A curious thing about the polls of Texas

You’ve probably seen this headline by now.

Colin Allred

For the first time in their hotly contested campaign, a poll shows U.S. Sen. Ted Cruz, R-Texas, trailing challenger U.S. Rep. Colin Allred.

The Morning Consult poll, conducted between Sept. 9 and 18, places Allred — a moderate North Texas Democrat — a point ahead of Cruz with likely voters, 45% to 44%. Allred’s lead is within the poll’s margin of error, however.

Cruz is running for a third term in the Senate after narrowly defeating former El Paso Congressman Beto O’Rourke in a closely watched 2018 race.

However, this time around, he’s been targeted by Democratic groups, who consider his seat one of the most flippable in the Senate. He’s also running with additional baggage, including taking a family trip to Cancun while hundreds of Texans died in Winter Storm Uri, one of state’s worst natural disasters.

While Cruz has racked up consistent leads in recent polls, a recent roundup of survey numbers by The Hill shows him having trouble crossing the 50% margin — generally viewed as troubling news for an incumbent weeks before an election.

“For the first time in this race, a new poll has us leading Ted Cruz by 1 point. I don’t know about y’all but I’m fired up and ready to WIN!” Allred tweeted in response to the new numbers. “We’ve got 47 days, let’s do this Texas.”

It is of course great to get even one statewide poll of Texas result with the Democratic candidate leading. We got a couple in 2018 with Beto and a few more in 2020 with Biden; the end result didn’t match the polls, but at least it showed it was possible. I always look for more context when I see a poll result, and so I will provide that here for this one:

It would have been much nicer for this poll to also show Kamala Harris leading in Texas, but alas. We get what we get and we don’t get upset. My point in bringing this all up is that I’m seeing a trend in which polls of Texas show a Senate race that is at least as competitive as the Presidential race.

More specifically, these polls have been showing a Senate race in which Colin Allred does as well as Kamala Harris, but Ted Cruz lags behind Donald Trump. I should note that this was often (not always, but often) true when it was Biden versus Trump as well. Looking back at a few results from the 538 Presidential and Senate trackers:


Date     Pollster     Biden  Trump   Allred  Cruz
=================================================
June 9   UT/TPP          39     46       34    45
June 20  UT-Tyler        43     48       42    45
June 27  Manhattan Ins   41     52       43    46
July 1   UH/TSU          40     49       44    47


Date     Pollster    Harris  Trump   Allred  Cruz
=================================================
Aug 16   UH/TSU          45     50       45    47
Aug 22   PPP             44     49       45    47
Aug 29   Lake            43     51       43    47
Aug 31   UT/TPP          44     49       36    44
Sep 5    Emerson         48     51       44    48
Sep 18   M Consult       46     50       45    44

In all cases, when there were multiple results, I went with Likely Voter models and a two-candidate matchup, so no “RFK Jr” crap polluting the earlier results. In the four polls with Biden, Allred led Biden in two, trailed by one point in a third, and by five in the other. Cruz trailed Trump in all four, by one to six points.

With Harris, Allred led Kamala Harris in one poll, tied her in two, and trailed in three. Cruz again trailed Trump in all six polls. Allred’s deficit versus Cruz was smaller than Biden’s versus Trump in three of four polls, and smaller than Harris’ deficit versus Trump in four of the five polls where Allred trailed.

What is curious about this, at least to me, is that Trump was easily the worst performing Republican in Texas in each of his two prior campaigns. He won by 8.99 points in 2016, 52.23 to 43.24, while every other Republican statewide won by margins of at least 13 to 16. Wayne Christian (53.07) was the next lowest-scoring Republican, while Dori Garza (41.18) was the next highest-scoring Dem. In 2020, he won by 5.58 points, 52.06 to 46.48, while every other Republican was winning by eight to ten points; Nathan Hecht (52.98) and Tina Clinton (45.18) were the runnersup.

What if anything does that say about this year? I think third parties will have a smaller effect on the Presidential race than in the Senate race – this was also true in 2020, when third parties took about a point and a half for President and a bit more than two and a half for Senate – and since the bulk of that goes to the Libertarians, that could keep Cruz below the Trump line. Undervoting is a relatively small effect in the Senate race since it’s next on the ballot after President – there were 160K fewer votes for Senate in 2020 than President, out of over 11 million ballots cast. John Cornyn actually had about 70K more votes in 2020 than Trump, while MJ Hegar had about 370K fewer than Biden,

Which brings up perhaps the biggest potential factor. I tend to believe that variances in polls like these are mostly due to people who are not yet willing to say they’re voting for the candidate they will eventually vote for. But there are crossovers, as noted by the Cornyn total from 2020. It’s easy enough to imagine a Biden/Cornyn voter in 2020. The question now is, which do you think is more likely, the Harris/Cruz voter, or the Trump/Allred voter? History suggests the former, but I don’t know how sure I am about that.

That’s all based on gut instinct and vibes, not on anything empirical. I can imagine both the “Trump does better than Cruz” and “Cruz does better than Trump” scenarios. The former is almost certainly the path to an Allred victory, however probable you deem that. I’m going to need to see at least one poll with Harris leading to fantasize about that possibility. As I’ve said before, what I mostly want to see is Harris outperforming Biden, as that would not only make the “Allred wins” case more probable, it would also surely boost downballot Dems, specifically in the legislative races, where a couple of victories could mean an insurmountable obstacle for vouchers. I would love to dream bigger than that, but let’s start there and see where it takes us.

Oh, and if you’ve been pining for an Allred-Cruz debate, your wish has been granted. Put October 15 on your calendar and start stocking up on the popcorn.

Related Posts:

This entry was posted in Election 2024, The making of the President and tagged , , , , , , , , , , , , . Bookmark the permalink.

2 Responses to A curious thing about the polls of Texas

  1. Pingback: Some Bexar County polling | Off the Kuff

  2. Pingback: Get ready for a lot more Allred and Cruz ads on TV | Off the Kuff

Comments are closed.