Here’s a figure about the 2022 midterm elections that might surprise you: Republicans won the national House popular vote by three percentage points — 51 percent to 48 percent. They still won by two points after adjusting for races in which only one major party was on the ballot.
Yes, that’s right: Republicans won the popular vote by a clear if modest margin, even as Democrats gained seats in the Senate and came within thousands of votes of holding the House.
If you’re looking to make sense of the 2022 election, the Republican lead in the national vote might just be the missing piece that helps fit a few odd puzzle pieces together.
The national polls, which showed growing Republican strength over the last month of the campaign, were dead-on. On paper, this ought to have meant a good — if not necessarily great — Republican election year.
Imagine, for instance, if the Republicans had run seven points better than Joe Biden’s 2020 showing in every state and district, as they did nationwide. They would have picked up 21 seats in the House, about the number many analysts expected. They also would have easily won the Senate, flipping Arizona, Nevada, Georgia, and holding Pennsylvania.
Yet for a variety of reasons, Republicans failed to translate their strength into anything like a clear victory.
Real Republican strength
The Republican win in the national House popular vote is not illusion. It is not a result of uncontested races. It is not the result of lopsided turnout, like Californians staying home while Texans showed up to vote. The Republicans would still lead even if every county or state made up the same share of the electorate that it did in 2020.
It is not just about one or two Republican shining successes, like Florida or New York, either. Republicans outran Donald J. Trump’s 2020 showing in nearly every state. The exceptions are all very small states with one or two districts, where individual races can be unrepresentative of the broader national picture.
Under a lot of circumstances, this Republican showing would be impressive. Consider, for instance, that Republican candidates won the most votes for U.S. House in all four of the crucial Senate states where Republicans fell short: Pennsylvania, Arizona, Georgia and Nevada.
Or, put differently: Republicans would have won the Senate, and fairly decisively, if only the likes of Dr. Mehmet Oz or Herschel Walker had fared as well as Republican House candidates on the same ballot.
The Aftermath of the 2022 Midterm Elections
A moment of reflection. In the aftermath of the midterms, Democrats and Republicans face key questions about the future of their parties. With the House and Senate now decided, here’s where things stand:
Republicans also won the most votes in Wisconsin, another state President Biden won in 2020. If this were a presidential election, House Republicans would have claimed 297 electoral votes. This doesn’t necessarily mean anything for 2024; it’s just another illustration that equivalent Republican strength could have looked quite impressive under slightly different circumstances.
Although the data is still fragmentary, it is clear that the Republican popular vote win was backed by a fairly sizable turnout advantage. These figures are all generally consistent with a decent Republican year, like the one evident in the state and national House popular vote.
It just didn’t show up on the scoreboard.
Republican robustness in the wrong places
If you look at the states where Republicans performed best in the House popular vote, there’s a pattern: With the exception of New York, almost all of them are in the South.
To the extent there was a so-called red wave this cycle, it largely sloshed into a relatively uninhabited area. Outside of New York, there was only one competitive House district in any of the states where Republicans outperformed Mr. Trump by at least nine points — the kind of margin that might feel like a wave. None of these states had a competitive Senate race.
Meanwhile, Democrats posted many of their best showings across the Northern tier, including much of New England, the Upper Midwest and the Northwest along with much of the interior West.
This is a recurring geographic pattern in American demographics and politics. While it shows up over and over again in American history, nowadays it roughly parallels where Mr. Trump fared well in the 2016 primaries and where he’d be likeliest to fare well again in 2024. Conversely, it also tracks with where we might expect relatively little support for abortion rights.
This state-level pattern also roughly parallels the geographic distribution of Black and Hispanic voters, who tend to be concentrated across the South and Southwest. Indeed, Republicans overperformed across the country in districts with large Black and Hispanic populations.
If we looked at two hypothetical kinds of districts — one all nonwhite, one all white — Republicans’ net gains would have been six points better in the nonwhite ones than in the white ones when compared with 2020 performance, after accounting for state and incumbency.
Black and Hispanic turnout also appeared to be much weaker than white turnout. Overall, turnout stayed near 80 percent of 2020 levels in more white areas but fell to around 50 percent of 2020 levels in areas where Black or Hispanic voters made up nearly all of the population.
