What Went Down At The California Recall Election: Live Results

We just got the first wave of very preliminary exit polls. Remember, there are still four hours of voting left, and exit polls need to be weighted to the final results before they can be truly called authoritative, so these numbers are almost certainly going to change. But there are still some good numbers for Newsom in here: For instance, only three in 10 California voters said his pandemic control measures were too strict, and 69 percent supported the state’s mask mandates for students. That doesn’t sound like an electorate ready to oust their governor over COVID-19.

Well, more like see you intermittently between now and 11 p.m. Eastern (polls close in California at 8 p.m. Pacific) — but that’s not really a headline now, is it?

Anyway, your indefatigable live blog crew is going to take a well-deserved break until the polls close in California, but we’ll keep an eye on exit polls and let you know anything useful to glean from them. Remember, exit polls change as the night progresses, and they shouldn’t be used to estimate who voters voted for. And because much of this election has been conducted by mail, the exit polls aren’t much better than a pre-election poll, as it’s hard to know whether people who responded to the exit poll actually voted.

We’re hopeful, though, that we’ll have a sense later tonight of where things are headed given where FiveThirtyEight’s final polling average of the first question on the recall ballot — whether to remove Newsom from office — ended up. (It showed that 41.5 percent of Californians want to recall him, while 57.3 percent want to keep him in office.)

But, of course, if there is a polling error — and as Nathaniel wrote earlier today, it’s certainly not out of the realm of possibilities — then we’ll likely have to wait several days for the results as the rest of the returns trickle in.

Stay with us. We’ll be here to track it all — whether that means just tonight or two days from now (although we hope it doesn’t come to that). In the meantime, please scroll back through the live blog and relive it in all its chronological glory. And if you have any questions, ping us @538politics and I’ll try to get your question answered on the live blog.

There is a perception that recall elections are strange American or Californian phenomena. This CNN article compares it to the Electoral College as among the top “quirks in U.S. democracy.” And while the Electoral College is an only-in-the-U.S. institution, the recall was neither created in the U.S., nor is it at all unique to our shores. Historians have traced it back to ancient Greece and Rome, and Switzerland’s use of recall elections has been cited as one reason for its revival in American politics during the Progressive Era, starting with Los Angeles’s prominent adoption of the procedure in 1903.

So perhaps more relevant is the usage of recall elections in recent years. And on this front, the U.S. is hardly a leader. Rather Peru — at least until a 2015 change in its law — has been seen as the world leader in recall elections. And arguably the most important recall vote of the 21st century was the 2004 recall of Venezuelan President Hugo Chávez. Chávez, of course, is not the only political leader to recently survive a recall. The former president of Romania narrowly survived an attempt during a constitutional crisis in 2012. And the United Kingdom adopted a recall law for members of Parliament in 2015 — it has been used three times since, with two ousted MPs. Taiwan and Japan have also seen high-profile recall elections in recent years. In fact, if we run through the past decade based on data I’ve collected, the mayors of Lima, Peru; Warsaw, Poland; and Duisburg, Germany all faced recall elections.

So, ultimately, however odd the Newsom recall feels to voters and commentators, it is truly a global phenomenon.

Just like in the 2020 election, California is holding this recall election predominantly by mail, which means that, just like in 2020, we’ll have to wait a while for full results. Expect the state to release a big chunk of votes in the hours after polls close at 8 p.m. Pacific (11 p.m. Eastern), but the rest of the returns will likely trickle in slowly over the next several days.

Can we be more precise than that? Well, maybe a little — if California’s counting of votes in 2020 is any guide.

By 11:11 p.m. Pacific Tuesday (2:11 a.m. Eastern on Wednesday) on election night 2020, around 64 percent of votes had been counted in California — but the state didn’t reach 70 percent reporting until Thursday evening and didn’t reach 80 percent reporting until Sunday. It took a week after the election to hit 90 percent reporting and two weeks after the election to hit 98 percent. That said, the timeline might be shorter for the recall, since ballots had a whopping 17 days to arrive after the 2020 election but have only a week this time around.

Still, the slow pace of vote-counting in California means we might not know whether Newsom has been recalled until later this week, especially if the election is close. And even if one side takes a clear lead in the initial results, proceed with caution. As we saw in 2020, it’s possible that the first ballots counted will be disproportionately Democratic- or Republican-leaning. Most counties in California release early and mail ballots cast by the weekend before election day first, then add in-person election day votes to those totals in the wee hours of Wednesday.

Finally, mail ballots that arrive at the last minute are slowly counted over the course of the next week. That could mean the results will initially look very good for Newsom, then get more Republican over the course of election night, then zag back toward Democrats over the next several days. That said, in 2020, Democrats didn’t gain as much ground in late-counted mail ballots as they have in the past, so don’t be too confident about which way the vote will shift.

Should Newsom get recalled, the candidate with the most votes on the second question will become governor, no matter how small a percentage of the overall vote. And as Elder leads with only 29.7 percent in FiveThirtyEight’s polling tracker, it seems the winner of the replacement vote will fall well short of a majority.

In fact, if Newsom loses, it’s not out of the question that Elder could win with the lowest vote share of any elected governor in modern times. As the chart below shows, no governor has won with less than 30 percent over the past half century or so under any electoral format. But Elder could.

