Our latest model run moves the presidential race in Texas to Toss Up and projects a narrow Democratic majority in the Texas House when the Legislature convenes in January. Another four Republican-held House seats are within 2 percentage points of potentially adding to that majority.

The new model run produces no ratings changes in the Senate, where we continue to project a one-seat gain for Democrats. The model projects a five-seat gain for Democrats in the congressional delegation, which would split it 18-18 when the new Congress convenes. An additional three Republican-held seats are projected to be within 1.2 percentage points of flipping.

President Trump is currently projected to win the state by 2 points over Democratic challenger Joe Biden, 50.5%-48.5%. He carried the state by 9 points, 52%-43%, in 2016. The projected 2020 margin is slightly tighter than U.S. Sen. Ted Cruz’s (R) 50.9%-48.3% victory over then-U.S. Rep. Beto O’Rourke (D-El Paso) in 2018.

A total of 20 races’ ratings moved one column toward the Democrats:

  • President to Toss Up from Lean Republican
  • CD2 (Crenshaw) to Toss Up from Lean Republican
  • CD3 (Taylor) to Toss Up from Lean Republican
  • CD31 (Carter) to Toss Up from Lean Republican
  • HD64 (Stucky) to Toss Up from Lean Republican
  • HD92 open to Toss Up from Lean Republican
  • HD93 (Krause) to Toss Up from Lean Republican
  • HD121 (Allison) to Toss Up from Lean Republican
  • HD66 (Shaheen) to Lean Democratic from Toss Up
  • HD67 (Leach) to Lean Democratic from Toss Up
  • HD112 (Button) to Lean Democratic from Toss Up
  • HD45 (Zwiener) to Likely Democratic from Lean Democratic
  • HD47 (Goodwin) to Likely Democratic from Lean Democratic
  • HD52 (Talarico) to Likely Democratic from Lean Democratic
  • HD102 (Ramos) to Likely Democratic from Lean Democratic
  • HD113 (Bowers) to Likely Democratic from Lean Democratic
  • HD129 (Paul) to Lean Republican from Likely Republican
  • HD150 (Swanson) to Lean Republican from Likely Republican
  • HD33 (Holland) to Likely Republican from Safe Republican; and
  • HD91 (Klick) to Likely Republican from Safe Republican.

The U.S. Senate inches closer to the Toss Up line but remains rated as Lean Republican along with the other statewide races.

Four races’ ratings moved one column toward the Republicans:

  • HD31 (Guillen) to Toss Up from Lean Democratic
  • HD34 (Herrero) to Lean Democratic from Likely Democratic
  • HD149 (Vo) to Likely Democratic from Safe Democratic; and
  • CD28 (Cuellar) to Likely Democratic from Safe Democratic

Rep. Ryan Guillen’s (D-Rio Grande City) district has been rapidly growing friendlier to Republicans, and the latest model run continues that acceleration, putting the district into play.

Our complete race ratings can be found here.

The nine Republican-held House seats projected to flip to the Democrats are HD26 open (Miller), HD64, HD66, HD67, HD96 open (Zedler), HD108 (Meyer), HD112, HD134 (S. Davis) and HD138 open (Bohac). The four within a point of flipping are HD92 open, HD93, HD94 (Tinderholt) and HD121. The Senate seat projected to flip to the Democrats is SD19 (Flores).

The four Congressional seats projected to flip to the Democrats are CD10 (McCaul), CD21 (Roy), CD22 open (Olson), CD23 open (Hurd) and CD24 open (Marchant). The three additional seats within 1.2 points of flipping are CD2, CD3 and CD31.


From 2000 to 2012, the presidential vote in Texas was consistently about 10 points redder than the nation as a whole, slightly redder when George W. Bush was on the ballot and just under 10 points redder with Barack Obama on the ballot. In 2016, with Donald Trump on the ballot for the first time, the state was just 5.8 points redder than the nation as a whole, measuring the Democratic and Republican candidates head-to-head. That puts the race within our Toss Up column based off of Fivethirtyeight’s current (as of 9 p.m. Sunday evening) national vote projection.

The race between U.S. Sen. John Cornyn (R) and Democratic challenger M.J. Hegar remains in the Lean Republican category, ever so slightly, and so do the remaining statewide races. Our latest model run projects the average statewide Democratic candidate to get 47.43% of the vote, measured head-to-head against the Republican, about 2 percentage points better than in 2018.

