Speaking of the midterms here’s a good primer from Ed Kilgore about what (and what not) to look for. Whether you obsessively refresh the 538 website or look away the minute the word “poll” appears on your screens, here are some things to keep in mind:
But there are some things we don’t quite know just yet that could wind up being as important as what we know (or at least think we know). Here are a few political suspense stories whose endings might shock or comfort us when it’s all said and done.
Early-Voting Patterns
By my rough calculation, early voting is underway in 31 states. Though polls can sometimes give a sense of how voting by mail is proceeding (along with harder data on mail-ballot requests and returns), the numbers you always here about shortly before any election involve in-person early voting, which is a bigger deal in some parts of the country (notably the South) than in others.
Sometimes the chatter is about overall early-voting levels as a sign of high or low overall turnout levels, as in a CNN report earlier this week:
Three weeks from Election Day, nearly 2.5 million Americans have already cast their ballots in the midterm elections, according to data from election officials, Edison Research and Catalist. In 30 states where Catalist has data for 2018 and 2022, pre-election voting is on par with this point four years ago — which was the highest turnout for a midterm election in decades.
In states with party registration, it’s often possible to discern which party’s voters are turning out early. And even without such data, some southern states collect racial data on early voters as part of a Voting Rights Act reporting requirement (one of the few features of the VRA still in place).
There’s been some excitement this week about very high initial early-voting numbers in Georgia, a state with highly competitive Senate and gubernatorial races. The data also show an especially high percentage of that vote has been cast by Black voters (39 percent, whereas Black voters only make up 29 percent of registered voters in the state).
Is that good news for Democrats, who really need high youth and minority turnout to over-perform expectations this year? Maybe, but we don’t know, as Sean Trende pointed out two years ago when there was even more excitement about early-voting numbers:
Unless you somehow know what is going to happen on Election Day, this argument is useless. To take an extreme example: Democrats could turn out every one of their voters early, and Republicans could still win the election by turning out more on Election Day.
Obviously, that isn’t going to happen. But we exist somewhere along that spectrum. Most, if not the overwhelming majority, of these early voters are people who would otherwise vote on Election Day. The fact that they decide to cast ballots early just isn’t all that interesting.
We don’t know in these states what share of Republicans, Democrats, or independents are voting for Republicans or Democrats, and we don’t know how many voters for any party are going to end up voting on [Election Day]. This is all speculation dressed up as news.
Understanding early voting in this particular cycle is additionally difficult because we don’t know if the early-voting habits many Democrats cultivated during the COVID-19 pandemic will stick, and how many Republicans are still averse to anything other than Election Day voting after Trump told them that’s what they should do in 2020. So it’s best to wait and see.
Potential Polling Errors
There was a lot of anxiety over polling errors in the 2016 and 2020 presidential elections, mostly involving under-sampling of non-college-educated white voters, which in turn led to underestimation of Trump’s vote. That could mean polls may be similarly off-kilter in the same direction again (even though most pollsters have tried to adjust methodologies to reduce under-sampling). But on the other hand, polls in the last midterm election were quite accurate. Then, as now, Donald Trump is not on the ballot anywhere. So what’s the appropriate precedent?
We probably won’t know that until the results are in. But there are some signs that polls with reputations of being more or less favorable to the two parties are beginning to converge as this particular election approaches. In Pennsylvania’s Senate race, for example, six of the last seven public polls, from a variety of outlets, showed John Fetterman two to four points ahead of Mehmet Oz. Similarly, in Nevada’s Senate race, the last ten polls have shown at most a five-point variation in a race narrowly favoring Republican Adam Laxalt. And in Arizona’s tense gubernatorial race, there’s only a four-point variation in nine polls dating back to mid-September, with most showing Republican Kari Lake with a slight advantage. If either candidate won narrowly in any of these races, no pollster is going to be completely humiliated, and there probably won’t be much discussion of polling errors.
That could all change, of course, before Election Day, and polls showing dramatic last-minute trends in key races will be hyped to the stratosphere by the campaigns and parties that appear to benefit. Until then the best bet remains looking at polling averages and not at individual polls. There’s enough confusion now over “best practices” in polling methodology that cherry-picking “better” pollsters is perilous.
