For much of the past decade-plus, Harris Co. experienced a phenomenon I called “partisan mood swings.” Democrats running countywide – potentially all of them – tended to win in presidential election years. Republicans running countywide – often all of them – tended to win in gubernatorial election years. The best, and only, explanation: straight-party voting.
More Democrats cast straight-party votes than Republicans in 2000, 2008, 2012 and 2016. More Republicans cast straight-party votes than Democrats in 2002, 2010 and 2014. The exceptions to this yoyoing pattern were:
- 2004 (Republicans by 45K), generally thought of as the nadir of Texas Democrats statewide (John Kerry lost by 23 points, and Democrats contested only three statewide offices and left scads of Republican-held offices unchallenged)
- 2006 (Democrats by 8K), the year Carole Keaton Strayhorn and Kinky Friedman’s independent gubernatorial bids siphoned off straight-party voters weary of Gov. Rick Perry (R) and indifferent to former one-term U.S. Rep. Chris Bell (D-Houston); and
- 2018 (Democrats by a record 105K).
Democrats swept all 42 countywide races in 2016 and all 87 of them in 2018. Republicans swept all 83 of them in 2014. Looking just at the countywide judicial races in 2012, Republicans won 21 and Democrats won 19, making it the only truly competitive year recently. In 2010, Republicans won 81 out of the 82 countywide races. The exception was Perry’s narrow loss to former Houston Mayor Bill White. In 2008, Democrats won all but eight of the countywide races.
If the county’s political mood was imagined as a yoyo, then the 2018 election broke the string. A record number of Democratic straight-party votes (516K) was cast that year, 32% more than in 2008. A record number of Republican straight-party votes (411K) was also cast in 2018, but the percentage increase since 2008 was a little more than half (18%) the Democrats’ gain, and the change since 2012 was negligible.
This significantly stronger growth in Democratic straight-party voting was spread across the county. The straight-party advantage shifted toward the Democrats in all but four of the county’s 25 state House districts, and all four of those districts were already held by Democrats.
And yet, all of that has resulted in just two seats flipping this decade, net of HD144, which flipped twice. In 2018, Democrats flipped Republican-held HD132 and HD135, ousting the incumbents. One more should have flipped (HD134), but its incumbent enjoyed sufficient crossover support to hang on. That district, and one more, are poised to flip this year.
On the Clock
HD134 (Lean Democratic): Rep. Sarah Davis (R-Houston) won re-election in 2018 despite the fact that former U.S. Rep. Beto O’Rourke (D-El Paso) defeated U.S. Sen. Ted Cruz (R) by 21 percentage points. Lt. Gov. Dan Patrick (R) lost the district to his Democratic challenger by 18 points, and even Gov. Greg Abbott (R) lost the district by 3 points.
Since 2012, the straight-party vote advantage in the district has shifted away from Davis’s party by more than 15K votes, the largest swing for any Texas House district held by a Republican at any time during the decade. In 2012, she enjoyed an 11,744-vote advantage in straight-party voting. It fell by 74% four years later to just over 3K votes. In 2018, she overcame a 3,359-vote deficit in straight-party voting by winning 65% of the vote from voters going down the ballot race-by-race (“full-ballot voters”). Cruz got 25% among the same group of voters. Put another way, about 12K voters in HD134 cast ballots for O’Rourke and Davis.
If Davis’s 2018 primary challenger been successful, the Democrat would almost certainly have won this seat by a double-digit margin. In 2018, HD134 was nearly 9 points bluer than the state as a whole. Six years earlier, it was nearly a full point redder than the state as a whole. The average Democrat won the district with 55% of the vote, measured head-to-head against the Republican, in 2018. In 2010, the most recent year of data before the district was drawn, the average Democrat received less than 39% of the vote.
Longtime Texas political observers will recall former Rep. Jim McReynolds (D-Lufkin), who served seven terms from 1997 until his defeat in the 2010 general election, a wipeout year for Democratic legislators. In an era where straight-party voting was far less prevalent than the current decade, McReynolds over-performed the rest of the Democratic ticket by as many as 19 points. In 2010, McReynolds over-performed his party by more than 12 points … and lost by nearly 14 points. There simply came a point when McReynolds’s ability to win enough crossover support was drowned out by a rising partisan tide.
