Texas is not the sort of place national candidates visit just before Election Day or where political ads play on a loop during popular TV shows.
And, yet, here we are: Texas has been declared “in play,” with some polls rating the long solidly red state as a tossup between President Donald Trump and Democratic nominee Joe Biden.
A key reason is decades of rapid growth, as a strong economy drew millions of residents from across the country to Texas, most of them to urban areas — and more residents means more voters.
Harris County alone has grown its voter roll by about 292,000 since the 2016 election, equivalent to absorbing all of registered voters in Galveston County and tacking on another 70,000.
To earn a shot at Texas’ 38 electoral votes, Biden would need to combine a blowout here with sweeps in the state’s other large metros to offset Trump’s expected dominance in rural areas. If that occurs, the Houston area’s newly minted voters likely will play a key role.
A Houston Chronicle analysis of precinct-level voting patterns shows the Houston area’s growth has moved in tandem with Democrats’ widening advantage here in recent elections.
You can read the rest, and I’ll get to some numbers in a minute. But first, the big question is who else is out there to vote?
With 9.7 million ballots already cast across Texas — more than 1 of 7 of them in Harris County — will anyone show up Tuesday?
Well, yes, political scientists and consultants agree, though their estimates of the expected turnout vary.
County voters have already matched the 58 percent total turnout of the 2004 election and are just short of the 61 to 63 percent turnout recorded in the last three presidential contests. Statewide, 57 percent of registered voters have cast ballots.
The Harris County Clerk’s Office expects 200,000 to 300,000 voters to turn out Tuesday. Two Democratic consultants expect about 350,000 — in line with the last two general elections. Two local political scientists think 400,000 is possible, though they differed on whether the higher tally would benefit Democrats or Republicans.
The lower estimate would produce a final tally of about 1.6 million votes, or 66 percent turnout. The higher figure would push turnout close to 75 percent, which would be a modern record, exceeding the 72 percent turnout posted in 1992, when Houstonian George H.W. Bush lost a tough reelection fight to Bill Clinton.
GOP political consultant Kevin Shuvalov expects a more modest mark, saying, “It’s going to be bigger, but we’re not going to have 2 million voters.”
Still, he added, “Looking at who’s left over, there’s a lot of reliable voters still out there who have participated in multiple general elections previously. And then you have a large chunk of voters they just really like to vote on Election Day. They’re traditionalists.”
Democratic consultant Robert Jara echoed that, noting that many older voters are in the habit of casting their ballots at the same nearby Election Day polling place.
“There’s a sense of community, really, of voting in your neighborhood,” Jara said.
Local Latino voters have also disproportionately voted on Election Day in the past, Jara said, noting that Democrats tend to get an Election Day boost from predominantly Latino eastside precincts.
We’ve talked about some of this before, including the propensity of Latino voters to vote later in the cycle. I’m on record saying that I expect 300K or more votes tomorrow – I think we end up over 1.7 million, but maybe not quite at 1.8 million. I’ll be more surprised if we fail to reach 1.7 million than if we exceed 1.8 million.
The thing about these big numbers is that they can, and very likely will, lead to big margins for Joe Biden, bigger perhaps than we might have expected. Remember, when I did my super-simple projection of the vote, one of the assumptions baked in was that turnout levels would be static. If counties like Harris, which will provide a big chunk of the Democratic vote, overperform their expectations, that changes the math. (The same is true for the heavily red counties, but the vast majority of those are small. There’s no equivalent of Harris, or Dallas, or Travis, on the Republican side.)
Here’s what that means in practice. As a reminder, Hillary Clinton carried Harris County by 162K votes in 2016, winning 53.95% of the vote to Trump’s 41.61%. Beto carried Harris County by 201K votes, getting 57.98% to Ted Cruz’s 41.31%. Let’s assume we hit the low end of turnout projections, with 1.7 million total votes. How would that affect Biden’s margins?
53.95% – 41.61% = 210K net votes for Biden
57.98% – 41.31% = 283K net votes for Biden
I think its safe to assume Biden will do better than Clinton’s 53.95% – among other things, the third party vote will be much smaller. Note how both Trump and Cruz were mired in the 41% range – other Republicans did do better, but these two uniquely disliked flag-bearers did particularly poorly. It’s not out of the question that Trump could fail to break 40% in Harris. If we assume a more maximalist final turnout of 1.8 million, and a 60-40 win for Biden, he’ll net 360K votes in Harris County. That’s a lot.
(For what it’s worth, months ago when I was discussing blue-sky scenarios with fellow Dems, I posited a 60-40 win with turnout of 1.5 million, which I thought was reasonably ambitious and assumed a voter registration goal of 2.4 million. Who knew I was actually being restrained?)
Now again, what happens in Harris is a piece of the puzzle, but it’s not determinative in and of itself. The number of counties that Trump will carry will far exceed the number of counties that Biden will carry, and while most of those counties will have a small number of voters, there are big ones out there like Montgomery. Republicans can make up the big numbers Dems will post in their few strongholds by posting a lot of 70-30 and 80-20 wins in small and medium-sized counties. Dems will need to at least hold those losses to what they were in 2016 and 2018 to have a manageable deficit to overcome. Longer term, by which I mean 2022 and 2024, Dems will need to figure out how to gain ground in places like Waco and Lubbock and Tyler and New Braunfels and Abilene and Amarillo.
That’s a discussion for another day. Here’s the final Derek Ryan email.
9,677,963 people voted in Texas! That’s 57.1% of all registered voters. To give you some perspective on how crazy that number is, turnout for the ENTIRE 2016 General Election was 59.4%. In 2012, it was 58.6%.
So where does that put us at the brief electoral intermission?
There are still 7.2 million registered voters who have NOT voted. Of those, 3.1 million have voted in a previous election in the last four election cycles (dating back to 2012). If 75% of these people vote on Election Day, that will get us to the 12 million figure I keep throwing out there.
Voters who most recently voted in a Republican Primary have a 432,000 vote advantage over those who most recently voted in a Democratic Primary. Again, it must be pointed out that 4.6 million people who voted early who have no previous primary history.
What has been the participation rate based on voters’ previous election history?
- 81% of voters with previous Republican Primary history voted early
- 82% of voters with previous Democratic Primary history voted early
- 59% of voters with previous General Election history (and no primary history) voted early
- 29% of voters with no previous General or Primary Election history voted early
So what should we expect on Election Night? Based on the data, here is what I expect to see. Typically, the first results that are released at 7:01pm are numbers from early voting. In many portions of the state, these results will likely favor Democratic candidates. Then, as results from Election Day trickle in, we will see data that likely favors Republican candidates. In previous election cycles, the opposite has been the case. If you are only following the statewide election results, this will certainly be the case. It is important to note that voting early is a bigger trend in urban and suburban counties more so than in rural counties. For example, the average turnout percentage in the top 20 counties was 58.1%, but in the remaining 234 counties, the average turnout percentage was 48.7%.
Why does that matter? In 2016, Hillary Clinton received 50.8% to Donald Trump’s 44.8% in the top 20 largest counties, but Donald Trump received 70.9% to Hillary Clinton’s 25.9% in the remaining 234 counties.
The report is here. Anyone out there who was waiting till today to vote?
UPDATE: I don’t have any better place to put this, and I only saw it on Monday even though it was published on Friday, but here’s a Chron interview with Chris Hollins that’s worth your time.