Friends, Neighbors, and Unfriendly Neighbors in Virginia’s Wild Democratic Statewide Primaries

KEY POINTS FROM THIS ARTICLE

— In a highly competitive Democratic lieutenant gubernatorial primary, state Sen. Ghazala Hashmi (D) was powered by strength in her home turf in greater Richmond, while a very poor showing in the city where he was recently mayor, Richmond, cost Levar Stoney (D) dearly.

— Meanwhile, former state Del. Jay Jones (D) won a narrow victory of his own in the attorney general primary by building on his coalition from his unsuccessful run four years ago.

— Looking ahead to the fall, there is still ticket-splitting among the three statewide elections in Virginia, but that ticket-splitting has generally been on the decline.

— Virginia will also elect all 100 members of the state House of Delegates this fall, where Democrats already hold a narrow majority.

Recapping last night’s results

In his classic book Southern Politics in State and Nation, political scientist V.O. Key Jr. made a famous observation about “friends and neighbors” boosting local candidates in state races. A friends and neighbors effect played a dramatic role in Virginia’s barnburner of a Democratic lieutenant gubernatorial primary—although in a bit of a twist, the deciding factor may have been that one candidate’s friends and neighbors weren’t so friendly to him.

With neither party having gubernatorial primaries last night, as former Rep. Abigail Spanberger (D, VA-7) and Lt. Gov. Winsome Earle-Sears (R) had no opposition on their respective sides, the race at the top of the statewide ballot was the Democratic contest for lieutenant governor (Republicans had no statewide downballot primaries at all, and we continue to rate the gubernatorial race as Leans Democratic). This open-seat primary was, on paper, a 6-way race, although only 4 Democrats raised serious money. According to data compiled by the Virginia Public Access Project, former Richmond Mayor Levar Stoney, along with state Sens. Aaron Rouse and Ghazala Hashmi, had all raised more than $1.5 million in the lead-up to the primary, while Babur Lateef, who serves on the Prince William County School Board and is a former member of the University of Virginia’s Board of Visitors, was not far behind.

Before Election Day, the crowded primary seemed entirely up in the air. Even as returns came in, the top three candidates would typically remain locked within a percentage point or so of each other. But Hashmi, who was first elected to the state Senate in 2019 as part of the Democratic takeover of the chamber and who talked up her Senate experience on the campaign trail, came out ahead. She took about 27.5%, while Stoney was at 26.6%, and Rouse was only a few tenths of a point behind—Lateef was a distant fourth, finishing in high single-digits, although he did carry his populous home county.

Sam Shirazi, an insightful tracker of Virginia elections, pointed out that one factor in Hashmi’s close win was that she seemed to have late momentum. While yesterday’s contest set a record for early voting in a statewide Democratic primary, close to 60% of the vote was still cast on Election Day. In fact, with the roughly 40% of voters who cast ballots ahead of yesterday (these would be early in-person and absentee by mail voters) Hashmi would have come in a relatively distant third: Stoney and Rouse each took about 28% with those voters, while she got 24%. But Election Day voters gave Hashmi a tick over 30% of the vote, while Stoney and Rouse fell down to just under 26% apiece.

As a bit of an aside, even though this week’s primary lacked a gubernatorial race, statewide turnout was still basically comparable to 2021’s Democratic primary, which featured a multi-way primary for governor, even if its result was lopsided (former Gov. Terry McAuliffe won with more than 60%). As of this writing, about 480,000 votes were cast in yesterday’s race, down just slightly from the 495,000 that 2021’s gubernatorial primary saw.

It would also be hard to break down the primary for lieutenant governor without getting into geography. Map 1 breaks down the vote by locality, with data as of early this morning.

Map 1: 2025 Democratic VA LG primary results

Rouse, who has represented Virginia Beach in the state Senate since 2023, dominated in his region of the state. That media market cast 21% of the state’s vote and gave Rouse 56%. So this is what a “friends and neighbors” vote is supposed to look like. However, outside of his home area, Rouse only carried a handful of smaller localities. Rouse’s notable second-place finishes came in the Danville, Roanoke, and Fredericksburg areas, though he didn’t have as much success in the larger Richmond and Northern Virginia metros.

Getting outside of the Norfolk media market, Hashmi paired most of the remaining localities around Interstate 64 (notably the Richmond and Charlottesville areas) with those that touch Interstate 81 (notably the Roanoke metro area and the Shenandoah Valley). Though Stoney carried Southside Virginia and about tied Hashmi in populous Fairfax County, what stood out to us was how poorly he did in his home turf—in other words, he seemed to get whatever the opposite of a “friends and neighbors” boost could be called. Despite serving as Richmond’s mayor for most of the last decade, he lost the city 58%-21% to Hashmi. While there is sometimes obvious racial patterns in Democratic primaries, this only seemed to get Stoney, who is Black, so far in his home city: Hashmi took smashing majorities in the city’s white liberal pockets and still easily carried several precincts that are heavily Black by composition. Richmond ended up being Hashmi’s strongest locality in the state.

