Off the Kuff Rotating Header Image

polls

Two from PPP (RV): Cruz 48, O’Rourke 45, and Cruz 49, O’Rourke 46

Fourth in a series from PPP.

Rep. Beto O’Rourke

A new Public Policy Polling (PPP) survey commissioned by Protect Our Care finds that 62 percent of voters in Texas say health care will be one of the most important issues they consider when casting their vote in November. What’s more, 44 percent are deeply concerned about Senator Ted Cruz’s work to repeal health care and nearly 60 percent oppose the Trump Administration’s lawsuit to end protections for those with pre-existing conditions, which Cruz has refused to oppose.

The poll shows that for Cruz, who has been among the GOP’s fiercest advocates for repealing the American health care law, the issue is a drag on his prospects for reelection. In the poll, Cruz is in a dead heat against Democrat Beto O’Rourke 48 (Cruz) to 45 (O’Rourke). The poll was conducted September 19th and 20th among 613 Texas registered voters. The survey has a margin of sampling error of +/- 4 percent.

“Ted Cruz thought he was going to score political points shutting the government down trying to repeal health care, but what he actually did was put his own reelection prospects in serious jeopardy. Ted Cruz’s constituents say health care is one of the most important issues to them this election, and as a result he’s taking on some pretty serious water in this race. Whether it is Cruz’s opposition to protections for people with pre-existing conditions or his vote for an age tax, Ted Cruz’s extreme health care views are rejected by his constituents.”

The full polling memo is here. That hit my inbox on Thursday. On Friday, I got this:

A new poll of likely voters in Texas, commissioned by End Citizens United (ECU) and conducted by PPP, shows that Beto O’Rourke continues to close in on Senator Ted Cruz. ECU’s latest poll shows O’Rourke behind Cruz by three points, 46-49 percent. Click here to see the full polling.

When voters learned of O’Rourke’s decision to reject all PAC money and Cruz’s reliance on special interests, O’Rourke takes the lead 48-46 percent.

This is the fourth poll ECU has conducted in the race with each poll showing O’Rourke increasingly closing the gap. In January, O’Rourke was within eight points of Cruz, 37-45 percent. By June, O’Rourke had moved to within six points, 43-49 percent. In July, he closed the gap to four points, 42-46 percent. The latest poll conducted this week has him within the margin of error.

“Beto and his message of reform continues to win over Texans,” said ECU President Tiffany Muller. “He’s spent the last 18 months visiting every county in the state, listening to voters, and inspiring people to get involved. With just a few weeks until Election Day, Beto has pulled even with Ted Cruz and is in position to win this race.”

PPP surveyed 603 Texas voters from September 19-20. The margin of error is +/- 4%.

Based on his record of fighting for reform and his decision to reject all PAC money, Beto O’Rourke was the first challenger ECU endorsed in the 2018 midterms. For O’Rourke, ECU was his first national endorsement. ECU’s grassroots members have donated over $350,000 to O’Rourke’s campaign, averaging just $13 per contribution. Beto is one of only a handful of members of Congress to reject all PAC money, a decision that has made him a leader of a growing trend, with 124 no corporate PAC candidates advancing to the general election.

That poll memo is here. I think these are two different polls – they have different sponsors, slightly different results, and slightly different sample sizes (613 for the first, 603 for the latter), though they were all done between September 19 and 20. The End Citizens United-sponsored polls done by PPP are the ones I have linked on my sidebar. Protect Our Care references earlier polls, which could be these ECU polls, but doesn’t provide links, so who knows. I will say that ECU’s characterization of this as “within the margin of error” is correct, and POC’s “in a dead heat” is wrong and should be avoided. Both of these polls, plus that Ipsos poll arrived after the Quinnipiac poll, to vastly less fanfare; at least RG Ratcliffe acknowledged the existence of the Ipsos poll. The 18-poll average is now 46.83 for Cruz, and 41.83 for Beto.

On a side note, I also received a press release from the Republican Party of Texas announcing that they had filed a complaint with the FEC against End Citizens United for “ECU’s failure to file a direct expenditure for public communications in support of – or as an in-kind contribution to – the Beto for Texas campaign.” You can search the Internet for the eyeroll GIF of your choice here. It’s the weekend, and I can hear a beer calling my name.

CD07 “live poll”: Culberson 48, Fletcher 45

Here’s another of those NYT-Siena “live polls” of interesting Congressional races.

Lizzie Fletcher

Houston Rep. John Culberson holds a narrow lead over Democratic challenger Lizzie Fletcher, a poll conducted by The New York Times Upshot found Tuesday.

Culberson’s 3-point lead is within the 5-point margin of error for the poll, which The New York Times conducted Sept. 14-18. Forty-eight percent of respondents said they preferred Culberson, and 45 percent sided with Fletcher.

The nine-term congressman had a 43 percent favorable rating in the poll, compared to Fletcher’s 28 percent. That difference could be attributed to name recognition — 49 percent of those polled answered that they “didn’t know” if they had a favorable opinion of Fletcher, a Houston attorney, compared to 24 percent for Culberson.

Election handicappers such as Cook Political Report, FiveThirtyEight, and Larry J. Sabato’s Crystal Ball have deemed the race a “toss-up,” while Inside Elections rated it a “lean Republican.”

[…]

Pollsters asked several other questions of district residents.

Of those polled, El Paso Rep. Beto O’Rourke held a 7-point lead over Sen. Ted Cruz, who is from Houston. President Donald Trump had a 51 percent approval rating. Fifty-five percent opposed funding for a wall at the U.S.-Mexico border, and 60 percent supported a ban on assault-style guns and high-capacity magazines.

The Upshot, in partnership with Siena College, called 40,665 people in the Houston district and received 500 responses.

The full NYT writeup is here. They had previously done a poll in CD23, which I wrote about here; a poll of CD32 is coming soon. The DMN story goes over the candidates’ fundraising and the ads they have running now – I’ve seen one of Fletcher’s, but not yet Culberson’s – but none of that interests me. There are two other things I want to talk about.

The first thing is the embedded image of where in the district they got their responses, and whether those responses were for Fletcher or Culberson. CD07 has a big piece inside Beltway 8, and a big piece on the farther west side of Harris County, outside Highway 6 and north of I-10 up to US 290. The original idea of this district was to pair the relatively small and mostly Democratic inner urban piece with the larger and more Republican outer suburban area. That has worked well for the decade, but as you can see, there are an awful lot of Fletcher-responding blue dots in the far-flung section of CD07. The outer areas of Harris County are the Republican stronghold, with sufficient population to counterbalance the Democratic city of Houston. If the unincorporated part of the county is trending bluer, that’s a big problem for the local GOP.

Which brings me to the second point. It’s important to remember that CD07 was a thirty-point Republican district in 2014. Kim Ogg, in her first run for DA in the special election against Devon Anderson, got 37.5% in CD07 in 2014. Ann Harris Bennett, then running for County Clerk, got 34.7%. Judith Snively, running for District Clerk, got 33.6%. It was a massacre.

Things were different in 2016, of course, but the fact that CD07 was carried by Hillary Clinton obscures the fact that CD07 was still fundamentally Republican. Ogg, in her rematch with Anderson, took 46.8% in the district. Vince Ryan got 46.2%, Ed Gonzalez got 45.0%, and Ann Harris Bennett, now running for Tax Assessor, just barely cleared 42%. The average Democratic district judge candidate got 43.5%.

Remember, though, that even in losing CD07 by 16 points (Bennett) or 13 points (average Dem judicial candidate), they all still won Harris County overall. Greg Abbott had a more modest 22-point win over Wendy Davis in CD07, and he needed all of that to win the county by less than five points. If CD07 is even close to being fifty-fifty, how does any Republican win countywide? If Beto O’Rourke is really leading in CD07 by seven points, then he’s going to crush it in the county, and that as much as any “likely voter” poll suggests a real, close race statewide.

Now of course all of this is predicated on this poll being accurate. As we discussed after the CD23 poll was published, this is a new and experimental approach to polling, and as always it’s just one result. We do have one earlier result, which happens to closely match this one, but two data points are only slightly better than one. My point is that this poll doesn’t have to be dead-on accurate to highlight the potential for a blue wave in Harris County. Even a modest shift from 2016 would point in that direction.

Differing views of likely voters

First we had this.

Rep. Beto O’Rourke

U.S. Sen. Ted Cruz, R-Texas, leads his Democratic challenger Beto O’Rourke by 9 percentage points among likely voters, according to a new poll from Quinnipiac University.

Released Tuesday, the survey found Cruz with 54 percent support and O’Rourke, an El Paso congressman, with 45 percent. Only 1 percent of those polled were undecided.

“The Texas U.S. Senate race between Sen. Ted Cruz and Congressman Beto O’Rourke, and Democratic hopes for an upset win there, have boosted talk of a Senate takeover,” Peter Brown, assistant director of the Quinnipiac University Poll, said in a news release. “These numbers may calm that talk.”

It’s the first time Quinnipiac has released a likely voter survey in the Senate race. Quinnipiac previously polled registered voters three times, finding Cruz ahead by 6 points in August, 11 in May and 3 in April.

Quinnipiac also surveyed the governor’s race in the most recent poll and continued to find a much less competitive contest, with Republican Gov. Greg Abbott leading Democratic opponent Lupe Valdez by 19 points.

I started writing a post about how like everything else this is one result, the first one we’ve had of just likely voters, then I got distracted by all the hot takes about how this means The Race Is All Over And It Was Never Really Close and the shitshow in SD19, so I didn’t get it finished. And then I woke up the next morning and saw this.

U.S. Rep. Beto O’Rourke, D-El Paso, leads Republican incumbent Ted Cruz by 2 percentage points among likely voters, according to an Ipsos online poll released Wednesday in conjunction with Reuters and the University of Virginia. O’Rourke has been closing the gap over the last several months, but this is the first poll that puts him ahead of Cruz.

Forty-seven percent of likely voters told Reuters they would vote for O’Rourke, while 45 percent said they would cast their ballot for Cruz. Three percent said they would vote for “Other,” and 5 percent said “None.” The margin of error on that portion of the poll was 3.5 percentage points.

A Quinnipiac poll released Tuesday put Cruz 9 percentage points ahead of O’Rourke among likely voters. That poll was based on phone interviews, while the Ipsos poll used an online survey. But it’s trying to predict who will show up on Election Day that shifts the numbers, said Ipsos Vice President Chris Jackson.

Ipsos is trying to gauge political enthusiasm on each side, said Jackson. The poll asked respondents to estimate the likelihood that they’d vote in the midterm elections on a scale from one to 10. “More Democrats are registering at the highest part of the scale, at the 10, than the Republicans,” Jackson said. And that’s what’s interesting, he said, because Republicans usually have the momentum advantage in Texas.

“It demonstrates how Democrats are mobilized,” said Jackson. “This election is going to be really competitive and its going be very hard fought.”

[…]

The poll also questioned voters about the Texas gubernatorial election and found that Gov. Greg Abbott leads his Democratic challenger, Lupe Valdez, by 9 percentage points among likely voters.

Well, well, well. Look, this too is just one result and we can’t really judge either of them until we see enough polls to get a feel for where these fit in. That said, this is 1) the first poll result of any kind showing Beto in the lead, 2) the first poll of any kind in at least a decade showing a Democrat with as much as 47% of the vote, and 3) an extremely satisfying quick corrective to all those hot takes from Tuesday.

So what do we make of all this? I think the DMN has it right:

Polling experts have long warned against putting too much weight in any one survey, particularly since different polls can take widely different approaches. Texas is also a special case, since the GOP’s longstanding dominance there means pollsters don’t usually pay it much attention.

Nate Silver, editor-in-chief of Five Thirty Eight, on Wednesday pretty well summed up the dynamic.

“Texas is a tough state to poll (lots of new residents, low turnout among certain voting groups, may be hard to reach Spanish-speaking voters),” he wrote on Twitter. “It’s probably a healthy sign that we’re seeing some disagreement.”

A closer look at the two surveys perhaps further proves the point. Both polls looked at likely voters, which is a key distinction from earlier surveys. The feelings of likely voters are supposed to give a better representation of Election Day results than those of registered voters. But that approach also means some assumptions have been made on who is likely to vote.

The Quinnipiac poll surveyed 807 likely voters earlier this month, tallying a margin of error of plus-minus 4.1 percentage points. It was conducted using live interviews over landlines and cell phones, which many experts say is the best way to approach the task. It had a sample that featured 35 percent Republicans, 26 percent Democrats and 33 percent independents.

The Ipsos poll, done in partnership with Reuters and the University of Virginia, surveyed nearly 1,000 voters earlier this month, garnering an error margin of plus-minus 3.5 percentage points.

It was conducted by way of online surveys. Its sample is also much different, reporting a significantly lower number of independents. So its breakdown is 47 percent Republicans, 43 percent Democrats and nine percent independents.

So live call versus online, and self-reported engagement levels versus whatever formula Quinnipiac used (they did not elaborate on that). Ipsos wound up with a sample that was slightly less Republican, which is consistent with the thesis that Dems are more engaged than usual. Who’s “right” and who’s “wrong” at this point is impossible to tell, though we may get a better feel for that as voting draws nearer. For now, be aware of the differences in methodology and look for any trends that emerge there. In the meantime and even though it’s mixing apples (registered voters) and oranges (likely voters), I’m updating the now-16 poll average to reflect 46.69 for Cruz, and 41.38 for Beto. Until we can say definitively otherwise, this is still a very close race. Washington Monthly and the Current have more.

CD23 “live poll”: Hurd 51, Ortiz Jones 43

Give this one a bit of side-eye.

Gina Ortiz Jones

Incumbent Republican Will Hurd is leading his Democratic challenger, Gina Ortiz Jones, in one of the country’s most competitive races in this year’s midterm elections, according to a new poll by The New York Times and Siena College.

The poll, which surveyed 495 people in the district by phone this week, shows Hurd with 51 percent  support compared with Ortiz Jones’ 43 percent. Seven percent of those surveyed were undecided, with a margin of error of plus or minus 5 percent.

The southwest Texas district that stretches from San Antonio to El Paso, long considered a “swing district,” is a prime target for Democrats who are looking to pick up House seats this November. Hurd, a former CIA officer, narrowly beat Democratic opponents in 2014 and 2016.

