Battleground Virginia

What Do the Campaigns See That the Experts Are Missing?

Michael Cohen
Soapbox
Published in
8 min readNov 7, 2016

--

Tuesday’s elections for President of the United States and control of the House of Representatives will probably not be decided based on what happens in the Commonwealth of Virginia and it’s tenth congressional district. In the closing days, however, all four campaigns and their party committees, are not acting that way.

Two Questions

The reason is simple: Virginia remains a battleground state. This year, Virginia can provide a partial answer to two questions that political scientists, professionals, and interested voters have been debating for many elections, but particularly this one.

  1. Is demography destiny?
    Do long-term shifts in who votes make the difference on Election Day? Increasing voter diversity and changes in where they live have moved Virginia to a purple state.
  2. Do events during campaigns matter?
    How much can an October Surprise influence a tightly contested election? Is the reopening of the FBI case against Clinton and Sunday’s re-closing, powerful enough to tip the balance between red and blue?

President: Demographic Advantage

For generations, Virginia voted solidly Republican. From 1952 to 2004, Democrats won the state only once, in the 1964 Lyndon Johnson landslide. Even when Democrats nominated southern governors and won, Virginia held the line. Jimmy Carter and Bill Clinton lost the Old Dominion on their way to the White House in 1986, 1992, and 1996.

Demographic change. Barack Obama’s first election to the White House in 2008 was the breaking point and his reelection in 2012 confirmed it was not an outlier. Obama was the beneficiary of two significant shifts in Virginia’s demography, both favoring Democrats. First, the Virginia electorate became more diverse. According to a report from the University of Virginia, the 1970 Virginia population was about 80 percent non-Hispanic white. By 2010, that dropped to 65 percent non-Hispanic white. As Pew found in their study of trends in party identification from 1992 to 2016, these shifts favor Democrats.

Courtesy of the Pew Research Center

Second, the electorate became more concentrated in Democrat-leaning areas. A rapid five-year growth of Northern Virginia counties saw shifts from GOP-leaning counties to more urban counties, which generally lean Democrat. This trend has slowed since 2010 but the general pattern remains.

Courtesy of University of Virginia’s Demographics Research Group

Political background. These trends have resulted in a state of politics where power is shared, reflecting today’s divided federal government reality. Democrats have leveraged the changing demography to win statewide governor, lieutenant governor, attorney general, and both U.S. Senate elections. As Public Policy Polling noted, Kaine’s job approval rating among Virginians actually increased despite joining the Clinton ticket.

Virginia’s Electoral College votes are awarded as a bloc making this a statewide election, favoring Democrats. This is the opportunity Hillary Clinton saw in choosing her running mate. Her selection of Tim Kaine, a popular former mayor, governor and current senator, seemingly removed Virginia from the battleground map.

The October/November Surprises. Despite this documented shift in demography, we once again learn that campaigns and events still matter. Like Trump, Clinton was never a popular candidate and doubts about her honesty and integrity have been long-term issues for her. The reopening of the FBI investigation brings those feelings back to the surface, particularly for Republicans, and it’s borne out in the RealClearPolitics averages for Virginia below. Sunday’s release of the letter closing the investigation, again, is unlikely to give Trump any more of a boost, which appears to have flattened, anyway.

Virginia Polling Averages | Courtesy of Real Clear Politics

Still, Virginia has resurfaced as a battleground state. Clinton’s lead on the evening of the final debate was about 10 points. The day before the election, the lead had been cut in half, almost within the margin of error of even the best publicly reported statewide political surveys. This is why Northern Virginia is getting such attention in the closing days.

To be fair, these averages are based on very limited data. One missing poll by the Hampton University’s Center for Public Policy conducted October 26–30, two days before and after the the FBI announcement, showed Trump ahead 44 percent to 41 percent. Also, we do not have access to either campaign’s internal polls, which likely also show the race close enough to send the principals back to hold rallies.

Finally, again, as I was writing this (hat tip to my old friend, Steve Malter, for the late news) the FBI decided that it would not change the conclusion it reached in July that Clinton should not face criminal charges based on the information shared from her private email server. It is likely that most Virginians, wrapping up their weekends, will learn about this on Monday and some might miss the news. The damage to Clinton is already done but it looks like the Trump surge already hit its ceiling.

Caution. Statewide Virginia polling has been suspect as recently as 2014, albeit during an off-year election. Polls leading up to Senator Mark Warner’s reelection underestimated how close the race was with former GOP chair Ed Gillespie. September polls had Warner up by more than 20 percent and the final RealClearPolitics average had Warner leading by about 10 percent, still a comfortable lead.

As POLITICO reported, “the race broke late. Exit polling showed that 10 percent of Virginians decided who to vote for in the final three days.” The race for president this year is also breaking late and we do not have enough data to say if it has stabilized or is still in flux. It’s anyone’s guess whether the likely voter models will hold up or if Democrat turnout operations will be able to close any gaps we’re seeing in polls.

