Tomorrow, November 5, Virginians will head to the polls to elect a new governor, lieutenant governor, and attorney general. In addition, each of the one-hundred delegate districts will be voting for their representative in the House of Delegates, and there will be a number of local referendums and offices on ballots across the state. Based on available polls and other information, I think there are a few possible scenarios that could play out tomorrow in the gubernatorial race. This piece will focus on that race, but I will also spend a bit of time at the end on the ‘down-ticket’ races being followed by Off on a Tangent this year.
Before diving into the scenarios, here is the bottom line. Below is a custom polling chart that I made using the tools made available by HuffPost Pollster. I started from the ‘official’ HuffPost model, but narrowed it to polls that consider ‘likely voters’ (as opposed to ‘registered voters’ or ‘all adults’). In addition, I excluded some pollsters that are affiliated with the Democratic and Republican parties, or have an obvious history of unreliability and partisan bias. I cross-referenced on my own with the RealClearPolitics poll aggregation, which paints a similar picture.
In general, I think this chart accurately reflects the state of the race . . . although I am not confident about the up-tick in Terry McAuliffe’s (D) numbers right at the end, which is based entirely on one poll. If I was in charge of the algorithm painting the blue line, I would have it bent slightly lower from where it is at the last point.
Scenario 1: McAuliffe Wins Comfortably (50 percent chance)
Given the nature of the race and polling trajectories, it is very likely that Terry McAuliffe (D) will defeat Attorney General Ken Cuccinelli (R-VA) and Robert Sarvis (L) by a comfortable margin.
This is unfortunate, in my opinion . . . in part because it means that McAuliffe will be rewarded for what was the most dirty, dishonest state-level campaign in recent Virginia history. The long-time Democratic strategist managed to paint Cuccinelli as an absurd caricature of himself, and the Republican attorney general was unable to effectively counteract the smear campaign. This bodes ill for the future of Virginia statewide politics, which had heretofore been relatively moderate and congenial compared to the hyperactive lies and spin of so many national campaigns.
It is also notable because Virginia, previously a moderate and pragmatic state, now seems inclined to choose modern progressivism—complete with its disdain for civil liberties and an intransigent fiscal immoderation—over the modern conservative alternative. Not long ago, it would have been unlikely for either Cuccinelli (a ‘real’ conservative) or McAuliffe (a ‘real’ progressive) to stand on the statewide ballot at all in a state that values bipartisanship and compromise (and a bit of cross-party gridlock now and then). And not long ago, if forced to choose between them, most Virginian’s would have picked the one who was more in favor of civil liberties, low taxes, and limited government. Instead, they seem poised to choose a candidate who would be more at home in a near-bankrupt left-wing utopia like California.
It is also telling that Robert Sarvis (L) is polling so well. This tells us that the old, moderate Virginia is still out there . . . and many of those Virginians don’t like either of the major party options. Sarvis, like most Libertarians, is more in-tune with traditionally Republican voters on fiscal and economic issues, while being more in-tune with traditionally Democratic voters on social issues. Recent polling indicates that Sarvis is pulling voters from both Cuccinelli and McAuliffe, but now seems to be drawing more heavily from Cuccinelli.
I suspect this is because moderates who tend Democratic are those who tend to prioritize a somewhat left-wing view on social issues, and so they are relatively happy with McAuliffe even if he goes a bit further than they would. Moderates who tend Republican, on the other hand, are more suspicious of the ‘party machine’—which has too-often supported corporate welfare and cronyism—than they have ever been before . . . and with good reason. These voters also tend to not recognize the fundamental right to life, and see opposition to abortion not as the basic human rights issue that it is, but as some kind of ‘radical,’ or even ‘theocratic’ stance (a perception that is misguided, at best).
I do suspect that the polls are slightly overstating McAuliffe’s lead. I expect that nearly half of the voters who say they will vote for Sarvis will defect to a ‘lesser of two evils’ major-party candidate. Among Sarvis defectors, I think they will break about two-thirds for Cuccinelli and one-third for McAuliffe, which will narrow McAuliffe’s lead somewhat. There are also about four percent of voters still undecided, and I expect them to break about even between the two major candidates.
