Many known unknowns remain in the US congressional elections, including the critical question of who will hold the majority in the Senate.

But it’s already clear that Republicans are going to perform far worse than the typical out-party in a midterm election. Democrats appear to be on track for a result that, while certainly not spectacular if viewed in isolation, is the best midterm performance for any incumbent party since 2002. There’s nothing like the massive “wave” elections of 1994, 2006, 2010 or 2018 here, or the steady opposition gains of 2014. In 1998, Democrats did break with precedent and actually gain seats in the House and Senate, despite holding the White House. But that was a question of shrinking existing Republican majorities.

That leaves 2002 as the only real example on record of a more successful midterm defense.

For those who remember it, that was a bizarre midterm year. The terrorist attacks of 9/11 just over a year before the voting hung heavily. George W Bush’s approval ratings shot up to a stratospheric level. Criticizing the incumbent administration was seen as dangerous and potentially even unpatriotic. Yet Bush and his political team were merciless in milking the “rally round the flag” effect for partisan gain.

Of course, 2022 is not going to go down as that year’s equal. But Democrats’ more modest success is nonetheless, in some respects, more puzzling. The 2002 outlier is easily explained by Bush’s freakishly high approval ratings. Of course, a wildly popular president is going to be hard for the out party to deal with. Biden’s approval rating, by contrast, is literally the worst on record for any postwar president at this point in his term, according to the polling site FiveThirtyEight.

How could a Democrat like Abigail Spanberger survive in a swingy district in Virginia in a climate like that?

Biden carried her traditionally Republican seat by a decent margin in 2020, but it swung back hard to the right in 2021 and voted to elect the Republican Glenn Youngkin as governor. Democrats never gave up on re-electing Spanberger, but it was clearly going to be an uphill fight the whole way, given she had been a reasonably loyal political ally of the unpopular president. And yet win she did. Michael Bennet romped home in a Colorado Senate race that been projected to be close. Maggie Hassan not only held her Senate seat in New Hampshire but, like Bennet, ran stronger than Biden did two years earlier. Those results weren’t replicated nationwide, but they were certainly visible across large swathes of the country – much larger than you normally see in a presidential midterm year.

It’s genuinely hard to know what would explain such a paradoxical result, but a good guess is that Democratic party campaign tactics worked. The Democrats raised lots of money and spent lots of money on running lots and lots and lots of ads, mostly about abortion.

This abortion-heavy strategy prompted a fair amount of naysaying and skepticism, for the very solid reason that most voters said it wasn’t the most important issue for them in the race, with inflation and the cost of living clearly taking the crown. But the logic of the abortion-first strategy’s advocates was that even though inflation mattered more, there wasn’t much Democrats could say or do to move voters on that topic. By contrast, driving up the salience of abortion really did change minds in Democrats’ ad-testing experiments.

So they tried it, and it seems to have worked – an inference further bolstered by the fact that Democrats seem to have held their own particularly strongly in places with large numbers of secular white people.

Of course, that’s not a strategy conjured out of thin air. What made it possible was the US supreme court’s decision to overturn Roe v Wade, a fairly predictable consequence of recent appointments but one that still seems to have shocked many Americans out of a sense of complacence.

It’s unusual for a party with concurrent governing majorities to face a policy setback on the scale of the Dobbs decision. But that unusual quality is likely why this election broke the pattern of midterms past. As one friend who works on reproductive rights quipped to me repeatedly this fall, “Dobbs is our 9/11” – a shocking and traumatic event that can suspend the laws of political gravity.

The Dobbs effect is also noteworthy in Florida, where Republicans did very, very well. Their statewide candidates Ron DeSantis and Marco Rubio romped against well-funded opponents, and all the down-ballot races went their way too.

Florida as a whole has been drifting more conservative for years, but the strong rightward break in a year when that didn’t happen elsewhere was striking. More than one factor was surely in play. But it is also noteworthy that even as DeSantis has attracted a national reputation as a pugnacious culture warrior par excellence, he trotted a very moderate course on abortion – backing a ban at the 15-week mark that would leave upwards of 95% of actual abortions untouched. That’s a good way of defusing reproductive rights backlash. And an interesting question for DeSantis’s future is: can he continue to hold that line while remaining a conservative darling, or will the base he’s courting for a potential presidential run want to see him go further?

But outside of the Sunshine State, Republicans have mostly been less cautious, and it has generated results for Democrats that are almost shockingly good given the state of the economy. That’s a testament to Democrats’ tactical savvy, and also a reminder of the huge political risks Republicans are running if inflation subsides over the next couple of years.

• Matthew Yglesias is a political commentator. He runs the SlowBoring Substack