Davis faces that same fate this year. Since 2012, the number of voters casting a straight Republican ballot has been largely flat. The number of Democratic straight-party voters doubled during the same period. Mitt Romney carried the district by nearly 15 points over then-President Obama in 2012. Four years later, Hillary Clinton beat Donald Trump here by more than 15 points – a nearly 30-point swing in just four years. As noted, O’Rourke added another 6 points to Clinton’s performance.
Our model gives Davis a full 5 points based on her average performance over her fellow Republicans during the last three election cycles. Even with this advantage, it projects her losing by nearly 5 points.
We say this without even acknowledging the challenger: Houston attorney Ann Johnson, the well-funded Democratic who lost to Davis in 2012, 55%-45%. Johnson raised $642K for that race, of which nearly $475K was raised after July 1 of that year. She has already raised $510K through June 30 of this year. The challenger’s competitiveness in campaign financing keeps the model from giving Davis any further credit.
In her first race against Davis, Johnson received 56% of the full-ballot vote, clawing back more than a third of the straight-party deficit she faced then. Davis has since improved her standing among full-ballot voters, taking nearly 65% of their vote in her 2018 race. There is no single-punch, straight-party option this year, but we expect Johnson would have a significant advantage this year, likely enough to overcome what we expect will be a smaller margin for the incumbent among traditional full-ballot voters.
Four years ago, Trump received 23% fewer votes in the district than Romney. Any further erosion in his numbers makes it even harder for Davis to survive a rising blue tide. This is the likeliest Texas House seat to flip, even with Davis’s history of over-performing, according to our model. As McReynolds and other over-performing Democrats learned in 2010, at some point the math simply doesn’t work anymore. We are at that point in this district.
HD138 open (Lean Democratic): According to our model, this seat is 4.6 percentage points less friendly to Democrats than HD134, but there is no incumbent performance to judge here, and thus no boost for over-performance. Rep. Dwayne Bohac (R-Houston) won re-election in 2018 by just 47 votes. Straight-party and full-ballot voters were both evenly split. Just 75 more people cast a Republican straight-party vote in 2018 than a Democratic one, and 28 fewer full-ballot voters cast their vote for Bohac than for his Democratic challenger.
The district’s voters were similarly evenly divided in their presidential preference in 2016. Clinton carried the district over Trump by a mere 36 votes. They weren’t so evenly divided in 2018. O’Rourke got 3K more votes than Cruz in a district Romney won by 20 points in 2012.
As with HD134, the number of Republican straight-party voters has been flat since 2012, while the number of district voters casting a Democratic straight-party vote has increased 60%. The Republican straight-party voting advantage has fallen every election year since 2010, dropping from more 7,600 votes to almost parity. Were there straight-party voting in 2020, the Democrats would likely have a four-figure advantage.
The district’s evolution has been fast and sharp. In 2014, the district was nearly 4 points redder than the state as a whole, and the average Democrat received south of 35%, measured head-to-head against the Republican. In 2018, the district was 4 points bluer than the state as a whole, and the average Democrat was just north of 50% head-to-head against the Republican. Going back to 2002, the district has shifted toward the Democrats by nearly 17 points relative to the state.
Bohac over-performed his fellow Republicans by 1.5 points in 2014 and half a point in 2018, the only two times this decade he has been challenged by a Democrat. His aspiring successor, Houston homemaker Lacey Hull, does not inherit that slight push in our model. She faces Houston attorney Akilah Bacy, whose convincing runoff win secured the nomination. Libertarian Grizzie Trojacek is also in the race. Hull has out-raised Bacy, $194K to $144K, for the election cycle, but they are about even in cash on hand with less than $50K between them combined. Bohac’s 2018 challenger raised $173K for the entire cycle.
We expect both major party candidates to raise and spend significant sums between now and November. According to our model, HD138 is the second likeliest seat to flip behind HD134 and just ahead of Republican-held HD108 in Dallas Co. The model projects a 4-point Bacy win head-to-head over Hull. Whether the Libertarian candidate’s presence will affect the outcome is an open question, but he likely won’t be a factor if the projected 4-point margin is realized.
Republican Claw-back Targets
HD132 (Lean Democratic): Rep. Gina Calanni (D-Katy) faces a rematch against former Rep. Mike Schofield (R-Katy), who she defeated by 113 votes in 2018. While this district represents Republicans’ top opportunity at regaining a lost seat, it has rapidly become friendlier to Democrats.
