Off the Kuff Rotating Header Image

PPP on the Texas Senate race

Public Policy Polling follows up its snapshot of the Republican GOP matchup for Governor with a poll of Senate possibilities.

We tested Lieutenant Governor David Dewhurst, Attorney General Greg Abbott, and state Senator Florence Shapiro on the Republican side against Houston Mayor Bill White and former Comptroller John Sharp on the Democratic side.

Abbott, who has not announced plans to seek the seat, appears to be the strongest initial candidate. 43% of voters in the state have a favorable opinion of him compared to only 25% that view him negatively. He leads Sharp 44-36 and White 42-36 in possible contests.

Dewhurst is almost as strong, with a 43/30 favorability breakdown. He leads Sharp and White by slightly more narrow margins than Abbott, 42-36 over the former and 42-37 over the latter.

It seems inevitable that one of those heavyweights will get into the race if there is indeed a vacancy, but we also tested Shapiro to see how competitive the contest would be if the GOP ended up nominating one of the less well known candidates who have already made their intentions to seek the seat known. Shapiro leads White 37-36 but trails Sharp 37-34, an indication the race could pretty much be a tossup if a more well known Republican doesn’t run.

It appears that Dewhurst or Abbott would be an early favorite, but it’s worth noting that Sharp and White have a lot more room to grow in terms of name recognition. 43% of the electorate has no opinion of White and 41% has none of Sharp, figures much higher than the 27% for Dewhurst and 32% for Abbott. That gives them an opportunity to define themselves positively with the voters who haven’t formed an opinion about them yet.

As Matt notes, the first problem with this is that it ignores most of the announced Republican candidates for this race. There’s no reason to believe Dewhurst or Abbott will jump in to this race, though there’s no reason to think they won’t, either. All we know now is that they haven’t given any indication. What I take from this is that the “Known R versus Unknown D” races show about a six or seven point lead for the R, while the “Unknown R (Sen. Shapiro is an announced candidate) versus Unknown D” races are basically tossups. Looking at the full report (PDF), the partisan breakdown is given as 44R/38D/18I, so in some sense this is just a recapitulation of party ID. And it does suggest that in a race between equally-funded candidates, Texas is indeed a competitive state.

Of course, that leads to the other problem with this, which as the first commenter on the PPP post noted is that we won’t have a two-person race, we’ll have a cast-of-thousands special election. That’s assuming KBH does resign at some point, whether before or after the November of 2010 election. There is still the possibility that she could lose the GOP primary, or remain in the Senate and then lose the general election for Governor. In which case, this election won’t be till 2012, and any poll taken now is even less useful than usual. So while this data point has some usefulness, I wouldn’t consider it predictive in any meaningful sense. It’s fun, and that’s fine as it is.

Related Posts:

  • No Related Posts


  1. cb says:

    I can not put any stock in such a poll. I am not a Bill White fan, but how can he have such high negatives when the vast majority of Texans outside of the Houston area probably know little about him and his politics. The same can be said with all the candidates. Abbott scores a little higher possibly because his name itself may generate a more positive response. The polls that will have meaning are those taken after we know who are truly in the race and after the public has had time to hear some of their campaign message.

  2. trowaman says:

    The reason for the higher negatives/favorables and lower unknowns is all candidates were as “your opinion on (party affiliation) (candidate).” There’s not really name ID occurring there, just partisan reflexes to party ID.

    I made some observations on this at Swingstateproject: