By
Earlier this year, the question wasn’t whether Democrats would suffer brutal losses in the midterm elections, but rather, how brutal the defeats would be. Those assumptions have changed quite a bit in recent months, thanks in part to circumstances, and in part to shifts in national polls.
After all, President Joe Biden’s approval rating has improved, and ample survey data suggest Democratic prospects in the Senate and House are considerably better than they were just three months ago.
It was against this backdrop that The New York Times’ Nate Cohn published a much-discussed report yesterday, making the case that Democratic optimism — or related Republican despondency — might very well be misplaced. The problem is that polling understated GOP support in the most recent election cycle, and that same problem may be understating GOP support now.
That warning sign is flashing again: Democratic Senate candidates are outrunning expectations in the same places where the polls overestimated Mr. Biden in 2020 and Mrs. Clinton in 2016.
The Times’ David Leonhardt wrote a related piece yesterday, emphasizing the same point.
Recent polls suggest that Democrats are favored to keep control of the Senate narrowly, while losing control of the House, also narrowly. But the Democrats’ strength in the Senate campaign depends partly on their strength in some of the same states where polls exaggerated Democratic support two years ago....
There is no doubt that both of these analyses are rooted in fact. In 2020, polling exaggerated Democratic prospects in battleground states such as Wisconsin and Ohio, and Democratic optimism about this year’s elections have been buoyed by data from states such as Wisconsin and Ohio.
You don’t need a PhD in political science to connect the dots here: It’s best not to rely too heavily on data from states where the data was recently wrong.
But — and you had to know a “but” was coming — there’s a detail that’s too easily overlooked: When looking at models and forecasts, it’s best to rely on apples-to-apples comparisons as much as possible.
The Times’ Maggie Haberman, for example, reflecting on Cohn’s report, noted that some pollsters have privately conceded of late that “in both presidential cycles [in 2016 and 2020], the races broke for [Republicans] after polling suggesting otherwise.”
That’s true. But it’s also true that this isn’t a presidential cycle; it’s a midterm cycle.
Republican arguments aren’t wrong about surveys understating Donald Trump’s support in both of the two most recent presidential elections: Pollsters, for a variety of complex reasons, missed a lot of the Republican’s voters, which led to plenty of surprises.
The fact remains, however, that midterm electorates tend to be qualitatively and quantitatively different from presidential election electorates — and Trump won’t be on the ballot in 2022 (at least not in a literal sense).
I can appreciate the appeal of looking at 2022 polling and saying, “Yeah, but polls were off in important ways in 2020.” The more relevant comparison, though, is not to the last presidential election but rather to the last midterm cycle — and as Cohn’s piece noted, “the polls in 2018 weren’t so bad.”
That’s no small acknowledgment: We could temper 2022 expectations by focusing on comparisons to 2020, but it’s ultimately more compelling to compare this year to the most recent set of elections in which Trump wasn’t on the ballot — a cycle in which polls in plenty of key states, including Wisconsin and Pennsylvania, proved to be pretty good.
To be sure, this isn’t a prediction. Election Day 2022 is eight weeks from today and I won’t pretend to know what’s going to happen. History is clearly on Republicans’ side, and it’s entirely possible the GOP will see significant gains. There’s every reason to believe that the party’s far-right base is energized, mobilized and ready to turn out in droves in the fall.
It’s also true that the landscape looks a lot less friendly to Republicans than it did a few months ago. The Supreme Court overturned Roe v. Wade, gas prices improved, job growth soared, Democrats have overperformed in several special elections, and internal divisions among GOP leaders have gone public.
It’s against this backdrop that polls — from many of the same outlets that showed Democrats reeling in the recent past — have also shifted in ways that have caused some Republican heartburn.
Was a lot of the 2020 data off? Yes. Is that especially persuasive in 2022? Not really.