The million-dollar question, who is going to vote in the 2020 Republican Primary Runoff?  In order to get the heart of this question, we at Campaign Engineering Strategies (CES) looked back over the last four primaries and primary runoffs from 2018, 2016, 2014, and 2010.  Here is what we see:

The first question we asked, what percent of the Primary Runoff voters also voted in the Primary?  As you can see from the following chart, an overwhelming majority of Primary Runoff voters also voted in the Primary.  We believe this is to be expected, but it is extremely helpful to put real numbers to the expectation.  Over the last four Republican Primary Runoff Elections, approximately 90% of the voters voted in the Primary.

We categorized these voters into four categories:

NNNN (0Rs) represents voters who didn’t vote in any of the past four Republican Primaries.

RNNN (1Rs) represents voters who voted in the 2018 Republican Primary, but did not vote in the 2016, 2014, or 2012 Republican primaries.

RRNN (2Rs) represents voters who voted in the 2018 and 2016 Republican Primaries, but not the 2014 or 2012 ones.

RRRN (3Rs) represents voters who voted in the 2018, 2016, and 2014 Republican Primaries, but not the 2012 one.

RRRR (4Rs) represents voters who voted in the 2018, 2016, 2014, and 2012 Republican primaries.

Our next goal was to understand how these Primary Runoff voters broke down in their voter models.   Interestingly, none of the models were very strong, but some were stronger than others.

For example, 4Rs returned to vote in the Primary Runoff at a rate of 40 percent in 2018 and 34 percent in 2016.

By comparison, 1Rs returned at 19 percent and 16 percent. 0Rs, first-time voters, returned at 16% and 4%.

We also looked at 4Rs who did not vote in the primary. They voted at a rate of around 6 percent and 4 percent. These strong Republican voters did not vote any better than the 0R voters who just voted in the primary.

So, what does all this mean in real numbers?

From our study:

NNNN (0Rs) had 7,064 voters vote in the primary, but only 1,111 returned to vote in the runoff, or 16 percent (representing 13 percent of the total runoff vote).

RNNN (1Rs) had 8,580 voters in the primary, but only 1,589 returned to vote in the runoff, or 19 percent (representing 18 percent of the total runoff vote).

RRNN (2Rs) had 2,090 voters vote in the primary, but only 578 returned to vote in the runoff, or 28 percent (representing 7 percent of the total runoff vote).

RRRN (3Rs) had 2,094 voters vote in the primary, but only 669 returned to vote in the runoff, or 32 percent (representing 8 percent of the total runoff vote).

RRRR (4Rs) had 7,714 voters vote in the Republican Primary, but only 3,078 returned to vote in the runoff, or 40 percent (representing 36 percent of the total runoff vote).

These 5 voter models, out of 81 available models, made up 82 percent of the entire Republican Primary Runoff.

So, what does this tell us?

  • A huge majority of the Republican Primary voters come from the voters who just voted in the Republican Primary.
  • The rate of return to vote is low to very low.
  • A majority of the Primary Runoff voters are from these five voter models.
  • Voters who did not vote in the Republican Primary, regardless of their voting propensity in other elections, have the very lowest propensity to vote in the Republican Primary Runoff.

Our goal at CES is to understand the value of every voter associated with your campaign.  Just a little bit of well-placed knowledge and sound campaign wisdom can go a long way to minimize campaign cost, reduce wasted efforts, and reach the right voters.

This is a commentary submitted and republished with the author’s permission. If you wish to submit a commentary to Texas Scorecard, please submit your article to submission@texasscorecard.com.

James Barnes

James is co-owner of Campaign Engineering Strategies.

Joel Starnes

Joel Starnes is the executive director and founder of the Organization for American Values and a co-owner of Campaign Engineering Strategies.

RELATED POSTS