There’s an outside chance that weakness among Black and Latino voters cost Democrats the House, given how close it was. Narrow losses by Democrats in some relatively diverse districts — like Arizona’s Sixth and First; California’s 13th and 22nd; and Virginia’s Second — might have been averted if their turnout and support among nonwhite voters had held up as well as it did among whites.
But on balance, the Democratic weakness among Black and Hispanic voters this cycle did more to hurt their victory margins than to cost them House and Senate races. Nonwhite voters are concentrated in relatively noncompetitive, urban districts; conversely, white voters represent an above-average share of the electorate in most of the key House races. Republicans happened to lose the Senate seats in some diverse states — Nevada, Arizona and Georgia — despite the relatively high turnout among whites.
Overall, the Republican lead in the national vote would fall to just under a point if every district represented the same share of the vote that it did two years ago. It would probably shrink even further if nonwhite voters, within each district, represented the same share of the electorate that they did in 2020.
Candidate quality
The red wave, to the extent it existed, may have come ashore in a relatively uninhabited area, but the red tide was still high enough to turn the House vote red in Georgia, Pennsylvania, Arizona and Nevada, even while the Democrats won the crucial Senate seats.
How did the Democrats survive? Perhaps the simplest explanation: On average, they had better candidates thanks partly, but not completely, to weak Republican nominees.
The “MAGA” Republicans — as characterized by The Cook Political Report, based on their backing from Mr. Trump in the primaries — ran far behind the mainstream Republicans. This alone does a lot to explain the Republican showing in key Senate races in which Mr. Walker, Dr. Oz, J.D. Vance and Blake Masters underperformed or lost.
But as tempting as it might be to assume that “bad Republican” nominees are mainly to blame, strong Democratic candidates probably made a difference, too.
Nationwide, Democratic incumbents enjoyed a modest incumbency advantage of a few percentage points — enough to stay standing in a red tide, even if they might have been submerged in a red wave. Almost by definition, incumbents are relatively good candidates (the bad candidates are less likely to become incumbents, after all), and they often enjoy additional advantages in fund-raising and name recognition.
Similarly, there were not many races where Democrats nominated progressives who might have alienated swing voters. Overall, progressive candidates — as defined again by The Cook Political Report’s primary score card — fared about a point worse than more typical Democrats. But there are few races where moderate Democrats can really argue that progressive nominees cost them victory.
Still some mysteries
All of this adds up to a fairly tidy explanation, but there are a few loose ends that give me pause about whether we’ve given enough credit to the Democrats.
Perhaps the most interesting cases are the House races where no Democrat was running for re-election and Republicans nominated mainstream candidates, like in Colorado’s Eighth, Pennsylvania’s 17th and Ohio’s 13th. Democrats often fared quite well in races like these, even though there wasn’t a MAGA Republican or a stalwart Democrat.
What’s the excuse for the Republicans there?
This was part of a broader pattern of Democratic strength in the battleground districts, especially in traditional battleground states. Yes, there were disappointing showings for them on both coasts, but there were very few outright poor showings — ones that look like a Republican +2 environment — in the competitive House districts in the key presidential or Senate battleground states.
Maybe Democratic strength in the battlegrounds can be attributed to good campaigns, with strong advertisements and fund-raising. Or maybe I could tell a story about how demographics, abortion and democracy help explain the pattern. But while threats to democracy and abortion rights were certainly more relevant in many battleground states than in the blue states, it is not a perfect pattern. It doesn’t make sense of Colorado, for instance.
Of course, national patterns will never perfectly explain every race. But there are enough examples like these to raise a basic question about the 2022 election: Should it be understood as an outright good Democratic year that was interrupted by a few isolated Republican waves (Florida, New York, Oregon) and obscured by low nonwhite turnout in solidly Democratic areas? Or was it a good but not great Republican year that the party didn’t translate into seats because of bad candidates and somewhat inefficiently distributed strength?
The national and state House popular votes lead me to adopt the “decent Republican year” frame. But there are just enough examples of unexplained Democratic resilience that I wouldn’t be too put off if someone preferred the framing the other way.