Ortiz Jones, a former Air Force intelligence officer, is hoping Democratic enthusiasm and opposition to President Donald Trump will propel her to victory in the district, which has garnered national attention and is on several “most competitive” lists.

Hurd, who is seen as a moderate Republican, has distanced himself from Trump on major issues like immigration and has criticized the president for his dealings with Russian President Vladimir Putin.

Here’s the full NYT writeup, which is worth reading. This is one of the districts The Upshot of the NYT is polling in real time, with the explanation “Our poll results are updated in real time, after every phone call. We hope to help you understand how polling works, and why it sometimes doesn’t.” Basically, when they get to 500 completed calls, they stop. That has raised some questions – which they openly acknowledge and discuss; you can follow Nate Cohn on Twitter for a lot of that – and if nothing else this is a pilot program. It’s ambitious and admirable, just (as they say with each result) not to be taken as the be-all and end-all.

In this case, I will note that in the three elections in CD23 this decade, the final numbers have been a lot closer than eight points, and no Republican has achieved a majority of the vote:

2016 – Hurd 48.29, Gallego 46.96
2014 – Hurd 49.78, Gallego 47.68
2012 – Gallego 50.31, Canseco 45.56

Even in the debacle of 2010, Quico Canseco only got 49.40% of the vote, though of course that was before this redistricting cycle. The idea that Will Hurd could get 51%, which would be a high water mark for Republicans in CD23, in a year like this seems unlikely to me. It’s very possible Hurd can win – he’s proven himself to be a strong candidate. It’s conceivable Hurd could top 50% – maybe he’s won enough people over, maybe Ortiz Jones isn’t so good on the campaign trail, who knows. I would be very, very surprised if he wins by as much as eight. We’ll see if there are any poll results out there for this district. In the meantime, The Upshot and Siena are working on CD07, while the DMN and the Times will be polling CD32, as well as statewide. Exciting times to come.

Crosswinds: Cruz 47, O’Rourke 44

It’s poll time again.

Rep. Beto O’Rourke

In a sign of just how crucial the millennial vote might be in the upcoming mid-term elections, a statewide poll released Tuesday shows Ted Cruz leading 47 percent to Beto O’Rourke’s 44 percent among likely Texas voters. Forty-nine percent of Texans between the age of 18-39 identify as supporting O’Rourke, while Cruz’s strongest support comes from voters ages 40 and above.

The Crosswind Texas Pulse Poll also hinted at some dissatisfaction toward Cruz from his own party: While the poll indicates an almost-equal party vote – with 81 percent Republicans favoring Cruz and 83 percent Democrats for his opponent – a surprising 15 percent of Republicans indicated their intent to vote for the Democrat. Forty-six percent of respondents who did not identify with either major party signaled their intent to vote for O’Rourke, versus 39 percent of non-affiliated or independent voters intending to pull the lever for Cruz.

Fifty-five percent of Hispanic voters and 57 percent of Black voters also expressed an intent to vote for O’Rourke, while 52 percent of white voters indicated their support for Cruz. However, 56 percent of voters who do not identify as white, Hispanic or Black also responded positively for Cruz. Only slightly more women overall support O’Rourke than Sen. Cruz, at 47 versus 42 percent.

“Texans are in for a nail-biter that has national implications,” said Crosswind CEO Thomas Graham. “O’Rourke is showing surprisingly strong support in traditionally red-state Texas, and Cruz has the edge in organization at this point, but clearly O’Rourke is gaining some ground.”

Meanwhile, Texans clearly favor incumbent Greg Abbott, who is holding steadily in his race against Lupe Valdez, although her 39 percent – to Abbott’s 52 percent – is likely to catch state GOP leaders by surprise. The poll results largely mirrored that of the senate race demographically and along party affiliations, although 45 percent of Texas women are showing more support for the current governor than for challenger Valdez’s 43 percent female support.

Perhaps most surprising is that in both races, a percentage of likely voters identifying as “conservative” seem ready to jump ship to non-conservative candidates: 14 percent for O’Rourke, and 16 percent for Valdez. Those identifying as “liberal” seem less inclined to go against those stated values, with just 3 percent for Cruz and 6 percent for Abbott.

The Crosswind Texas Pulse Poll, a periodic survey of Texans’ opinions on a variety of cultural, economic and political issues, was conducted by Crosswind Media & Public Relations from September 6-9, 2018. The survey included 800 likely voters in Texas. The margin of error is +/- 4 percentage points with a 95 percent level of confidence.

Of the 800 likely voters surveyed, 39 percent identified Republican and 27 percent identified Democrat, with 34 percent unidentified.

And that brings our 14-poll average to 46.29 for Cruz, and 40.71 for Beto. According to RG Ratcliffe, pollster Crosswinds Media and Public Relations is “a national public relations firm based in Austin and leans Republican”. Maybe this result will finally get Chris Wilson to quit whining about how everyone is overestimating Beto’s numbers. Instead, you can add this to the reasons why Republicans are freaking out about Cruz.

Yes, Republicans really are worried about Ted Cruz

Their actions speak volumes.

Not Ted Cruz

With a string of polls showing GOP Sen. Ted Cruz’s lead slipping, Texas Lt. Gov. Dan Patrick showed up in Washington on July 25 to deliver an urgent plea to White House officials: Send President Donald Trump.

Patrick, who chaired Trump’s 2016 campaign in the state, made the case that a Trump visit was needed to boost turnout for Cruz and the rest of the Texas Republican ticket. The lieutenant governor soon got his wish: Trump announced on Twitter late last month that he was planning a blowout October rally for Cruz, his former GOP rival.

The previously unreported meeting comes as senior Republicans grow increasingly concerned about the senator’s prospects in the reliably red state, with some expressing fear that an underperformance could threaten GOP candidates running further down the ballot. Cruz’s Democratic opponent, Rep. Beto O’Rourke, has raised barrels of cash, closed the polling gap and emerged as a cause célèbre of liberals nationwide.

Trump’s rally is just the most public display of a Republican cavalry rushing to the senator’s aid. Cruz remains a favorite to win another term, and some senior GOP figures insist the concern is overblown. Yet the party — which has had a fraught relationship with the anti-establishment Texas senator over the years — is suddenly leaving little to chance. Behind the scenes, the White House, party leaders and a collection of conservative outside groups have begun plotting out a full-fledged effort to bolster Cruz.

Texas Sen. John Cornyn, who’s planning an October fundraiser for Cruz at Washington’s Capital Grille restaurant, said he had a simple directive to GOP givers.

“We’re not bluffing, this is real, and it is a serious threat,” Cornyn, the No. 2 Senate Republican, said in an interview. “If Ted does his job and we do ours, I think we’ll be fine. But if we have donors sitting on the sidelines thinking that, ‘Well, this isn’t all that serious,’ or ‘I don’t need to be concerned,’ then that’s a problem.”

What caught my eye in this story was the timing of Dan Patrick’s schlep to DC to beg for help. Here’s what the five most recent polls looked like as of that July 25 date:

Cruz +9, Cruz +5, Cruz +8, Cruz +6, Cruz + 6 – Average Cruz lead = 6.8

And the five polls since then:

Cruz +2, Cruz +6, Cruz +4, Cruz +4, Cruz +1 – Average Cruz lead = 3.4

So at the time that Danno made his pilgrimage, Cruz had a solid if unspectacular lead in the publicly available polls. Since then, he’s had a much narrower, albeit still consistent, lead. On the (I hope) reasonable assumption that Patrick is not clairvoyant, it makes one wonder what he and his cronies were seeing in the polls back then that made them so worried. I mean, it could just be an abundance of caution, though that’s wildly inconsistent with Texas Republicans’ public braggadocio about their own prowess and the supposed conservatism of the state’s electorate. Since when do Texas GOPers need help from the outside to win elections? Especially in a year where the national party has about a thousand endangered Congressional seats to protect, not to mention a non-trivial number of governors, and they’d much rather be spending money to oust Democratic Senators, asking for the spigot to be tapped in support of Ted Cruz sure seems like a lot.

Unless, of course, their own data at the time was sounding an alarm for them, not just for Cruz but for however many downballot Republicans that could get left exposed by a low tide for the junior Senator. And if that was the case for them then – and maybe it was and maybe it wasn’t, we just don’t know – then what is is saying now? Maybe the public data has caught up to where their own data was, and maybe things have shifted further. Again, we don’t know. That doesn’t stop us from speculating, as we wait for the next batch of poll results. My point here is simply to highlight that Republicans are aware of the political environment they’re in. It’s on us to prove they were right to be so concerned. Slate has more.

On negative ads and name recognition

I confess, I’m amused by this.

U.S. Sen. Ted Cruz is going after Democrat Beto O’Rourke for cursing during some of his campaign speeches.

The Cruz campaign released a digital ad on social media that shows O’Rourke cursing at various campaign events over the last year.

“So he’s showing up across Texas sharing his wit, his wisdom and his character,” an unidentified narrator says as clips of O’Rourke cursing are bleeped out.

The ad closes by saying O’Rourke is “showing the #@%* up.”

That’s…bad? Doesn’t that imply that Ted Cruz isn’t “showing the #@%* up”? I’m not exactly sure what I’m supposed to make of that, but then it seems that Ted Cruz isn’t very good at negative ads. Maybe if he had some accomplishments he could tout or something like that. I’m just spitballing here. What does it say about an incumbent when all his first moves out of the gate are to attack his opponent rather than brag about his record?

It’s clear that it’s going to be all mud from here on out, not just from Cruz but also from the big money outside agitators that prop him up as well. All of which leads to a bit of musing from RG Ratcliffe on the state of the race.

The increased attention that O’Rourke’s growing celebrity is drawing has some Democratic stalwarts worried that O’Rourke has not built a campaign that has the ability to quickly trade jab for jab. This may be problematic because poll after poll—which show the race for Senate is close—suggests that a large number of voters still do not know O’Rourke. This low voter ID allows Cruz and his allies to help define his opponent leading into Election Day.

You might want to put in a mouth guard to prevent injury as you grind your teeth at the quotes from unnamed “Democratic operatives”, but never mind that for now. My first inclination in times like these is always to look to the data, so I went through all the recent polls to see what I could find about Beto O’Rourke’s favorability and name recognition. Here’s what I got:

UT/Trib, June 25 – 16% Neutral, 24% Don’t Know

Quinnipiac, July 31 – 43% “Haven’t heard enough”

PPP, August 1 – “O’Rourke’s name recognition has grown since January as well as his favorability. In January, only 39% of voters had an opinion of him, and his favorability was 20% while 19% had an unfavorable opinion of him. Now 57% have an opinion of him with 31% having a favorable and 26% having an unfavorable opinion.”

NBC News, August 22 – 36% Unsure/Never heard among RVs

ECPS, Aug 27 – 27% Neutral, 11% Never heard

Not all polls asked about Beto’s favorability (though they nearly always asked about Cruz’s, and nearly everyone has an opinion on him), and those that did were not consistent in their question wording or their categorizations. Still, even with the variability, it’s clear that a decent number of people don’t have a firm opinion about Beto O’Rourke, and thus we get the pearl-clutching.

And to be fair, it’s a very reasonable point to make. If you don’t already have an opinion about a candidate – maybe even if you do – that means your perception is up for grabs. If you’re a politician with plenty of money – and while O’Rourke has greatly outraised Cruz, he still has lots of dough and his buddies in the conservative PAC business have bottomless coffers – you can have an effect on that. Thus the old adage about defining yourself before your opponent does it for you.

That said, I think it’s also worth contemplating how much effect negative ads, even competent ones, may have this year, especially on a high-charisma candidate like Beto O’Rourke. For one thing, as we have recently observed, not all candidates are vulnerable to negative information about them. For another, people who dislike Ted Cruz (of which there are many) and are undecided about Beto O’Rourke may be less likely to believe or be swayed by an attack on O’Rourke by Cruz. Most of all, in a year where so many people are highly motivated to deliver a message to Donald Trump, negative ads just may not mean much to them.

I’m not saying that Cruz’s barrage can’t or won’t have an effect. It probably will, though at this point it’s impossible to say how much of an effect it may have. I am saying that this is a weird year, with unusual dynamics, and it’s worth thinking this sort of thing through. I do hope Beto has a strategy for weathering the attacks, even if that strategy is “keep on keeping on”. We’re sure to get a lot more polling data over the next two months, so whatever effect there is, I’m sure we’ll see it.

The Atlantic on CD07

I have three things to say about this:

Lizzie Fletcher

On a Saturday morning in Houston, the high was 94 degrees with a chance of rain. It was hardly friendly weather for canvassing—the door-knocking, yard sign–delivering, get-out-the-vote efforts that define a politician’s grassroots network. Yet dozens of Seventh District residents, sporting lizzie fletcher for congress T-shirts, had happily crammed into a small office room on Richmond Avenue, awaiting their marching orders.

Fletcher stood on a step stool at the front of the room. The 43-year-old cuts an unconventional profile in the Seventh—female, liberal, inexperienced. Any one of those descriptors should be a nonstarter in this district, which a handsome blue blood named George H. W. Bush first turned Republican in 1966. That Bush has had only two successors in nearly five decades—both white, conservative men—appears testament to that fact.

But in a nod to the vast strangeness of 2018, Democrats see the Seventh as one of their best shots at taking the House. Indeed, Texas is changing. Across the state, Republican incumbents including Representative John Culberson here in the Seventh; Representatives Pete Sessions and Will Hurd; and even Senator Ted Cruz are struggling to fend off Democratic challengers. Suddenly, the idea of a progressive woman, a political outsider, unseating an 18-year incumbent like Culberson doesn’t feel so far-fetched.

On this Saturday in August, wearing a campaign T-shirt, a black miniskirt, and flip-flops, Fletcher prepped her volunteers by invoking the anniversary of Hurricane Harvey. It was exactly one year before that Harvey had dumped as much as 51 inches of rain on Houston, killing 75 people in Texas, and the trauma still ran deep. “For so many of us, Harvey was really a low point and a high point of our lives in Houston,” she said. The low point was obvious. But the high point, she said, was that in this community, “if you could help, you did.”

She didn’t have to adopt a hyper-partisan caricature—rallying for Donald Trump’s impeachment, say, or decrying his big tax cut for the wealthy—to energize the room. Rather, she compared volunteer efforts in the aftermath of Harvey to that day’s canvassing. “We are in a crisis in our country,” she said, her slight Southern lilt elongating her i’s. “And the best way—the best way—to do something about it is to do what y’all are doing today: Just show up.”