I believe that this scenario, or some lightly-modified version of it, has a fifty percent chance of playing-out tomorrow. My best guess for the actual outcome would be something like: 50 percent for McAuliffe, 45 percent for Cuccinelli, and 5 percent for Sarvis. It is worth noting that Sarvis, although he will almost certainly be made into a scapegoat by some bitter Republicans in the aftermath, is not actually a ‘spoiler’ for Cuccinelli in this scenario. Even if all Sarvis voters would have supported Cuccinelli in a two-way race, it would only be enough to turn the election into a virtual tie. And, in reality, we cannot assume that any more than two-thirds of the remaining Sarvis voters would have gone for Cuccinelli. Some of them would have voted for McAuliffe in a two-candidate race, and some are ‘true believers’ who either wouldn’t vote, or would have written somebody in, had Sarvis not been in the race.
Scenario 2: McAuliffe Wins Narrowly (45 percent chance)
As discussed in my review of the first scenario (above), I think that the polls are somewhat overstating Terry McAuliffe’s (D) lead over Attorney General Ken Cuccinelli (R-VA) in this race. But even after compensating with an assumption that Robert Sarvis (L) won’t pull anywhere near ten percent, and assuming that his defectors tend to the Republican side, McAuliffe still comes out with a comfortable lead.
However, this outcome is based on a number of other assumptions. We assume that the remaining four percent of voters who are undecided will split about fifty-fifty between the two major candidates, and we assume that the pollsters are accurately identifying the ‘likely voters’ who will show up at the polls on Tuesday. These are a lot of assumptions, especially considering that turnout in off-year elections is notoriously hard to predict. And since off-year turnout is generally quite low compared to presidential election years, small things—like rain storms—can cause shifts in who goes to the polls and who doesn’t and thereby have surprising impacts on the race outcome.
A race like this one—where one candidate seems to be solidly ahead, but not by so much that you might consider the election a ‘blow-out’—is also prone to significant narrowing on election day. This happens because some supporters of the leading candidate think that the election is a ‘done deal’ and that their vote won’t matter, so they don’t put a lot of effort into getting to the polls. Meanwhile, some supporters of the trailing candidate get motivated to make a last-minute push, or decide that they’ll go to the polls (when they might have planned to skip) in hopes of pushing their preferred candidate over the top.
In this second scenario, we start with the first—assuming that many Sarvis voters will defect, and that they will break more for Cuccinelli than for McAuliffe. But in addition, we add undecided voters breaking somewhat more for Cuccinelli than for McAuliffe, and assume that the race will experience a narrowing like that described above. Even with all of these factors added into the race (conservatively weighted), we still end up with a McAuliffe win.
I believe that this scenario, or some lightly-modified version of it, has a forty-five percent chance of playing-out tomorrow. My best guess for the actual outcome in this case would be something like: 48 percent for McAuliffe, 47 percent for Cuccinelli, and 5 percent for Sarvis. In this scenario, some bitter Republicans will again point to Sarvis as a ‘spoiler,’ but here they would have a stronger argument than in the first scenario. Assuming that this second scenario comes to pass, it is possible (though not assured) that a two-way race would have ended up going narrowly for Cuccinelli, assuming that Sarvis’s remaining five percent tend more to the conservative side of the spectrum.
Scenario 3: A Cuccinelli Upset (5 percent chance)
I cannot analyze this race without acknowledging the possibility—very small, but distinct—of an unexpected win for Attorney General Ken Cuccinelli (R-VA). The fact of the matter is that Terry McAuliffe (D) is deeply unlikable, and has run an extraordinarily dirty and negative campaign . . . a kind of shameless politicking that still resonates badly with many Virginians. It is possible that some of his supporters will, at the last minute, choose not to vote at all.
Low turnout tends to benefit Republican candidates, as Republican-leaning voters tend to be more passionate about the voting process and the exercise of their civil rights (and civic duties). Additionally, particular to this race, some conservative voters—perhaps over-dramatically—see this election as a pivotal moment in Virginia history where we can choose the paths of either liberty or statism. There are progressive voters who view the race in similar terms, particularly among those who irrationally conflate abortion with ‘women’s rights’ (newsflash: roughly half of abortion’s victims are women) . . . but they seem fewer and further between. It is unclear whether McAuliffe’s over-the-top ‘war on women’ rhetoric will be as effective as he thinks it will be in the Old Dominion, where we have traditionally tended to see through nonsense.
