SurveyUSA Mayoral runoff poll: Whitmire 42, SJL 35

We finally have a second pollster.

State Sen. John Whitmire maintains a 7 percentage point lead over U.S. Rep. Sheila Jackson Lee in the Houston mayoral race, with a fifth of likely voters having yet to make up their minds just weeks before the December runoff, the latest poll by SurveyUSA shows.

Researchers interviewed 805 likely Houston voters from Nov. 13 to 18 and found that 42% of the respondents supported Whitmire and 35% backed Jackson Lee. These figures are nearly identical to their vote shares in the general election, where Whitmire had 42.5% and Jackson Lee 35.6%. The two advanced to the Dec. 9 runoff because neither garnered a majority of votes.

Twenty-two percent of likely voters remain undecided, which is not unusual in city elections typically marked by low turnout, said Brandon Rottinghaus, a University of Houston political science professor and co-author of the poll.

While the undecided electorate represents an opportunity for Jackson Lee to potentially overtake Whitmire’s lead, convincing these voters to support her could be a “Herculean task” given the limited time frame, Rottinghaus said. In past Houston mayoral contests, the first-place finisher in the general election has won every runoff since 1977.

Among all likely voters, the ones most certain to cast a ballot in the runoff lean toward Whitmire, whereas those less sure about voting in December appear to favor Jackson Lee. For the congresswoman to secure a victory, her path hinges on quickly energizing these less committed voters, according to Rottinghaus.


In line with previous survey results, the latest poll reveals Houston’s voter bases are markedly divided along demographic lines.

Whitmire is projected to have a 43 percentage point advantage among white voters (63% to 20%) and a 20-point lead among Latino voters (43% to 23%). Jackson Lee, on the other hand, holds a strong lead among Black voters (63% to 15%).

More male likely voters said they would cast their ballot for Whitmire (51% to 34%), while women voters are evenly split between the two contestants (35% to 35%).

Though Houston city elections are nonpartisan, voters’ party affiliations continue to play a notable role in the latest poll: Whitmire garnered 68% of Republican support to Jackson Lee’s 12%, while Democrats backed Jackson Lee by 55% to Whitmire’s 25%.

The polling memo and data can be found here. It was a joint effort between Houston Public Media, the Chron, and UH, and as such HPM also has a story.

Roughly two-thirds of white voters favor Whitmire, while roughly two-thirds of Black voters support Jackson Lee. But University of Houston political scientist Brandon Rottinghaus noted that Black turnout in the first round of the mayoral election was down about 20% compared to eight years ago.

“For whatever reason, Black voters are not as enamored with Sheila Jackson Lee as they were with Sylvester Turner,” Rottinghaus said. “You would need to see tremendous turnout in the African American community in order for Sheila Jackson Lee to overcome the support that John Whitmire gets among older Anglos, who are much more likely to vote in municipal elections.”

UH political scientist Jeronimo Cortina said the Latino vote is likely to prove critical, and there too, Whitmire appears to have the edge.

“We’re talking about 43% versus 23%,” Cortina said. “And partially, this is because Senator Whitmire got very important endorsements during his campaign (among others, Congresswoman Sylvia Garcia and State Senator Carol Alvarado), and Jackson Lee, perhaps, did not get as many as those endorsements. However, she got the endorsement of (Harris) County Judge Lina Hidalgo, so that also is important.”

Whitmire also leads overwhelmingly among Republican and conservative voters. His endorsement by former Houston City Councilmember Jack Christie, who openly campaigned as a Republican in the first round, should help Whitmire pick up support among undecideds.

“Typically, those undecideds are distributed to the most Republican member of the candidates who are running. Now, that’s not perfect here, because John Whitmire is obviously a Democrat, but he seemed to be much more conservative than she is in this poll,” Rottinghaus said. “Sheila Jackson Lee is seen to be very liberal by many, and that’s something that can hurt her, especially among voters who are likely to turn out in a municipal race like this.”

Jackson Lee has the advantage among voters under 50, while Whitmire does better among those over 50. Jeronimo Cortina says that balance, too, works in Whitmire’s favor. “Older voters tend to be more reliable in getting out to vote,” Cortina said.

I said my piece about the runoff dynamics in the precinct analysis post. I think I’m more or less in alignment with the two professors here.

The thing that stood out to me about this poll was the large number of undecided respondents. I’m not sure how you can truly be a “likely” voter and not have your mind at least mostly made up. That thought, which made me question how good the sample used in this poll was, led me to take a trip through my archives to look at previous Mayoral runoff polls. Here’s what I found:

2003, Bob Stein/Richard Murray poll – Bill White 53, Orlando Sanchez 35. This post also references a SurveyUSA poll that had White leading 58-40, but I apparently didn’t have a separate post for that and the poll link is broken.

2009, KHOU/KUHF/Rice U – Annise Parker 37, Gene Locke 34.
2009, Chron/Zogby – Parker 42, Locke 36.
2009, KHOU/KUHF/Rice U – Parker 49, Locke 36.

2015, HRBC – Bill King 48, Sylvester Turner 43. Also notes an internal Turner poll that had him up 47-40.
2015, KHOU/KUHF/Rice U – Sylvester Turner 38, Bill King 38.