[…]

Today the district claims one of the most ethnically and economically diverse populations in Houston. It is 38 percent white, 31 percent Latino, 12 percent African American, and 10 percent Asian. To drive through the Seventh is to glimpse a vast number of takes on American life. The district touches some of the ritziest parts of Houston—the flashy mansions of River Oaks, the designer-stocked Galleria. Track a few miles southwest and you’ll find Gulfton, where Indian and Pakistani restaurants line the so-called Gandhi district and a single street might host Ethiopian and Guatemalan churches. Spin back up I-10 and there’s the Barker Reservoir, behind which many upper-middle-class homes were destroyed by Hurricane Harvey.

As the state undergoes a demographic transformation with the political shifts to match, the question for some political analysts has become not if Texas will turn blue, but when. So it has with the Seventh: The decades-long Republican stronghold swung for Hillary Clinton in the 2016 presidential election, and Democrats have since zeroed in on it as a linchpin of their map to secure the House majority. “Any blue wave from Texas to Washington, including California, is going to start with this race,” the longtime Democratic lobbyist Scott Eckart told me. “If Culberson loses, I think all the others will follow.”

So far, polling suggests that, for Democrats, the Seventh is in fact within reach. Both Fletcher’s and Culberson’s internal polling clocks the race within the margin of error, according to three sources to whom the numbers have been relayed. Which means the pressure is on for Fletcher to run the perfect campaign not just for her own sake, but for House Democrats writ large.

“The political momentum here has shifted, and Lizzie is the ideal person to capitalize on that,” said Mustafa Tameez, a Democratic strategist based in the district. “She’s a progressive woman, she’s young, she’s smart. She checks off every box.”

[…]

This is in part why her campaign is less a collection of partisan talking points and more a commentary on local issues such as flood relief: She’s long been personally privy to the cyclical trauma of flooding in Harris County. Culberson “has been my rep since he was first elected in 2001,” Fletcher told me. “That year, we had Tropical Storm Allison. And I was working downtown at the time, and downtown flooded, my building flooded, people died. It was just this really incredible event that kind of snuck up on us.

“So he’s been on notice since he took office that this was something we needed to deal with,” she continued. “I didn’t ever agree with his positions in the first place … but what we are dealing with, in terms of flooding, is a years-long problem, and Culberson has been completely missing from the discussion.”

For Fletcher, it makes one of the key pro-Culberson arguments—that he’s a senior member of the powerful Appropriations Committee—unconvincing. “As a senior member of the Appropriations Committee … in the majority, why is it that two Democrats in our community are bringing the bill to fund all of our flood-mitigation projects, and he won’t sign on?” she said. “I think if you ask anybody, they’ll say we haven’t seen him use that to benefit our community, in all the years he’s been on it.”

1. The subhed on this article is “The Republican incumbent John Culberson has held the minority-majority Seventh Congressional District for almost two decades, but the Democrat Lizzie Fletcher hopes to finally turn out progressives and minorities.” So naturally I wanted to look at historic turnout numbers:


Year   CD07   County  Ratio
===========================
2016  67.04    61.33   1.09
2014  39.05    33.65   1.16
2012  67.72    61.99   1.09
2010  49.42    41.67   1.19
2008  70.61    62.81   1.12
2006  40.65    31.59   1.29
2004  66.87    58.03   1.15
2002  37.37    35.01   1.08

So turnout in CD07 is always higher than turnout in Harris County as a whole, ten to fifteen percent more in Presidential years and fifteen to thirty percent more in most non-Presidential years. That’s probably due to non-Presidential year turnout being generally lower in more Democratic areas. There’s still plenty of room for turnout to improve here. The goal of course will be to make sure that the reason for the bump in turnout is primarily due to voters who are friendlier to Fletcher than to Culberson.

2. As I’m sure you can guess, the prospect of poll data in CD07 is irresistible to me. We do have one publicly released poll that showed a two-point lead for Culberson. My guess is that the others mentioned in the story are all around that same margin, most likely all with Culberson in the lead. It’s all consistent with the larger picture. I do wonder, if the current slump in Trump’s approval ratings persists, if we’ll start to see more polls of Congressional districts being made public.

3. I do like the idea of turning Culberson’s tenure on the Appropriations Committee against him. If he couldn’t or didn’t deliver when his district and much of the rest of the region suffered such catastrophic floods as Allison and Harvey, then what good is he and his vaunted seniority and position of influence? It’s an argument that has a chance of catching on with people who aren’t congenital Democrats, and a good argument to make in an anti-incumbent year. Doesn’t mean it will work, or that it will be enough even if it does work, but it’s a good place to start.

NBC News: Cruz 49, O’Rourke 45

It’s been three weeks since our last poll result.

Rep. Beto O’Rourke

In a head-to-head match up, Cruz held a 4-point lead over O’Rourke. Forty-nine percent of respondents backed Cruz, compared to 45 percent who supported O’Rourke. Six percent of respondents remain undecided. The poll has a margin of error of 3.8 percentage points.

Cruz has maintained a fairly strong favorability rating, with 49 percent of those surveyed viewing him favorably and 41 percent viewing him unfavorably. O’Rourke is far more unknown. Forty-one percent of respondents viewed him favorably while 23 percent of those surveyed had an unfavorable view. Thirty-six percent were either unsure of their opinion of O’Rourke or hadn’t heard of him.

[…]

The poll also showed Texas Gov. Greg Abbott with a daunting 19-point lead over former Dallas County Sheriff Lupe Valdez, similar to other public polling of the race.

Additionally, President Donald Trump is just above water in the state: 47 percent of registered voters approve of his job performance, against a 45 percent disapproval rating.

You can see more details here. There are two things I want to note about this poll, which brings our 12-poll average to 46.9 for Cruz and 40.75 for O’Rourke. One is that O’Rourke’s 45% is the highest level he’s reached in any poll so far (he’s gotten a 44 from Quinnipiac and a couple of 43s before now; Bill White reached 44 once and 43 once in 2010) and the second highest of any Democrat in any poll since I’ve been tracking them, trailing the 46 Hillary Clinton got in two different weird WaPo/Survey Monkey polls in 2016. I had just been saying that I’d like to see some results with Beto above 43%, and lo and behold we have one. Now let me say that I’d like to see more of this, and we’ll see if my wish gets granted again.

The other point has to do with the difference in the Senate race and in the Governor’s race. Not all of the polls we have seen so far have included results for the Governor’s race, but some have. Here’s how they compare:

NBC News, Aug 21

Cruz 49, O’Rourke 45
Abbott 56, Valdez 37
Cruz -7, O’Rourke +8

Quinnipiac, Aug 2

Cruz 49, O’Rourke 43
Abbott 51, Valdez 38
Cruz -2, O’Rourke +5

Lyceum, Aug 1

Cruz 41, O’Rourke 39
Abbott 47, Valdez 31
Cruz -6, O’Rourke +8

Gravis, July 10

Cruz 51, O’Rourke 42
Abbott 51, Valdez 41
Cruz 0, O’Rourke +1

UT/Trib, June 25

Cruz 41, O’Rourke 36
Abbott 44, Valdez 32
Cruz -3, O’Rourke +4

Quinnipiac, May 30

Cruz 50, O’Rourke 39
Abbott 53, Valdez 34
Cruz -3, O’Rourke +5

Quinnipiac, April 18

Cruz 47, O’Rourke 44
Avvott 49, Valdez 40
Cruz -2, O’Rourke +4

Average differences: Cruz -3.3, O’Rourke +5
Average differences minus NBC and Lyceum: Cruz -2, O’Rourke +3.8

I think we all agree that Beto O’Rourke will do better than other Democratic candidates in November. If he does, there are two possible reasons for it. One is that some number of people will vote for him and then not vote in other races, and the other is that some number of people who otherwise vote Republican will cross over to vote for him. I don’t think we’ll really know how this shakes out until we see results, but I would guess that at this time, the poll results mostly reflect the higher profile of the Senate race, and to a lesser extent the potential for crossovers. Hillary Clinton got 300K to 400K more votes than most of the other downballot Dems in 2016, which translated to her doing four to seven points better than they did, while Bill White got about 400K more votes than his downballot colleagues in 2010. That translated to a 14 or 15 point improvement for him, as that was a much lower turnout election.

The distance between Beto O’Rourke and Lupe Valdez is similar to the distance between Hillary Clinton and other Dems in 2016, though as you can see there are two polls including this one that show a wide gap while the other five show much narrower differences. In a non-Presidential election like this, we could be talking a net 300K or so swing towards Beto if the polls are accurate. As we’ve seen too many times before, that’s only a big deal if the base Democratic vote is enough to put him close to the base Republican vote. The fundamentals have always been the same, we just have more data now. I for one would hesitate to make any projections or draw any conclusions beyond the basic observation that O’Rourke is polling better than Lupe Valdez, and will almost surely outperform her. We don’t know enough to say more, and if you’re inclined to take this one data point as destiny, you’re doing it wrong.

Final EV turnout for the flood bond referendum

Lower than initial estimates, though I think the initial estimates were on the optimistic side. But really, we were all guessing.

Tuesday is the final day of early voting for Harris County’s proposed $2.5 billion flood bond, and as residents continue to trickle to the polls, the county clerk has downgraded his turnout estimate by a third.

When early voting began Aug. 8, Harris County Clerk Stan Stanart estimated 10 percent to 13 percent of the electorate would turn out, totaling between 230,000 and 300,000 voters. He lowered that estimate Monday afternoon to 170,000 to 180,000 voters, around 7.5 percent.

Put another way: that’s less than one vote per Harris County home or apartment building flooded by Hurricane Harvey. Stanart pleaded with Harris County’s 2.3 million registered voters to take the time to cast a ballot.

“There’s no lines at all. Just come in and vote, we’re waiting on you,” Stanart implored. “You get the government you vote for, so here’s your chance.”

[…]

Robert Stein, a Rice University professor who studies elections, said he expects most ballots to be cast during early voting. Though Commissioners Court members chose to hold the vote on the one-year anniversary of Harvey in the hopes of raising turnout, Stein said he is doubtful voters will rush to the polls on Saturday.

Some Republicans, including state Sen. Paul Bettencourt, have called for an end to summer elections on tax-increasing items, such as bonds, because they historically have low turnout.

Stein said poor voter participation should be cause for concern, but the date of the election was unlikely to change the public’s level of support for the bond.

“For the health and welfare and democracies, we should have more people voting,” Stein said. “But I don’t think the outcome would have been radically different if we had it in November.”

I agree with Professor Stein on all points. I will also reiterate my position that going with a November election for this would have been the safer choice, all things being equal. This one is on a road to passage because basically no one has argued against it. Having it in August was a choice made for reasons symbolic and strategic, and one can agree or disagree with those reasons. It could have mattered, but in the end I’m pretty sure it won’t have mattered.

Anyway, here are the final EV numbers. Tuesday was the last day, and like other last days of early voting it was the busiest, with 13,680 in person and absentee ballots being cast. That brings the EV total to 92,691 overall. I have no idea what anyone expected, but I’m sticking with my final turnout estimate of around 150K. We’ll see.

Crosstab versus 538

You are familiar with the Congressional race projections from G. Elliott Morris at The Crosstab, which I’ve noted here and here. He uses a probabilistic model for each district. Which as it happens is also the approach taken by FiveThirtyEight in their model. You can see all those projections here. You know what that means: Let’s compare the two!


Dist   XTab    538
==================
CD02  12.9%   7.8%
CD03   5.5%   0.8%
CD06  18.2%   6.6%
CD07  52.0%  49.3%
CD10  18.1%   2.6%
CD14   5.2%   1.8%
CD17   3.5%   0.5%
CD21  18.1%  17.8%
CD22  17.4%  14.2%
CD23  84.6%  72.4%
CD24  25.3%   4.4%
CD25   9.1%   7.9%
CD27   5.6%   0.5%
CD31   9.3%  20.1%
CD32  41.4%  11.7%

Overall, 538 is a bit more pessimistic about the individual Texas races. Where the Crosstab sees CDs 02, 06, 10, and especially 24 as lower-tier possibilities, 538 basically views them as nearly impossible. The one place 538 is more bullish than the Crosstab is in CD31; I’d love to understand the particulars behind that difference. But by far the most striking difference is in CD32, which the Crosstab has as slightly leaning red, while 538 sees it has almost a lock for the Republicans, behind not only CD31 but also CDs 21 and 22 on the Dem pickup list. That’s well out of line with the view of other national prognosticators, and at least one poll. Doesn’t mean they’re wrong, just that any time there’s this big a difference of opinion it’s notable. I’ll check back on this later to see if anything changes.

CD32 poll: Sessions 47, Allred 45

Another internal poll, another close race.

Colin Allred

Texas Rep. Pete Sessions’ re-election race is looking increasingly competitive, with Democrat Colin Allred polling close to the longtime Republican lawmaker, according to a new internal Democratic survey.

The Dallas-area 32nd District is traditionally GOP territory. But this year’s race is considered competitive, in part because the 32nd is one of three Republican-held districts in the Lone Star State that Hillary Clinton carried in 2016. Inside Elections with Nathan L. Gonzales rates the race a Toss-up.

Sessions led Allred 47 percent to 45 percent in the initial head-to-head matchup, according to the polling memo shared first with Roll Call.

The poll, conducted by GBA Strategies, also showed both Sessions and President Donald Trump with low favorable ratings. Forty-one percent of those surveyed viewed Trump favorably while 38 percent had a favorable view of Sessions.

Twenty-three percent viewed Allred favorably, but only 15 percent viewed him unfavorably, signaling that respondents might not have strong opinions of the Democratic nominee or know much about him.

Standard disclaimers about internal polls apply, and as always it’s just one data point. The main thing I take away from these and other Congressional district polls is that they offer corroborating evidence for the statewide polls we have seen and the closer-than-usual environment we are in. I hope we get enough of these to get a feel for what the trendlines might look like as well.

Day Seven flood bond EV totals

The word of the week is “slow”.

Fewer than 46,000 ballots have been cast in the first week of early voting on Harris County’s $2.5 billion flood bond referendum, but county officials on Monday said they expect many more voters leading up to the Aug. 25 anniversary of Hurricane Harvey’s landfall.

According to the county clerk’s office, 2,692 voters went to the polls in person Monday. Combined with 575 mail-in ballots returned Monday, the first six days of early voting have seen a total of 45,517 ballots.

