Getting Deeper into Texas and North Carolina’s Primaries

KEY POINTS FROM THIS ARTICLE

— In Texas, Rep. Wesley Hunt’s (R, TX-38) presence in the GOP Senate primary likely hurt Sen. John Cornyn (R-TX) on balance.

— However, Hunt’s performance was not distinctively strong anywhere in the state, as he even came in third place in the district he represents.

— Overall, Democrats outvoted Republicans in Texas by 3 points, making 2026 the first midterm cycle since 2002 where Democrats cast more ballots in Texas.

— In North Carolina, Democrats outvoted Republicans by an even larger 57%-43%, even as they have fallen behind both Unaffiliateds and Republicans in their share of the state’s overall voter registration.

More from Texas and North Carolina

Last week, as we were publishing analysis of Texas and North Carolina’s primaries, news had just broken that Donald Trump was going to, finally, weigh in on the Lone Star State’s Republican primary for Senate. As of this writing, that seemingly imminent endorsement has not materialized. Trump could be holding off out of frustration—apparently, he did not like how reports that he’d endorse Sen. John Cornyn (R-TX) seemed to preempt an official announcement.

But while we wait to see what Trump does—or does not—do, Cornyn, who beat expectations to finish in first place with about 42%, finds himself locked in a runoff with state Attorney General Ken Paxton, who took just under 41%.

Since the primary, Paxton has said that, even if Trump acquiesces to the wishes of national Republican leadership and backs Cornyn, he would remain in the runoff. Paxton has offered to consider suspending his campaign if the SAVE Act passes the Senate, although Senate Majority Leader John Thune (R-SD) sounds skeptical that the GOP-backed elections bill would get the necessary bipartisan support to break a filibuster (or to do away with the filibuster altogether in service of passing this legislation).

Taking a step back, the Republican Senate runoff was precipitated in the first place by the presence of two-term Rep. Wesley Hunt (R, TX-38) in the primary. Although the runoff will be open to all registered voters who did not cast a ballot in last week’s Democratic primary, Hunt’s voters could still play a large role in picking the Republican nominee. Hunt himself has not endorsed in the runoff.

Statewide, Hunt took 13.5% of the vote, leaving him a distant third behind Cornyn and Paxton, who both got a little more than 40%. He was most competitive in the Houston metro area; not surprising, given that his district is based there. But he still finished with less than 20% of the vote in Houston’s Harris County, which was his best large county in the state.

Hunt polled above his statewide share in 68 of the state’s 254 counties—this group represented about 46.5% of last week’s GOP electorate. Overall, Paxton carried those counties with 41.6% to Cornyn’s 39.0%. Meanwhile, the 186 counties where Hunt punched below his statewide number backed Cornyn by 4.5 points, as he took 44.4% to Paxton’s 39.9%. That Cornyn clearly fared better in areas where Hunt was weaker suggests that the latter’s presence was a net drain on the incumbent, at least to some degree.

Zeroing in on the Houston area, Map 1 shows the Republican Senate primary vote in TX-38, Hunt’s district. While several districts in Harris County were impacted by last year’s new congressional map, TX-38’s borders saw only minor changes. In the GOP Senate primary, TX-38’s vote was nearly identical to Harris County’s overall picture, even though the former is deep red in general elections and the latter has a mild blue lean. Cornyn carried TX-38 with 38% as Paxton took 36.7% and Hunt took just over 22%.

Map 1: TX-38 vote in 2026 Republican Senate primary

Even in Hunt’s own district, the Republican primary was, essentially, a two-way race: Hunt only carried two precincts and came in second place in just a handful of the 168 precincts in the district.

Within the district, Cornyn’s most dedicated constituency was voters located south of the Katy Freeway. That area, which is dark blue on the first image, is home to some of the Houston area’s most established inner suburbs—per Redistricter, this part of the district also has a roughly 70% college attainment rate. The majority of the district’s residents, though, live in an area north of State Highway 6, which includes Cypress and Tomball; Paxton generally polled somewhere in the 40s in those precincts. If there was a north-south dynamic between the two stronger Republicans, Hunt’s strength was eerily consistent throughout: precincts that backed Cornyn gave Hunt, collectively, 22%, while he took the same share in Paxton’s precincts.

Though Hunt tended to run strongest in the Houston area, his support elsewhere in the state was not concentrated in any clear geographic bloc, potentially making those voters unpredictable in the runoff. A poll from Public Policy Polling, a Democratic firm, that was fielded shortly after the primary showed Paxton leading Cornyn 45%-42%—Paxton’s showing included a 14-point edge with Hunt voters.

