Alexandria Ocasio-Cortez, a Democratic congressional candidate from New York, speaks during a campaign stop in Michigan for Democratic gubernatorial candidate Abdul El-Sayed.Paul Sancya / AP

The future of the Democratic Party looks a lot like Alexandria Ocasio-Cortez. Once it was the party of patrician liberals like Franklin Roosevelt; now women, people of color, and voters in big cities are the demographics at the heart of the party.

The question is whether the future of the Democratic Party votes like Ocasio-Cortez, a self-described Democratic socialist. Her June victory over incumbent Joe Crowley in a Democratic primary for a U.S. House seat in New York City was perhaps the most heralded example of what has been described as a burgeoning leftist shift in the Democratic Party. For progressive activists, it’s a boon decades in the making; for moderate Democrats, it’s a political headache; and for Republicans, including President Trump, it’s both a worrying sign of creeping socialism and an effective bogeyman for rallying supporters.

“Coincident with the 200th anniversary of Karl Marx’s birth, socialism is making a comeback in American political discourse,” an October 23 report from the White House’s Council of Economic Advisers warned. In a statement decrying the white paper, the Democratic Socialists of America said, “DSA agrees with the White House ‘socialism’ report in one respect — ‘socialism is making a comeback in American political discourse.’”

Despite all the noise, it’s hard to get a good grasp on what’s happening. It’s not just progressives who are ascendant. Before Ocasio-Cortez’s victory, the previous Democratic it-candidate for the cycle had been Conor Lamb—a moderate Democrat from rural Pennsylvania. Ocasio-Cortez and Lamb are likely to be colleagues in the House starting in January. Given the widely varying specifics of individual races, it’s hard to come up with useful comparisons to measure the leftist moment. Here’s what we can say: There’s clear leftward movement among Democratic voters on a range of issues, and there are more progressive candidates running than ever. But this doesn’t amount, at least yet, to the socialist groundswell that advocates pine for and critics fear. The actual policy positions, and number of leftist officeholders, will remain limited—at least for now. What happens in 2020 could be more telling.

With Democratic enthusiasm at high levels this year, there’s a heated debate about what’s motivating voters. The Harvard political scientist Theda Skocpol recently told The Washington Post that members of the anti-Trump resistance “are going to revitalize the roots of the Democratic Party, and they are going to feminize it, but they are not going to turn into Bernie Sanders.” Time’s Charlotte Alter found much the same dynamic. “It’s not that the Democrats are being pulled left. It’s more that Democrats are being pulled local,” she reported. “And while ideas like ‘Medicare for All’ and ‘Abolish ICE’ have spread far beyond the party’s left flank, the anti-Trump resistance movement is ultimately more results-driven than ideological.”

Much of the strongest opposition to Trump does seem to emanate from the center-left, especially in cities and suburban areas. Many of its movers and shakers are people who have voted Democratic, but were never actively involved in politics—until the 2016 election activated them. According to a Brookings survey, 60 percent of Democratic primary voters cast their votes to express opposition to the president. Of course, that leaves almost 40 percent who voted for another reason. Although charismatic politicians like Sanders and Ocasio-Cortez get the most attention, the most significant shift in the Democratic Party is among voters, not candidates.

According to Pew data, 46 percent of Democrats and Democratic-leaning voters now identify as liberal—up from 28 percent 10 years ago. Meanwhile, the percentage who say they’re moderates has dropped from 44 to 37. The number of conservatives continues to drop, too. But these changes most likely reflect the exodus of right-leaning Democrats as both parties become more ideologically homogeneous. It doesn’t necessarily mean that there’s been huge growth on the party’s left wing.

Then there’s the question of what it means to be “liberal.” Among progressive pundits, there’s a debate between “liberals” (for example, Obama Democrats) and “leftists” (progressives with more socialist inclinations). But Pew, which maintains the best longitudinal data, doesn’t subdivide that way, and it’s hard to know what voters mean when they self-identify as liberal. For example, support for LGBT rights was once a litmus test for American liberalism, notes Amy Walter, national editor of the Cook Political Report. Now that view is the consensus within the Democratic Party—and gay marriage is largely accepted among Republicans, too. So what is liberalism now?

“Is it the Bernie Sanders view on economic issues? Is it views on social issues?” Walters asks. “Can you have progressive views on social issues, but if you don’t agree with Bernie Sanders’s opinion on government size, are you not a liberal?”

Still, digging into Pew’s data on specific positions can provide a good sense of how Democrats are moving leftward on certain issues, especially immigration, economics, and race. Most astonishing is immigration. As my colleague Peter Beinart has reported, leaders in the Democratic Party have undergone a dramatic shift toward unalloyed support of immigration, including to a certain extent illegal immigration. But voters have moved as well. In 1994, just 32 percent of Democrats said that immigrants strengthened the country. Now 84 percent do.

On economics, three-quarters of Democrats say that the government doesn’t do enough to help poor people, up from half in 1994. Two-thirds say that government should regulate business more, again up from half in 1994. Conversely, in 1994, two-thirds of Democrats believed that people could get ahead if they were willing to work hard. Now only half do. The percentage of Democrats who believe that corporations make too much money is up 12 points. But the movement is not uniform. While the portion of Democrats who say that the government should do more to help the poor, even if it requires taking on debt, rose from 58 percent in 1994 to 71 percent in 2017, it is still below the peak of 77 percent, in 2007.

There’s also dramatic movement on race, which may more than anything reflect the exodus of conservative whites as the Democratic Party becomes more minority-heavy. The percentage of Democrats who say that the government needs to do more to fight racism has risen from 57 to 81 since 2009. In 1994, four in 10 Democrats said that racial discrimination was the main reason black people couldn’t get ahead; in 2017, more than six in 10 did. White voters have moved especially dramatically, as Thomas Edsall notes: On both of these indicators, white Democrats are actually further left than black ones.

This is in large part because the whites who are members of the Democratic Party have changed. Non-college-educated whites have shrunk as a portion of the base by 20 percent, versus big gains from voters with at least a four-year degree, both white and nonwhite. The better educated voters are, the more likely they are to be more liberal. Assuming the party continues to change demographically the way it has, it’s likely to get only more liberal.

“I think there’s a significant shift,” says Robert Borosage, president of the Institute for America’s Future and an adviser to Jesse Jackson’s 1988 presidential campaign. “You can see it in the war of ideas, where more and more Democrats at least nod their heads at Medicare for all. You even have President Obama saying it’s time for big ideas like Medicare for all and a jobs guarantee.”

If anything, the party’s candidates still appear to be to the right of the base—at the least candidates who win nominations. Brookings’s Primaries Project recorded the astonishing growth in the number of Democrats running for House seats this year—1,077, up from 646 in 2014 and 700 in 2016. (In contrast, the number of GOP candidates rose from 755 four years ago to 874 this year, a much more gradual increase.) Brookings’s Elaine Kamarck and Alexander R. Podkul categorized these Democrats as either “progressives” or “establishment” Democrats and found a roughly equal share of candidates. (Moderates compose a small and shrinking third wheel.)

That meant that a greater number of incumbent Democrats, many of them more establishment, faced primaries—45 percent, up from fewer than 28 percent in 2014. Yet despite the understandable attention paid to wins by Ocasio-Cortez and Ayanna Pressley, who defeated Representative Mike Capuano in a Boston-area district, most of the challenges fell short. According to the Primaries Project, “Compared to 2016, primary elections this cycle were actually slightly less competitive for incumbents.” Kamarck notes that efforts by the Democratic Congressional Campaign Committee, while widely derided by progressives, had their desired effect of nominating establishment-backed candidates in close districts, with a 39-for-41 record.

Furthermore, Ocasio-Cortez and Pressley will both replace reliably progressive votes in the House, so while both are women of color and therefore more in step with the Democratic Party’s evolving demographics, they’re likely to end up largely the same on the issues. Otherwise, Brookings found, progressive candidates were more likely than their establishment colleagues to win nominations in more strongly leaning Republican districts—meaning that anything short of a blue tsunami is likely to leave them high, dry, and at home.