“Bond elections don’t usually get voters excited, but there are plenty of days of early voting,” Harris County Clerk Stan Stanart said.

Last week, Stanart estimated that 230,000 to 300,000 voters would cast ballots on the bond referendum. By Monday, he had dropped his projection to 150,000 to 200,000 total votes by the end of the election, even as he expects turnout to increase closer to the one-year anniversary of Harvey, when media coverage and advertisements in support of the flood bond will increase publicity.

[…]

Rice University political scientist Robert Stein said he is skeptical the number of voters will increase come Aug. 25, but he added that low turnout does not necessarily signal a lack of support for the bond plan. He predicted the bond would pass with at least 60 percent of the votes cast.

A University of Houston poll last week put support for the bond around 62 percent.

Stein said low voter turnout is a “free rider” issue for residents who assume their vote does not matter.

“The public believes this (flood control bond) will pass and want it to pass,” he said. “But the assumption is perfectly reasonable that, ‘I’m not going to vote. Someone else will do it.’”

See here for more on that poll. I tend to agree with Professor Stein on both counts here. I suspect that the bulk of the ballots will be cast early, and I don’t see much in the way of opposition, at least not at a level to push people to the polls.

I suspect Stanart’s initial optimism was based on the number of mail ballots sent out. There were about 68K of them sent out for this election; by comparison, there were about 89K mail ballots sent out for the November 2014 election, of which about 71K were returned. More people vote by mail these days, and an election like this is going to be especially heavy with older voters, but that’s still a significant enough number to suggest a level of turnout that’s a decent fraction of a regular November off-year election. It’s just that the in person EV totals have not been consistent with that.

In any event, here are the EV toitals after one full week. If there’s an uptick coming, it has not yet arrived. After seven days, 16,277 people have voted in person and 34,388 by mail, for 54,665 in total. I do think we will see an upward trend in the last few days, as we usually do, but for now we are just toddling along. And as Campos notes, the original idea was for this to have modest-at-best turnout, so I suppose we are more or less where we should have expected to be. Have you voted yet? I figure I will on Friday.

CD06 poll: Wright 48, Sanchez 39

Via Patrick Svitek on Twitter, I learned of a recent PPP poll in CD06. Here’s the polling memo, and here’s the information you’re most interested in:

Q1 Do you approve or disapprove of President Donald Trump’s job performance?


48% Approve
46% Disapprove
 5% Not sure

Q2 If the election for U.S. House of Representatives were held today, would you vote for the Democratic or Republican candidate in your District?


45% Democratic candidate
49% Republican candidate
 6% Not sure

Q3 If the candidates for U.S. House of Representatives this fall were Democrat Jana Lynne Sanchez and Republican Ron Wright, who would you vote for?


39% Jana Lynne Sanchez
48% Ron Wright
13% Not sure

Not exactly sure why there’s a dropoff from the generic Democrat to Jana Sanchez, but that’s not a terribly unusual event in polls. Smokey Joe Barton won in 2016 by 19 points, and he won in 2014 by 25 points, so whichever result is closer to the truth represents a much tighter race than we’ve seen recently. As noted in other contexts, this is consistent with statewide polling showing narrower than usual margins. I hope we see more Congessional-level polls in the state going forward.

The range of Republican anxiety

Some folks are a little scared about all this “blue wave” talk and poll numbers and what have you.

Not Ted Cruz

As Ted Cruz took questions at a Republican women’s event [in Smithville] Saturday evening, Bastrop retiree Ronnie Ann Burt wanted to know: Should she really trust the growing barrage of chatter online that the senator’s re-election bid is in peril?

Cruz’s response: Believe it.

“It’s clear we have a real and contested race where the margin is much too close for comfort,” said Cruz, who’s facing a vigorous, massively funded challenge from U.S. Rep. Beto O’Rourke, D-El Paso.

Cruz’s stop in this small Central Texas town was part of a return to the campaign trail Saturday in which the incumbent cranked up his long-building warnings that Democratic enthusiasm in the era of President Donald Trump should not be discounted, even in a state as red as Texas.

The timing couldn’t have been more fitting: A trio of polls came out this week showing Cruz’s race tightening and a national political forecaster shifted the contest in O’Rourke’s favor. Meanwhile, Cruz launched his first TV ads Friday, including three targeting O’Rourke, and the challenger moved quickly to turn them into a fundraising boon for him.

Appearing Saturday afternoon at the conservative Resurgent Gathering in Austin, Cruz delivered a nearly 10-minute assessment of the uncertain political landscape he faces in November.

“The biggest challenge I have in this race … is complacency,” Cruz said. “People say all the time, ‘Oh, come on, it’s a Texas re-elect. How could you possibly lose?’ Well, in an ordinary cycle, that might be true. But this is not an ordinary cycle. The far left is filled with anger and rage and we underestimate that anger at our peril.”

Cruz added that there is reason to be skeptical of the polls — his campaign has criticized their methodologies — but the trendline “ought to be a cause for concern for everyone.”

[…]

Cruz’s remarks at events Saturday came a day after Gov. Greg Abbott offered a more reassuring forecast for November while addressing the Resurgent conference. He dismissed the idea of a “blue wave” in November as media hype that “sells papers” and reminded the audience that he ended up defeating his much-ballyhooed Democratic opponent, Wendy Davis, by over 20 points in 2014.

“Texas is going to stay red,” said Abbott, whose Democratic opponent, Lupe Valdez, has not caught traction in the way O’Rourke has against Cruz.

Cruz did not sound as sure as Abbott on Saturday — and his supporters appeared to get the message.

Cruz and Abbott are two sides of the same coin here. Cruz is quite right that complacency is a big potential problem for him, for the simple reason that if Republican turnout is less energetic than it has been in recent elections, Democrats have a smaller hill to climb to catch them. I’ve talked multiple times about how I’m hoping for Republicans to have a 2006-style year for turnout, as that would mean some 200K to 300K fewer votes than they got in 2014. This is Cruz’s main concern as well, and his message is simply “Don’t take this for granted”.

Abbott, on the other hand, is not wrong to observe that even with the recent polls, Cruz is still in the lead, and that other Republicans (most notably himself, not that he’s bragging or anything) are doing better than Cruz, that one UT/Trib poll result for Ken Paxton aside. Until such time as we start seeing poll results with one or more Dems in the lead, one can quite confidently say that the Republicans are ahead and thus favored to win. While that may run a bit counter to Cruz’s “we have to have a sense of urgency” message, Abbott is aiming at the media (to get them to run something other than a positive story about Beto O’Rourke and Democratic enthusiasm) and also at Dems, to say basically “don’t bother getting your hopes up, you still can’t win”. I don’t think he’s going to demoralize anyone, but it can’t hurt to try.

Finally, a word on the polls. Republican pollster Chris Wilson complained bitterly about that Lyceum poll, saying they had the samples all wrong. I don’t know if he has the same complaint about Quinnipiac and PPP and everyone else who has put out a result on this race, but I do know that he himself hasn’t published a result lately. Maybe he’s just lying low to let us all fall into a false sense of security, I don’t know. The average of all these poll suggests a six-point race, more or less, so go argue against that if you want to. It is certainly possible that pollsters are misreading the electorate this year, and thus skewing the numbers because they’re not polling the right mix of people. It’s also possible that Chris Wilson is one of those misguided pollsters.

Poll shows flood bond referendum in good shape

Standard caveats apply.

A majority of Harris County voters say they will support a $2.5 billion flood infrastructure bond at the polls later this month, according to University of Houston research released Monday.

Sixty-two percent of residents who said they are certain to vote said they will support the bond, compared to 55 percent of all respondents. Just 10 percent said they oppose the bond, while one-third remain unsure.

“People see flooding as a Houston and Harris County problem, not a problem affecting only certain neighborhoods or people,” Jim Granato, executive director of UH’s Hobby School of Public Affairs, said in a statement. “They believe the region’s future will be decided, at least in part, by how we respond.”

[…]

Residents who sustained property damage from Hurricane Harvey were slightly more likely to support the bond than those who remained dry, 60 percent to 52 percent.

Partisanship appears to play little role in residents’ views on the bond, as 58 percent of Republicans and 63 percent of Democrats support the proposal, which has wide support elected officials from both parties.

Sixty-nine percent of college-educated residents said they’ll vote yes, while the poll found residents 65 and older support the bond to the tune of 58 percent. Seniors whose homes are worth less than $200,000, as well as residents who are disabled, would not see their taxes rise because of the bond.

Just 18 percent of the youngest polling cohort, residents 18 to 25, said they support the bond, though 59 percent professed they remain unsure.

Harris County residents found most agreement when asked whether Austin should help the region’s recovery by tapping into the state’s Rainy Day Fund. Almost 88 percent said the Abbott administration should.

You know the drill: It’s one poll, polling local races is always tricky because the turnout model can vary wildly, nobody knows who is going to show up for a weirdo August election. That said, the fact that 55% of all adults were in favor of the bond, with the number climbing to 62% for the self-proclaimed likely to vote, is a positive sign. At the very least, it suggests that the people who are paying more attention are also more likely to favor the bond. The low numbers for those who are against it, much lower than those with no opinion, also augur well. I think this poll is probably correct about the outcome, though getting the exact numbers right is anyone’s guess. Early voting starts Wednesday – you know, tomorrow – so we’ll know soon enough. How are you voting on this? Since I was asked in an earlier post, I’ll state that I am voting for it. What about you?

PPP: Cruz 46, O’Rourke 42

Once, twice, three times a poll result.

Rep. Beto O’Rourke

A new Public Policy Polling survey finds that the Texas Senate race between Republican incumbent Ted Cruz and Democratic candidate Beto O’Rourke is competitive as O’Rourke continues to close the gap. In the initial matchup in January, Cruz led O’Rourke 45-37, but the results for August show a closer race with O’Rourke only 4 points behind Cruz, 46-42.

Texans believe that special interest money in Texas elections is a problem and would rather elect Democratic candidate Beto O’Rourke to the U.S. Senate due to his campaign being entirely funded by individuals, while Republican incumbent Ted Cruz has accepted $1.3 million from corporate PACs. After learning that O’Rourke is not taking a dime from political action committees or special interest group PACs, voters in January supported him over Cruz 43-41, and O’Rourke retains and expands this lead in August, 46-43.

A plurality (48%) of voters believe special interest money in Texas elections is a major problem, and 24% believe it is a minor problem. Also, a majority of voters (56%) would be more likely to support a candidate who has pledged to not take any money from corporate special interests, which is good news for O’Rourke.

Key findings from the survey include:

– O’Rourke’s name recognition has grown since January as well as his favorability. In January, only 39% of voters had an opinion of him, and his favorability was 20% while 19% had an unfavorable opinion of him. Now 57% have an opinion of him with 31% having a favorable and 26% having an unfavorable opinion.
– A plurality of Texans (44%) think Cruz is more responsive to his big campaign donors than to ordinary Texans.

PPP surveyed 797 Texas voters from August 1-2, 2018. The margin of error is +/- 3.5%. 75% of interviews for the poll were conducted over the phone with 25% interviewed over the internet to reach respondents who don’t have landline telephones.

As noted above, this is the third PPP poll of the Texas Senate race, with all three being done on behalf of End Citizens United. There was no other info about this on the PPP webpage, so what you see here is everything I know about it. Adding this into the other 10 results and the average of the 11 polls so far (all but that WPA poll from last December) is 46.7 for Cruz and 40.4 for Beto. It remains a close race, but it sure would be nice to see 1) some more results that will bring down the average difference between the two; a result or two with Beto in the lead would not suck, either; and 2) some results with Beto above 43%. Beto is unquestionably doing better in the polls than any previous Dem since I’ve been tracking this stuff. But “doing better” and “in a position to be called the favorite” are two different things. Here’s hoping.

Quinnipiac: Cruz 49, O’Rourke 43

Two polls in one week.

Rep. Beto O’Rourke

A new poll released Wednesday morning suggests a tightening race between U.S. Sen. Ted Cruz and U.S. Rep. Beto O’Rourke.

The newly released poll from Quinnipiac University gives Cruz a 6-point lead: 49 percent of registered Texas voters reported backing the Republican incumbent while 43 percent said they support O’Rourke, an El Paso Democrat. The poll’s margin of error is 3.5 percent. The results are closer than a poll Quinnipiac released in late May, which showed Cruz holding an 11-point lead over his opponent.

“U.S. Sen. Ted Cruz has a slight, by no means overwhelming, lead,” said Peter Brown, assistant director of the Quinnipiac Poll. “Congressman Beto O’Rourke has done a good job making the race competitive. With three months until Election Day, he is clearly in contention. A Democratic victory in the Lone Star state would be a serious blow to GOP hopes of keeping their U.S. Senate majority.”

The poll found 50 percent of Texas voters had a favorable view of Cruz while 42 percent had an unfavorable view. O’Rourke, on the other hand, had a 33 percent favorability rating, with 43 percent of voters not knowing enough about the congressman to form an opinion of him.

You can see the Quinnipiac writeup of the poll here. They also show Greg Abbott leading Lupe Valdez in the Governor’s race 51-38. This Q-poll lands right in the middle of the other two. Remember when the second one, showing an 11-point lead for Cruz, was considered the “correct” poll? Averages are more useful than single results, and with this result our averages are 46.5 for Cruz and 40.2 for Beto. A persistent lead for Cruz, but not a big one.

And thus the continued closeness of this race has caused it to draw national attention.

Democrat Beto O’Rourke of El Paso had a very good day Wednesday. Publicly, two new polls showed him in striking distance of incumbent Republican U.S. Senator Ted Cruz. Privately, his Wednesday might have been even better. That’s because Senate Minority Leader Chuck Schumer, of New York, and about a dozen other senators in the Democratic leadership suddenly became interested in the Texas race. That was followed by a serious look at Texas from officials of the Democratic Senatorial Campaign Committee, whose sole function is to help finance Democrats to win seats in the upper chamber of Congress. The DSCC hasn’t shown much interest in Texas for a least a generation.

The senators held a briefing on Capitol Hill Wednesday morning with at least four political polling companies about the time that the first poll, the Texas Lyceum 2018 Poll, showed Cruz with a two-point lead over O’Rourke: 41 to 39 percent. Hours later, a Quinnipiac University Poll had Cruz leading O’Rourke by a margin of 49 to 43 percent. This is the third poll that Quinnipiac has released of likely Texas voters this election season and the six-point difference in Wednesday’s poll represents momentum by O’Rourke who was down by 11 percentage points when its last poll was released May 30.