Republicans running in the district, including former Rep. Bill Callegari (R-Katy), enjoyed a nearly 9K-vote advantage in straight-party voting in 2012. Romney carried the district by 20 points over Obama, and Cruz won the district by 21 points that year. The number of Democratic straight-party votes plunged 52% in 2014, before more than tripling to 22K in 2016. In that year, the Republican straight-party advantage was just over 4K, and Trump won the district by 4 points.
In 2018, an additional 4,600 people cast a Democratic straight-party ballot, more than covering the additional 1,200 Republican straight-party ballots, cutting the Republican margin to just 717 votes. Calanni overcame that deficit by taking 54% of the full-ballot vote, a big increase over the 36% share won by Schofield’s 2014 opponent. O’Rourke won the district by 3.5 points over Cruz, a nearly 25-point swing in six years.
The district was 3.4 points bluer than the state in 2018, up from 1.8 points in 2016 and a flip from 4 points redder than the state in 2014. Over the past two decades, the Democrats have practically doubled their vote share in the district, rising from 25% of the vote in the precincts currently comprising the district in 2002 to 49.7% in 2018.
Schofield defeated Angelica Garcia, Abbott’s endorsed candidate, in the Republican primary by 6 points. She out-raised him, $234K to $128K, but he outspent her, $311K to $79K. Calanni raised $183K for the election cycle through June 30 and enters the general election season with a modest $46K edge in cash on hand.
Our model places this district within a third of a point of being a Toss Up but still leans it toward Calanni. This assumes a modest further shift towards Democrats generally in the district and a half-point boost for Calanni based on her over-performing other Democrats – and Schofield underperforming other Republicans – in the district in 2018. Absent a shift toward Republican voters, Schofield would need to reverse his 7-point loss among voters who have not historically, or at least recently, voted straight-party in order to complete his comeback. It’s certainly doable, but it’s likelier that the district will break however the presidential race goes. If Democratic nominee Joe Biden carries the district over Trump, then Schofield’s road back to the House gets that much tougher.
HD135 (Lean Democratic): Our model projects this district to be Republicans’ third best opportunity to reclaim a seat in the state House, following previously discussed HD132 and HD65 in Denton Co. Rep. Jon Rosenthal (D-Houston) ousted longtime Rep. Gary Elkins (R-Houston) by nearly 2K votes, buoyed by a 523-vote advantage in straight-party voting and a 56%-44% split among full-ballot voters.
Indeed, Elkins suffered significant erosion of both straight-party and full-ballot support. In 2012, Elkins won by 20 points behind an 8,500-vote advantage in straight-party voting, accounting for more than three quarters of his margin of victory. That advantage dropped by more than 5K by 2016, when Elkins won re-election by nearly 6K votes. He received 58% of the vote from people not casting straight-party ballots that year. Two years later, he lost the full-ballot vote by 12 points, and the Republicans’ edge in straight-party voting was fully erased.
Elkins out-raised Rosenthal, $199K to $111K, for the election cycle and outspent him nearly 6-to-1, but Elkins ended the election year with $278K on hand. Perhaps the campaign could have done more to make a difference, perhaps not. Historically, Elkins performed on par or slightly better than other Republicans even as the district shifted underneath him. In 2010, the average Democrat received just 33% of the vote here, measured head-to-head against the Republican. By 2018, Democrats were getting 51% of the vote, making the district nearly 5 points bluer than the state as a whole.
Rosenthal faces Jersey City Mayor Justin Ray, who easily turned aside a lone primary challenger, 84%-16%. Libertarian Paul Bilyeu is also in the race. The incumbent has a modest edge in total contributions ($210K to $172K) and in cash on hand ($87K to $61K), but this is the kind of race we expect the Republican to end up better funded. It may not be enough.
Trump won the district over Clinton by just 2 points, well below Romney’s 19-point win four years earlier. O’Rourke defeated Cruz here by 6.5 points in 2018, representing a 27-point swing from Cruz’s 2012 margin.
As with HD132, Republicans cannot win this district without an Trump at least matching his performance in 2016. In just four years, the district has grown more than 7 points friendlier to Democrats – and 7 points unfriendlier to Republicans – to the point that 10 statewide Democratic candidates carried the district, including the challenger to George P. Bush.