So why did Stoney fare so poorly with the voters who knew him best? Beyond the reality that being the mayor of a major city is ultimately a challenging job, a major part of the answer probably has to do with the city’s recent water troubles. Though the city’s water problems began flaring up shortly after he left office, it seems evident that voters still blamed Stoney to a large degree—while the water crises made national news in January, the city was put under a boil water advisory in late May, which could have represented a case of unfortunate timing for the former mayor, as it kept the issue fresh in voters’ minds. Even if Stoney had a visible role as mayor, Richmond’s voters were also quite familiar with Hashmi. Though redistricting removed some of her holdings in the capital city and pushed her out further into neighboring Chesterfield County, the state Senate seat that Hashmi held for the first 4 years of her time in the chamber included nearly half of Richmond, so she got a friends and neighbors vote instead of Stoney (Hashmi also carried the suburban counties around Richmond, although Stoney was generally more competitive in those). With that, Hashmi’s 10,500-vote margin out of Richmond more than accounted for her 3,600-vote statewide lead.

The Democratic primary was a two-way race that also produced a close result. Former Del. Jay Jones defeated Shannon Taylor, the Commonwealth’s Attorney for Henrico County, by a 51%-49% margin. Not surprisingly, both candidates pledged to fight the Trump administration if elected. Taylor, who took office in 2012—when Henrico County was not as Democratic as it is today—emphasized her experience, while Jones brought up his support from former Gov. Ralph Northam (D). Jones may have also benefitted from the fact this contest represented his second statewide campaign: four years ago, he challenged Mark Herring (D), who was then the incumbent state attorney general, and lost by 13 points.

Unlike the race for lieutenant governor, there was not an especially salient differential by voting method, as Jones took slight majorities of the early in-person, mail-in, and Election Day vote.

Geographically, Jones simply seemed to expand on his 2021 primary coalition. Four years ago, Jones carried the Hampton Roads area and Southside—his family has long been active in Norfolk-area politics—but Herring, who is from Loudoun County, put up landslide margins in Northern Virginia. Though Fairfax County gave Taylor a 7-point margin, Jones fought to roughly a draw in Loudoun and Prince William counties, losing them only narrowly to Taylor. In the Richmond area, Taylor carried her home base of Henrico County by a 60%-40% vote but her strength tapered off once one moved south of the James River, as she narrowly lost Richmond and Chesterfield counties. Map 2 shows how Jones built off his 2021 performance by adding about two dozen localities that he lost four years ago.

Map 2: 2025 Democratic VA AG primary results

Jones enters the general election against current state Attorney General Jason Miyares (R), who is seeking a second term. Hashmi is set to face radio host John Reid (R) in the lieutenant governor’s race. Outgoing Gov. Glenn Youngkin (R) unsuccessfully tried to push Reid out of the race after opposition researchers linked Reid, who is gay, to a blog featuring male adult content; Reid denied that link and declined to leave the race. There is some question about how much, if at all, the GOP statewide ticket will campaign together. Speaking of, let’s look at the recent ticket-splitting trends in Virginia’s statewide elections.

Ticket-splitting on the decline in statewide VA races

It’s common for there to be different margins among the three Virginia statewide races. But, as with ticket-splitting in other kinds of races across the nation, the general historical trend is toward less ticket-splitting in these contests.

Table 1 shows the election results for the three Virginia statewide elected offices since 1969. We chose this starting point because 1969 is a key dividing line in Virginia history, as Linwood Holton (R) became the first modern Republican to win the governorship, starting an era of two-party competition in the state following a long run of conservative Democratic dominance.

Table 1: Statewide elections in Virginia, 1969-2021

Source: Our Campaigns, Virginia Department of Elections, Dave Leip’s Atlas of U.S. Presidential Elections

Notice that the 2021 election featured a historically-low amount of ticket-splitting. There was hardly any difference among these three races, with Republicans winning the governor’s race by 1.9 points, the lieutenant governor’s race by 1.5, and the attorney general’s race by 0.8. That’s the smallest gap between the best Democratic and best Republican margins of any election in this timeframe.

Over the past five elections, the difference between the best margin for each party hit double digits only once, in 2013. That year, Ralph Northam (D) won the lieutenant governor’s race by a little over 10 points against a weak Republican nominee. In the same election, the aforementioned Mark Herring (D) beat Mark Obenshain (R) by less than a tenth of a percentage point for attorney general. Obenshain, who is still in the state Senate, is a member of a famous Republican political family. In the other recent elections, the differences among these races were smaller.

Attorney General Miyares, as the only incumbent seeking reelection to his current job this year, is likely the safest bet to finish as the best-performing Republican. Four years ago, Herring (D) was the only incumbent on the 2021 Democratic slate, and he turned in the best performance, though it was only subtly different from the two other races and was not enough to hold on against Miyares.

