What Went Down During The 2021 Elections

Although today is Election Day, voting has already been underway in Virginia for several weeks. According to The Virginia Public Access Project, 1,142,308 people had already voted (either early in person or by absentee ballot) as of Oct. 31. And according to Democratic data firm TargetSmart, approximately 53 percent of these voters are Democrats, 31 percent are Republicans and 16 percent are independents.

But don’t read too much into this early voting data. It does not tell us much about who is going to win.

First, while a sizable number of Virginians have voted early, a substantial number will vote on Election Day. In fact, a recent Washington Post-Schar School poll found that 56 percent of likely voters plan to vote in person on Election Day, with these voters favoring Youngkin by 19 points. So although Democrats have likely taken a strong lead with early votes, Republicans can easily close the gap with strong turnout on Election Day.

Second, early voting data tells us who has voted, but it does not tell us how they voted. Although “it all comes down to turnout” is a popular axiom, vote choice among swing voters matters. And a major reason the Washington Post poll found a competitive race was that independent voters favored Youngkin by 18 points, a turnaround from favoring Biden by double digits in 2020. It’s possible, too, that some of these voters voted early.

And finally, we don’t really have a good comparison point to benchmark this year’s early voting statistics. In the 2017 gubernatorial election, early voting was far more restricted, while in the 2020 presidential election, many voters who would normally vote in-person voted absentee because of the pandemic.

So while it is encouraging that many voters have already cast ballots, we’ll have to wait until election night to find out who’s going to win.

Our colleagues at ABC News, Kenneth Moton and Beatrice Peterson, are on the ground in Virginia today, and they said in an email that voters were lined up to cast their ballots early this morning in Fairfax County — one of the state’s most populous counties.

In an effort to better understand how voters were feeling in this closely contested race, they talked to several women earlier this morning, and at least two women outside of the suburban polling location told them that they voted for Youngkin. For Susan Moore, education was top of mind, while for her friend Lynne Kirk, the big-ticket issue was abortion.

Kirk told them on camera that McAuliffe “had the opportunity to be governor before, and I don’t feel like he did a great job at that time.” Kirk added, “It’s time to give somebody else another chance.”

But Moira Skelley told them that she voted for McAuliffe, telling ABC News, “I just feel like if Youngkin was to become governor, I just don’t know what would happen to Virginia. I would really not like to see it turn red.”

No matter what happens in Virginia, the outcome will be oversold as evidence that the 2022 midterms will go a certain way. A Youngkin win will be described as a doomsday for Democrats, whereas some will be point to a McAuliffe win as a sign that Democrats will avoid crushing losses next year.

The truth is more complicated, of course, as one election can never tell us everything about future electoral conditions. But just how predictive is Virginia’s gubernatorial race of the next midterm election?

CNN analyst and former FiveThirtyEighter Harry Enten looked at this question back in 2017 and found that Virginia hasn’t been a terribly strong predictor of the next midterm since the early 1990s. Updating his analysis, I found this largely remains true. From 1993 to 2018, the average difference between the swing in Virginia’s gubernatorial election from the state’s partisan lean and the margin in the midterm national popular vote for the House of Representatives has been just shy of 7 percentage points.

However, that doesn’t mean Virginia gubernatorial contests haven’t told us anything. The correlation between the swing in Virginia’s governor race and the House vote margin during that timespan is .56, which suggests there’s a moderately positive relationship between them (although we’re operating off a small sample size).

Simply put, when the Virginia result has swung more Republican, the national House vote has tended to go for the GOP, too, and vice versa for Democrats.

More broadly, then, the Virginia election can’t predict the national House vote a year from now, but it can potentially signal the direction of that vote. Heading into tonight, the polls suggest Virginia has a very close race despite being 5 points more Democratic than the country as a whole, according to FiveThirtyEight’s partisan lean metric. And perhaps more importantly, the factors that have created this GOP-leaning environment, such as Biden’s poor approval rating, could push voters across the country in a similar direction next year.