Off the Kuff Rotating Header Image

Early voting, Day 9: Who are these people?

The question keeps getting asked, who is it that has been voting so far?

An unprecedented number of Texans cast their ballots during the first week of early voting, but it is impossible to predict whether that surge will benefit Republicans or Democrats because more than 25 percent of the voters have no primary election voting history, an analysis of data from the Secretary of State shows.

People whose voting records provide no clue of their party affiliation cast 27.8 percent of the ballots in the 15 most populous counties in Texas, according to the analysis by Republican consultant Derek Ryan.

About one-third of the early voters in those counties had voted in a Republican primary in the past; for Democrats, it was 30 percent. Those percentages are consistent with early voting totals from the last midterm primary, in 2014, Ryan said.

But the 2018 numbers leave too many unknowns to draw conclusions, Ryan said.

“Unless somebody’s out there polling those people and calling them, there’s really no way necessarily to know if those people are voting Republican or Democrat,” Ryan said. “The same goes for the people that have primary history. Just because somebody voted in a Republican primary, it doesn’t always necessarily mean that they’re a Republican or that they are voting for all the Republicans on the ballot.”

In Harris County, 30 percent of early voters had no primary voting history. Thirty-three percent of early voters in the county most recently voted in a Republican primary, compared to 28.6 percent who most recently voted in a Democratic primary.

In Bexar County, 28.5 percent of early voters had no primary voting history. For those who have cast ballots in primary elections before, 29.3 percent most recently voted in a Republican primary, compared to 32.6 percent who most recently voted in a Democratic primary.

The 15-county analysis also found an increase in voters with Hispanic surnames. Those voters have cast 19 percent of the ballots in early voting so far; in 2014, 15.2 percent of early voters in Texas had Hispanic surnames.

In the 2018 election, People aged 60 to 69 made up 21 percent of early voters so far, the largest age group, the 15-county analysis shows. Voters aged 50-59 made up the second largest group at just under 20 percent, and voters aged 40-49 percent made up the third largest group at about 15 percent. Early voters aged 20-29 made up about 8 percent. This breakdown was consistent with totals for the 2014 midterm elections.

One point to bear in mind when pondering the people with no primary history: In 2016, 2.8 million people voted in the Republican primary in Texas. That means that the no-primary-history people are not from that group. The comparable figure from 2016 for Dems is 1.4 million people. It’s true that in 2008, some 2.8 million people voted in the Democratic primary, but that was five election cycles ago. There are a lot of people who have voted in Texas elections since then who could not or did not participate in the 2008 primary.

I don’t want to draw any broad inferences from that. There were still about two million people who voted mostly Republican in November of 2016 but not in March, and a bit more than that on the Democratic side. The people with no primary history are mostly evidence of a larger electorate, for which I think we can all agree we already have evidence. There is evidence of more younger voters and of unlikely voters. I’ll say that benefits Democrats, but remember that Dems can do a lot better in 2018 than they did in 2014 and still fall short.

So. Here are the totals for Tuesday, and here are the daily totals from 2010, from 2014, and from 2016, as well as a spreadsheet with totals from 2004, 2008, 2012, and 2016. The running tallies:

Year    Mail    Early    Total
2010  45,219  250,066  295,285
2014  60,400  191,432  251,832
2018  80,279  557,264  637,543

2008  47,413  443,267  490,680
2012  59,304  491,349  550,653
2016  86,456  626,627  713,083

A little less than Monday, but still 62K in person and 64K overall. By tomorrow, barring a complete dropoff, we will surpass the entire final turnout for 2014. By Friday, even if there isn’t the usual end-of-early-voting surge and we stay on the same pace as now, we’ll surpass the entire final turnout for 2010. Have I mentioned that we were breaking records and the only real question was by how much? This is what I mean. Things are pretty brisk in Dallas County, too. Have you voted yet?

Related Posts:


  1. Manny Barrera says:

    I don’t have any doubt this are voters that are coming and voting for Democrats, the only question is how far down the ballot are they going.

    The way the Trump/Republican Party is acting indicates they believe the same. Stanart using their Jewish Bogey Man, Soros, to try to get his haters to come out an vote. That is but one example there are others.

    Hillary which was attacked by the Trump Party for eight years carried Harris County by 12%. She is not on the Ballot but Trump is. I expect that Beto will carry Harris County by 15 to 20%.

  2. Bob in Champions says:

    I like that you put in perspective the 2008 primary. In that year, there was a conscious effort (started by people like Rush Limbaugh) to flood the Democratic primary with Republicans voting for Sen. Clinton to delay Obama clinching the nomination. While it delivered Texas for HRC, it only delayed Obama by about a week (I think Obama still ended up getting more delegates from TX than Clinton did, anyway)
    But, that was a decade ago- most of those who are politically engaged to the level of switching over to vote for Clinton have had 10 years to vote in another primary and be reclassified (aside from the subset that would have passed away).
    That said, I feel as though the most telling statistics are the surge in voting by: the young, Latinos, those who have voted once or never.
    That last group is interesting as it will include: new citizen, young people who just aged into voting, and those who weren’t engaged because they formerly felt like politicians were “all the same”… until either Trump or Beto changed their mind.

  3. Tom in Lazybrook says:

    I think way too much is read into trying to extrapolate early voters primary voting history into the results in a general election, especially when one is using data from cycles when there was no/little competition in the Dem primary vs competitive GOP primaries. Many Dem leaning voters voted in GOP primaries in 2014, 2010, and 2012. Id argue that 2008 and 2016 are the only cycles that should be used to try to determine voter intent…and even that is imperfect.

  4. Northsider says:

    Question, and I’m just asking. Isn’t most polling done with “likely voters?” Would the 27.8% of voters that there is no clue as to their party affiliation, and/or the ones with no primary voting histories, etc., would those be considered “likely voters” for polling purposes? Or do we have a fat chunk of people voting that were not counted in any polls?

  5. C.L. says:

    You know when we’re going to know who voted and for whom ? November 7th, and not a day before.

  6. Northsider – All polls have to assume some voters will be new. Populations change over time, and that’s especially true in a state like Texas. Pollsters have to make assumptions about what the electorate looks like, and while they are generally in the ballpark, that’s one place where errors can occur.

  7. brad says:

    These early voter people are me.

    99% of the time I vote on election day. I like the vibe of voting on the day.

    I only voted early so I could bank the vote to avoid any potential issue for getting my vote in, despite my extreme flexibility to vote any time, and importantly to free up time to drive voters to the polls on the day.

    I am an independent who voted for 49 DEM candidates, 42 REP candidates (mostly judges, except for County Judge and Clerk), 2 Libertarians, and 1 abstention for an uncontested Justice of the Peace race.

  8. Manny Barrera says:


    Very difficult to poll people that have never voted before as one would have no idea as to who they are. They could call newly registered voters, people turning 18 or who recently became citizens.

    If one votes in a general election but not in a primary then they don’t have a primary history, those are bout 30-40% of the voters.

  9. Bill Daniels says:

    I think polling is fatally flawed because virtually all of it is done via landline Most of the younger generation has no landline to call.