Numbers such as these, combined with O’Rourke raising more money than Cruz in two consecutive quarters, had at least one of the pollsters present at Wednesday’ meeting declare to Schumer and his colleagues that there was a potential path to victory for Democrats in Texas, according to at least one witness present. Schumer took great interest in that assessment and began digging deeper, the witness told Texas Monthly.

There are two things that are interesting about this. One is that as we know, Democrats have a lot of seats to defend in red states this year. Conventional wisdom was that they’d be spending nearly all their resources on defense, plus Nevada and Arizona. For them to even have a meeting to talk about Texas means they’re feeling pretty good about their position, and can look to spend some money elsewhere. It also means they think there’s a chance that Texas could be in play. Maybe not a great chance, but a non-trivial one, which could grow in magnitude if the national environment gets friendlier. I suspect this is more of a move to explore options in the event of such an upgrade to the national atmosphere than anything approaching a commitment to Beto and Texas, but even that tells you something.

Lyceum: Cruz 41, O’Rourke 39

Good result, though the others with it could be better.

Rep. Beto O’Rourke

A new poll released Wednesday suggests that U.S. Sen Ted Cruz and U.S. Rep. Beto O’Rourke, D-El Paso, are in a dead heat.

The poll from Texas Lyceum shows Cruz holding a slim margin over his Democratic challenger in the U.S. Senate race. Among likely voters, Cruz carries 41 percent of the vote compared to O’Rourke’s 39 percent. Nineteen percent of voters said they were undecided.

That lead falls within the polls 4.67 percent margin of error.

“O’Rourke continues to nip at Cruz’s heels, but it’s a long way to go until Election Day,” Josh Blank, Lyceum Poll Research Director, said in a news release. “If this race looks different than the rest, that’s probably because it is because a strong Democratic challenger raising prolific sums of money and tons of earned media.”

All the information about the 2018 Lyceum poll is here. Here’s the press release, the executive summary, the toplines, and the crosstabs. Here also are the results for the four races they polled:

Registered voters:

Senate – Cruz 36, O’Rourke 34
Governor – Abbott 44, Valdez 25
Lt. Governor – Patrick 32, Collier 23
Attorney General – Paxton 32, Nelson 20

Likely voters:

Senate – Cruz 41, O’Rourke 39
Governor – Abbott 47, Valdez 31
Lt. Governor – Patrick 39, Collier 29
Attorney General – Paxton 35, Nelson 25

I’ve generally gone with RV totals in these polls, but you can make your own choice here. I’m including the LV totals in the polling average for Senate, which now stands at 46.2 for Cruz and 39.9 for O’Rourke. The Lyceum did its 2014 polling in October, which is a bit annoying as that makes it less directly comparable. At the time, their numbers in the Abbott-Davis race looked not too bad, but that was the last time one could make that assertion. What makes me want to pull my hair out is that they did generic ballot polling for Congress and the Lege in 2014, giving Republicans a 46-35 lead in the former and a 38-31 lead in the latter, but apparently didn’t ask that same question this time around. Argh! That sure would have been a nice little data point to have.

I’ve spent a lot of my time on this blog nitpicking polls and questioning assumptions and samples and whatnot, oftentimes for reasons that in retrospect don’t look that great. So it is with a certain measure of grim satisfaction that I read this:

The newest poll is sure to draw skepticism from Cruz supporters. Even before it was released, Cruz’s pollster Chris Wilson published an article on Medium questioning whether it would be accurate.

“Dating back to 2008 the Texas Lyceum has generously given Democrats a massive house effect boost of seven (7!!!) points,” he wrote, add that the poll has historically overestimated the share of the Hispanic vote.

I feel your pain, buddy. But just for the record, here are some previous Lyceum results:

2016 – Trump 39, Clinton 32 (LVs)
2014 – Abbott 49, Davis 40 (LVs)
2012: Romney 58, Obama 39 (LVs)

They definitely underestimated Abbott in 2014 (though they did show a wider lead 47-33 lead for Dan Patrick over Letitia Van de Putte), but the total for Davis was spot on. They were pretty close on the other two. Take your “house effect” complaint to the nerds at 538 (which doesn’t have the Texas Lyceum poll in its pollster ratings). Texas Monthly has more.

The Trib looks at the AG race

There’s case that this is the second-most interesting statewide race on the ballot.

Justin Nelson

Three years ago almost to the day, a Collin County grand jury indicted Texas Attorney General Ken Paxton for securities fraud. As the state’s top lawyer turned himself into a jail in his hometown of McKinney and smiled for his mug shot, Democrats couldn’t help but feel optimistic. The last time Texas elected a Democrat for attorney general was over two decades ago. Paxton’s legal troubles could potentially serve, they hoped, as the springboard to breaking that streak.

What perhaps no one could have foreseen back in 2015 was the dizzying array of twists and turns the legal case against Paxton would undergo. Three summers later, there is still no trial date in sight and one is unlikely to emerge before Election Day.

Yet despite avoiding a challenge from within his own party this year – arguably the biggest political threat for a statewide official in deep-red Texas – the political fallout from Paxton’s indictment remains to be seen. In November, he’ll face his first actual test of it at the ballot box: a challenge from Democrat Justin Nelson. The well-credentialed Austin lawyer is framing the race as a crystal-clear referendum on the charges that have dogged Paxton for the vast majority of his first term.

“The question that voters will have is who voters want to hire as a lawyer for Texas, for all Texans,” Nelson said in a recent interview. “Voters will be able to choose me, someone who has clerked for Justice Sandra Day O’Connor on the Supreme Court, who has taught at the University of Texas law school, Texas Super Lawyer, a partner at a successful law firm, versus my opponent, who it’s embarrassing that he’s indicted for fraud in one of the most heavily Republican counties in Texas.

“And I think that when voters see that contrast, it will be integrity versus indictment.”

[…]

In Nelson, Democrats have ample reason to be optimistic. He brings an impressive resume to the race as a clerk for former U.S. Supreme Court Justice Sandra Day O’Connor and current partner at Houston-based litigation powerhouse Susman Godfrey. That network that comes with that — as well as Nelson’s own wealth — has allowed him to build a bigger war chest than any other Democratic statewide candidate beside U.S. Rep. Beto O’Rourke, who is running a far more high-profile race to unseat U.S. Sen. Ted Cruz. Still, Paxton maintains a wide cash-on-hand advantage.

In addition to Nelson’s fundraising, Democrats have been buoyed by a pair of public polls that suggested the race may be close. A recent University of Texas/Texas Tribune poll found Paxton leading Nelson among registered voters by just 1 percentage point, but 26 percent of voters had yet to pick a candidate.

“I feel really good about his race,” said state Sen. Sylvia Garcia, D-Houston, ranking Paxton and Agriculture Commissioner Sid Miller as ripe targets for Democratic upsets in November. “[Nelson’s race] is particularly compelling because … I’ve seen people react when you talk about an indictment and an attorney general — I mean, this is our lawyer, and they can’t get their head around the idea that our own lawyer is under indictment. They’ll easily concede, ‘Yeah, he’s innocent until proven guilty’, but it’s the whole image — it’s the whole cloud over the work — it just doesn’t sit easy when you hear about that.”

To be sure, Nelson is not building his challenge entirely around the incumbent’s legal troubles. Nelson also is campaigning on issues like ending gerrymandering, which was the topic of a pub crawl he led last month in Austin that touched three congressional districts in a five-block radius. During an interview at the last stop — at Easy Tiger in downtown Austin — Nelson said he would use his platform as attorney general to fight partisan gerrymandering in particular through written opinions, through the litigation process and at the Legislative Redistricting Board, where the attorney general has a seat. Nelson and Paxton were already on opposite sides of the issue before their race began, having signed on to dueling amicus briefs in the Wisconsin gerrymandering case that the U.S. Supreme Court declined to decide on last month.

Nelson has raised (and loaned himself) a few bucks, he’s got the national Democratic Attorney General Association in his corner, and there’s that one-point poll result; another poll from the Texas Lyceum that includes the AG race will be out next week. And then there’s the wild card of the Paxton trial, and how much the publicity for that has eroded Paxton’s natural advantage as a Republican. Going in on redistricting as an issue is a good idea as well. The main questions as always are how much does the average voter already know about this stuff, how much do they care, and how effectively can Nelson get his message out? Justin Nelson is a very appealing candidate, but he has to overcome the tide. That’s what this boils down to.

Republicans and Independents

Something to ponder.

The good news for President Trump in the latest NBC News/Wall Street Journal poll — half of which was conducted before and the day of the Helsinki presser with Putin, half of which was conducted afterward — is that his standing with the GOP base is stronger than ever.

Eighty-eight percent of Republican voters in the poll approve of Trump’s job — the highest of his presidency — and 29 percent of all voters strongly approve of his performance, which is another high for him. “The more Trump gets criticized by the media, the more his base seems to rally behind him,” says Democratic pollster Fred Yang, who co-conducted the NBC/WSJ poll with the Republican team from Public Opinion Strategies.

Trump’s approval rating in the poll is 45 percent among all registered voters (up 1 point from June), while 52 percent disapprove, including 44 percent who do so strongly.

The bad news for the president is that his standing — plus the GOP’s — is now worse with independents than it was a month ago. Just 36 percent of independents approve of Trump’s job (down 7 points from June). What’s more, independents prefer a Democratic-controlled Congress by more than 20 points, 48 percent to 26 percent. In June, the Dem lead among indies was just 7 points, 39 percent to 32 percent.

As you know, I’ve been looking for signs of Republican disapproval with Trump as a potential catalyst for lower turnout among GOPers this year. That does not appear to be happening, though voter enthusiasm (as noted in this poll as well) continues to tilt towards Dems. However, there is a potential alternate explanation for the durability of Trump’s support among the Rs:

Voters have to identify themselves with a political party, and that identification isn’t stable; it ebbs and flows with events and circumstances. Trump might win high marks from most Republicans, but the pool of Republican voters might be smaller than in the past. Far from standing tall over the entire GOP, Trump’s base may have eroded significantly from where it was at the beginning of his administration.

According to the Pew Research Center, Republican Party identification fell 3 points, to 26 percent, from 2016 to the end of 2017. The number of self-identified independents increased at the same time, from 34 percent to 37 percent, while the number of Democrats remained steady. Gallup shows a similar change: From November 2016 to November 2017, there was a 5-point drop in the number of people who called themselves Republicans, from 42 percent to 37 percent. Democratic self-identification remained unchanged at 44 percent.

The sheer size of the United States makes it easy to find vocal support for anyone and anything, and Donald Trump has his vocal supporters. But their staunch commitment overshadows the reality: a shrinking base for a president who won by the skin of his teeth, reliant on a small group of voters in just a handful of states. His scandals and outrages—controversies and improprieties—have had an effect. Even rank-and-file GOP reactions to Helsinki are revealing; according to CBS, 21 percent of Republican voters disapproved of the president, a striking number given typical partisan loyalty.

Charles Franklin had a Twitter thread about this, for which the short version is that this isn’t really a sign of long-term decline in the number of Republicans compared to Democrats. But the data is volatile, so when there is a dip in the cycle it could have an effect on Texas. I return now to that Gardner Selby piece about the Civiqs polling data:

O’Rourke’s camp didn’t offer a comment about the poll’s claim. But Chris Wilson, who conducts polls for Cruz’s campaign, drew on a Moulitsas-tweeted illustration breaking out the demographics of the results to suggest by phone that Republican respondents to the poll outnumbered Democrats by insufficient percentage points. [director of Civiqs Drew] Linzer separately told us the results imply that 31 percent of respondents identified as Democrats, 36 percent as Republicans and 33 percent as independents.

Wilson said that considering Republicans’ prevalence in statewide races since 1994, any poll projecting fall results should query more Republicans — perhaps making the sample 40 percent Republican, 30 percent Democratic and 20 percent independent.

Linzer said that the poll reflected the partisan mix of Texas registered voters.

I don’t know how good anyone’s state-levevl data is, but it is the case that some of the poll variance we’ve seen is rooted partially in the partisan mix the pollster used. Beyond that is another question I bring up a lot. How much do the national trends affect Texas? It sure seems like the answer is “in a proportionate fashion”, as we saw in the Dem direction in 2006 and 2008 and in a Republican direction in 2010 and 2014, but every year is its own universe. If there is a trend towards fewer self-identified Republicans, to what extent is that the case here? Or is it the case that the Texas GOP has some level of insulation from these slings and arrows? Obviously, the answers to those questions affect not only the assumptions one makes when polling, but ultimately the final result. I just want to make sure we’re thinking about that.

What’s the deal with that Civiqs poll?

Last week, Markos at Daily Kos posted this:

I didn’t want to believe Politico’s, “Beto-mania sweeps Texas.” But a look at our Civiqs Texas data (which we poll for our own use, not for any client), and holy shit, it’s real!

Yup. That’s the odious and treason-enabling Ted Cruz leading by just TWO POINTS.

For context, our numbers in the Texas governor’s race look more as you’d expect, with Republican incumbent Gov. Greg Abbott leading Democratic challenger Lupe Valdez 51-41. So where is Cruz struggling?

There’s the typical 2018-style gender gap, with Cruz leading among men 54-41, and losing women 43-51. Cruz is winning white people 59-36, while losing Latinos 70-24, and black voters 79-14. O’Rourke is winning 18-34 voters 55-37, while losing the 65+ set 58-38. The future looks bright, but we can’t afford to wait for the future! A lot will be riding on O’Rourke’s ability to get out the young vote. Do-or-die, pretty much.

In the end, with both candidates solidifying the support of their base, everything will turn on O’Rourke’s ability to squeeze out the most votes from self-styled independents.

There’s another graph showing Beto leading Cruz 49-44 among indies, but that’s not the point I’m working on here. The point is that Civiqs has Cruz up only two on O’Rourke, which is the closest the race has been shown to be, and it’s got Republicans – perhaps sincerely, perhaps performatively – freaked out, as a recent fundraising email shows.

