Positive Thinking, Minus the Positivity: Texas Data Points for the Week in Politics, September 18, 2020

Governor Abbott started the weekend on Thursday by announcing that based on a new criteria, much of the state would be able to relax (though not remove) restrictions on  some business and public activities. We also got to put a debate between Senator John Cornyn and challenger M.J. Hegar on our calendars this week (October 9), even as a report issued by a non-profit suggested that the pandemic resulted in a drop in new voter registration, at least in the spring. In other equally unsurprising voting data, very few people are voting in the Senate District 30 special election to replace Senator Pat Fallon. National Democrats' late summer confidence got a little shakier this week, especially as they pored over the ever-difficult-to-intrpret poll numbers among Latinos. And for much of the week, Donald Trump kept the show going, raising the usual questions. See below for data and discussion of some of the developments and low points of the week in politics.

1. Unhappy hour. Governor Abbott on Thursday introduced a new threshold for loosening coronavirus restrictions from 50% to 75% of capacity on a range of businesses and activities that include restaurants and retail outlets – but not bars, which means one of the loudest sources of complaint against Abbott will keep shouting (and Tweeting). In the June 2020 UT/Texas Politics Project, Texans pretty clearly felt safer in restaurants than bars, including Republicans, as the graphics below illustrate. Per the order (GA-30, if you’re keeping track), restrictions will be loosened for a wide range of businesses and activities in any of the state’s hospital regions where coronavirus patients make up less than 15% of all hospitalizations.This will partially open up all but three hospital regions - the Rio Grande Valley, Laredo, and Victoria areas, which underlines the ongoing recognition that class position and ethnic identity are multipliers of the severity of the impact of the pandemic. Hospitalization has gained weight in Abbott's decision-making scheme (thought it's been part of the discsussion prior to Thursday). This shift comes after a week in which one of the former metrics oft-discussed and used by Governor Abbott, the testing positivity rate, underwent a substantial revision that rewrote common understandings of the trajectory of the pandemic in the spring and summer. This was likely confusing to any casual observer trying to make sense of the data and whete the state stands, and is also likely to feed the motivated skepticism of those pressing for more rapid reopening, including the fringe deniers of the reality of the virus. To illustrate the difficulties for the news consuming public, the media coverage of the new metrics went in both directions. Shawn Mulcahy’s story in The Texas Tribune underlined how the new calculations produced much higher positivity rates in May, June, and July than those reported at the time using the now-replaced calculation. But in The Houston Chronicle, Cayla Harris and Jeremy Blackman led with the revised August positivity rates being lower with the new calculation. So it was worse than we thought when Texas was opening up, and it was better than we thought as we held the line in August - based on this week’s construction and interpretation of the data, anyway. We’re coming up on two weeks from the Labor Day weekend as the Governor opens up – based on the COVID-19 hospitalization share. The changes in the positivity measurement and reporting didn't get any attention at the press conference. All of which might make one think that we need to be a bit more specific than simply calling for basing decision making on data – some data, any data. But the bar for such considerations has, of course, been lowered in 2020 by the standard set by the White House. In the meantime, fire up Open Table – if you feel lucky.

Loading chart...

Loading chart...

Loading chart...

Loading chart...

2. Is there a door on that debate stage? Incumbent U.S. Senator John Corynyn and Democratic challenger M.J. Hegar announced this week they would debate on October 9th in Austin at the Bob Bullock Texas State History Museum – a funny location for two such decidedly un-Bullockian political leaders.The race is finally showing some signs of life after being a very sleepy affair for most of 2020. This situation has suited the Cornyn campaign just fine, as events have really not broken Hegar’s way. After fighting through a crowded primary field that required a run-off with State Senator Royce West, which was then delayed due to the pandemic, she now finds herself at substantial disadvantage in available campaign funds – she ended the run-off season with just over $900,000 in the cash on hand, compared to Cornyn’s bank balance of about $14.5 million. Going into Labor Day, she still struggled to broaden her name recognition: In the late August/early September Dallas Morning News/UT Tyler poll, about half of the respondents said they had neither a favorable or an unfavorable impression of her (21%) or didn’t know enough about her to say (28%), which was about 14 percent more people with a view of her than we saw in the UT/Texas Politics Poll in mid-June. She stands to benefit from the intense Democratic dislike of Donald Trump, Cornyn’s relative weakness among his own partisans, and the possibility of a rising Democratic tide that would lift all boats - though that tide is looking decidedly less high now than it did a couple of months ago. Cornyn has a history of comparatively mediocre poll numbers, but he also had a history of winning elections. Hegar trails Cornyn in public polling since June by a range of 7 to 13 percentage points, and she needs to generate some positive and prominent coverage in the debate against the more experienced Cornyn. It's a big night for Hegar to try to gain some attention, and for Cornyn to prevent her from getting either of them in th national news. If this race becomes more prominent in people's minds, it could also figure into the prospects for down ballot candidates, espcially Republicans. It seems unlikely, though, that Senator Cornyn's profile will suddenly develop enough gravity to pull voters away from the political black hole force of Donald Trump.