“Six months ago, a lot of Democrats were worried that the process was going to make the task of regaining the majority in the House more difficult by nominating some candidates in swing districts who were too progressive for those districts,” says Bill Galston, a colleague of Kamarck’s both at Brookings and in the Clinton White House. That hasn’t been the case, he says. “Progressive victories have been scored in places where almost any Democrat would be the odds-on favorite to win,” like in Ocasio-Cortez’s district.

Borosage dismisses the poor record of progressives challenging incumbents as beside the point. “This is the way movements work. Of course they would lose more,” he says. “The fact that they challenge puts people on notice. There’s no sitting congressperson who’s got any brain that doesn’t make a calculation: Am I vulnerable on my left? What do I have to do?”

The best place to measure leftward shift might be at the state legislative level. Not only are there more than 6,000 races—a more meaningful sample than the few high-profile contested U.S. House primaries—but it’s much easier for progressive, outsider candidates to run at lower levels, because there’s less money and institutional support required. And since most candidates for statewide and national office begin at lower levels, these races should give a good sense of where the Democratic Party’s grass roots are heading.

The problem is that with so many races, and so many specific circumstances in each race, grasping all of them is effectively impossible, but some ways to measure state-level action do exist. As I reported in August, hundreds of progressive first-time candidates are running for state legislatures around the country, many of them espousing strongly left-wing stances, and often emerging not from traditional Democratic Party structures but from social-activist groups and movements.

Some of the effects are visible even before the midterm ballots are counted. Virginia holds its legislative and gubernatorial elections in odd-numbered years, and in 2017 Democrats made a strong showing in races for the General Assembly, losing out on control of the House of Delegates only through a random drawing that resolved a tie. Many of the new members are strong progressives. A baker’s dozen won while refusing to take donations from Dominion Power, typically a major power broker in the state. The new crew also helped to push through an expansion of Medicaid under the Affordable Care Act, which had been stalled for years. In New York, the “Independent Democratic Caucus,” a group of renegade Democrats who caucused with Republicans, pulled the nominally liberal state’s governance to the center. Though the group disbanded earlier this year, six of its eight former members lost their primaries to more progressive challengers.

It’s impossible to precisely and concisely diagnose the cause of the Democratic Party’s current leftward movement. A good starting point is the 2008 financial crash and the resulting recession, which shook many Americans’ sense of security. The crash is often cited as a contributor to Donald Trump’s supposedly populist movement, but it also helped drive Sanders’s support. So did Barack Obama’s relatively moderate, corporate-friendly response, which created an opening for alternative visions of what the party could be. Meanwhile, the moderate and conservative Blue Dog Democrats were largely washed out of office in 2010. That cleared the intellectual field in the party and made intramural debates starker.

“The New Democrats,” says Borosage, referring to Hillary Clinton, Joe Biden, and other establishment party leaders, “are still there, but they are remarkably bereft of any compelling ideas. It’s not a compelling ideological alternative. They’re just a little less.”

To be fair, some of the ideas that the new progressives are offering aren’t all that new. Medicare for all has a sheen of novelty to it, but as Kamarck notes, universal health care has been a stated goal of the Democratic Party for decades. But others are fresher, from marijuana legalization to a $15 minimum wage to a universal basic income.

Some of the shift may also be the result of a feedback loop between voters and candidates. For example, voters may have been attracted to Sanders in 2016 less for his avowed leftism and more for his bluntness, lack of polish, and perceived honesty as compared with Hillary Clinton. The political scientists Christopher H. Achen and Larry Bartels analyzed data from the 2016 American National Election Study and concluded that Sanders voters weren’t necessarily any more progressive than Clinton backers. That could change, though.

“Over time, some of the ideology rubs off on the adherents,” says Larry Sabato, a political scientist at the University of Virginia. “They know they like Bernie Sanders, and over time they learn what Bernie Sanders stands for and they like that. There’s some lingering effect.”