You’d think a poll like this would set the world on fire, but other than the GOP email there hasn’t been much out there. This is probably because Civiqs uses some non-standard methodology. Here’s Gardner Selby doing an explainer/fact check thing on it:

To our inquiry, the director of Civiqs, Drew Linzer, told us the poll showing O’Rourke just behind Cruz, available to Civiqs subscribers, was accurately portrayed by Moulitsas.

By email, Linzer further said the poll, not commissioned or sponsored by a client, relied on responses since February 2017 from respondents engaged over the web as volunteers for Civiqs survey panels who self-identified as Texas residents and registered voters.

Civiqs says in the methodology section of its website that it maintains a growing list of Americans who have agreed to take polls at civiqs.com. Panelists reside in all 50 states and Washington, D.C., and match the contours of the United States along key geographic, demographic and ideological lines,” Civiqs says.

We asked Linzer, a former political scientist at Emory University, to talk about relying on responses collected over more than 17 months. “The lengthy timeframe,’Linzer replied, “enables Civiqs to perform long-term daily tracking in a sustainable, methodologically rigorous manner.” Civiqs, he said, applies “specialized trendline fitting statistical models” to ensure the results match current public opinion.

Linzer said by phone that for the Texas poll, Civiqs emailed 40 to 50 respondents a day, asking each one about their Texas residence and registered voter status and stance on the Senate candidates, with those results over time feeding the trend line.

As of July 4, the latest date shown on the trend line posted by Moulitsas, “the most recent day’s worth of data has the greatest impact on the result,” Linzer said, though responses from previous days mattered a lot, too, he said. Linzer said responses gathered in 2017 also contributed to the result, but only “infinitesimally so.”

“It’s complicated. It is a little different from what a traditional pollster does” in reaching one set of respondents over a few consecutive days, Linzer said.

Make of that what you will, basically. FiveThirtyEight doesn’t include Civiqs in its pollster rankings, perhaps (I’m hypothesizing) because Civiqs’ methodology is such that they don’t meet the 538 definition of a “pollster”. Whatever the case, I’m glad Selby wrote about this because I had no idea what to make of this. I still don’t, but at least I feel like I understand the dimensions of the question better. Also, I’m going to get a second post out of this, which is always a good thing from my perspective. You’ll have to wait till tomorrow on that one.

We are going to get some Congressional polling soon

Nate Cohn on Twitter:


In case you can’t tell from the picture, CDs 07, 23, and 32 are included among the districts that will be – actually, probably are already being – polled. We have one poll from CD07 done by the DCCC a month ago, and it showed a two point lead for John Culberson. I’m not aware of any polling in either of the other districts, but I’ll be very interested to see what we get here. Cohn mentions that he thinks the writeups of these polls will be out next week. I can hardly wait.

Gravis: Cruz 51, O’Rourke 42

Here’s a new poll of four statewide races in Texas, for which there may or may not be any news coverage. The executive summary:

Gravis Marketing, a nonpartisan research firm, conducted a random survey of 602 likely voters across Texas. The poll was conducted from July 3rd through July 7th and has a margin of error of ±4.0%. The totals may not round to 100% because of rounding. The survey was conducted using an online panel of cell phone users and interactive voice responses. The results are weighted by voting demographics. The poll was paid for by Gravis Marketing.

Rep. Beto O’Rourke

As there is no news story to excerpt, I’ll just go straight to the results:

US Senate: Ted Cruz 51, Beto O’Rourke 42

Governor: Greg Abbott 51, Lupe Valdez 41

Lt. Governor: Dan Patrick 46, Mike Collier 44

Attorney General: Ken Paxton 45, Justin Nelson 41

There are a bunch of approval and issue questions in the polling memo, so feel free to browse through it. I will note two things. One is that Gravis is rated as a C+ pollster by FiveThirtyEight, better than some but worse than many others. Like Quinnipiac, they have no record in Texas prior to this year that I’m aware of. Two, while I haven’t spent any time critiquing subsamples in the polls we’ve seen so far, I have to say that the subsamples in this poll are nuts. Somehow, Gravis found the most ridiculously and unbelievably Republican group of 18-29 year olds and Hispanics I’ve ever seen, as well as the least hostile-to-Democrats Anglos. I have no explanation for this, and to some extent it doesn’t really matter. It is what it is, and what it is is another data point. And that data point brings the Senate poll average to 46.9 for Cruz, and 40.0 for O’Rourke.

I heard about this poll via a campaign email from Mike Collier, who for obvious reasons wanted to tout this result. (The TDP subsequently posted about it.) The low “don’t know/no answer” rate for the Lite Guv and AG questions is suspicious, but maybe that’s a function of their “likely voter” screen. Collier trails Patrick 50.4 to 42.8 among white voters, which is why he is so close in the race despite trailing 43.8 to 36.4 among Hispanics and leading by a mere 57.9 to 34.9 among blacks. Did I make my incredulity about this polls’ subsamples clear enough? You see some wacky stuff sometimes when the subgroups are small, but good Lord. As I’ve said, it’s a data point. Don’t make any more of it than that.

UT/Trib: Cruz 41, O’Rourke 36, part 2

We pick up where we left off.

Republican Ted Cruz leads Democrat Beto O’Rourke 41 percent to 36 percent in the general election race for a Texas seat in the U.S. Senate, according to the latest University of Texas/Texas Tribune Poll.

Neal Dikeman, the Libertarian Party nominee for U.S. Senate, garnered 2 percent, according to the survey. And 20 percent of registered voters said either that they would vote for someone else in an election held today (3 percent) or that they haven’t thought enough about the contest to have a preference (17 percent).

In the governor’s race, Republican incumbent Greg Abbott holds a comfortable 12-percentage-point lead over Democratic challenger Lupe Valdez — the exact same advantage he held over Democrat Wendy Davis in an early-summer poll in 2014. Abbott went on to win that race by 20 percentage points. In this survey, Abbott had the support of 44 percent to Valdez’s 32 percent. Libertarian Mark Tippetts had the support of 4 percent of registered voters, while 20 percent chose “someone else” or said they haven’t made a choice yet.

[…]

The June UT/TT Poll, conducted from June 8 to June 17, is an early look at the 2018 general election, a survey of registered voters — not of the “likely voters” whose intentions will become clearer in the weeks immediately preceding the election. If recent history is the guide, most registered voters won’t vote in November; according to the Texas Secretary of State, only 34 percent of registered voters turned out in 2014, the last gubernatorial election year.

The numbers also reflect, perhaps, the faint rumble of excitement from Democrats and wariness from Republicans who together are wondering what kind of midterm election President Donald Trump might inspire. The last gubernatorial election year in Texas, 2014, came at Barack Obama’s second midterm, and like his first midterm — the Tea Party explosion of 2010 — it was a rough year for Democrats in Texas and elsewhere. As the late social philosopher Yogi Berra once said, this year could be “Déjà vu all over again.”

Accordingly, voter uncertainty rises in down-ballot races where even previously elected officials are less well known. Republican incumbent Dan Patrick leads Democrat Mike Collier in the contest for lieutenant governor, 37 percent to 31 percent. Kerry McKennon, the Libertarian in that race, had the support of 4 percent of the registered voters surveyed, while the rest said they were undecided (23 percent) or would vote for someone other than the three named candidates (5 percent).

“As you move down to races that are just less well known, you see the numbers drop,” said Daron Shaw, a government professor at the University of Texas at Austin and co-director of the poll. “They drop more for the Republicans. Part of that reflects the visibility of those races, and of those candidates.”

Henson said Patrick and other down-ballot incumbents work in the shadow of the governor, especially when the Legislature is not in in session. “That said, he’s still solid with the Republican base, though he lags behind Abbott and Cruz in both prominence and popularity,” he said. “There’s nothing unusual about that.”

And indecision marks the race for Texas attorney general, where Republican incumbent Ken Paxton has 32 percent to Democrat Justin Nelson’s 31 percent and 6 percent for Libertarian Michael Ray Harris. Four percent of registered voters said they plan to vote for someone else in that race and a fourth — 26 percent — said they haven’t chosen a favorite.

Nelson and Harris are unknown to statewide general election voters. Paxton, first elected in 2014, is fighting felony indictments for securities fraud — allegations that arose from his work as a private attorney before he was AG. He has steadily maintained his innocence, but political adversaries are hoping his legal problems prompt the state’s persistently conservative electorate to consider turning out an incumbent Republican officeholder.

“If you’ve heard anything about Ken Paxton in the last four years, more than likely you’ve heard about his legal troubles,” said Josh Blank, manager of polling and research at UT’s Texas Politics Project. Henson added a note of caution to that: There’s also no erosion in Ken Paxton support by the Republican base. This reflects some stirrings amongst the Democrats and Paxton’s troubles. But it would premature to draw drastic conclusions for November based upon these numbers from June.”

Shaw noted that the support for the Democrats in the three state races is uniform: Each has 31 percent or 32 percent of the vote. “All the variability is on the Republican side, it seems to me,” he said. When those voters move away from the Republican side, Shaw said, “they move not to the Democrats but to the Libertarian or to undecided.”

Trump is still getting very strong job ratings from Republican voters — strong enough to make his overall numbers look balanced, according to the poll. Among all registered voters, 47 percent approve of the job the president is doing, while 44 percent disapprove. Only 8 percent had no opinion.

See here for yesterday’s discussion. Before we go any further, let me provide a bit of context here, since I seem to be the only person to have noticed that that Trib poll from June 2014 also inquired about other races. Here for your perusal is a comparison of then and now:


Year    Office  Republican  Democrat  R Pct  D Pct
==================================================
2014    Senate      Cornyn   Alameel     36     25
2018    Senate        Cruz  O'Rourke     41     36

2014  Governor      Abbott     Davis     44     32
2018  Governor      Abbott    Valdez     44     32

2014  Lite Guv     Patrick       VdP     41     26
2018  Lite Guv     Patrick   Collier     37     31

2014  Atty Gen      Paxton   Houston     40     27
2018  Atty Gen      Paxton    Nelson     32     31

So four years ago, Wendy Davis topped Dems with 32%, with the others ranging from 25 to 27. All Dems trailed by double digits (there were some closer races further down the ballot, but that was entirely due to lower scores for the Republicans in those mostly obscure contests). Republicans other than the oddly-underperforming John Cornyn were all at 40% or higher. The Governor’s race was the marquee event, with the largest share of respondents offering an opinion.

This year, Beto O’Rourke leads the way for Dems at 36%, with others at 31 or 32. Abbott and Ted Cruz top 40%, but Dan Patrick and Ken Paxton are both lower than they were in 2014, with Paxton barely ahead of Justin Nelson. Only Abbott has a double-digit lead, with the other three in front by six, five, and one (!) points.

And yet the one quote we get about the numbers suggests that 2018 could be like 2010 or 2014? I must be missing something. Hey, how about we add in some 2010 numbers from the May 2010 UT/Trib poll?


Year    Office  Republican  Democrat  R Pct  D Pct
==================================================
2014    Senate      Cornyn   Alameel     36     25
2018    Senate        Cruz  O'Rourke     41     36

2010  Governor       Perry     White     44     35
2014  Governor      Abbott     Davis     44     32
2018  Governor      Abbott    Valdez     44     32

2010  Lite Guv    Dewhurst       LCT     44     30
2014  Lite Guv     Patrick       VdP     41     26
2018  Lite Guv     Patrick   Collier     37     31

2010  Atty Gen      Abbott Radnofsky     47     28
2014  Atty Gen      Paxton   Houston     40     27
2018  Atty Gen      Paxton    Nelson     32     31

There was no Senate race in 2010. I dunno, maybe the fact that Republicans outside the Governor’s race are doing worse this year than they did in the last two cycles is worth noting? Especially since two of them were first-time statewide candidates in 2014 and are running for re-election this year? Or am I the only one who’s able to remember that we had polls back then?

Since this cycle began and everyone started talking about Democratic energy going into the midterms, I’ve been looking for evidence of said energy here in Texas. There are objective signs of it, from the vast number of candidates running, to the strong fundraising numbers at the Congressional level, to the higher primary turnout, and so on. I haven’t as yet seen much in the poll numbers to show a Democratic boost, though. As we’ve observed before, Beto O’Rourke’s numbers aren’t that different than Bill White or Wendy Davis’ were. A bit higher than Davis overall, but still mostly in that 35-42 range. However, I did find something in the poll data, which was not in the story, that does suggest more Dem enthusiasm. Again, a comparison to 2010 and 2014 is instructive. In each of these three polls, there’s at least one “generic ballot” question, relating to the US House and the Texas Legislature. Let’s take a look at them.

If the 2010 election for [Congress/Lege] in your district were held today, would you vote for the Democratic candidate, the Republican candidate, or haven’t you thought enough about it to have an opinion?

2010 Congress – GOP 46, Dem 34
2010 Lege – GOP 44, Dem 33

If the 2014 election for the Texas Legislature in your district were held today, would you vote for the Democratic candidate, the Republican candidate, or haven’t you thought about it enough to have an opinion?

2014 Lege – GOP 46, Dem 38

If the 2018 election for [Congress/Lege] in your district were held today, would you vote for [RANDOMIZE “the Democratic candidate” and “the Republican candidate”] the Democratic candidate, the Republican candidate, or haven’t you thought about it enough to have an opinion?

2018 Congress – GOP 43, Dem 41
2018 Lege – GOP 43, Dem 42

Annoyingly, in 2014 they only asked that question about the Lege, and not about Congress. Be that as it may, Dems are up in this measure as well. True, they were up in 2014 compared to 2010, and in the end that meant nothing. This may mean nothing too, but why not at least note it in passing? How is it that I often seem to know these poll numbers better than Jim Henson and Daron Shaw themselves do?

Now maybe the pollsters have changed their methodology since then. It’s been eight years, I’m sure there have been a few tweaks, and as such we may not be doing a true comparison across these years. Even if that were the case, I’d still find this particular number worthy of mention. Moe than two thirds of Texas’ Congressional delegation is Republican. Even accounting for unopposed incumbents, the Republican share of the Congressional vote ought to be well above fifty percent in a given year, yet this poll suggests a neck and neck comparison. If you can think of a better explanation for this than a higher level of engagement among Dems than we’re used to seeing, I’m open to hearing it. And if I hadn’t noticed that, I don’t know who else might have.

UT/Trib: Cruz 41, O’Rourke 36

Well, what do you know?

Republican Ted Cruz leads Democrat Beto O’Rourke 41 percent to 36 percent in the general election race for a Texas seat in the U.S. Senate, according to the latest University of Texas/Texas Tribune Poll.