Loading chart...
ApproveDisapproveNeither/Don't Know
November 201527%34%38%
February 201627%32%41%
June 201624%35%41%
October 201628%36%37%
February 201730%34%36%
June 201728%41%30%
October 201728%42%30%
February 201829%38%33%
June 201827%38%34%
October 201839%34%28%
February 201936%35%29%
June 201937%34%29%
October 201935%34%31%
February 202036%39%25%
April 202038%36%26%
June 202036%40%24%
October 202039%39%22%
February 202132%42%26%
March 202133%42%25%
April 202131%43%25%
June 202134%41%24%
August 202128%44%28%
October 202129%44%27%
February 202231%35%34%
April 202232%39%29%
June 202224%50%26%
August 202229%42%29%
October 202232%42%27%
December 202235%40%25%
February 202330%43%27%
April 202333%38%29%
June 202333%39%28%
August 202328%42%30%
October 202330%39%30%
December 202328%42%29%
February 202434%39%28%
April 202436%38%28%

Loading chart...
Approve strongly2%22%62%
Approve somewhat3%14%24%
Neither approve nor disapprove2%12%5%
Disapprove somewhat6%16%3%
Disapprove strongly87%34%5%
Don't know0%2%1%

3. There's got to be an easier way. Alex Samuels had a story in The TExas Tribune earlier this week drawing on a report issued by the Center for Election Innovation and Research suggesting that the registration of new voters slowed during the spring of 2020 as a result of the pandemic. Texas registrations recovered to 2016 levels in June and July, though there’s not much time to recover lost ground (the last day to register to vote in Texas is Monday, October 5). It seems that online voter registration would make this a lot easier, especially during a pandemic. As of the June 2018 UT/Texas Tribune Poll, 57% of Texans agreed – including a plurality of Republicans. Maybe someone will poll soon to find out how much this had changed given the incessant messaging from GOP leaders at the state and national level about the osensible dangers of making it easier for people to vote, even if the narrow margins among GOP voters two years ago illustrates that Republican voters' dark suspicions about the process pre-date the pandemic scare tactics around voting by mail.

Loading chart...
Don't know/No opinion20%

Loading chart...
Don't know/No opinion15%33%22%

4. Meanwhile, in the Senate District 30 special election... Early voting got underway Monday in the election to replace current incumbent Pat Fallon, who is the presumptive replacement in John Ratcliffe’s seat in the solidly Republican 4th Congressional district. While the general public has no idea this is happening, the six-way context is drawing lots of attention from obsessives in the political trade. A senate seat is of course a significant prize in state politics, but I suspect there's also no small amount of schadenfreude at work here among those watching the dynamic between the two most well-known candidates in the race, State Representative Drew Springer and salon owner and self-styled pandemic refusenik Shelly Luther, with the comparatively late entry of former Denton Mayor Chris Watts another wrinkle in a race that was initally heavily handicapped to favor State Rep Springer. (Welcome to 2020, Rep. Springer!) For those watching the race, here is a table with the trickle of early votes as of Friday morning broken down by counties in the district. You can follow the updated tallies here at the early voting page at the Texas Secretary of State website. There was a preview of the race in the August 29 edition of Texas Data Points. Election Day is September 29, thanks to Gov. Abbott’s experiment in snap elections (which some might find suspiciously European).

Senate District 30 Special Election Early Votes 
(through Saturday September 19*)

County RV total (2020) 2018 total votes 2020 in-person to date 2020 by mail to date Total 2020 vote to date County share of
total vote to date
County share of 2018 SD30 vote 
Collin 625221 57632 1716 677 2393 12% 18%
Denton 544958 46336 1654 1559 3213 16% 15%
Parker 100416 53879 2116 14 2130 11% 17%
Grayson 84196 42608 2958 1087 4045 20% 13%
Wichita 82240 33563 1894 117 2011 10% 11%
Wise 44341 22718 1714 197 1911 9% 7%
Cooke 26566 14329 931 329 1260 6% 5%
Erath 23288 12407 497 384 881 4% 4%
Palo Pinto 18594 9340 303 11 314 2% 3%
Montague 13728 769 593 115 884 4% 2%
Young 11614 6329 457 0 457 2% 2%
Clay 7816 4240 205 119 324 2% 1%
Archer 6421 3584 237 15 252 1% 1%
Jack 5125 2767 178 32 210 1% 1%
TOTALS 1594524 317043 15629 4656 20285 100% 100%

Source: Texas Secretary of State
* Table updated Sunday September 20 with data through September 19, and county vote share calculations.