Based on the final polls showing Clinton’s lead stabilizing, and that the FBI will not prosecute her, events will not overtake her potential demographic changes. While this may have been a toss up a few days ago, it’s now more likely that Clinton will win Virginia.

Congress: Fierce Battle in District 10 Toss-Up

Virginia’s sprawling District 10 is close enough to the nation’s capital that we know many who work in undisclosed locations and just far enough away to know that their commutes stink. Ranked the best place to live in Virginia by Money Magazine, our corner of exurbia is a happy, high growth, high-income, highly educated, child-driven experience.

Demography. Voter demographics have shifted over the past ten years. Residents are now more diverse than the state and its main counties, Loudoun and Fairfax, lead state growth and are becoming less rural, more suburban and urban, which leans towards Democrats.

Political background. Viewed as a genteel moderate, Republican Frank Wolf hadn’t had opposition since the early 1980s and happily retired from 34 years of service on Capitol Hill in 2014. The open seat was won by Barbara Comstock, a former aide, opposition researcher, and member of the Virginia House of Delegates, who routed Democrat John Foust, 57 percent to 40 percent. Foust got close but then gender gaffed “I don’t think she’s even had a real job” and it was pretty much over from there.

Opponent. This year, Comstock faces a much better opponent in a much more competitive political climate. LuAnn Bennett is a real estate executive and has served as an appointee of then-Governor Tim Kaine on various commissions including climate change, health care, and children’s issues. A first-time candidate herself, she was once married to northern Virginia Congressman Jim Moran, who retired in 2014. While she’s never been elected to office it would be a mistake to characterize her as a political neophyte.

Activity. While outside groups have been very active, incumbent Comstock out-raised Bennett $4.6 million to $2.2 million. Still, living here you couldn’t tell the difference. Both flooded our airwaves with negative television ads, including quite a few during the World Series. My (real) mailbox was full of recycling bin-worthy negative ads for several weeks. Our doors were visited about equally by canvassers. Even worse, our social media feeds have been littered with paid and neighborly suggestions on why Comstock is Trump-lite or Bennett is Clinton-lite.

Social media. While Comstock had more followers entering Election Day, Bennett was growing quicker, suggesting a much greater enthusiasm for her candidacy. Both were trending up and the last few days indicate that the slope of that growth was about equal.

Still, if a relative unknown could go toe-to-toe with someone who is well-known, you give the advantage to the challenger. This is the Sanders scenario I described in this PEORIA Project study for the Graduate School of Political Management at George Washington University.

Courtesy of Crimson Hexagon | Barbara Comstock’s Follower Rate
Courtesy of Crimson Hexagon | LuAnn Bennett’s Follower Rate

Polling. Very limited data is available but what we have shows an incumbent in trouble. Only two polls are available and both are from the same Democrat-aligned firm. A Expedition Strategies poll of 400 likely voters conducted September 19–21 had Comstock up two points, 47 percent to 45 percent, well within the poll’s 4.9 percent margin of error. A poll by the same firm October 10–12 found that Bennett was two points ahead, 46 to 42 percent. Republican pollsters have been conspicuously quiet.

In 2014, the Tarrance Group released results from a poll taken September 23–25 showing Comstock with a 12 point, 46–34 percent, lead over Foust. A poll taken October 17–18 sponsored by Citizens United Political Victory Fund had Comstock with a 14 point lead, 51–35 percent. Bottom line: if Comstock were winning the 2016 election against Bennett, it’s likely we would have heard about it by now.

The race to represent the tenth district is now rated by Cook Political Report as a toss up where just three weeks ago it was rated as leaning Republican. University of Virginia’s Sabato’s Crystal Ball also moved their rating to toss-up from leans Republican on October 20 only to move it back (!) on November 7. It is unclear if the recent volatility in the presidential race has affected the congressional campaign. All we know is that both sides are fully engaged.

Based on improving polling, higher social media growth, and demographics, both major campaigns visiting Northern Virginia in the final days, and the FBI setting aside its case against Clinton, Bennett likely has a slight edge over Comstock entering Election Day.

After the FBI’s October 28 surprise, both presidential and congressional campaigns saw the race tightening and started to act and plan, accordingly. Democrats saw recent demographic trends that were favorable and Republicans saw recent events moving in their direction. Now that the FBI has removed the remaining cloud hanging over Clinton, polls showing Clinton maintaining her lead will likely hold. Democrats are in position to win Virginia and the tenth congressional district for the first time since 1980.

If you enjoyed this article, click the💚 below so other people will see this here on Medium. Follow me on Twitter @michaelcohen. You can follow our research on this website or on Twitter @PEORIAProject, which is funded by a generous grant from Mark R. Shenkman. To learn more about the Graduate School of Political Management visit our website or follow us on Twitter @GSPMgwu.

--

--

Michael Cohen
Soapbox

Founder of Cohen Research Group. Publisher of Congress in Your Pocket. Lecturer at Johns Hopkins. Author of Modern Political Campaigns