Neal Dikeman, the Libertarian Party nominee for U.S. Senate, garnered 2 percent, according to the survey. And 20 percent of registered voters said either that they would vote for someone else in an election held today (3 percent) or that they haven’t thought enough about the contest to have a preference (17 percent).

In the governor’s race, Republican incumbent Greg Abbott holds a comfortable 12-percentage-point lead over Democratic challenger Lupe Valdez — the exact same advantage he held over Democrat Wendy Davis in an early-summer poll in 2014. Abbott went on to win that race by 20 percentage points. In this survey, Abbott had the support of 44 percent to Valdez’s 32 percent. Libertarian Mark Tippetts had the support of 4 percent of registered voters, while 20 percent chose “someone else” or said they haven’t made a choice yet.

[…]

The June UT/TT Poll, conducted from June 8 to June 17, is an early look at the 2018 general election, a survey of registered voters — not of the “likely voters” whose intentions will become clearer in the weeks immediately preceding the election. If recent history is the guide, most registered voters won’t vote in November; according to the Texas Secretary of State, only 34 percent of registered voters turned out in 2014, the last gubernatorial election year.

The numbers also reflect, perhaps, the faint rumble of excitement from Democrats and wariness from Republicans who together are wondering what kind of midterm election President Donald Trump might inspire. The last gubernatorial election year in Texas, 2014, came at Barack Obama’s second midterm, and like his first midterm — the Tea Party explosion of 2010 — it was a rough year for Democrats in Texas and elsewhere. As the late social philosopher Yogi Berra once said, this year could be “Déjà vu all over again.”

Accordingly, voter uncertainty rises in down-ballot races where even previously elected officials are less well known. Republican incumbent Dan Patrick leads Democrat Mike Collier in the contest for lieutenant governor, 37 percent to 31 percent. Kerry McKennon, the Libertarian in that race, had the support of 4 percent of the registered voters surveyed, while the rest said they were undecided (23 percent) or would vote for someone other than the three named candidates (5 percent).

“As you move down to races that are just less well known, you see the numbers drop,” said Daron Shaw, a government professor at the University of Texas at Austin and co-director of the poll. “They drop more for the Republicans. Part of that reflects the visibility of those races, and of those candidates.”

Henson said Patrick and other down-ballot incumbents work in the shadow of the governor, especially when the Legislature is not in in session. “That said, he’s still solid with the Republican base, though he lags behind Abbott and Cruz in both prominence and popularity,” he said. “There’s nothing unusual about that.”

And indecision marks the race for Texas attorney general, where Republican incumbent Ken Paxton has 32 percent to Democrat Justin Nelson’s 31 percent and 6 percent for Libertarian Michael Ray Harris. Four percent of registered voters said they plan to vote for someone else in that race and a fourth — 26 percent — said they haven’t chosen a favorite.

Nelson and Harris are unknown to statewide general election voters. Paxton, first elected in 2014, is fighting felony indictments for securities fraud — allegations that arose from his work as a private attorney before he was AG. He has steadily maintained his innocence, but political adversaries are hoping his legal problems prompt the state’s persistently conservative electorate to consider turning out an incumbent Republican officeholder.

“If you’ve heard anything about Ken Paxton in the last four years, more than likely you’ve heard about his legal troubles,” said Josh Blank, manager of polling and research at UT’s Texas Politics Project. Henson added a note of caution to that: There’s also no erosion in Ken Paxton support by the Republican base. This reflects some stirrings amongst the Democrats and Paxton’s troubles. But it would premature to draw drastic conclusions for November based upon these numbers from June.”

Shaw noted that the support for the Democrats in the three state races is uniform: Each has 31 percent or 32 percent of the vote. “All the variability is on the Republican side, it seems to me,” he said. When those voters move away from the Republican side, Shaw said, “they move not to the Democrats but to the Libertarian or to undecided.”

Trump is still getting very strong job ratings from Republican voters — strong enough to make his overall numbers look balanced, according to the poll. Among all registered voters, 47 percent approve of the job the president is doing, while 44 percent disapprove. Only 8 percent had no opinion.

Our seven-poll average now stands at Cruz 46.3, O’Rourke 39.7. Remember when that second Q poll, the one that had Cruz up by 11, became the One True Result? we now have four polls since then, and all of them are in the five-to-eight points range, which is to say all right arounf the polling average. Imagine that. This result, one of the better ones for O’Rourke, occurs in the context of good approval numbers for Donald Trump. In fact, Trump’s numbers have been mostly above water lately, yet Beto remains competitive. Here’s a summary:

UT/Trib, February 2017, 46 approve/44 disapprove
UT/Trib, June 2017, 43 approve/51 disapprove
UT/Trib, October 2017, 45 approve/49 disapprove
UT/Trib, February 2018, 46 approve/46 disapprove
Quinnipiac Senate poll, April 2018, 43 approve/51 disapprove
Quinnipiac Senate poll, May 2018, 47 approve/47 disapprove
PPP Senate poll, June 2018, 49 approve/46 disapprove
CBS/YouGov Senate poll, June 2018, 50 approve/50 disapprove
UT/Trib Senate poll, June 2018, 47 approve/44 disapprove

Not too surprisingly, Beto’s best showing was in that first Q poll. The fact that he’s consistently within single digits despite Trump being even or better in approval is encouraging, and suggests things could really get interesting if his numbers ever soften.

What about the Governor’s race? There have been eight polls of the Senate race so far, but this is only the third poll to include the Governor’s race. Here’s how those compare:

Quinnipiac, April

Cruz 47, Beto 44
Abbott 49, Valdez 40

Quinnipiac, May

Cruz 50, Beto 39
Abbott 53, Valdez 44

UT/Trib, June

Cruz 41, Beto 36
Abbott 44, Valdez 32

So Cruz runs two or three points behind Abbott, while Beto runs four or five points ahead of Valdez. Some of the latter may be a function of name recognition, but overall I’d be comfortable saying Beto would do a few points better overall than Valdez. I hesitate to draw broad conclusions, but it seems clear Beto is on a path to outperform Valdez, and quite possibly the rest of the Dem ticket. By how much is an open question, and I would remind everyone that other than Bill White in 2010, the statewide results in both 2010 and 2014 landed in a pretty narrow range. Keep an eye on this, but don’t spend too much time thinking about it yet.

I have more to discuss with this poll, but this post is already long. I will pick things up tomorrow.

CBS/YouGov: Cruz 44, O’Rourke 36 (RVs)

Time for another poll.

Rep. Beto O’Rourke

In Texas’ Senate race, incumbent Republican Ted Cruz has a 10-point lead over Democrat Beto O’Rourke among likely voters. Cruz benefits from strong support from his own party and has an advantage among independents as well. O’Rourke is supported by Democrats, leads with Hispanics and has an edge with women. Cruz performs well with whites and men.

Cruz also has an overall job approval rating of 54 percent in Texas among registered voters, a bit higher than President Trump’s (50 percent) in the state.

On the matter of separating families specifically, both Cruz and O’Rourke get net positive ratings (largely driven by support from their own parties), although three in 10 voters do not have an opinion about O’Rourke on this, as he may be less known to voters than Cruz.

Poll data is here. They also did Arizona and Florida’s Senate races, if those interest you. For the Texas Senate race (question 6), the result from the full 1,025-person sample of registered voters was 44-36 as indicated in the headline. It was in the smaller (821 respondents) “likely voter” group that Cruz was up 50-40. I’m skeptical of likely voter screens at this early point in time, and all of the other poll results I have on the sidebar are for RVs, so for comparison purposes that’s the one I’m going with. The average of the six polls I’m using (all but the WPA one from January 5) now has Cruz at 47.2, with 40.2 for O’Rourke.

SD10 poll: Powell 46, Burton 42

From the Trib’s email newsletter:

Beverly Powell

State Sen. Konni Burton’s Democratic challenger, Beverly Powell, has a 4-point lead over the Colleyville Republican, according to a new poll from Powell’s campaign.

The survey of 600 likely voters found Powell, a former Burleson ISD trustee, receiving 46 percent of the vote and Burton 42 percent, with 11 percent undecided. Powell expanded her lead to 9 points — 53 percent to 44 percent — after respondents were read positive descriptions of both candidates.

Burton’s District 10 is regarded as the most competitive Texas Senate district in November, and Powell’s campaign says the survey shows it’s “in a strong position to win.”

“I think the results make clear that Beverly’s commitment to education and her pro-business background resonates with voters in the district,” Powell campaign manager Garry Jones tells us. “And I think it shows Konni Burton has really ignored voters in SD-10, taking her marching orders from [Lt. Gov.] Dan Patrick and Empower Texans for the past two sessions instead of listening to the business interests and constituents of Tarrant County.”

The poll also asked likely voters in the battleground district about the U.S. Senate race and found the Republican incumbent, Ted Cruz, trailing Democratic challenger Beto O’Rourke by 4 points, 49 percent to 45 percent. Six percent were unsure.

Democratic pollster Keith Frederick conducted the survey from May 14-21 using phone interviews, 38 percent of which included cell phones. The margin of error was plus-or-minus 4 points.

The polling memo is here, though it doesn’t tell you much more. It does indicate that the sample self-identified as 40% Republican, 36% Democrat, and the rest Independent. We’ve discussed the reasons to be cautious about internal polls before, and those reasons apply here. Powell won the primary in March so it’s not unreasonable to think this is not the first poll her team has commissioned, and the “informed voter” part of it is surely aimed at potential funders. This has been a contentious race from the get-go, in part because it’s the one truly swingy Senate seat. Even in the wipeout of 2014, it wasn’t that red – Greg Abbott and Dan Patrick won it by about eight points, while downballot Republicans won it by about ten. In a context where the statewide split is something like 55-45 instead of the 60-40 it was four years ago, this district is basically 50-50. If nothing else, this result is consistent with the US Senate polls we’ve seen. Link via the Lone Star Project, which also teases an encouraging poll in SD16, which I’m trying to learn more about.

GQR: Cruz 49, O’Rourke 43

Would you like another Senate poll result? Of course you would.

Rep. Beto O’Rourke

A new poll of likely voters, commissioned by End Citizens United (ECU) and conducted by Greenberg Quinlan Rosner, shows a single digit contest in the Texas Senate election with Representative Beto O’Rourke trailing Senator Ted Cruz by just six points, 43 to 49 percent with eight percent undecided. Click here to read the full polling memo.

“This poll is another indication of the real energy behind Beto’s campaign and his call to unrig Washington. Now he’s in striking distance of Senator Cruz,” said ECU President Tiffany Muller. “Beto is running a campaign centered around real conversations with the people of Texas, and the more people learn about him, the stronger his campaign grows. Beto is the first challenger ECU endorsed this cycle because of his determination to end the corrupting influence of Big Money in politics and give people – not special interests – the most powerful voice in Washington.”

Greenberg Quinlan Rosner surveyed 1,000 likely voters from May 29 – June 5. The margin of error is +/- 3.1%.

The polling echos an ECU poll from earlier this year, which had O’Rourke within eight points of Cruz. It also found that 63% of Texans are more likely to support a candidate who rejects corporate special interest money.

As noted, the polling memo is here, though there’s basically a bar graph depicting each candidate’s total, and that’s it. No poll questions, no crosstabs, no complementary results, nothing. GQR is a good pollster, so I assume this is a reasonably well-done survey, but we’ll have to accept that this is all we know about it. The earlier poll referenced is that January PPP poll, also done for End Citizens United, that had Cruz up 45-37. This poll is right in line with the average, which now stands at Cruz 47.8, O’Rourke 41.0. A clear and consistent lead for Cruz, but a closer race than what we’re used to seeing.

Looking back at 2010 and 2014

I’ve talked a lot about polls in the past week, so I thought I’d take a minute and look back at the polling data that we had as of this time in the 2010 and 2014 elections, to see if we can learn anything. The polls those years were about Governor’s races while this year is focused on the Senate race, but that’s all right. I’m not intending for this to be a straight apples-to-apples comparison, just more of a general feel. So with no further ado:

PPP, June 2010: Perry 43, White 43
UT/Trib, May 2010: Perry 44, White 35
Rasmussen, May 2010: Perry 51, White 38
Rasmussen, April 2010: Perry 48, White 44
UT/Trib, Feb 2010: Perry 44, White 35
PPP, Feb 2010: Perry 48, White 42

Avg: Perry 46.3, White 39.5

Boy, were we optimistic in the early days of 2010. Bill White was a top-notch candidate, coming off a successful tenure as Mayor of Houston with high popularity numbers and a strong fundraising apparatus. The polls supported that optimism, with that June result showing a tied race. Rick Perry, in the meantime, was coming off a 39% re-election in 2006 and a bruising primary win over then-Sen. Kay Bailey Hutchison. There were lots of reasons to think that people had gotten tired of Perry and his schtick after a decade in office, and the enthusiasm from the 2008 election was still felt and seen as a harbinger of things to come.

We know how this movie ended. The thing was, it wasn’t apparent that it was headed that way till the final days. Polls from September and early October continued to show a tight race. It wasn’t really until early voting had started and the last polls were published that we began to see the downward trends. It wasn’t a lack of Democratic enthusiasm that doomed White and the rest of the ticket – turnout was up from 2006, not that that was saying much – but Republican turnout was off the charts, swamping Democratic boats across the country and wiping out large swaths of the Democratic caucus in the Legislature. We didn’t know it in June, but there was a very ill wind about to blow.

UT/Trib, June 2014: Abbott 44, Davis 32
PPP, April 2014: Abbott 51, Davis 37
Rasmussen, March 2014: Abbott 53, Davis 41
ECPS, March 2014: Abbott 49, Davis 42
UT/Trib, Feb 2014: Abbott 47, Davis 36

Avg: Abbott 48.8, Davis 37.6

There are a lot of ways in which 2014 was like 2010 – initial excitement and optimism, high-profile candidate who drew national attention and had good fundraising chops, all ending in a gut-wrenching wipeout. One major way in which things were very different is that the early polls did not support that initial optimism in 2014. I distinctly remember writing a lot of words about why 2014 was going to be different and not at all like 2010. We were so young and innocent then. We also had a lot more warning about the impending doom we faced, as the next poll result after this one had Abbott up by 16, and in only two of the last seven polls was Davis within single digits. I was right about one thing – Republican turnout was in fact down from 2010. It’s just that Democratic turnout was as best flat from 2010, despite the endlessly-hyped presence of Battleground Texas, and that all added up to roughly a 2002-style outcome.