Youth Turnout
Back as recently as 2014, you could confidently predict that any party depending on young voters was in trouble during midterm elections, because The Kidz did not vote much in non-presidential elections, for a variety of reasons mostly having to do with personal mobility and complicated lives and work schedules. But something remarkable happened in 2018: Youth turnout more than doubled. Combined with high voting preferences for Democratic candidates, this youth-turnout boom helped Democrats win back the U.S. House and win some key governorships that year. Youth turnout remained high and solidly Democratic in the presidential year of 2020, too.
If the large and diverse millennial and Gen-Z cohorts show up similarly on November 8, they could save a lot of Democratic bacon. Objective indicators of youth engagement with voting this year are high. But there’s significant disgruntlement with Joe Biden among young voters, who are also very much cross-pressured by economic concerns they feel acutely, and a liberalism on cultural issues like abortion on which they feel strongly.
Even fairly small variations in youth turnout and voting preferences could be crucial in close races. And young voters obviously aren’t the only demographic category that should be watched closely. Republicans are counting on maintaining and if possible increasing the inroads they made in 2020 among Latino and certain Black voters.
Contested Elections
Given the extraordinary number of Republican candidates this year who have bought into Donald Trump’s stolen-election fables from 2020, there are obviously reasons to fear that some of these election-deniers may deny their own defeats and cast the results in doubt. A survey by the Washington Post identified 12 Republican candidates in high-profile statewide races who would not affirm they would accept the results, win or lose. So barring a GOP sweep, we can expect some contested elections in the courts, in the court of public opinion, or unfortunately even in the streets.
Democrats might have some issues of their own given the wave of restrictive voting laws Republicans have enacted in many states, along with the voter intimidation efforts of MAGA “poll watchers” that will appear across the country.
With control of the the U.S. House and Senate, and many key state positions at stake this year, you can expect post-election contests over close elections to become larger and more divisive than ever. With one of our two major parties more or less completely subscribing to doubts about “election integrity,” it’s only going to get worse.
The Wave Factor
Some of the talk about “waves” and “winds” and “breezes” in this election represents a meteorological metaphor for perceived momentum and predictions of the results. But as Amy Walter recently pointed out, there is a tendency in most elections for close contest to break in one direction or the other:
[S]ome of the races that many are expecting to go in different directions — like Pennsylvania toward Democrats and Nevada toward Republicans — may not turn out to be the case. Instead, we shouldn’t be surprised to see Pennsylvania not as an outlier but part of a trend. For example, if Republicans are winning Pennsylvania on Election Night, we should expect to see the lion’s share of those other Toss Up seats go that way. A Democratic win in Pennsylvania would suggest that Democrats are going to win a disproportionate share of the closest contests and hold onto the majority.
“Waves” are more predictable in House races where national trends frequently dwarf whatever individual candidates are doing. But we’ve seen Senate waves too: Democrats won eight of ten toss-up Senate elections (using the Cook Political Report’s authoritative ratings) in the otherwise very close 2012 cycle. Republicans won eight of nine toss-up Senate races in 2014. And I’m old enough to remember the elections of 1980, when Republicans netted 12 Senate seats — winning virtually every competitive race — and took control of the upper chamber for the first time since the Eisenhower administration.
Late trends can move a lot of elections, in other words, particularly at a time when partisan polarization has made all elections more or less national.
Another Overtime in Georgia
Lastly, one other imponderable is the possibility that Senate control could come down for the second cycle in a row to a post-November runoff in Georgia. That state eccentrically requires majorities for general-election victories, and the Raphael Warnock–Herschel Walker Senate race looks close enough to make the expected 3 to 4 percent minor-party vote an off-ramp to a December 6 runoff. The two combatants might even be joined by bitter gubernatorial rivals Stacey Abrams and Brian Kemp. It could be lit.
Politics are not predictable these days. Polls are unreliable and voters are closely divided in enough states that any one election can mark a huge shift in power. It’s not just the fact that the right has gone batshit crazy, the volatility of it all makes it even more stressful.
Here’s hoping that Dems have a huge turnout. It’s all we’ve got.