Turning to the boffo early voting turnout and new voter registrations, particularly in counties carried by former U.S. Rep. Beto O’Rourke in 2018, we have removed some of the brakes our model placed on a continued blue-ward shift in the electorate headed into this year (Technically, we slightly re-weighted the change in partisan lean between 2016 and 2018 across all districts.) that left us more conservative than other prognosticators. As a result, we now project the Democrats will flip the nine seats needed to obtain a nominal majority going into the 2021 legislative session, and another four seats are within a percentage point of adding to that majority.

At the center of all these assumptions and projections is the apparently deteriorating position of Trump nationally as Election Day draws nearer. Already, more than 27.5 million people – including more than 4 million Texans – have cast their ballots, which means Trump is running out of opportunities to change the trajectory of the race.

Morning Consult’s recent poll found Trump leading Biden, 49%-47%, overall but trailing Biden, 52%-35%, among independent voters. The CBS Battleground Tracker poll conducted in October 2016 found Trump leading among independents, 48%-29%, and the Univ. of Texas/Texas Tribune poll from October 2016 found an ever bigger 46%-19% Trump advantage. At that time, around one eighth of likely voters were either undecided or expressed support for some other candidate. This year, far fewer voters are undecided or voting for someone else. Trump does not appear to be winning those voters over.

Since 2018, we’ve said that the biggest issues voters will consider when voting in 2020 are Trump, Trump and Trump. That was before COVID-19, an economic collapse, mass social unrest and catastrophic natural disasters. Despite all of this, Trump maintains the strong support of around the same 40% of the electorate that has always been strongly behind him. The problem for him, and Republicans generally, is that his administration appears to have made little progress toward growing that 40% base.

The big question is, do Republican and independent voters turned off by Trump limit their defections to his line on the ballot, or does their dissatisfaction go deeper down the ballot. Ironically, the elimination of the single-punch straight-party option means voters must now judge each Republican candidate on their own instead of blindly accepting the slate with a single check mark. Indeed, years of consistent straight-party voting, especially among voters who do not participate in primary elections, means a lot of Republican officeholders may be virtually unknown to a large swath of voters who have never marked the box by their names.

This of course also applies to Democratic candidates and, to a much lesser extent, minor party candidates. However, voters tend to be more motivated to vote against something than for it, and there is much to vote against this year.

Echoes of 2010

This year’s election is looking more and more like 2010, when voters took their dislike of the Obama presidency out on Democrats up and down the ballot. That election created the largest Republican legislative majority in state history just one election cycle after Democrats were a couple of mail pieces away from a 75-75 tie in the House. Twenty-one Democratic legislators were defeated in 2010.

If that precipitous fall from near parity was a one-act play, then 2018-20 may be a two-act play. Democrats flipped a dozen seats in the House in 2018. If they get the nine they need to win a majority, then the two-cycle shift will be 21 seats.

In 2018, Texas voters reacted to the first two years of Trump’s presidency by ousting two Republican members of Congress, two Republican state senators and eight Republican legislators. Four additional open Republican seats flipped to Democrats. It was the first time since 2006 that Democrats gained seats in either chamber in a gubernatorial election cycle. Turnout, measured as percent of registered voters casting ballots for the top office on the ballot, was the highest for a gubernatorial election since 1970 and exceeded 50% for the first time since 1994, the last year a Democrat won statewide.

In October 2010, the Univ. of Texas/Texas Tribune poll found Obama’s net job approval rating was +34 among Democrats, -52 among independents and -89 among Republicans. Last week, the UT/TT poll found Republicans’ net approval of Trump was a more tepid +20, while independents were slightly more disapproving (-56) and Democrats’ were roughly even (-86) with Republicans’ views 10 years earlier.

In the past week, we’ve seen an upswing in advertising supporting Biden and a significant increase in negative advertising aimed at Democratic House and congressional candidates, including Hegar. These trends make tangible a fear that Trump may now be a liability.

Dissatisfaction with the president, uneasiness about the economy and given an opportunity to change course, Texas voters in 2010 overwhelmingly chose a different direction. In 2018, an indication of that course change flowed through the U.S. Senate race to the state House to courtrooms and county offices. The stage is set for further course correction. Our model is beginning to come around to that possibility. We will run it again in a week or so.

©2020 Texas Election Source LLC