against the incumbent Senator, with 35 percent of Texas adults saying they support him over Ted Cruz at 31 percent.
“Ballot tests conducted this far in advance of an actual election are, at best, useful in gauging the potential weaknesses of incumbents seeking re-election,” said Daron Shaw. “But the substantial percentage of undecided respondents—coupled with the conservative, pro-Republican proclivities of the Texas electorate in recent years—suggest a cautious interpretation.”
Patrick vs. Collier
Meantime, Lieutenant Governor Dan Patrick’s Democratic challenger, Houston area accountant Mike Collier, comes within the margin of error if that 2018 race were held today. 27 percent chose the little-known Collier compared to 25 percent who chose Lieutenant Governor Patrick. But again “not thought about it” outpaces both candidates at 46 percent in that race – which is also 18 months away.
Right Track/ Wrong Track
Compared to last year, fewer Texans believe the country is on the wrong track at 52 percent compared to to 63 percent in 2016. However, party and race drive much of the results, with 84 percent of Democrats saying the country is on the wrong track, and 73 percent of Republicans expressing that things are moving in the right direction.
President Trump’s job approval numbers line up by party
More Texans disapprove than approve of the job Donald Trump is doing as President (54 percent to 42 percent), but the results vary significantly by party. 85 percent of Republicans give the President positive marks compared to 86 percent of Democrats who disapprove of his job performance. Same goes for young Texans – 73 percent of 18-29 year olds are not enthused with the President’s job performance along with 61 percent of Hispanics. Meantime, he is viewed positively by 60 percent of Whites.
The press release for Day Two, from which I am quoting above, is here, and the Day Two Executive Summary is here. My post on the Day One poll is here, and the Lyceum poll page for 2017 is here. As you might imagine, I have a few thoughts about this.
1. For comparison purposes, the UT/Trib poll from February had Trump’s approval ratings at 46/44, which is to say slightly more approval but considerably less disapproval than the Lyceum result, with both polls showing a strong split between Dems and Republicans. What explains the divergence of the results, given the similar partisan dynamic? Two likely reasons: First, the Trib poll is of registered voters, while the Lyceum surveys adults, of whom 11% are not registered. It’s probable that the broader the sample, the less Republican-leaning it is. We don’t know what the partisan mix is of the Lyceum poll so this is just a guess, but it is consistent with the numbers. Two, the Trib result showed that independents were basically evenly split on Trump, at least in February. The Lyceum poll doesn’t say how indies felt about Trump, but if it is the case that they were sufficiently against him, that would have tilted the numbers into negative territory. Again I’m just guessing, but either or both of these things being true could explain the difference.
2. I’m not sure what the “cautious interpretation” of the very early horse race numbers Daron Shaw has in mind is, but my cautious interpretation is that these numbers kind of stink for Ted Cruz and Dan Patrick. Not because of what the Democrats got, though I’ll speak to those figures in a minute, but because there was so little support expressed for Cruz and Patrick. A key feature of many super early polls is that a lot of people haven’t given the matter any thought, and of those who have many don’t yet have an opinion or don’t feel strongly enough about it to express an opinion. With challengers, there’s often a name recognition factor as well, so the generally low number that a newbie will get reflects little more than some raw partisan preference. But here we are talking about two incumbents who are the highest-profile politicians in the state. For Cruz to top out at 31 percent and Patrick at 25 percent, with both trailing lesser-known opponents, suggests that there’s not a whole lot of love for these guys. It’s hardly a time for panic, but I’d be at least a little bit concerned about such limp numbers if I were them.
3. By the same token, even a 35% support level for Joaquin Castro at this point in time, and even before he’s a candidate (if indeed he becomes one), is not too shabby. Remember, most people haven’t given this any thought or don’t have a strong opinion if they have one, yet Castro is already almost at the level of support that actual 2014 statewide Democrats received that year. That suggests at least the possibility of a higher than usual level of engagement and interest. For another point of comparison, the November 2013 UT/Trib poll for the Governor’s race had Greg Abbott leading Wendy Davis 40-35; this was not long after the summer of the Davis filibuster and the the HB2 special sessions, when enthusiasm for Davis was about as high as it ever was to get, as well as being seven months farther along in the calendar. It’s one result and I don’t want to over-interpret, but given all the other evidence we have about Democratic levels of engagement this year, it feels like we’re starting out in a different place. Beto O’Rourke’s thirty percent against Cruz is closer to what I’d consider the normal default level for Dems in a very early poll, but in this case the difference between himself and Catro may just be a reflection of a higher level of name recognition for Castro.
4. Again, it is important to remember this is a poll of adults, eleven percent of whom in this sample are not registered to vote. I don’t know how the numbers break down by registered/not registered, but the point here is that it is likely a significant number of the people in this poll will not participate in the 2018 election, and as such their opinions just don’t matter. That said, a huge piece of the puzzle for Democrats, especially next year, will be to get lower propensity voters to the polls, as we saw happen in the recent Congressional special elections in Kansas and Georgia. This one poll doesn’t tell us much, but future polls may paint a picture of how or if that is happening for Democrats, and for Republicans too – if they are less engaged, then they will have trouble.
5. Which brings me back to the Presidential approval numbers, as they are likely to be the best proxy we will have for voter enthusiasm going forward. As noted before, Democrats and Republicans have roughly similar levels of disapproval and approval of Donald Trump, which means that any change in the overall level of approval for Trump will come from either independents turning against him and/or Republicans abandoning him. This poll suggests the possibility of #1 happening, but as yet we have not seen evidence of #2. If we ever do, that’s going to be a big deal, and potentially a big problem for the Republicans. RG Ratcliffe, TPM, and the Trib have more.
Wow, all I got to say is that “Texas is in Play”.
“I’m not sure what the “cautious interpretation” of the very early horse race numbers Daron Shaw has in mind is, ”
it means: he think the numbers are meaningless, but he can’t say that because he makes a living by publishing them.
Trump will be the deciding factor in the race.