Some Thoughts on Virginia
With polling places opening for Election Day in Virginia in less than 48 hours from when I write this, I wanted to share my thoughts on what I’m seeing in the various data indicators we have at our disposal.
The short (and overly trite) version is that… it’s likely to be close, and it will come down to Election Day. I recognize that’s not cutting edge analysis, but let me share how I’ve reached that conclusion by sharing some thoughts on what each side should feel optimistic about at this point:
Dems Should Feel Good About
Historical Trends and a Good Cushion — In much of this analysis I’ll compare the 2021 early vote to 2020 and 2017. But in the end, Dems don’t need an electorate composition as favorable as either year! Let’s keep in mind that Biden carried the state by 10 points. Similarly, the 2017 electorate produced a a 9 point Dem margin of victory. Lagging behind either year, in terms of turnout shares for traditionally Democratic constituencies, isn’t necessarily a harbinger of doom.
Relatively Strong Black Turnout — While Black voters aren’t reaching the early vote share levels they reached in 2020, they are exceeding 2017 vote shares by a wide margin.
Late Surging Vote — As recently as last week the early vote turnout looked relatively bleak for Democrats, with key northern Virginia counties lagging badly. Since then we’ve seen a late surge with much of the gap closing. Democrats have reason to hope that their voters are simply coming out later than usual this year.
GOPs Should Feel Good About
An Older Electorate — The youth vote (under age 30) is lagging badly behind previous benchmarks. Sure, we would expect the youth vote to lag the vote shares we saw in the historic 2020 presidential election. What is more disconcerting for Dems is how far behind 2017 benchmarks the youth vote currently lags. Yes, 2017 had much stricter rules around voting early, but one would assume the liberalization of early voting rules for this election relative to ’17 would result in a younger early vote electorate.
A More Rural Electorate — The rural early vote share is well ahead of where it was at this point in ’17, and slightly ahead of 2020, while the urban and suburban vote shares lag. Republicans are expected to dominate the vote count in most rural communities.
A Potential Enthusiasm Advantage- The modeled partisanship of early voters who are casting a ballot in their first election has a GOP margin that is 16 points greater than the margin among “super voter” early voters — those who vote regularly. This might point to an early GOP enthusiasm gap.
Mixed Indicators
Modeled Partisanship — Our TargetEarly site breaks down the early vote by modeled partisanship, a helpful indicator of the potential party self ID of those who have already cast a ballot in Virginia. Since voters aren’t able to register with a party in Virginia, this is our best indicator of relative partisan turnout. That said, we must keep in mind that modeled partisanship is a moving target. Each year, our team updates the partisan models, reflecting our best estimate of each voter’s political leaning. As Virginia has trended more Democratic in recent years, the model has also seen more movement towards the Democratic side of the scale. So, while the TargetEarly data shows a stronger Dem advantage among early voters than we have seen in recent elections, at least part of that is almost certainly due to movement in the model.
Polling -The 538 average of polls is an incredibly useful tool for getting a sense of where any race stands, and where the trend line has been moving. That said, it is a function of its inputs. In this case, we have seen the 538 average move towards Republicans in recent days, now showing a slight Youngkin advantage. That said, the Republicans have effectively flooded the zone with polls from GOP consulting firms in the past week, all showing a Youngkin lead. The fact is that there is only one poll in the past month that shows a Youngkin lead that was not conducted by a Republican consulting firm, and that poll was conducted by FoxNews, and had its own apparent sampling issues. And of course, we can’t talk about polls in Virginia without noting that the poll average in 2017 was a bigger miss of the actual result than the presidential polls in ’16 or ’20.
The Bottom Line
This race will almost certainly be close. At a minimum, closer than the Dem statewide wins in ’20 and ’17. Election Day will be decisive. I’ll have more to share on election night! Feel free to follow me on Twitter.
Note: most of the graphics in this post came from our TargetEarly site. Check it out, and share anything interesting you dig up!