In our preview of the Texas primary, another thing that we flagged, at a broader level, was the composition of the electorate. Though Democrats have sometimes done better in presidential years, 2002 was the most recent midterm cycle where more primary voters pulled a Democratic ballot than a Republican one. Early voting patterns seemed to suggest Democrats were on track to outpace Republicans, and that ended up being the case, as Table 1 shows. Democrats outvoted Republicans by 3 points, while turnout this month was higher than it’s been in any recent election.

Table 1: Texas primary vote composition in recent years

On the Democratic side specifically, we saw exuberant posts here and there pointing to Democrats’ high participation rates in some traditionally Democratic, but Trump-won, counties in South Texas. True, that area looked quite blue last week, but there were probably more local factors at play. To help quantify this, Map 2 considers the Democratic House vote in each county as a percentage of the Senate vote.

Map 2: House votes cast as a percentage of Senate votes cast in Texas Democratic primary

While all the counties that cast more votes in the House races were in South Texas, Starr and Zapata counties led the way, each casting more than 120% of their Senate primary votes for the House. This could suggest that some of those primary voters might not go on to vote Democratic for Senate in the fall general election. Both of those border counties are in TX-28, held by Rep. Henry Cuellar (D). Though we noted Cuellar seemed to underperform in his primary, winning renomination with less than 60%, he was close to 80% in both Starr and Zapata. As Split Ticket’s Armin Thomas observed, these two counties seem to have retained a parochialism that some of the other more populous South Texas border counties might lack or have dropped.

Last week, North Carolina voters were also settling on nominees in a major Senate race. In the Democratic primary, former Gov. Roy Cooper (D) was nominated with more than 90% of the vote. Though the Republican primary featured more competition, former Republican National Committee Chairman Michael Whatley, who ran with Trump’s endorsement, won with close to two-thirds of the vote.

Unlike Texas, which lacks party registration altogether, North Carolina’s primary system is semi-closed. While voters can register with the two major parties—and, more recently, some minor parties have become options—Unaffiliated voters can choose to participate in a primary. Last week, even as Democrats had the less eventful primary for Senate—which is the highest office on the ballot—about 57% of primary ballots were cast on the Democratic side to 43% for the Republicans.

In Table 1, we looked at the voter makeup of the Texas primary electorate over the past couple of decades. Table 2 does, essentially, the same thing with North Carolina. Table 2 also looks at the makeup of the electorate’s voter registration at the time of each respective primary.

Table 2: North Carolina primary vote composition in recent years

Though Democrats made up 45% of registered voters in 2008, their share of the voter registration pool has steadily dropped. However, as Republicans have remained relatively stagnant, the Democratic drop has been to the benefit of the Unaffiliated bloc. During Joe Biden’s term, Unaffiliateds surpassed Democrats, and, more recently, Democrats actually fell very slightly behind Republicans.

While Republicans have undoubtedly enjoyed watching the Democratic share of voter registration sink, this still didn’t stop Democrats from having their best midterm primary electorate share on Table 2. In fact, given the lower share of Democratic registered voters, their strong showing last week was necessarily more reliant on Unaffiliated voters choosing to participate on their side. Of course, for some perspective, we would have loved to see how 2018, the last Trump midterm, would have fit into Table 2, but, unfortunately for election nerds, it was what Catawba College professor Michael Bitzer calls a “blue moon” year in the state, as it lacked a presidential or Senate race.

Though we are holding our general election rating for the Senate race at Toss-up, there are some encouraging signals for Cooper. For instance, let’s consider the two most recent midterm cycles on Table 2: while 2026’s turnout was slightly higher than what 2022 saw, Republicans went from outvoting Democrats by more than 55%-45% to being outvoted by about 14 points. Both cycles also hosted Senate races where Republicans had a livelier primary.

Another analysis by CBS elections analyst Kabir Khanna, which considered recent Senate and House races in the state, also suggested positive news for Democrats.

Taken together, these primary turnout patterns suggest that Democrats may have enjoyed stronger engagement from both their base and unaffiliated voters this year—an encouraging sign for Cooper as he enters what will be a high-spending slog to November.

J. Miles Coleman is the associate editor of Sabato’s Crystal Ball, the University of Virginia Center for Politics’ authoritative, nonpartisan newsletter on American campaigns and elections. He also serves as the Center’s Media Relations Coordinator. Follow him on Twitter @jmilescoleman.

See Other Political Commentary by J. Miles Coleman.

See Other Political Commentary.

Views expressed in this column are those of the author, not those of Rasmussen Reports. Comments about this content should be directed to the author or syndicate.