So if we take all of the assumptions in the second scenario and bump them just a hair further, this race suddenly becomes a similarly-narrow Cuccinelli upset. If the turnout shifts a bit more to the ‘human rights and civil liberties’ side, or if Sarvis loses more of his voters than I expect him to, or if more McAuliffe’s supporters balk at the last minute, this race is still winnable—just barely—for the Republican candidate.
I believe that this scenario, or some lightly-modified version of it, has only a five percent chance of playing-out tomorrow. My best guess for the actual outcome in this case would be something like: 49 percent for Cuccinelli, 48 percent for McAuliffe, and 3 percent for Sarvis.
At a glance, the race for lieutenant governor has similar dynamics to the governor’s race, but there are a number of details that shift it more favorably for state Senator Ralph Northam (D-VA 6th). I consider it very unlikely that Reverend E.W. Jackson (R) would be able to eke out a win here, even if the voters go narrowly for the Republican candidate at the top of the ticket.
Although there is no Libertarian (or any other third-party) candidate to siphon off voters from the major party candidates, which would seem to benefit Jackson, Northam is a much more likable candidate than his up-ticket party-mate which significantly blunts any benefit. Northam has also shied away from the kind of dishonest, negative campaigning so prevalent among Democratic Party candidates in this election cycle, which will help him with moderate voters.
No matter which gubernatorial scenario comes to pass, I consider Northam’s election to be as much of a sure bet as you can get in off-year politics. My best guess for an outcome is something like: 55 percent for Northam and 45 percent for Jackson . . . although good, detailed polls are hard to come by, so I have relatively low confidence in this estimate.
Like the lieutenant governor race, the race for attorney general seems to have similar dynamics to the race at the top of the ticket . . . but here, the details shift it somewhat more favorably for state Senator Mark Obenshain (R-VA 26th). That said, state Senator Mark Herring (D-VA 33rd) could also still win this race easily. I would call this race a true toss-up.
Herring has run a negative, dishonest campaign much like that of the gubernatorial front-runner, but Obenshain has done a better job of refuting Herring’s misrepresentations. More fundamentally, many Virginians understand that the attorney general is the top law enforcement officer in the state—not a ‘general purpose’ politician—and so many of them focus on the relevant civil rights and law enforcement elements of the candidates’ campaigns. The ‘hot button,’ hyper-partisan issues are less in-play here.
I would say that, if the first gubernatorial scenario comes to pass, the attorney general’s office will likely go to Herring by a narrow margin. If the second gubernatorial scenario comes to pass, it will probably go to Obenshain by a similarly narrow margin. In the unlikely event of a Republican win at the gubernatorial level—the third scenario—Obenshain will likely win in this race by a more comfortable margin.
My best guess for an outcome is basically a tie, but I think that Herring is the [very slightly] more likely winner. I would guess something like: 50.5 percent for Herring and 49.5 percent for Obenshain . . . but I have a very low level of confidence in this estimate.
House of Delegates, 87th District
Reliable poll data is very hard to come by for Virginia’s state legislature races, but the most recent poll available for the 87th District seat shows incumbent Delegate David Ramadan (R-87th) in a 47-47 toss-up race with retired Air Force Major John Bell (D).
I have very little to go on in making a prediction here. My hunch is that Ramadan, who won election in 2009 by the slimmest of margins, will win by the slimmest of margins again and hold his seat. Although Ramadan is probably to the ‘right’ of the average 87th District voter, he has also been an extraordinarily strong advocate on important local issues, particularly relating to transportation. This likely blunts some of the ideological opposition and hurts Bell’s chances in an otherwise center-left district.
My best guess here is another tie, but I do think that Ramadan will eke out a perilously narrow win. I would guess something like: 50.5 percent for Ramadan and 49.5 percent for Bell . . . but, like in the attorney general race, I have a very low level of confidence in this estimate.
Loudoun County Bond Referendums
It’s almost not worth the time to write these up, since bond referendums pretty much always pass by a comfortable margin, but I might as well say some words in the interest of completeness. I expect a similar pattern to what we have seen in past years, with each bond referendum passing with a solid super-majority of the vote.
Most likely, the Transportation and Fire & Rescue bonds will each pass with about 75 or 80 percent of the vote, give or take a few points. I expect the Schools and Parks bonds to pass as well, but only with about 65 or 70 percent of the vote, give or take. These are based mostly on past performance, not on any poll data relating to this specific election, so I have a low level of confidence in the exact numbers.