How Corey Stewart Carried #VA10

After the returns settled from the Virginia primary last week, one of the more surprising results was in the GOP primary. While the overall closeness of the contest was notable, it was rather unexpected that Corey Stewart beat party favorite Ed Gillespie in the 10th Congressional district. VA-10 is one of the suburban districts in northern Virginia, near Washington D.C.:

This was surprising because Stewart was known for his fervently anti-establishment campaign, and he became most known for his defense of Confederate monuments. Traditionally, this type of Republican has not done well in VA-10. For the past few decades, it was represented by Frank Wolf (R) – a rather mainstream Republican who was known for tending to the local needs of the district, and was often able to attract a significant amount of crossover support in his elections. In 2014, Wolf retired, and was succeeded by Barbara Comstock, who is considered one of the more moderate House Republicans.

Further, VA-10 is essentially the Virginia version of GA-06. This is one of the wealthiest and most educated districts in the country, so it didn’t exactly seem like fertile ground for a populist-type candidate.

The result in VA-10 was very close, but Stewart carried it by just under 2%:

A good point of comparison for this was the 2016 Presidential primary. Stewart ran as an unapologetic Trump supporter. This made his win in VA-10 more surprising, because the President did not do well there. In the primary, he lost it by 8% to Sen. Rubio:

Here’s a comparison between these two maps. Stewart was able to flip many precincts that supported Rubio, mostly in Loudoun County, and western Fairfax County. Gillespie, fared well in the district’s northern Fairfax arm; this area is wealthier, and he retained much of the anti-Trump vote there:

Stewart was clearly able hold Trump’s base in the district, as well as win over a decent chunk of Rubio’s voters. In fact, precincts in VA-10 that supported Stewart would have, aggregately, supported Rubio with 34% to Trump’s 32%. Rubio carried Gillespie’s precincts by a more comfortable 40/27:

Another factor that contributed to Stewart’s strong showing here was likely Virginia’s open primary system. Virginia doesn’t have party registration, so voters can chose to cast a ballot in either primary.

In 2016, Secretary Clinton was a clear favorite in the primary over Sen. Sanders, while the GOP was looking competitive. As a result, more voters chose to participate in the Republican. In the 2016 primary, 61% of VA1-0 voters chose a Republican ballot to 39% for the Democratic primary:

This year, the opposite dynamic was in play. Having consolidated most of the party’s institutional support, Gillespie was seen an obvious frontrunner on the GOP side. The Democratic contest, between Lt. Gov. Ralph Northam and Congressman Tom Perriello, was seen as much more competitive. This time, the partisan split was 58/42 in favor of Democrats:

Every precinct in the district saw an uptick in the percentage of voters choosing to vote in the Democratic primary:

The region that ‘swung’ the hardest to the Democrats was eastern Loudoun County and northern Prince William County; these areas had many Rubio -> Stewart precincts. Likewise, the swung was relatively small in the western part of the district, where Trump and Stewart held up well, suggesting many Republicans there stuck with them.

Finally, one factor that hurt Gillespie was the candidacy of State Sen. Frank Wagner. He ended up being something of a third wheel, but took a critical mass of votes from Gillespie’s left. He supported Kasich in 2016. Kasich took 12% in VA-10, and Wagner took a similar 11%, which almost certainly would have otherwise gone to Gillespie. Wagner also got the endorsement from the Washington Post, which likely also helped with moderate Republicans.

Overall, it seems like many moderate and swing voters flipped to the Democratic primary this time, which pushed the GOP electorate further rightward. This was helped a candidate like Stewart, who had a hard base of support, and hurt Gillespie, who was seen a more ‘squishy’ establishment insider.

With Virginia’s open primary, its probably best not to read too far into these results. As Stewart is from Prince William County, which is partiality in VA-10, he started out with some personal vote there; maybe if he was from a different part of the state, he wouldn’t have done as well in the district. Still, as Dave Wasserman pointed out, many formerly moderate Republicans in Virginia may be identifying more with the Democratic Party overall, so trends in places like VA-10 are very much worth watching.