Dating back to 1968, the lowest vote share for an elected governor was 33 percent, which Democrat Buddy Roemer took as he won Louisiana’s all-party primary for governor in 1987, as fellow Democrat Edwin Edwards declined to contest a runoff against Roemer after no candidate won a majority. In regular general elections during that period, the lowest vote share that won was Angus King’s 35 percent in Maine’s 1994 gubernatorial race. And if Elder were to win with 30 percent or less, he’d actually have the lowest winner’s vote share in any gubernatorial race since 1900, surpassing Democrat Ernest Lister’s 31 percent mark in Washington’s 1912 election.

The conventional wisdom surrounding recall elections is that petitioners often have an advantage because turnout drops, especially if it is held in an off-year or on a special election date. And at the local level, recall turnout can really be all over the place. Without going into all the numbers, we regularly see lower turnout in state legislative recalls that don’t occur in a general election.

But that theory is blown up when looking at the three previous gubernatorial recall elections in U.S. history. It is exactly the opposite. While Newsom may very well be different, the earlier examples show turnout shooting up.

The 2003 Davis recall, also in California, held in October in an off year, saw a massive voter boost. In the 2002 election, when Davis was reelected, only 7.7 million people voted. But they came out in force for the circus-like atmosphere of Davis’s recall, with 9.4 million votes cast for the first question. Looking ahead at Schwarzenegger’s re-election race in 2006, we see just how impressive the recall numbers were: Voter interest dropped, with just 8.7 million coming to the polls.

Meanwhile, the Walker recall in Wisconsin took place in an election year, but it was on a special election date in June so there was every reason to think the votes wouldn’t be that impressive. Yet, Walker won the 2010 gubernatorial election, in which 2.1 million people voted, but in 2012, 2.5 million turned out for the recall. The numbers dropped back down in 2014, with 2.4 million people casting their ballots.

The Frazier recall in 1921 in North Dakota, our third example, is a bit different. Frazier had a two-year term, so his original election was in the higher-turnout presidential election year in 1920. We can’t compare the numbers in his previous election in 1918 (because it took place before women’s suffrage), but his reelection bid in 1920 saw a turnout of 229,606. Meanwhile, the 1921 recall saw 218,76 voters come out, which is a very small drop-off from the 1920 election. Notably, Frazier’s reelection bid in 1922 saw an even sharper drop off -- turnout dropped to 191,369.

The Frazier recall doesn’t quite fit the mold of our other two examples, and our sample size of gubernatorial recall elections is small (just three elections excluding Newsom’s), but it’s still a bit of a misconception that irregularly scheduled special elections drive down voter turnout. If anything, when a governor recall election is on the ballot, voters seem to turn out more. We’ll have to see if that’s the case for Newsom this evening.

It feels like we’ve all been looking ahead to the recall for so long, but now that it’s upon us, we haven’t spent that much time thinking about what comes after. If Newsom is recalled, his replacement (Elder, in all likelihood) would take office on Oct. 22. But the new governor would have only a year left in his term until California’s regularly scheduled gubernatorial election — and if the new governor is a Republican, Democrats would have a good shot at taking the office right back in 2022.

If Newsom wins, on the other hand, it might make his 2022 reelection a foregone conclusion. As Joshua has written, failed recalls have only served to strengthen their targets politically; think of how the attempted recall of former Wisconsin Gov. Scott Walker in 2012 made him a leading contender for the GOP presidential nomination in 2016 (though, of course, we all know how that turned out). With Biden in the White House (and Newsom’s political BFF Harris his heir apparent), Newsom may not become president any time soon, but he could still achieve the status of national Democratic hero and Republican dragon-slayer.

It will definitely be interesting to see how “yes” fares among Asian American and Hispanic voters, where we saw a lot of swing toward Trump in 2020. A few weeks ago, Larry Elder rolled out a high-profile endorsement from Gloria Romero, the former state Senate Democratic majority leader. He’s also been running Facebook ads featuring testimony from Betty Chu, a pioneering Asian American attorney, and campaigning with Yorba Linda Mayor Peggy Huang.

Republicans believe that 2020 signaled that voters of color are moving rapidly toward them. The recall — in one of the most diverse states in the country — is the first big test of whether that’s actually the case. It’s not a clean test because of the format of the election, but it’s something to watch nevertheless.

I’ll keep an eye on Imperial County, too. Biden won this county in 2020, but it definitely shifted right since 2016. Also, the county has a large number of Hispanic residents, and Hispanic voters are really going to be key to watch in tonight’s race.

Like Nathaniel, I'm keeping an eye on Orange County, just south of Los Angeles. It was the traditional home of the California GOP, but the rise of Trump sent it over into the Democratic column, as it went for Hillary Clinton by about 9 points in 2016 -- the first time since 1936 the county went Democratic in a presidential election (Biden also carried it by 9 points). But despite this, Newsom carried Orange by less than 1 point in the 2018 gubernatorial election, and last year Republicans flipped two U.S. House seats there. Republicans could get some "yes on recall" votes, it would seem, but of course, for the recall campaign to have a chance, it's going to need to win Orange by a few points at the very least.

Los Angeles. There's a big north/south divide in turnout in mail-in ballots, with the significant northern counties all over 45 percent. Los Angeles and San Bernardino are both 35 percent and under. In 2003, the Davis campaign lost 1 million votes from 2002 to 2003. Almost half of those votes were in L.A./Orange/San Diego. Beyond this election, what does it mean that Democrats didn't get mail-in ballots up in L.A.?

One thing I'll be watching out for is how the deepest-blue parts of the state are looking -- I'm talking about the Bay Area and Los Angeles. To state the obvious, big raw vote totals there would be good news for Newsom, and his strength in those areas could be an early indicator of how quickly this will be wrapped up.