That said, incumbency is not always predictive of the top finisher on one side or the other: Herring, running for reelection to a second term as AG, himself finished a couple of points behind Northam’s margin in 2017 as the then-lieutenant governor performed the best of the three statewide Democratic candidates. In 2009, Republicans won all three offices in a romp, but the one incumbent running for reelection to his post—then-Lt. Gov. Bill Bolling—actually won by the smallest margin. So there aren’t necessarily hard-and-fast historic rules here.

Our main takeaway here is that we might not see the three races as closely-tied as they were in 2021, although we also shouldn’t expect the big, double-digit differences we often saw in the late 20th century.

An early look at the House of Delegates

Back in 2023, Democrats retained control of the Virginia Senate and flipped the Virginia House of Delegates from Republicans, preventing Gov. Youngkin from enjoying a governing “trifecta” for the last two years of his governorship.

The Senate is not up until 2027, so we know that the best Republicans can do in November is preserving a form of divided government. It is worth noting that the lieutenant governor can break some ties in the state Senate (although not on the all-important budget), and the Democratic majority is only 21-19. So if John Reid won the lieutenant governor’s office and Republicans later managed to flip a current Democratic seat in a special election—this is all just hypothetical because it would require a Senate vacancy—they could bring it to 20-20 with a Republican tiebreaker. The state Senate majority actually was on the line in a special election earlier this year, but Democrats easily held a clearly Democratic-leaning seat in Northern Virginia.

The bottom line here is that Spanberger and the Democrats have a clear path to a governing trifecta in November, whereas Republicans really don’t because of Democratic control of the Senate.

While the Senate is not on the ballot, all 100 seats of the House of Delegates are. Democrats also hold a slim majority in that chamber, 51-49. But the map itself, at least at first blush, suggests Democrats have a good opportunity to do better than that—especially if Spanberger wins the governor’s race going away.

Republicans got as close as they did to winning the majority in 2023 by carrying the most marginal races: They won the five closest races, which were all decided by less than 2 points. The closest Democratic victory, by Del. Josh Thomas in the Prince William County-based HD-21, came by 3.6 points. Thomas also helps illustrate how blue-leaning the 51 current Democratic seats are: Kamala Harris won HD-21 by 5.2 points, making it the only currently Democratic-held district that voted to the right of the state for president last year (Harris won the commonwealth by 5.8 points). Of the Democrats’ 50 other seats, all of them voted for Harris by double digits except for HD-65 (Harris +9.5), a Fredericksburg-based district held by Del. Joshua Cole, and HD-97 (Harris +7.8), a Virginia Beach-based district held by Del. Michael Feggans. (We used State Navigate, a great new site focusing on state-level elections, for these figures, along with Redistricter.) On Tuesday, Republicans nominated Gregory Gorham to face Thomas in HD-21; Gorham had just $146 in his campaign account as of the most recent reporting period compared to Thomas at about $192,000 (in the other two races mentioned above, the Democratic candidates have cash edges over their Republican opponents, too, although the gap isn’t as stark).

One thing that could give Republicans hope in the race for the state House is that presidential district-level numbers in Virginia state legislative races have typically represented more of a hard-to-reach ceiling for Democrats as opposed to an easy-to-reach floor in recent years. Back in 2023, Democrats in state House and Senate races featuring two-party competition ran about 8 points on average behind Joe Biden’s 2020 two-party district-level margins. When compared to Harris’s results, Democrats ran roughly 4 points behind her on average in those same 2023 races (that makes sense—Harris also ran a little bit more than 4 points behind Biden’s 2020 showing). If Earle-Sears wins outright or loses only narrowly, she likely would end up carrying some Harris-won districts (for instance, Youngkin narrowly carried those three aforementioned most competitive Harris-won Democratic districts in 2021). It also is probably the case that some areas that have been realigning toward the Democrats at the presidential level retain some residual Republican loyalty down the ballot in state races.

It is worth noting, though, that in 2017—the recent election year most comparable to this one, taking place concurrently with the gubernatorial election and with a Republican in the White House—Democrats came closest to hitting the most recent presidential numbers, with Democrats running, on average, just a point behind Hillary Clinton’s 2016 two-party margin. If Democrats perform like they did in 2017—when they swept the three statewide elections by at least 5 points apiece—one could easily see them netting seats in the House of Delegates, as Republicans are defending eight districts that Harris won and a few others that Donald Trump carried but that could feature competitive races.

We’ll take a closer look at the House of Delegates closer to the election: The main thing we wanted to note is that Spanberger, if elected, would have a good chance of being elected with friendly state legislative majorities, while Republicans would not have the Senate and perhaps not even necessarily the House if Earle-Sears won.

P.S. Unlike in the Senate, where the lieutenant governor presides and can act as a tiebreaking vote, the 100-member House has no such presiding officer and thus no tiebreaker.

Kyle Kondik is a Political Analyst at the Center for Politics at the University of Virginia and the Managing Editor of Sabato's Crystal Ball.

J. Miles Coleman is an elections analyst for Decision Desk HQ and a political cartographer. Follow him on Twitter @jmilescoleman.

See Other Political Commentary by Kyle Kondik.

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.