California’s primary was a test run for November. It’s not looking good for Democrats.

Come November, California Democrats better hope it’s 2012 all over again.

Republicans did well in that year’s vote, which was the last time the state’s “top-two” congressional primaries coincided with a GOP presidential race.

And just as in 2012, Republicans saw strong numbers in last month’s California congressional primaries. They outpolled Democrats in all but one of the 10 districts identified by the two parties as general election targets, according to results of the elections certified by state officials last week.

This isn’t just a California story, but the Golden State’s unusual primary system — it’s one of only four states that don’t hold partisan primaries — makes it a great test run eight months before the general election.

Unlike in partisan primaries, voters of all stripes choose between candidates from both parties, and the results have historically been fairly predictive of the general election. The average shift by district between the primary and the general elections is less than a percentage point toward Democrats, according to a POLITICO analysis of California primaries dating back to the advent of the “top-two” system 12 years ago.

And California is already set to play a major role in determining which party controls the House next year. Unlike the next two largest states, Texas and Florida, California’s redistricting commission drew a map that includes more than a half-dozen competitive districts, most of which are held by Republicans, who are defending seats in places where then-President Donald Trump lost in 2020.

While the average shift over the past six election cycles is only modestly toward Democrats, the 2012 example stands out. Like 2024, Republicans had a presidential race to drive their voters to turn out, while the Democratic incumbent was seeking reelection. And after Republicans ran up huge margins in the concurrent congressional primaries, the average district swung by a whopping 5.4 points toward Democrats come November.

But there are also reasons to doubt that could happen again. Since then, California has implemented all-mail voting, designed to increase turnout. In the past two elections, it’s actually Republicans who have gained on average between the primary and the general election, according to POLITICO’s analysis.

Inside the swing districts

While about 10 of California’s 52 House districts are competitive, five are at the center of the 2024 map. All are held by Republicans: Reps. John Duarte and David Valadao in the Central Valley, and Reps. Mike Garcia, Ken Calvert and Michelle Steel in Southern California.

These five GOP incumbents are among Democrats’ top targets anywhere in the country, but POLITICO’s analysis of the primary results found the party has a long way to go to knock them off. Combined, Democratic candidates captured 45 percent of the two-party vote share in four of the five districts, and garnered 47 percent in Calvert’s Palm Springs-based seat.

Applying the swing from the 2022 elections in each of these districts would result in a GOP sweep. The greatest primary-to-general-election swing came in Steel’s Orange County seat, where the Democratic vote share increased by 4.4 points — but that still wouldn’t be enough to oust Steel after Democratic candidates earned only 45.1 percent of the two-party primary vote last month.

In three of the districts, Democrats are relying on retread candidates who also ran in 2022. Duarte is again facing former state Assemblymember Adam Gray, whom he defeated narrowly, 50.2 percent to 49.8 percent. Former state Assemblymember Rudy Salas is pursuing a rematch against Valadao after losing by 3 points two years ago. Calvert will once again square off against Will Rollins, a former federal prosecutor, whom he defeated by 5 points.

But Democrats hope their two new candidates — former Virgin Galactic CEO George Whitesides against Garcia in the northern Los Angeles suburbs, and attorney Derek Tran against Steel — can improve the primary numbers.

The reaches mostly seem to be just that

After those core five seats, both parties are targeting districts that are best characterized as reaches: Democrats say they are seeking to oust Rep. Kevin Kiley in a vast district that stretches along much of the state’s border with Nevada, from the Sierra Nevada to Death Valley, and Rep. Young Kim in Orange County.

Republicans, meanwhile, have Rep. Josh Harder’s Stockton-based seat on its target list, along with two Southern California seats currently held by Rep. Mike Levin and outgoing Rep. Katie Porter.

Of these seats, only an open one in Orange County appears to be in play. Kiley took 55.9 percent of the primary vote last month, and Kim won 56.4 percent. Democrats would need anomalously huge swings in the general election to come close to beating either.

Republicans, meanwhile, will likely struggle to oust Levin, who got 51 percent in his district, which snakes along the Pacific Coast from southern Orange County to northern San Diego County. And even though Harder captured a tick under 50 percent (49.7 percent), the slightest swing toward Democrats in the general election would be enough to pull him across the finish line.

The open seat Porter is vacating could be a different story. In a fractious Democratic primary, which featured millions of dollars in outside spending, the party’s candidates collected 48.7 percent of the two-party vote. That — along with Democratic nominee Dave Min’s DUI arrest last year — could present an opening for GOP nominee Scott Baugh, who lost to Porter by 3 points in 2022.

Why 2012 matters

If the results swing significantly toward Democrats this November, 2012 will be an instructive case. That year, Mitt Romney had just become the presumptive nominee days before the California primary, compared to Donald Trump crossing that threshold a week after this year’s Super Tuesday primary.

There’s evidence the GOP presidential primary was a significant turnout driver that year. More than 1.9 million votes were cast in the Republican presidential primary, just shy of the 2.1 million on the Democratic side — comparable numbers despite the state’s strong Democratic lean.

So it’s not surprising that — by the time November rolled around — the congressional races swung heavily toward Democrats. Beneath that 5.4-point average swing was enough movement to flip four seats (the map had also just been redistricted following the 2010 census).

Democrats captured only 43.7 percent of the two-party vote in the race for then-GOP Rep. Dan Lungren’s Sacramento seat, but a 7.9-point swing catapulted now-Democratic Rep. Ami Bera to Congress. Similarly, a 6.7-point swing helped now-Rep. Julia Brownley to overtake her GOP opponent in the race for retiring Rep. Elton Gallegly’s seat in Ventura County. And a massive, 11-point swing helped Democrat Raul Ruiz oust then-Rep. Mary Bono (R-Calif.) in the Imperial Valley.

Has mail voting changed everything?

But there are reasons to doubt such a swing toward Democrats will happen this fall.

In the past two elections, it’s been Republicans who’ve seen better numbers in the general election than in the primary: a 2.2-point average swing in 2020, and a negligible, 0.1-point uptick in 2022.

Democrats were the party with the still-ongoing presidential primary in 2020 — Sen. Bernie Sanders of Vermont actually defeated now-President Joe Biden that year — so it’s not surprising that Republicans did better in November.

But it’s the lack of movement in many districts in 2022 that could provide a blueprint for this year. And the expansion of mail voting may be the best explanation.

Election officials in California mail a ballot to every registered voter — a practice that began on a temporary basis in 2020 but was codified into law before 2022. And while mail voting has always been popular in California — a narrow majority, 51 percent, of votes in the 2012 general election were cast by mail, now it’s near universal. Nearly 88 percent of votes in the 2022 general election were cast by mail.

Mail voting has flattened the turnout curve. In the 2012 primary, 31 percent of registered voters cast ballots. Last month, turnout was higher: 35 percent of registered voters.

And the differential turnout seen in 2012 — the number of votes in the GOP presidential primary nearly equaled Democratic votes that year — did not appear last month, when far more votes were cast in the Democratic primary than the GOP race.

November is a long way away, but observers looking for tea leaves about the battle for the House may not have to wait that long. Washington state uses an identical primary system to California’s, and their August primaries have historically been predictive of the November result.

Jonathan Lai contributed to this report.