I’m really curious to see what turnout in the Bay Area looks like. It’s a reliably blue — and reliably pro-Newsom — region that’s home to just about one-fifth of the state’s registered voters. While it’s hard to imagine the voters in these deep-blue counties electing to recall the governor, a big turnout from Newsom’s home county of San Francisco — or Santa Clara, Alameda or Contra Costa counties, which actually have more voters — could offset any losses inland or in Southern California.

I'll be watching Orange County. Once California's quintessentially Republican county, it has zoomed to the left in this era of suburbs turning blue and has voted for Biden by 9 percentage points in 2020 — BUT it still held onto its GOP roots down-ballot, sending Republicans Young Kim and Michelle Steel to Congress. In this recall, how will it vote relative to the state as a whole? As if this is a presidential race? Or as if it's a down-ballot one?

Ryan just wrote about how there won't really be bellwether counties to watch tonight -- but there will still be a number of noteworthy parts of the state to watch as results come in. What areas are people paying special attention to tonight?

We have a question from Michael A. on YouTube about which “bellwether counties” (or counties that vote for the winner of an election correctly) to watch tonight.

We’re sorry to say, Michael, but that’s not really possible. Last November, we were able to get clues about statewide results by looking at counties where greater than 98 percent of the expected vote was already counted. For example, I found on election night that Biden had improved on Clinton’s performance in Lackawanna County, which boded well for his chances of winning Pennsylvania overall.

Unfortunately, though, this concept of looking at bellwether counties doesn’t work for the California recall election this year because no counties will have complete counts tonight. Even in a non-recall situation, California elections can take a really long time to get final results. Eric Ting of SFGate reached out to the 12 most populous counties to find out when they’re planning to report various batches of votes. And what he’s learned is that the general pattern we can expect is that mail ballots will come in first (which are very Democratic), followed by election day votes (which are more Republican relative to the state as a whole) and finally late-arriving mail ballots (which may be somewhere in between). Late-arriving mail ballots include those that voters postmark on election day and arrive at election centers within seven days.

Therefore, we can probably expect that the statewide margin will start out overwhelmingly Democratic (when mail ballots are counted) before dropping somewhat as election day ballots are added to the count. The margin at daybreak on Wednesday might still move a few points in either direction as late-arriving mail ballots get counted over the next few weeks.

While it’s not clear that recalls are more prevalent (a study of recall elections in the 1970s suggests that it may not be true on the local level), they definitely are on the state level. Partisanship and a more aggressive environment are usually cited as the cause, but the 1930s and the 1960s were not exactly free of partisan strife. So why are there more recall elections now?

It could be that the single biggest driver is technology.

The challenge is not winning recalls. It is getting on the ballot. And the internet, email and social media have all allowed voters to be alerted, angered and drawn into a brewing battle more easily. Demographic data is also more readily available for petition organizers and easily parsable, smartphones have also eased signature gathering efforts, and campaigns are cheaper and easier thanks to the ubiquity of personal printers. Think of how difficult it would have been to make signs in the past, versus the ease of today.

Fundraising has gotten much easier as well. The ability to easily collect small money donations through targeted emails and ads have reshaped all levels of politics. My favorite example is South Carolina Rep. Joe Wilson, who yelled “you lie” during Obama’s joint session in Congress in 2009. Without any effort, and in a relatively non-competitive seat, he raised $2 million in a week, and his opponent raised over $1.5 million. And this last election cycle saw Democratic donors pour over $100 million into the South Carolina Senate race and nearly as much into Kentucky’s Senate race even though both contests turned out to be easy wins for the Republican incumbents.

The reality is that technology, perhaps more than anything else, will help fuel future recall elections.

The FiveThirtyEight polling average finds the effort to recall Newsom failing by 15.8 percentage points (an average of 57.3 percent oppose the recall, an average of 41.5 percent support it). But how does that translate into the likelihood that the recall will succeed?

We’re not issuing a probabilistic forecast of the race, as we do for presidential and midterm elections, but here’s how we’re thinking about it. A 15.8-point lead is healthy. Newsom is definitely favored, and it would be a significant polling error if he lost. But this is a very difficult race to poll: The odd timing of the election (September of an odd-numbered year) and the fact that it is being conducted primarily by mail make it hard to determine who is a likely voter. According to erstwhile FiveThirtyEighter Harry Enten (now at CNN), polls of the 2003 California recall election were off by 9 points, and the “true” margin of error of polls of special elections for U.S. House (which, like this recall, are oddly timed) is around 13 points.

So, if ever there was going to be a major polling error, it would be in a race like this one. Nevertheless, 16-point polling errors have only happened once in a blue moon, so Newsom very likely to still be the governor of California when all this is said and done.

California most recently went through a gubernatorial recall election in 2003, when voters chose to recall Democratic Gov. Gray Davis by about 11 percentage points. To replace him, they chose Republican Arnold Schwarzenegger with about 49 percent of the vote. But there are some pivotal differences between the 2003 and 2021 recalls -- distinctions that help make Newsom’s recall less likely than Davis’s was all those years ago. First, while Newsom has faced criticism as governor, he is not nearly as unpopular as Davis was leading up to the 2003 recall vote. As the chart below shows, Newsom’s job approval rating has hovered around 50 percent among likely voters in surveys conducted by PPIC ahead of the recall, whereas the same pollster found Davis below 30 percent approval among the likely electorate before the 2003 vote.

