The Democrats’ recent upset win in a Wisconsin state Senate race has led to dire predictions of a blue wave election this fall. Even Wisconsin’s Republican Gov. Scott Walker expressed his concern.
Without exit polling data, it’s hard to know exactly what happened in the Badger State. But given the mostly data-free speculation about implications for this year’s midterms, a deeper dive into Alabama and Virginia, where we do have actual exit polls, seems overdue.
The best way to look at election data is to approach it like a calculus problem — eliminate the constants and compare what changed from one election to another. Context and perspective are crucial in analyzing election data. By comparing exits polls from past races with 2017 data, it’s clear that current narratives surrounding the Alabama Senate and Virginia governor’s races have missed the most important factor behind the Republican losses.
One key voter group determined the outcomes in both races and it wasn’t the party bases, or women in Virginia or African-American women in Alabama. It wasn’t even larger turnouts per se. It was independents — in Alabama, where they dramatically changed their behavior upsetting the Republican status quo, and in Virginia, where they were consistent with past voting behavior that favors Democrats.
The exit polls for both races provide some insight into what really happened — how independents drove the outcomes but for different reasons.
Watch: Scenes From Doug Jones’ Election Night Rally
Start with Alabama. Turnout was higher than one might expect for a special election; but the makeup of the turnout, which is critical, was similar to a presidential election, the most recent data being the 2012 exit poll. Yet in 2017, Democrat Doug Jones beat Republican Roy Moore by 2 points while Barack Obama lost Alabama to Mitt Romney by 22 points in 2012. What changed?
First, here’s what didn’t. Republicans made up 43 percent of the electorate in both the 2017 and 2012 elections. They were a constant. Jones got 8 percent of the GOP vote; Obama 1 percent. A slight pickup for Jones but not enough to explain the huge vote differential between him and Obama. It also wasn’t a surge in the African-American turnout, which accounted for 29 percent of the electorate in 2017 and 28 percent in 2012, and their contribution to the outcome was roughly the same.
Also, claims that a big increase in female voters, especially African-American women, was behind the Jones victory don’t hold water. Women made up 51 percent of the vote in 2017 in contrast to 55 percent in 2012. African-American women in 2012 made up 18 percent of the electorate. In 2017, they made up 17 percent. Female voters remain a challenge for Republicans, but their percentage of the electorate did not represent a surge.
Jones sits in the Senate today largely because he won independents, 51 percent to 43 percent; Obama lost them 23 percent to 75 percent. That’s a monumental shift. Specific voter groups each added a piece to the victory, but only the independents represented a substantial change.
What drove that change? Roy Moore. The takeaway from Alabama is that the quality of a candidate matters and with a candidate like Moore, it can produce atypical vote behavior with independents, even in one of the reddest states in the union.
The exit polls showed that independents had a 54 percent favorable to 44 percent unfavorable opinion of Jones. Moore, on the other hand, had an atrocious 31 percent to 67 percent favorable-unfavorable rating. After losing five winnable Senate seats in 2010 and 2012, thanks to extremely poor candidates, Republicans need to finally learn from past mistakes.
Now, for Virginia. In contrast to Alabama, here the results of the off-year governor’s race reflected the outcome of the 2016 presidential contest. Republican Ed Gillespie lost to Democrat Ralph Northam 45 percent to 54 percent; Trump lost to Hillary Clinton 44 percent to 50 percent.
It also reflected what has now been a nine-year trend as the commonwealth has shifted to become a blue state for all practical purposes. The last time a Republican won statewide was 2009.
What were the constants in this race?
Turnout was up 4 percent but the makeup of the electorate again was similar to previous presidential-year elections. Democrats accounted for 41 percent in 2017; 40 percent in 2016. Republicans slipped slightly, going from 33 percent in 2016 to 30 percent in 2017.
African-American turnout was about the same — 20 percent in 2017 and 21 percent in 2016. Contrary to the prevailing narrative, women actually made up a smaller part of the turnout decreasing from 53 percent in 2016 to 49 percent a year later.
As in Alabama, Northam’s strength with independents won the day but for a different reason. Virginia has become a blue state. Gillespie was unable to expand the Republican reach to independents though he won them 50 percent to 47 percent. For context on how Virginia has changed, former Republican Governor Bob McDonnell in 2009 won independents 66 percent to 33 percent.
What the race boils down to is this — it didn’t reflect anything out of the ordinary for Virginia these days. Clearly, the polls showed Republicans need to step up their game with independents as well as with younger voters.
So what’s the calculus for this year? Is a blue wave about to hit the GOP head on? While a higher turnout usually favors Democrats as we saw in Virginia, Republicans have won key House and Senate races in high-turnout years. 2016 is a good example. However, Democrats don’t necessarily need a presidential year to win either, as we saw in 2006. But whatever the turnout this fall, both parties need independents to win.
In fact, a look at the turning-point congressional elections of the last 30 years reflects decisive margins among independents nationally. In 1994, Republicans won them by 14 points; in 2006, Democrats won them by 18 points; and in 2010, Republicans won them by 19 points.
The biggest takeaway from Alabama and Virginia is that neither Republicans nor Democrats have a base that is large enough to deliver a national majority coalition without independents.
Is it time for Republicans to head for higher ground? Recent major public polls show a significant closing of the generic ballot and the public warming to the tax reform bill.
That improving environment, combined with the Democrats’ base strategy, can open the door for Republicans.
Correction 10:45 a.m. | Due to an editing error, an earlier version of this column misstated Donald Trump’s percentage of the 2016 presidential vote in Virginia. It was 44 percent.
David Winston is the president of The Winston Group and a longtime adviser to congressional Republicans. He previously served as the director of planning for House Speaker Newt Gingrich. He advises Fortune 100 companies, foundations, and nonprofit organizations on strategic planning and public policy issues, and is an election analyst for CBS News.