2019, KHOU/HPM – Sylvester Turner 56, Tony Buzbee 34.

With the exception of that first Parker/Locke poll from 2009, all of these were done in December, often right before Runoff Day, while early voting was taking place. This one as noted was conducted in mid-November, not that long after the initial election. (For years where there are multiple polls, they are in order of when they occurred, so the oldest ones are listed first.) Only the 2015 KHOU/KUHF poll, which did correctly predict the tightness of the race, had as many undecideds among them. I guess most years more people who really are going to vote have their minds made up later on in the calendar. Perhaps if there’s another poll next week we’ll see a result with more definitive answers.

One more thing from the HPM story:

The poll also examined the runoff for Houston city controller between former Harris County Clerk Chris Hollins, a Democrat, and former Harris County Treasurer Orlando Sanchez, a Republican. Jeronimo Cortina said Hollins is ahead, but Sanchez could still pull out a win.

“Chris Hollins has a little bit of a lead, 36% versus 25%,” Cortina said. “And here, the most interesting thing is that the percentage of undecided increases. It goes from 22% to 40% undecided, so there is a lot of room for either Hollins or Sanchez, to improve their numbers.”

I’d say that’s more than a little bit of a lead. The higher level of undecided voters for this race is unsurprising, given that far fewer people actually vote in Controller races versus Mayoral races. Six thousand people skipped the Mayor’s race in November, for an undervote rate of 2.36%. Forty-one thousand skipped the Controller’s race, for a 16.25% undervote rate. Whether or not you think there “should” be any undecided “likely” voters in the Mayor’s runoff, there definitely will be more of them for the Controller’s runoff.

Related Posts:

This entry was posted in Election 2023 and tagged , , , , , , , , , , , , , , , , , , , . Bookmark the permalink.

6 Responses to SurveyUSA Mayoral runoff poll: Whitmire 42, SJL 35

  1. Manny says:

    Buzbee’s internal poll showed him behind by about 10%. Turner won by 12 percentage points.

    My personal feeling is that this election is up for grabs. I feel that SJL has an edge, especially if she can do as Brown did;

  2. David Fagan says:

    805 sounds like a small sample size for a city the size of Houston. 1500 would give a smaller margin of error. With 805, the margin of error has got to be in at least 3% range, possibly the 5% range. Which would look like:
    Whitmire 39%-45%
    Lee: 32%-38%

    Which, if more voters vote for lee, therefore less for whitmire, the much closer statistics would apply. If the margin of error is 5%:

    Whitmire: 37%-47%
    Lee: 30%-40%

    This scenario gives a lot of overlapping numbers, and if statistics are going to be printed all over the news, I think the voters would get a better idea about what it is expressing rather than the numbers looking like one has all the possibilities and the other doesn’t have a chance.

    The argument that the audience doesn’t want to understand all that boils down to “what is the media for?”. it’s impossible to say the media is not there to ededicate, because with the publication of numbers like these, that’s exactly what they are doing, educating the public. But, the representation of the information needs to be more accurate. Kuff said it himself, he’s going on what he’s been given, do does everyone else, so a more accurate representation needs to be expressed by the media, the mediad themselves expresses their responsibility to bring information, bring more accurate information.

  3. In the Mayor’s race, I haven’t seen Whitmire run a single negative commercial attacking any of his opponents. From day one, all of Whitmire’s ads have been positive, with him talking about his experience and what he intends to do to help ALL Houstonians. In this day and age, can a politician really take the high road and still win? My brain and guts say no, but my heart sure hopes so.

  4. Manny says:


    Whitmire can’t afford to piss anyone off; if he thought negative would work to his benefit, he would be doing that. The most powerful voting bloc in Houston is the Black vote. Does he want to piss them off? Republicans, he may only win with Republican votes. FYI, the recording was not negative? That was part of the Whitmire campaign designed to hurt at the best time. They had that planned for months. The recording is 14 years old, so it just happened to appear this year.

    Latinos are not returning to vote, and those who do will probably split their vote similarly.

    It would not surprise me to find that Michele or Obama make an appearance to help her or someone of almost similar importance.

    We just had President Clinton endorse SJL.

    Think Gregg Abbott or Dan Patrick, or maybe the orange man will come and give Whitmire a boost?

  5. Manny, the Whitmire campaign denied any involvement in the recording’s release, but I’m sure it was leaked by people who prefer Whitmire over SJL. That isn’t Whitmire’s fault. Frankly, if that recording is anyone’s fault, it’s SJL for cussing out and demeaning her own staff. Remember, this wasn’t an isolated incident. SJL has once voted the meanest member in the House of Representatives and, over a long period, she had the highest staff turnover rate in Congress (see links below).

    I think Whitmire has not run negative attack ads against SJL for several reasons: (1) Respect – despite her faults, SJL has been a champion on many Democratic issues, (2) Whitmire is an old-school politician who still values a certain amount of civility in politics, especially when competing against a fellow Democrat, (3) Whitmire knows that, if elected, he has to be able to bring all Houstonians together (including people who supported SJL), and (4) Whitmire believes by staying positive and focusing on the needs of Houstonians, he will be rewarded by the majority of voters. We shall see.

  6. Manny says:

    I believe in the tooth fairy, forgot that one, GS.

Comments are closed.