Not only is he more popular than Davis was, but Newsom also has the benefit of running in a California that is noticeably bluer than it was nearly two decades ago. One year after Davis was recalled, Democrat John Kerry carried California by 10 points against then-President George W. Bush. By comparison, Biden won the Golden State by a whopping 29 points in 2020. Correspondingly, Democrats held only a 44 percent to 35 percent party registration advantage over Republicans ahead of the 2003 recall election, whereas that edge is now about 47 percent to 24 percent. And while the share of unaffiliated voters has increased in that timespan -- rising from about 16 percent to 23 percent -- that voting bloc is also Democratic-leaning: Earlier this year, PPIC’s polling found that 52 percent of California independents leaned toward the Democrats, while 36 percent leaned toward the GOP.

I think Nathaniel’s distinction between solidly Democratic and solidly progressive is spot-on. The Democratic supermajority in Sacramento has been drafting progressive policies, but very few are actually being passed into law. Even with a weakened GOP unable to stop the passage of most bills, there are moderating forces in the state that haven’t gone anywhere — lobbyists, business interests, swing-district representatives, a general nervousness around raising taxes in a state where taxes are already high.

Yeah, I agree with Nathaniel -- the last time Republicans won a statewide election in California was 2006. Democrats have won almost every statewide election in California in the past decade by double digits (with a couple of exceptions in the red wave year of 2014). Ballot propositions can be a different story, but I'd note that they often differ from a state's partisanship, not just in California but across the country. Illinois voters rejected a proposal that would have increased taxes on the rich, while many Southern states have approved minimum-wage increases. Voters can often have different preferences than party leaders.

California has also grown more Democratic in recent years, and slightly less Republican. In 2017, 44.8 percent of registered voters in the state were Democrats. Now, 46.5 percent are, according to the Public Policy Institute of California. In 2017, 25.9 percent of registered voters were Republican, and that has dropped to 24.1 percent.

Yeah, Nathaniel, I've always found it interesting that California is expected to be a haven for progressivism -- because the state is definitely blue -- but Democrats of all stripes live there, and there's a lot of disagreement on how to approach the issues facing the state. Part of that, I think, has to do with the state's overall diversity -- not just in terms of race but also wealth. It's also a very expensive state to run any kind of political campaign, which could help explain why progressives sometimes find it surprisingly hard to get traction.

The distinction I'd draw, Sarah, is that California is definitely solidly Democratic, but that doesn't mean it's super progressive. The results of its 2020 ballot measures were pretty conservative: The state declined to reinstate affirmative action for public hiring and college admissions; it voted down a measure that would have increased taxes on commercial and industrial properties; and it passed a measure backed by Uber and DoorDash that classified gig workers as independent contractors (although that measure was recently overturned by a court).

California was the cornerstone of the Republican Party -- the post-1968 party ascendancy was based here (5 of 6 presidential elections). No surprise that this took place in the decade that California became the biggest state in the nation. Once the GOP lost California, it lost one-fifth of the Electoral College vote needed to win the presidency. While there may not be a typical California voter, the Republicans have drawn increasingly inward in the state -- which may be why Kevin Faulconer failed while Larry Elder succeeded. They used to be able to win the independents. That's over. Until they figure it out, they won't be going anywhere in California.

Maya just published a long history of the GOP in California and how the state wasn't always favorable for Democrats. Lest you forget, California is the state that helped boost Reagan's political career (and he's not the only prominent GOP politician for whom this was the case).

But now, as Maya pointed out, Republicans are increasingly a minority in the state. And as Emma has also written on the live blog, there is no one "Californian voter," and many progressive ballot measures have failed there despite the state's blue hue. What do we know about the GOP in California? And is California's progressive streak overrated or in trouble?

This may come as a bit of a surprise, but California’s reputation as a liberal bastion is a fairly recent phenomenon. The state has elected only four Democratic governors since 1900 (not including Newsom), and in that time, it has also produced a number of influential Republican politicians — two eventual U.S. presidents, Ronald Reagan and Richard Nixon, as well as Supreme Court Chief Justice Earl Warren. The state also incubated politics that would become hallmarks of the conservative movement, ranging from Nixon-era anti-Communism to the anti-tax movement of the Reagan years to anti-immigration policies that reemerged during the Trump era.

But in recent years, the California GOP has shrunk according to almost every measure possible. There are currently 19 Republicans in the state’s 80-seat Assembly, half as many as there were in 1994 (the last year the party won control of the chamber); and the state’s 53-person congressional delegation includes just 11 Republicans — up from the eight left after the 2018 midterms, but significantly less than the 20-plus Republicans sent to the House throughout the 1980s, 1990s and into the 2000s. Moreover, just 24 percent of the state’s voters are registered Republicans, a decrease of 11 percentage points from 2003.

What explains the party’s struggles? Part of it is due to the state’s changing demographics and rapid diversification. According to the U.S. Census Bureau, California’s Hispanic population increased by 11 percent between 2010 and 2020 and now comprises over 39 percent of the state — a plurality. Meanwhile, the state’s Asian population increased by 25 percent, and now represents over 15 percent of the state (and this is not taking into account Californians who identify as two or more races, a group that, mirroring national trends, grew as a whole by more than 217 percent over the last decade). This is significant because majorities of likely Hispanic and Asian voters are, according to a 2020 PPIC survey, Democrats. White voters in California, though, are more closely split, with 40 percent registered as Democrats and 34 percent as Republicans.