But the 2016 candidate who did the most to push the Democratic Party leftward may not have been Sanders but Trump. Although the Pew data suggest a gradually more leftist Democratic electorate on a slew of issues, some of the most dramatic changes, as I have previously written, emerged shortly after Trump’s entry onto the political scene. Take immigration and race, two of the issues on which Trump was most outspoken as a candidate. The number of Democrats who say that racial discrimination is the main challenge to African Americans today jumped from 47 percent in late 2015, early in Trump’s candidacy, to 64 percent by June 2017. The portion saying that immigrants strengthen the country leapt from 66 percent a month before Trump kicked off his campaign to 84 percent by July 2017.

The scholars Norm Ornstein and Thomas E. Mann have concluded that although American politics is highly polarized today, the polarization has been asymmetric. While both parties have become more homogeneously conservative, in the case of the GOP, or liberal, in the case of the Democratic Party, there are more very conservative voters in the Republican Party than there are very liberal voters in the Democratic Party. One result of that rightward shift has been an increasingly gridlocked Congress, where the far right has become a powerful obstructionist faction—much to the chagrin of Republican Speakers John Boehner and Paul Ryan.

Democrats have not polarized as radically, but if they continue to become more liberal, could they produce a similar effect? Could the Congressional Progressive Caucus act as a mirror image to the House Freedom Caucus? Kamarck thinks that won’t happen, betting that the difference in ideology means even more progressive Democrats will be willing to compromise. Conservatives are willing to obstruct action in Congress because it’s congruent with their view that government action is suspect. Progressives are less able to justify stonewalling.

“Because Democrats are the party of government, they will not get too crazy. They will want to actually pass things,” Kamarck says. “You may get some radicals, but in the end if they really want health care for all, they’re going to be hard put to explain to their people why they voted down something that gave health care for, say, 83 percent.”

In other words, they’re going to have to compromise. The test of her theory will come after the election. In fact, some of the best questions about the future of the Democratic Party will only begin to be answered starting November 7. Those include considering who got elected; seeing how they behave in office; and then watching the way the 2020 presidential primary, which will start almost immediately, begins. Should Democrats fail to win control of the House, it will likely trigger a vast tremor throughout the party, with unpredictable results. (The Senate appears out of reach, though a surprise is possible.)

If candidates like Andrew Gillum, the very progressive Democratic nominee for Florida governor, win, it will provide proof that a left-wing candidate (and an African American one at that) can win in the South, tempering the “electability” argument that has sometimes lifted center-left Democrats. If Gillum loses—even if the race is close, and despite local factors—it could turn voters away from candidates like him.

And if Democrats do win the House, they’ll have to decide how to wield their newfound control. Most progressive legislation emanating from the House would be largely symbolic, since a Republican Senate and president would block any action. A Democratic majority would likely include a good number of seats won by Democrats in historically Republican districts, and certainly in districts that Trump won in 2016; it would also probably include a larger Congressional Progressive Caucus. If the more progressive wing of the party puts forward strongly progressive bills, will freshman Democrats who represent suburban districts where formerly Republican white women have flipped their votes balk, or will they go along? And if they go along, will they end up pulling their districts leftward with them, or simply doom themselves to one-term status? (An Axios analysis found that among 44 Democrats with a good chance to flip GOP districts, very few rule out voting for a Medicare for all bill, and some specifically affirm their support for it.)

Meanwhile, the scramble for the presidential nomination will be underway nearly from the moment the polls close on Election Day. There is already a crowded field of possible contenders, from lefties like Sanders and Warren to old-school hopefuls like Joe Biden, including everything in between. There are signs that some of the younger, less ideologically committed figures who are interested in the race are betting on progressive ascendancy. Borosage notes that Senators Kirsten Gillibrand, Kamala Harris, and Cory Booker, all of whom tended toward the center earlier in their careers, have already begun taking up progressive stances like universal basic incomes and abolishing ICE.

“As a national entity, the Democratic Party comes to life during the presidential nominating process,” Galston says. “To make a judgment about the party as a whole, one would have to make a prediction about the balance of power in that process.” And even then, he says, a shift is ratified only if a president wins reelection.

In other words, the Democratic Party is moving left, but it might not be clear quite how far left for two or even six years. The revolution is coming more slowly than its champions hope and its critics fear.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.