Mellman: Election 2021 aftermath: How big was that swing?

A few brave souls have tried to argue that last Tuesday was a good night for Democrats.

That’s a very difficult assertion to defend.

But take the analysis much deeper and you’ll find evidence for a variety of contradictory propositions in the data from Virginia.

For example:

The swing against Democrats was mammoth.

Comparing the margin in the 2020 presidential race to the gubernatorial contest this year reveals a 12-point swing to the GOP in Virginia.

The swing to the GOP wasn’t that big.

Comparing Terry McAuliffe’s margin in 2021 to his own previous outing in 2013 shrinks the swing to just over 4.5 points — less than the swing against former President Trump from 2016 to 2020, which pundits derided as puny.

In the races for lieutenant governor and attorney general, the swing from 2017 to this year was about 7 points, just slightly bigger than the anti-Trump swing.

It was all about turnout.

Analysts typically turn first to exit polls for such conclusions, though they’re poor tools for assessing turnout.

For years there was only one and we regarded it as canonical.

Now there are two — what I’ll call the traditional exit poll and Votecast, run by the prestigious University of Chicago, National Opinion Research Center — but neither are actually exit polls anymore.

They are both large, late, surveys, employing careful but different methodologies that sometimes converge nicely and sometimes produce strikingly different results.

According to the NORC Votecast survey, the partisan composition of Virginia’s electorate changed significantly. Democrats held a 4-point advantage in 2020 but faced a 6-point deficit in 2021—more than enough to make a decisive difference in the outcome.

 Turnout didn’t really matter.

The traditional exit poll found the partisan composition of the electorate exactly the same in 2020 and 2021, with Democrats enjoying a 2-point plurality. No relative falloff in Democratic turnout, no surge in Republicans casting ballots.

Independents made a decisive difference for Youngkin.

According to the traditional exit poll, independents, who comprised 30 percent of the electorate in both 2020 and 2021, gave Biden a 19-point advantage, but preferred Glenn Youngkin over McAuliffe by 9, a colossal 28-point swing.

Independents helped McAuliffe, but just a little.

Votecast found independents made up just 8 percent of the electorate and actually swung toward McAuliffe, albeit by a mere 2 points.

All the swing toward Youngkin came from non-college white voters.

The traditional exit poll tells us non-college educated white voters chose Trump by 24 points, but went for Youngkin by a massive 52-point margin. By contrast, margins among college-educated whites, as well as among both college and non-college minorities, barely budged.

The swing toward Younkin was across the board in ethnicity/education categories.

Votecast found movement across all categories. By its reckoning, non-college whites moved toward the GOP, but so did college whites, by similar numbers, and minorities.

Youngkin did better consolidating Republicans than Trump.

And now for something completely different (to borrow from Rocky, Bullwinkle and Monty Python), both polls agree on Youngkin posting bigger margins with fellow GOPers than Trump.

Parental rights in education had a significant impact on the race.

Votecast asked about the importance of five factors in the election. The two most salient for the largest number of voters? Schools’ handling of COVID-19, which McAuliffe won by 26 points and the teaching of critical race theory, which Youngkin won by 43.

Parental rights in education played little role in the Virginia outcome.

Both polls find little difference in the vote between those with and without children at home and similar swings in both segments.

This last isn’t a data difference, but rather about analytic strategy. I don’t find the absence of difference between those who do and do not currently have a child at home persuasive, but serious public opinion analysts have advanced the reasoning.

All-in-all though, if you think you know what really happened in this election, guess again.

Mellman is president of The Mellman Group and has helped elect 30 U.S. senators, 12 governors and dozens of House members. Mellman served as pollster to Senate Democratic leaders for over 20 years, as president of the American Association of Political Consultants, and is president of Democratic Majority for Israel.