Another factor is the suburbs’ broader shift toward the Democratic Party. This change has been hastened by the state’s housing crisis, which has expanded already sprawling suburbs and driven residents who might’ve once settled in San Francisco or Los Angeles into historically conservative areas around the Central Valley or Inland Empire.

There’s also a structural element to the party’s challenges (at least in statewide elections). California switched to a jungle primary system in 2012, allowing all candidates to run in one primary and advancing the top two vote-getters to the general election, regardless of party affiliation. With so many reliable Democratic votes, Republicans have been shut out of the state’s last two elections for U.S. Senate.

The California GOP has also struggled to balance conservative priorities with the interests of a young, diverse electorate. California residents have long expressed concern about climate change, and state leadership — regardless of party — have generally responded. But even with a Republican governor in Arnold Schwarzenegger pushing progressive climate legislation, the party itself struggled to get on the same page when it came to climate policies. In recent years, the state GOP has split on issues like extending the state’s cap-and-trade program.

Finally, the state GOP is still figuring out how to win in the shadow of Trump, who was immensely unpopular in California (even among Republicans, by the end of his term). In the recall election, that’s manifested in a struggle to coalesce behind a single candidate. There are Trumpy candidates running in the recall election, but the state party declined to endorse.

In the shadow of Trump — who was immensely unpopular in California, and eventually even among Republicans — it’s been hard for Republicans to find a foothold in the state. The state party hasn’t quite figured out how to balance the Trumpier side of the party against its more centrist vein. This summer, for instance, it declined to endorse a candidate in the recall, reportedly because members were nervous that endorsing one candidate would dissuade voters who favor a different one. There are some indicators, though, that there’s room for the California GOP to regain ground: After losing seven U.S. House seats in the 2018 blue wave, Republicans took back four last year.

But with numbers largely stacked against California Republicans, there’s not much space for intraparty dissent, and at this point, it seems like the California GOP hasn’t coalesced behind any one candidate — or idea.

One good reason for any growth in recalls is their success rate. Most recall attempts die on the vine -- somewhere in the neighborhood of two-thirds to three-quarters fail to get enough signatures based on my analysis. But if you get on the ballot, watch out.

From 2011-2020, I counted 1,002 recalls that went to an election in the United States (when there are multiple officials facing recalls in one jurisdiction at once, I count each official separately as you must collect signatures for each individual). Additionally, I found that 198 officials resigned in the face of recall threats or a recall election, with four other officials removed by other means before the election could be held. These recalls took place in 29 states and the District of Columbia, plus an additional three states had officials resign in the face of recall threats.

The results: 601 officials were kicked out by the voters, 401 survived the vote – just about a 60 percent removal rate.

You may think this success is due to the use of special elections. But I found that recalls that take place on a special election day are slightly less likely to result in an ouster (57 percent) than those that take place in conjunction with a general or primary election day (64 percent).

How does California stack up? Of the 110 officials I counted who faced recall votes during this time period, 86 of them were removed, for an ouster rate of 78 percent. Nineteen others resigned and two were voted out by their city council before the recall vote could be held. Recall elections have been, on net, pretty effective.

Do you have a question about the California recall election? Let us know in the community tab on YouTube and our reporters might make a video to answer your question.

Going back to Sarah's comment about Elder not getting over 30 percent in the polls: He may ultimately appear to have a higher share of the vote in the official results, where "blank" votes are removed from the total number of votes on the replacement question. For example, if Elder got 30 percent but 30 percent of voters didn't vote for a replacement candidate, he would get 43 percent of the replacement vote. It'll be interesting to see how media organizations handle this and which numbers they decide to highlight in their reporting about the results.

A look at the state legislative recalls in California (as well as in Colorado and Michigan) shows that the Democrats were wise to skip the replacement race. In practically every race, we see that the other party would win the replacement, frequently by very high amounts. Note that Davis got 44.6 percent of the vote and the Democrats' standard-bearer, Bustamante, got only 31.5 percent. The yes vote on Davis was 55.4 percent. The combined Republican vote was over 62.5 percent.

I think not having a prominent Democratic replacement candidate was the right move for the party. The problem with the backup option is basically that a ton of Democratic voters would have blanked the replacement question regardless. In 2003, 8 percent of voters blanked the replacement question (probably disproportionately Democrats) even though the lieutenant governor was on the ballot. To still win if the recall passed, Democrats would need a very specific scenario with (i) the recall barely passing, (ii) a somewhat divided Republican field and (iii) almost all of their voters aligning to vote for the same replacement candidate. That third scenario is particularly tricky since many voters don't follow politics all that closely. Potential gains from this scenario would likely be outweighed by losses from a messaging perspective -- Newsom and his team don't want to give Democrats "permission" to vote yes on the recall because there's another Democrat waiting in the wings.

When the polls looked closer on recall, it definitely did seem risky, Sarah. But in the end, the move to discourage a big name from running on the Democratic side may have made it easier for Newsom to make Elder the face of the recall campaign and energize Democratic voters to show up at the polls (or in most cases, mail their ballots in).

That's a good point, Geoffrey, and what I thought was an interesting -- and maybe risky -- strategy from Newsom's team at the time. It definitely could help explain why Elder's support isn't higher than 30 percent in our polling average, though.

