Going into Tuesday’s midterm elections, the consensus among pundits was that Democrats would get washed away by a red tsunami. The question was not whether the GOP would win, but by how much. Yet outside of Florida and New York, that wave failed to materialize. 

Democrats have held onto their Senate seats in Nevada and Arizona, Georgia is headed toward another runoff and John Fetterman has flipped the open seat in Pennsylvania. Most forecasters expect Republicans to win a narrow majority of around 220 seats in the House, far below expectations of 240 or more. Still, a margin of error of plus-or-minus nine seats leaves the possibility of a razor-thin Democratic majority open. We will most likely have to wait for mail ballots to be counted in several key races in California before we know the final outcome.

So how did the experts miss this? On one level, it’s a pretty surprising outcome. Historically, the party in the White House averages around 47 percent of the popular vote in midterm elections and loses an average of 26 seats in the House and two in the Senate. Add in inflation sitting at eight percent, falling real wages, and Joe Biden’s approval rating stuck in the low 40s, and it seems like the perfect recipe for a shellacking. On Election Day in 2010, former President Barack Obama’s approval rating was sitting at 45 percent; Democrats would go on to lose 63 seats in the House and six Senate seats. Right now Democrats seem to be on track to lose only a handful of House seats and pick up a Senate seat.

On another level, these results weren’t surprising at all — they’re exactly what the polls predicted. The final New York Times/Siena College surveys showed Democratic incumbents leading by six points in Pennsylvania’s eighth congressional district and 14 points in Kansas’ third, with tied races in New Mexico’s second and Nevada’s first congressional districts. In the Senate, the Times had Mark Kelly with a six-point lead over Blake Masters in Arizona, John Fetterman up five in Pennsylvania, Raphael Warnock up by three in Georgia and Catherine Cortez Masto in a dead heat in Nevada. 

The thing is that most forecasters plainly didn’t believe the polls. And in their defense, polls infamously underrated Republican candidates in 2016 and then missed by even more in 2020 — the final NYT/Siena poll of Wisconsin had Biden up by 11, he carried the Badger State by just 0.6 points. 

Polling averages had Wisconsin governor Tony Evers tied with his Republican opponent, Tim Michels, but betting markets gave Michels a 70 percent chance of victory. On Election Day Evers won 51-47.

If you look at the numbers for 2022, there is  no evidence of a systemic bias against one party. There are misses in both directions, which is what you would expect. 

There are a couple of possible explanations for why the polls were more accurate this year. Perhaps weighting by education solved the problem in 2018 and then COVID-19 messed things up in 2020, perhaps it’s something about Trump himself being on the ballot. But the more interesting question is how Democrats managed to do so well despite a weak economy and an unpopular party leader — and there’s a strong case that it comes down to Roe v. Wade being overturned. 

In 2021, Democrats lost the governor’s race in Virginia and barely held on in New Jersey. Before the Dobbs vs. Jackson ruling was handed down, Democrats had been consistently running behind Biden’s 2020 margins in special elections. All signs were pointing to a red wave.

But after Roe was overturned, Democrats started overperforming. In August a ballot question to remove the state constitution’s protection for abortion rights lost by 18 points in Kansas, a state that Donald Trump won by double digits. On Tuesday, pro-choice referenda won in four states ranging from conservative Kentucky to purple Michigan to deep-blue California.

Many analysts assumed that the backlash to Dobbs was temporary and would be overshadowed by economic concerns by November. That’s why they found polling showing Democrats leading or locked in tight races incredulous and dismissed their overperformance in special elections. But it seems that they overlearned the lesson of the last war. 

There is a substantial body of evidence in political science showing that voters often dislike major policy changes. That’s a major reason why the president’s party usually loses the midterms — voters choose a divided government to force compromise and moderation. But the single biggest policy change of the Biden presidency wasn’t a move to the left —it was the conservative majority on the Supreme Court taking away a right that American women have had for 50 years. 

Voters didn’t take kindly to that, and that’s why Tuesday’s results showed that the Dobbs decision was “bad for Republicans,” as a certain former president privately remarked to his friends. 

MILAN SINGH