A Year or Two Too Soon
HD126 (Lean Republican): Republicans’ straight-party advantage in 2018 was 4,606 votes, representing a 62% decrease since 2012. Over that period, Republican straight-party voters declined by nearly 3K while Democratic straight-party voters increased by nearly 5K. Democrats improved their share of the full-ballot vote in 2018 (47%) versus 2016 (38%), but these factors still mean Rep. Sam Harless (R-Spring) is favored to win a seat he won by 5K votes two years ago.
The district was just a half a point redder than the state as a whole in 2018, and Harless over-performed other Republicans by another half a point. Trump carried the district by 10 points over Clinton, and Cruz won it by nearly 4 points. Both are far removed from Romney’s 25-point win in 2012, but there should still be enough Republican voters here to keep the seat in Republican hands.
Harless faces a rematch against Houston management director Natali Hurtado. The challenger has out-raised Harless, $154K to $107K, for the cycle. She raised $121K since the March primary, well above the $29K Harless raised since January 1. Harless entered July with a nearly $60K advantage in cash on hand. Considering that Hurtado raised $15K for her entire 2018 campaign, she has proven that she can be competitive with Harless in terms of campaign resources.
However, the district may not be quite ripe for a flip, barring a significant further erosion in support for Trump atop the ticket. It will be redrawn before further demographic and political shifts put it squarely within range.
HD150 (Likely Republican): This district is only nominally friendlier to Republicans than HD138, but the difference here is a challenger who has raised less than $1K. Cruz narrowly edged our O’Rourke in this district, 51%-48%, and the other statewide Republicans won their races here by 5-10 points (Abbott by 13). Rep. Valoree Swanson’s (R-Spring) 2018 challenger raised less than $5K for the race and lost by 16. Swanson has over-performed her fellow Republicans by about 2 points in each of the last two races, another distinction separating this seat’s Likely Republican rating from HD138’s Lean rating.
The district was 12 points redder than the state in 2012. It was just 6 points redder in 2018. Given a few more years, these precincts could produce a highly competitive race. But not this year, barring a complete collapse of the Trump brand here. The Democratic challenger, university student Michael Walsh, cannot win by separating himself from the Democratic pack. He needs the pack to win.
Out of Reach
HD129 (Likely Republican): While true that Republicans’ straight-party advantage shrunk 43% since 2012, it was still north of 9K votes in 2018. The difference between 2012 and 2018 is entirely driven by increased Democratic straight-party voting.
Rep. Dennis Paul’s (R-Webster) district is still redder than the state but has been steadily getting closer to the state average, going from 8.1 points redder than the state in 2012, when Romney carried the district by more than 30 points, to 3.4 points redder in 2018, when Cruz carried the district by 9 points.
Paul faces Seabrook IT analyst Kayla Alix, who he has out-raised by more than $200K. She reported having less than $1K on hand as of June 30.
According to our model, Democrats have a better shot at flipping 20 other Republican-held seats (including aforementioned HD150) than this one.
HD133 (Likely Republican): Nearly half of the Republicans’ straight-party advantage in this district has been erased since 2012 by increased Democratic straight-party voters, but Rep. Jim Murphy’s (R-Houston) advantage was still five figures in 2018 in a district Cruz carried by 9 points.
Trump won it by 13 in 2016, only a third the size of Romney’s 38-point win in 2012. Democrats have made gains here, but they still have a long way to go. The $8K raised so far this cycle by Houston licensed professional counselor Sandra Moore, Murphy’s Democratic challenger, won’t make her much of a factor outside of her party label. That’s before accounting for the incumbent’s greater than $500K advantage in cash on hand.
HD144 (Likely Democratic): The seat that changed partisan hands twice during the decade currently sits at the far outer edge of the Likely column, not even a quarter of a point away from the Safe column.
In 2018, HD144 was 14 points bluer than the state as a whole, and O’Rourke won it by 23.5 points. Clinton won it by 19, a big increase over Obama’s 3-point win 3 in 2012. In between those two presidential elections was an anomalous, very low-turnout election in which a Republican surprised Rep. Mary Ann Perez (D-Houston). She won the seat back two years later and avoided an upset in 2018.
Republican challenger Tony Salas, a Houston transportation association director, has raised less than $5K so far. Perez had a roughly 4,500-vote straight-party advantage in 2018. While not as large as other Likely seats, turnout here typically runs 15 percentage points below the state’s turnout, making those 4,500 votes that much harder to overcome.
According to our model, Republicans have a better shot at flipping 19 other Democrat-held House seats than this one.
©2020 Texas Election Source LLC