PPP, June 2018: Cruz 48, O’Rourke 42
Quinnipiac, May 2018: Cruz 50, O’Rourke 39
Quinnipiac, April 2018: Cruz 47, O’Rourke 44
PPP, Jan 2018: Cruz 45, O’Rourke 37

Avg: Cruz 47.5, O’Rourke 40.5

I discussed these last week, when that PPP poll hit. I’m dropping the Wilson Perkins result from this calculation, as it was done in the latter days of 2017, but if you insist on including it the averages change to Cruz 48.4, O’Rourke 39.2. That’s not as good as the 2010 average – if you just take these four polls, it’s basically even with 2010 – but it’s about two points better than 2014, three points better without the outlier. We don’t know how this one will end, of course, and it remains to be seen where the polls go from here. I just wanted to provide some context, so there you have it.

From the “Only negative results apply” department

There’s one paragraph in this story about Beto O’Rourke finishing a quest to visit all 254 counties in Texas that really makes me grind my teeth.

When he’s not behind the wheel, O’Rourke has proven to be a formidable fundraiser, regularly outperforming his more famous opponent. In the first quarter of 2018, he raised $6.7 million, more than any other Democratic candidate for U.S. Senate raised over the same period. But his performance in the Democratic primary in March was disappointing: Even as he coasted to a 38-point victory over challenger Selma Hernandez, O’Rourke lost several key counties along the Mexican border to the Houston activist. And a Quinnipiac University poll released last month found Cruz leading O’Rourke by 11 percentage points.

Yes, that Quinnipiac poll happened. It was also preceded by another Quinnipiac poll that gave Cruz a three-point lead, and followed a week later by a PPP poll that had it at six points. There are also clues from other polls, as well as from Congressional forecasting models that indicate a closer-than-expected state environment. But hey, mentioning that one poll showing the widest spread is good enough, because it provides a sense of “balance” or something. Pardon me for a minute while I bang my head on the desk.

As for the rest of the story, it’s fine. The subject of O’Rourke’s journey around Texas and his more in-person campaign style has been told before and will surely be told again. And as I’ve said before, we don’t really know if this is a more-effective strategy than what has been done before, but it’s not like the standard practice has a stellar track record, and this seems like as good a year as any to try something a little different. It also may be the case that this is the best method, but it is still destined to fall short. I just want us to learn the right lessons from it, whatever the outcome.

PPP: Cruz 48, O’Rourke 42

Hey, look, another poll.

Rep. Beto O’Rourke

Democrat Beto O’Rourke trails U.S. Sen. Ted Cruz, R-Texas, by 6 percentage points, according to a new poll commissioned by Giffords, the gun control group started by former U.S. Rep. Gabrielle Giffords, D-Ariz.

The survey, done by the left-leaning firm Public Policy Polling, found Cruz leading O’Rourke 48-42 percent, with 10 percent undecided. O’Rourke’s deficit narrowed to 1 point — 44-45 percent — after respondents were read a series of statements about his and Cruz’s positions on guns, including O’Rourke’s support for an assault weapons ban.

Cruz has seized on that support to criticize O’Rourke as too liberal for Texas, saying he’s campaigning on “aggressive gun control.” But the poll found a majority of voters — 51 percent — were much or somewhat more likely to get behind O’Rourke after hearing that he wants to outlaw assault weapons.

[…]

The poll surveyed 861 Texas voters from May 21-22 using automated telephone interviews. It had a margin of error of +/- 3.3 percentage points.

You can see the poll data here. As a reminder, here are the other polling results we’ve had so far:

WPA, Jan 5: Cruz 52, O’Rourke 34
PPP, Jan 27: Cruz 45, O’Rourke 37
Quinnipiac, April 19: Cruz 47, O’Rourke 44
Quinnipiac, May 31: Cruz 50, O’Rourke 39

Remember how those two Q-polls gave everyone whiplash, going from “The race is statistically tied!” to “So much for that so-called ‘blue wave’!” in what seemed like minutes. Maybe it would be helpful to point out that if you take the average of those two polls – which is to say, treat them as a combined sample rather than two separate and independent data points – you get Cruz 48.5, O’Rourke 41.5, or something very close to this result. If you average all five polls, you get Cruz 48.4, O’Rourke 39.2, and the main reason Beto’s total is that low is that one early Republican firm’s poll, which I think we can all agree now looks a bit like an outlier.

One other point to make is that in this PPP sample, Donald Trump has an approval rating of 49 approve and 46 disapprove. He was at 47/44 in that second Q-poll, the one with Cruz up by 11. Trump’s approval has bounced around in various polls, not all of which included horse-race questions, and not too surprisingly where there are race questions the Dems tend to do better the worse Trump does. This result is an exception to that; indeed, that earlier PPP poll showing Cruz up by 8 did so in the context of Trump’s approval being negative by three points. In some ways, I find this the most encouraging part of the PPP result.

As for the “now that you know this about the candidates’ views on gun control” stuff, you can take that as you want. At least O’Rourke will have the wherewithal to make sure people know about that aspect of Cruz’s candidacy, but beyond that I’m back in “it’s just another data point” territory. I want to believe, but I need further convincing.

Checking in on the Congressional forecast

Now that our November lineups are finalized, I thought I’d check in once again on the 2018 Congressional race forecast, from G. Elliott Morris of The Crosstab. I last wrote about this in December, at a time when the generic ballot preference was consistently showing a double-digit lead for Democrats. The polls are closer now but the Dems still have a sizable lead. Here’s how things project in Texas, according to this model:


Dist  Flip%  Margin  16 Marg  14 Marg
=====================================
CD02  14.3%   -10.6    -18.6    -33.7
CD03   7.4%   -14.4    -25.1    -37.1
CD06  19.2%   - 8.7    -16.0    -21.3
CD07  49.1%   - 0.2    -11.5    -31.4
CD10  19.0%   - 7.5    -16.1    -22.6
CD14   5.5%   -13.8    -20.7    -22.8
CD17   4.6%   -14.7    -22.4    -28.9
CD21  19.3%   - 8.6    -18.6    -26.0
CD22  18.6%   - 7.7    -16.0    -33.3
CD23  86.8%     9.7    - 0.5    -15.5
CD24  26.1%   - 5.5    -16.4    -30.9
CD25  11.3%   -10.5    -21.1    -22.5
CD27   4.3%   -17.1    -23.6    -30.3
CD31  10.8%   -10.7    -19.5    -27.7
CD32  39.9%    -2.2    -12.1    -23.7

These data points are from Sunday; there are daily updates, which move things a bit one way or the other. “Flip% is the probability that the Democratic challenger will win that district. “Margin” is the difference between the projected Republican share of the vote and the projected Democratic share, so a positive number is a Democratic win and a negative number is a Republican win. (Obviously, that’s a point within a range, not a gospel truth, hence the Flip% probability.)

“16 Marg” and “14 Marg” are my additions, as earlier versions of this table had similar values. As with the Margin column it’s the difference between Republican and Democratic performance. However, while Margin compares Congressional candidate percentages, we can’t reliably do that for 2016 and 2014, since some of these races were unopposed. As is my custom, I used Court of Criminal Appeals races – CCA3 for 2014, CCA6 for 2016. This provides another illustration of my point from that post about the CD07 poll. You can’t have tighter Congressional races up and down the ballot and not have tighter statewide races. It may be that Morris’ model is wrong, and it may be that the totality of statewide polling data will make it clear that he’s being too bullish on the Dems. All I’m saying is that stuff like this has to be taken into account as well.

The differences in the margins fascinate me. For the 2014 to 2016 shift, most of that reflects the kind of turnout pattern we have been used to seeing in Presidential versus non-Presidential years lately. The effect is much more pronounced in urban areas, and in this case it was greatly enhanced by the Trump effect, with a side of demographic change and voter registration efforts. Projected shifts from 2016 to 2018 are nearly all about the national atmosphere. It’s kind of amazing to me that the district projected to be the most flippable outside the top three is CD24, which has gotten maybe one percent of the attention that even some of the second-tier districts have gotten. Maybe that’s a blind spot in reporting, and maybe it’s a non-optimized opportunity on the Dems’ part. CDs 06, 10, and 22 all had smaller 2016 margins than CD24, so maybe they’ll catch up when all is said and done.

I’ll check in on this again in August or so. In the meantime, here’s a story about G. Elliott Morris, the guy who’s doing these projections. One way or another, his work will be closely scrutinized on November 7.

DCCC poll: Culberson 47, Fletcher 45

Game on.

Lizzie Fletcher

The U.S. House race between GOP incumbent John Culberson and Democratic challenger Lizzie Pannill Fletcher is generally expected to be closer than most in this traditionally Republican enclave of west Houston and the Harris County suburbs.

Now an internal Democratic poll of the 7th Congressional District shows it to be a statistical tie. The poll of district voters, released Friday by the Democratic Congressional Campaign Committee, found Fletcher within 2 points of Culberson, 45 percent to 47 percent. That is within the poll’s 4.9 percent margin of error.

[…]

The DCCC poll shows Fletcher leading Culberson by 8 points among women (50 percent to 42 percent), 20 points among independents (52 percent to 32 percent), and by 28 points among voters under 50 (57 percent to 29 percent).

Further proof that that the district could be in play: The poll found that a generic Democrat is within striking distance of a generic Republican – 46 percent to 47 percent. That’s tighter than the difference between Fletcher and Culberson, but still within the margin of error.

The Democratic poll also gave Culberson a net-negative favorability rating, with 32 percent of voters having a favorable view of the congressman, compared to 39 percent who don’t. Similarly, the poll found that 35 percent of voters approve of Culberson’s job performance, while 39 percent disapprove.

Meanwhile, Trump also remains underwater in a district, which he lost by 1.4 points in 2016. In the DCCC poll, 50 percent of Seventh District voters disapprove of his job performance, while 42 percent approve.

I first heard about this poll via G. Elliott Morris’s Twitter feed, but this story adds some details. Internal polls are generally treated with skepticism – scroll down to see the responses to that tweet for a couple of examples – and I want to talk about why that is first. The main reason why internal polls are looked at differently is because when an internal poll is released, you have no way of knowing how many other polls that particular campaign or committee might have done that they did not choose to release. In other words, the poll that gets released may be the most favorable of the bunch, cherry-picked to present a sunny view of the situation. Media and tracking polls are public, with all their results out there to be seen, so when there’s an outlier it tends to stand out. You just don’t know if an internal poll is an outlier or not.

The other reason why internal polls are different is that they are sometimes used for specific purposes like testing a message or attracting financial support. Polls that take the measure of a race, then “inform” the respondents about one of the candidates and re-ask the original question again at the end, are a common example of this. The Justin Nelson poll from December is in this category. There’s nothing wrong with this – it’s a valuable campaign tool – but since the result comes from an idealized scenario – in a real campaign, both candidates get to “inform” voters, assuming they have the resources to do so – these polls are not very useful as predictive tools.

For those reasons, and because full poll data is often not available, poll aggregators and election modelers tend to give internal polls less weight. All that said, this poll is an example of one we can probably take more seriously. For one thing, given that the runoff was less than two weeks ago, there very likely have not been any other polls done by the DCCC since Fletcher became the nominee. There’s (again, probably) nothing to cherry-pick from. The DCCC, which has now added Fletcher to its Red to Blue group, generally doesn’t try to convince funders to invest in a particular race, and for them to want to include CD07 as a race to target they’ll want accurate horse-race numbers. None of this means that they couldn’t have made optimistic assumptions about turnout or the makeup of the electorate – we don’t have the internal poll data, so who knows what they sampled from – but all pollsters have to make those judgments.

All things considered, I believe we can take this poll more or less at face value. Which is to say, it’s a data point, and we hope to see more of them to get a fuller picture of what may be happening. Given that, the way to think about this is not just for this race, which we believe will be close and competitive, but for how it fits into the bigger picture. For one thing, Democrats swept Harris County in 2016 while John Culberson was winning in CD07 by 12 points. If we’re in an election year where CD07 is truly a tossup, then that strongly implies an even better year for Democrats in the county. Even more than that Lina Hidalgo poll, this should be encouraging for Dems, and downright terrifying for Republicans.

But it’s not just Harris County. There are two big reasons why CD07 is and has been seen as a top pickup opportunity. The main reason is because Hillary Clinton carried the district in 2016, but as we have discussed here before, some of that was because of crossover voters. Like I said above, Culberson still won the district 56-44. The other, equally important, reason is that the national atmosphere is one that favors Democrats and strongly indicates that the Republican advantage in districts like CD07 will be greatly diminished. Put another way, we expect that more Democrats and fewer Republicans will vote than in other similar election years. And that’s not just true in CD07, and in other battleground districts like CD23 and CD32. It’s true across the board, and it’s factored into every election prediction model, like the Morris model. Scroll down to the “Forecasts for every House seat” section and compare his projected margin in each Congressional district to the actual margins from 2016 and 2014.

This is something that I don’t think has been absorbed by media outlets and pundits in this state, all of which comes very much to the fore when a statewide poll like the second one from Quinnipiac comes out. Greg Abbott, who carried Harris County by five points in 2014, carried CD07 by a 60-38 margin in 2014; Culberson won that year by a 63-35 score. Again, if we are in an election where CD07 is a tossup, then the effect of that will be felt statewide, not just countywide. More to the point, if we are in that election, then the same effect will be felt in every Congressional district in Texas. It will be felt more in some districts than in others, and in specific races with specific candidates with strengths and weaknesses that may counter or enhance the national mood. But it will be felt.

The point I’m making is that a poll like that second Quinnipiac poll may be right, and polls like the DCCC CD07 poll and the Hidalgo Harris County poll may be right, but they can’t all be right. If the Q-poll is right, the other two are almost certainly too optimistic about Democratic chances, and if the latter two are right, then that Q-poll is almost certainly understating Democratic statewide support. I wish the people who write about these things would take that into consideration when they do. We don’t know yet which view is right. The fact that these conflicting polls exist is almost certainly because everyone has a different idea of what that national atmosphere will be like, and how big its effect on Texas will be. If you’re skeptical of any effect here you need to explain why. For now at least, all I’m saying is that polls like these don’t exist in a vacuum. Don’t evaluate one without taking into consideration the others.