The number of votes in Denton County is kind of interesting, no? #spoileralert 

5. Democrats worry. I know, I know, you can't believe it. The general confidence among national Democrats a month ago seemed to flag a little among Democrats this week and last. A major theme of this week's stressing stemmed from criticism of Biden's lack of effective outreach among Latinos. The discussion had its origin in reports of less -han-stellar polling among Latinos in Florida last week, but was raising questions among reporters and observers in Texas about Biden's efforts here, too. These questions, in turn, activate the large, interesting, and difficult to resolve question about polling of Latinos in Texas and elsewhere writ large. In the recent Hispanic Public Policy Foundation Poll (conduced August 4 through August 13), which oversampled Latinos, there wasn't a favorability item broken down by race, but Biden led Trump among Hispanics 47.4% to 37.95.  In he June UT/Texas Politics Project Poll trial ballots, Trump with 39%, Biden with 46% among Latinos; the the April UT/Texas Tribune Poll, the same item delivered Trump with 40% and Biden with 50%. In September, the DMN/UT Tyler poll found Biden 53%, Trump 26% after their likely voter screen, and Biden 53%, Trump 26% in their overall sample. Biden's glass is about half full here, and outside the trail ballot, Biden's numbers among Latinos in our polling have been pretty unimpressive, though these have also moved around a lot. His favorability ratings in April 2020 were 41% favorable and 38% unfavorable; in June, they were 37% favorable, 47% unfavorable. There are likely some age and gender dynamics within these relatively small samples that contribute to the variation in these results, a thought suggested by Eric McDaniel's observations of male Latino voters' atttitudes around threats to White masculunity in a blog post on the Texas Politics Project site last week. As voters overall become more focused on the election, there will be more movement among Latino voters move as they become more focused on the election, and the less attentive and undecided land somewhere (including, for many, on not voting at all). The ratcheting up of tension in Texas triggered by Latino attitudes in Florida raises two reminders. The first should be obvious to anyone enaged in this subject by now, but the discussion in the last couple of weeks suggest it's not: The pan-ethnic cateigories of Latino/Hispanic/LatinX aggregate different ethnic groups and/or countries of origins, which are not homogenious in their attitudes and political dispositions. There are consequential differences, for example,  between the composition of Latinos as a group in Florida, with its sizable Cuban-origin and growing Puerto Rican population, and in Texas, where Mexican Americans predominate. Second, the Democratic party isn't guaranteed a share of these voters comparable to the ideal of their share among Black voters. Democratic opinion leaders and spokespeople will be quick to cluck and say OF COURSE they know that, but the assumption of entitlement and bad subconscious modeling often creeps into their expectations, public and private. There's a lot more work to be done here by objective researchers to get more polling focused on Latinos, and to keep the fundametal assumptions that should be operative here straignt among both practiioners and media. 

Loading chart...
Donald Trump59%13%39%
Joe Biden36%79%46%
Haven't thought about it enough to have an opinion5%8%14%

6. This week in Trump.  Take your pick. In no particular order, he participated in a town hall on ABC that demonstrated his unwillingness to listen to voters (points for consistency, anyway); held an indoor rally in Nevada chock full of Trump followers; berated the FBI on Twitter.  IOW, just another week – and I know I left at least two or three other instances of aberrant politics (and behavior) off the list. Trump will continue to pose strategic and tactical difficulties for Republican candidates in competitive races that very few will confess to in public as Republican candidates, consultants, and interest groups attempt to mobilize their base while also attempting towoo the thin layer of voters who remain persuadable, especially independents of a conservative bent. Looking at aggregated data in our archive of polling data, Trump's job approval among independents weakened noticably in 2020. The average of his approval ratings among independents went from an net positive average of 9 percentage points across three UT/Texas Tribune polls in 2017 to a net negative of 13 percentage points across three polls in 2020 (so far – keep an eye out).  (And, as always, these are "true indepdendents," excluding self-described "leaners" toward one party or another.) And by the way: for anyone looking retrospectively at Trump in Texas, he's tagged in 130 items in our data archive, and 148 entries in our blog archives.

Donald Trump Job Approval Among Texas Independents,
2017- June 2020

Year Approve Disapprove Net DK
2020 35% 48% -13 16%
2019 44% 42% +2 13%
2018 44% 42% +2 14%
2017 47% 38% +9 22%

Source: University of Texas/Texas Tribune, University of Texas/Texas Politics Project Polls, 2017-2020.

Loading chart...
PollApproveDisapproveNeither/Don't know
February 201739%37%24%
June 201747%41%11%
October 201755%35%10%
February 201849%37%13%
June 201843%45%12%
October 201839%43%17%
February 201946%37%17%
June 201946%39%15%
October 201941%51%8%
February 202036%47%17%
April 202034%47%18%
June 202036%50%14%
October 202031%53%15%

Loading chart...
PollApproveDisapproveNeither/Don't know
February 201750%40%10%
June 201745%48%7%
October 201746%48%6%
February 201849%44%7%
June 201850%44%7%
October 201850%46%4%
February 201950%45%5%
June 201951%43%5%
October 201948%47%5%
February 202044%51%6%
April 202049%46%7%
June 202048%49%4%
October 202051%46%3%


Subscribe to the Texas Politics Project Email List

* indicates required