Well, in some ways it's interesting to consider who didn't run in the replacement candidate election: any Democrat of note. The most well-known Democrat appears to be Kevin Paffrath, a YouTube celebrity who dispenses financial advice in videos, and he's polling at about 6 percent in our polling average. Unlike in 2003, when Democratic Lt. Gov. Cruz Bustamante ran in the replacement election -- which was awkward with fellow Democrat Gray Davis facing recall -- there wasn't a big-name Democrat for party members to flock to. This was partly intentional, as California Democrats and Newsom's campaign tried to discourage any notable Democrats from running, such as former Los Angeles Mayor Antonio Villaraigosa, who lost the 2018 gubernatorial primary to Newsom and was the subject of speculation as a possible replacement candidate earlier this year.

California’s political landscape has also changed pretty drastically over the past decade, and the share of Republicans in the state has declined. That makes it a lot harder for any GOP candidate to really make a splash in the recall, too.

Kevin Faulconer, the former San Diego mayor, stands out to me. He was a popular mayor — one of the only large-city GOP mayors in the country — and his politics and disposition are relatively moderate. Before the recall got going, he was the guy a lot of Republicans in Washington and Sacramento wanted to be their standard-bearer in the 2022 gubernatorial contest.

But he never really got off the ground in the recall election. He was overshadowed by John Cox and the bear, then by Jenner, and now by Larry Elder. He’s polling in the mid-single digits, behind Elder and YouTuber Kevin Paffrath. A poor showing in 2021 doesn’t preclude him from running again in 2022, but it does weaken his image as a credible statewide candidate who can motivate Republicans and independents alike.

Yeah, Sarah, I was a little surprised that Jenner didn’t make a bigger splash because as we’ve previously reported, celebrity candidates usually benefit from things that have been proven to sway voters, like charisma, physical attractiveness and media attention. That said, from the outset, it wasn’t super-clear that Jenner was respected in Republican circles. She experienced pretty harsh anti-trans harassment at the Conservative Political Action Conference in Dallas in July, and she never polled above single digits.

Both Curtis and Nathaniel just posted about the different candidates who've tossed their hats in the ring (46 in total, and as you might imagine, candidates vary significantly in quality). Which candidates have caught your eye in the lead-up to the election? I, for one, am surprised that Caitlin Jenner didn't make more of a splash. Maybe it's her celebrity status, but when she announced her bid, I thought we might have another Arnold Schwarzenegger on our hands.

This is a point Geoffrey made in his and Nathaniel's writeup yesterday, but I guess I'm really surprised that no candidate cracked 30 percent support. Is that unusual?

As Nathaniel noted, it’s tough to stand out in a field of 46 candidates, so if you don’t have a syndicated radio show like Elder or 1.7 million YouTube subscribers like Paffrath, you might have to find unorthodox ways to get noticed. Here are a few standouts who caught our eye but are unlikely to make waves today.

• Angelyne: Before Paris Hilton and Kim Kardashian were famous for being famous, there was Angelyne, California’s ’80s-era billboard queen. Making her second run for governor (she first ran for governor during the 2003 recall of Gray Davis), Angelyne has a splashy platform that includes inaugurating an “annual masquerade ball” where everyone dresses up as the governor (pets are invited, but must be costumed), an annual Bubble Bath Day on which something must get cleaned, and an ominous promise to “send in the otters.”
• Dan Kapelovitz: The Green Party candidate (one of two on the ballot) has one question for California voters: Can you dig it? The criminal defense attorney styles himself as “the only candidate of the broad progressive left who can win and govern effectively,” while also opposing the recall altogether. His main gripe is with the Newsom campaign’s strategy of telling voters to leave the ballot’s second question blank; he’d like to offer himself as an alternative should Newsom be recalled.
• Adam Papagan: A self-proclaimed “regular guy,” this independent candidate is hoping to demystify the often convoluted political process (something we try to do here as well!) and has been documenting his run for governor via TikTok. With just over 550 followers, Papagan has recorded himself turning in the signatures required to get on the ballot, highlighting paperwork he’s had to file, and getting viewer suggestions on a short and snappy candidate statement (every word, he pointed out, costs $25). His final message of choice: “Love U.” His platform may be sparse, but at least you know how he feels about you.
• John R. Drake: As the youngest candidate on the ballot, the recent community college graduate from Ventura knows how to cause a stir on social media. A viral TikTok garnered over 2 million views after someone stumbled upon the Democratic candidate’s … unorthodox merch, which all feature an illustration of him with cat ears and fangs. We won’t judge you if you don’t click.

There are 46 candidates hoping to replace Newsom as governor of California, but only a few of them are worth noting. Here they are in order of how well they’re polling in the final FiveThirtyEight average:

• Larry Elder (29.7 percent): The conservative talk-radio host has ridden his name recognition to front-runner status in the replacement election, but he has also faced allegations of sexual harassment, and his ex-fiancée has said that he waved a gun at her during an argument.
• Kevin Paffrath (6.1 percent): A finance influencer with 1.7 million YouTube subscribers, Paffrath has become the race’s leading Democrat almost by default. That’s despite some unusual policy proposals, such as building a water pipeline from the Mississippi River to California and using the National Guard to round up California’s homeless population and put them in “emergency facilities.”
• Kevin Faulconer (5.1 percent): Having served as mayor of San Diego for almost seven years, Faulconer has the most traditional political credentials of anyone in the race. Yet the moderate Republican, once touted as the California GOP’s best shot at winning statewide office again, has failed to catch on with voters, according to the polls.
• John Cox (4.5 percent): This businessman and perennial candidate lost three campaigns in Illinois, waged a quixotic presidential campaign in 2008, moved to California and then was the Republican candidate against Newsom in 2018. He deserves a mention only because of the massive amount of his own money he has spent on the campaign (more than $9 million), some of which he used to tour the state with a 1,000-pound bear.
• Honorable mention: Caitlyn Jenner (1.0 percent): Yes, the reality TV star is still running, but her campaign is not a serious one. Fifteen candidates, including several you’ve probably never heard of, are polling higher than she is.

