2026 compared to other dramatic midterm election reversals of the past half-century

As I said a few days ago, you don’t need me anymore to do your fancy post-election data analyses. So instead I can offer speculation (Could the 2026 election results lead to a decrease in political polarization?) and meta-commentary (NYT columnist offers epically stupid take on 2026 midterm elections).

But we can also learn from some really simple analyses.

Comparing 2026 to previous midterm swings

How big of a deal was the 2026 national election?

To look at consequences, you can count the number of seats won by the two parties. To look at the aggregate opinions of voters, it makes sense to look at the national popular vote for the House of Representatives, where the Democrats received 56.2% of the two-party vote, a swing of 7.5 percentage points from their House vote share in 2024.

Is 56.2% with a swing of 7.5pp a little or a lot? One way to judge this is to compare to the midterm stunners of recent decades:

2010 is the election that put the brakes on the Obama government: the Tea Party movement convinced the voters that government expansion was going too far. In that election, the Republicans won 53.5% of the two-party vote, a 9.0 percentage point swing from 2008.

2006 was the year that George W. Bush’s post-9/11 high finally wore off. The Democrats’ share of the two-party vote in that election: 54.1%, a 5.5pp swing from the previous election.

1994 was the year that Newt Gingrich and the Republicans offered a clear alternative to the liberal policies of the Clinton administration, taking control of Congress with 53.5% of the two-party vote, a 6.1pp swing in their favor. 53.5% is lower than the other numbers I’ll be showing today, but it arguably understates the Republicans’ accomplishment in 1994, in that this was their first stint in the majority in forty years.

1982 isn’t remembered much now, because the Republicans roared back two years later, but, with the worst economic crisis since the Great Depression, and President Reagan having very low job approval, the Democrats did very well at the ballot box, winning 56.0% of the two-party House vote, a 4.1pp swing from the previous election.

1974 was the post-Watergate election, bringing many fresh faces into Congress. The Democrats capitalized on economic stagnation and discontent with the criminal doings of the Nixon administration to gather 56.9% of the two-party vote, a 4.2pp swing.

In summary, the Democrats’ performance in 2026—a 7.5 percentage point swing yielding 56.2% of the two-party vote—was indeed exceptional, at or above the level of previous repudiations of the incumbent party during the midterm elections.

This can be understood in various ways.

One way that 2026 was different is that this time the incumbent party was really the incumbent party. Unlike in those previous campaigns, this time the Republicans had unified control of the federal government—the legislative, executive, and judicial branches—and the incumbent party was itself highly unified (as can be seen, for example, in the Senate confirmation of some extreme cabinet officers). So, voters who wanted party balancing had a clear choice in a way that wasn’t so obvious in previous years where there was divided government.

The other way is to compare to historical baselines. The Democrats’ victories in 1974 and 1982 cemented their status as the default majority party, but that’s not as impressive as the Republicans’ feat in 1994, and then in 2010, of moving from the minority to gain firm control of Congress. The Democrats achieved this in 2026.

What comes next?

In most of the above cases, the midterm winner did not follow up with a decisive win at the next presidential election. 2006, followed by Obama’s victory two years later, was the only exception. Consider the other cases:
After 1974, the Republican party was left for dead: it had been the majority party for awhile and then was wiped out in the congressional elections. But, two years later, the not-particularly-popular Gerald Ford came within 2 percentage points of being elected president.
After the repudiation of Republicans in 1982, the economy came back to life—as the president quipped, now that things are going better, “they don’t call it Reaganomics anymore”—and the presidential election was a landslide.
And the Republicans’ congressional victories in 1994 and 2010 were followed by Clinton and Obama doing just fine in their subsequent reelection campaigns.

That said, the Republicans’ successes in 1994 and 2010 represent one possible template: the triumphant return of a party that had been given up for dead. In both cases, the victorious congressional party did not follow up with a presidential election win, but the party remained strong enough to determine much of the political agenda going forward.

The take-home message based on this history: the Democrats should value their congressional win for itself and not assume they’re on a glide path for 2028. Conversely, the Republicans should figure out how best to work with the Democratic majority without thinking of themselves as being about to lose all power.

Nationalization of politics

Along with partisan polarization has come a nationalization of the vote, which has led to two things. First, the vote swings in presidential elections have been much smaller. After their losses in 1982, the Republicans rebounded a bit in 1984, upping their House vote share to 47.4%, a gain of 3.4 percentage points. In that same year, Reagan was reelected with a two-party vote share of 59.2%!

That sort of thing doesn’t happen anymore, in two ways. First, with partisan polarization, you wouldn’t expect either party’s presidential candidate to come close to 60% of the vote. Second, congressional elections are nationalized, and you wouldn’t expect to see a stunning 12 percentage point gap between a party’s congressional and presidential vote share.