The answer depends on where you are in the state. Where I’m sitting right now in the Bay Area, registered Democrats outnumber registered Republicans by about 9 to 1 (63 percent to 7 percent in San Francisco County), but drive inland less than 100 miles to Calaveras or Mariposa counties and there are nearly twice as many registered Republicans as Democrats. And it’s the northeast part of the state, Lassen and Modoc counties, that are the state’s reddest areas, although they are much less densely populated than the most Democratic places like the Bay Area and Los Angeles County.

Overall, though, California has gotten more blue in the past few years. Democrats currently make up 47 percent of registered voters, while registered Republicans make up just 24 percent. The share of independent voters, however, has stayed relatively constant, although a slight majority say they lean Democratic when asked to pick one of the two parties.

But even though a place like San Francisco (where Newsom was formerly mayor) is solidly blue, that doesn’t mean there is overwhelming support for the governor. Take the latest Berkeley/IGS poll. Despite it showing most Californians wanting to keep Newsom in office, more likely voters from the Bay Area agreed than disagreed that Newsom’s own actions did not align with the strict pandemic policies he wanted others to follow (48 percent to 43 percent).

Likewise, even though California is becoming more diverse and Democratic-leaning, progressive candidates and policies still aren’t always elected or enacted. In 2020, for instance, Biden won the state by nearly 30 points, but over 6 million Californians still voted for Trump (this was more votes than any other Republican presidential candidate has received in the state). And on that same ballot, Californians rejected propositions that would treat rideshare drivers as employees eligible for benefits (59 percent to 41 percent) and repeal a ban on the use of affirmative action in public university admissions and public employment (57 percent to 43 percent).

Understanding who your average Californian voter is complicated. And even though the state leans Democratic overall, a lot of the politics in the state comes down to where you are and who turns out to vote.

FiveThirtyEight Politics podcast host Galen Druke discusses the final polls leading up to the California recall election and explains why they’ve changed so much in the past few weeks.

Even if Newsom is recalled and Elder wins the governorship, Elder’s ability to make major changes to California policy would be limited by Democrats’ supermajorities in both chambers of the state legislature. Yet should the situation arise, there’s one potentially seismic action that Elder could take without being overruled by another body, at least for now: filling a vacancy in the U.S. Senate. That could shake up the state of play in Washington, where Democrats hold a narrow majority in the 50-50 Senate via Harris’s tiebreaking vote.

The worry for Democrats is that something will befall Sen. Dianne Feinstein, who is 88 years old. There have been questions regarding her health, and some observers have wondered if she might even resign her seat (her current term ends in January 2025). Further stoking Democratic fears, Elder pledged to appoint a Republican in Feinstein’s place should her seat become vacant, which would give Republicans a 51-49 edge in the U.S. Senate and upend Democrats’ full control of the federal government. Still, if Elder were to win, California Democrats might quickly use their legislative supermajorities to adjust the state’s Senate vacancy rules to prevent this scenario from happening.

While the California recall didn’t begin with COVID-19, it’s certainly been influenced by the pandemic. In particular, efforts to recall Newsom gained momentum last fall when the governor was spotted dining with friends at uber-spendy restaurant The French Laundry -- in direct opposition to his own public gathering policies. Then there was the fact that Newsom’s kids attended an in-person private school at a time when California’s public school families were struggling through distance learning.

Whether or not he ends up being recalled, Newsom is going to end up as a mascot for the ways COVID-19 has made some class divides in the United States very conspicuous. At the height of the fall and winter surge in 2020, for example, a survey of more than 2,000 parents with kids in K-12 classes found that 60 percent of private schools received in-person learning, while only 24 percent of public schools were. High-income students, specifically, were also more likely to have access to in-person classes. Now, that’s not to say it would have been a great idea for America’s public schools to have been open for normal operation last winter, when thousands were dying daily and the disease was spreading rapidly. But if you were a parent who watched your kid struggling with online school or who had to scramble to do your work and be a classroom assistant at the same time … well, you probably feel some twinge of jealousy.

And school isn’t the only place these inequalities have popped up. Early in the pandemic, polls showed that affluent households were more easily able to socially distance and without suffering as serious a financial impact as the poor. The rich were able to stock up on supplies and work from home, while the poor worked in service industry jobs and were more likely to have medical complications that put them at higher risk for severe cases of the coronavirus. Vaccination has also been divided along class lines, with poorer people and people with less education more likely to remain unvaccinated.

Today, a majority of Californians think Newsom is making the right calls on COVID-19 policy. But the contradictions between “what we have to do” and “who gets hurt” could help explain how animosity turned a niche, partisan effort into a real live recall.

Regardless of whether Newsom wins or loses, don’t be surprised if Democrats try to change California’s recall laws to make it harder to oust people from office. They already have a track record of doing so to suit their own interests: In 2017, with a recall effort underway against state Sen. Josh Newman, the legislature significantly lengthened the recall process such that Newman’s recall election would coincide with the higher-turnout 2018 primary election (Newman lost anyway, but then regained his seat in 2020). Then, earlier this year, they effectively undid some of those changes so that this year’s recall would take place sooner, which pundits believed would benefit Newsom.

As baldly political as the move would be, however, there’s a lot that’s ripe for change in California’s recall process. As I wrote earlier, the two-part recall question can be confusing and could produce an anti-majoritarian outcome. And as Maya and Jasmine wrote on Friday, California is the easiest state for a recall to succeed in due to a lack of restrictions on who can be recalled, the relatively low number of signatures required and the long signature collection period. So don’t be surprised if Democrats try to limit future recalls by bringing California’s recall law more into line with those of other states.

Only four governors in the history of the U.S. have faced recall elections. Two of them were from California. Why is it that Golden State governors seem more susceptible to recall than their peers in other states?

Turns out that there are a number of structural reasons that make California’s recall process more lenient than most. In addition to making almost every public office eligible for recall, California doesn’t limit the reasons for which an elected official can be recalled. It also has a relatively low signature requirement, and it allows recall petitions to circulate for a longer period of time.

All of this adds up to create an environment that makes it comparatively easier to both seek a recall and get it on the ballot in California.

California’s reputation as a solid sea of Democratic blue obscures a lot of ideological diversity — and genuine opposition to Newsom.

Granted, Democrats have supermajorities in both chambers of the state legislature; yes, Newsom’s 2018 victory represented the third consecutive Democratic victory in gubernatorial races; and sure, Newsom did win an average of 59 percent of the two-party vote in his three statewide elections. But there are regional differences across the state, and just as Newsom rose to power on the strength of his support along the coasts, he could (theoretically) be undone by his lack of support in the state’s interior regions. (A state as large and as diverse as California can be dissected a million different ways. For the sake of simplicity, I used the same regions that the state did when tracking COVID-19 last year.)

Newsom came up through San Francisco City Hall and has consistently had a strong base in the heavily Democratic Bay Area. In his three statewide elections, he has never earned less than 81 percent of the two-party vote in San Francisco County. In the broader Bay Area, that figure is just over 70 percent. (While I have to insert an obligatory “party registration does not equal party self-identification” here, it is worth noting that eight of the 10 counties with the highest share of registered Democrats are considered part of the Bay Area.) He also consistently won the two-party vote share in Southern California, a dense and strongly Democratic region that contains three of the state’s four largest urban areas, and gained enough ground in the Greater Sacramento area to win the region in 2014 and 2018 after narrowly losing it in 2010. In fact, the only region in which Newsom did not see a net increase in his vote share between his first and most recent statewide election is Northern California, an 11-county region north of the San Francisco Bay that’s home to only about 1.9 percent of the state’s registered voters.

But the region in which Newsom has managed to grow his support the most over the last decade is the San Joaquin Valley, the state’s largest agricultural region. Eight years after winning just 39 percent of the two-party vote, Newsom took nearly 47 percent of the vote in his gubernatorial campaign. It’s also a region that’s heavily impacted by drought, wildfires and state water policy, and Newsom has not exactly endeared himself to residents in this regard. According to the most recent PPIC poll, the Central Valley (a broader geographic region that combines Greater Sacramento and the San Joaquin Valley with the southernmost counties of Northern California) is the only surveyed region in which Newsom’s approval rating is underwater. Likely voters in the region are virtually split on whether to recall Newsom, a far cry from the poll’s topline of 39 percent in favor of recalling the governor, 58 percent opposed. So keep an eye on the results coming out of Fresno and Kern counties (the two most populous counties in the San Joaquin Valley region) -- should Newsom see his support flagging, it could present a new challenge for a politician who’s made his name on the coasts.

Should a majority of voters choose to recall Newsom, the replacement candidate with the most votes on the second question would become governor. Based on the polls, it looks likely that Elder will attract the most support of any replacement candidate. Coming into the election, he is polling at about 29 percent in FiveThirtyEight’s polling tracker. No other candidate is even in the double digits.

Elder has a clear lead despite allegations of sexual harassment and claims made by his ex-fiancee that he brandished a gun at her during an argument. Yet the fact he’s polling under 30 percent is indicative of how the 46-person candidate field lacks a contender with broad appeal. A contributing factor is the absence of any well-known Democratic contender, as California Democrats discouraged big names from running to make the idea of Newsom’s recall less palatable.

Still, keep in mind that Elder may very well win more than 29 percent because the pool of voters who cast a replacement vote (the second question) could be notably smaller than the overall recall electorate (the first question), so Elder’s vote total might represent a bigger slice of just the replacement vote pie. That’s because many Democratic-leaning voters may skip the second question entirely in response to explicit encouragement from Newsom’s campaign to vote “no” on recall. In 2003, there were about 4 percent fewer votes cast for the replacement candidates than on the recall question, but it seems likely the undervote will be much larger this time around.

Former President Donald Trump is also picking up on the idea that Newsom is going to win. He just put out a statement saying that Newsom “will probably win.” That comes on the heels of a statement from Trump yesterday spreading false claims of fraud and cheating in the election and calling it “just another giant Election Scam.” This is the kind of rhetoric that Republicans blamed for their loss in the Georgia Senate races but that has since become much more mainstream within the GOP. That said, it’s hard to know how many voters are staying up-to-date on the latest Trump proclamations now that he’s banned from social media, so the impact yesterday and today’s statements have on turnout is up in the air. They’ll certainly get less oxygen than a nationally covered election night rally in rural Georgia.