With almost a dozen candidates still competing to become the Democratic presidential nominee, no one can predict who will be chosen. Nor can anyone predict how long it will take until we can confidently identify the nominee.

But a look at the rules and structure of the 2020 Democratic primaries suggests that — unless the handful of states that vote in February deliver a consensus — we may be in for a long few months of uncertainty. Here’s why.

How the U.S. primaries work

The U.S. presidential nomination involves separate primaries and caucuses that take place in each state in staggered order, between February and June. Citizens in each state vote to select that state’s delegates. Officially, the party has no nominee until those delegates vote — which this year will happen at the Democratic National Convention in Milwaukee July 13 to 16.

But we often see a presumptive nominee emerge much earlier. That happens when a candidate wins a majority of the total delegates or when the other major competitors withdraw from the race.

In 2020, when can a candidate secure a majority of delegates?

In the 2000 Democratic race, Al Gore became the presumptive nominee March 9 when Bill Bradley withdrew. Gore captured the needed majority of delegates March 14, when six states voted. Given the structure of the 2000 primary calendar, that was the earliest possible date to secure a majority of delegates. In some other years, such as 1980 and 2008, the Democratic nominee was not certain until June.

This year, it is mathematically possible for a candidate to capture a majority of Democratic delegates March 17, when Arizona, Florida, Illinois and Ohio vote. But don’t expect to be sure of the Democratic nominee that early. That’s both due to the structure of the nominating process and the fact that there’s no strong front-runner.

The Democratic Party allocates delegates using proportional representation, based on the outcomes in the primaries and caucuses, as long as a candidate gets at least 15 percent of the votes in a state or district. This means that candidates who get at least 15 percent of the vote will get delegates in rough proportion to their vote totals. For instance, if Candidate A gets about 20 percent of the vote total in a state, she’ll have about 20 percent of the state’s delegates at the convention. That means no state’s primary is winner-take-all; each state’s delegation will be mixed. As a result, as long as several candidates perform relatively well, not single candidate is likely to win big at any one moment in the voting.

The 2020 Democratic nomination calendar is relatively front-loaded, with contests early in the calendar. Fourteen states (along with American Samoa and a delegation called Democrats Abroad) will vote on Super Tuesday, March 3, when roughly 34 percent of the delegates will be elected. That’s the day primaries will be held in Alabama, Arkansas, California, Colorado, Maine, Massachusetts, Minnesota, North Carolina, Oklahoma, Tennessee, Texas, Utah, Vermont and Virginia.

Here’s what makes that day especially unpredictable: Candidates who are strong in one region may not perform well in another. Bernie Sanders and Elizabeth Warren are better received among more-progressive Democratic voters in the Northeast; Joe Biden is favored in states with more racial diversity, including South Carolina. It’s possible that no single candidate will dominate Super Tuesday.

We may know more, however, after the “early carve-out states” vote in February.

What are ‘early carve-out states’ and why do they matter?

Those are the four states that, by Democratic National Committee rules, are allowed to vote first. Iowa held its caucuses Monday, New Hampshire has a primary Tuesday, Nevada hosts caucuses Feb. 22 and South Carolina holds the first-in-the-South primary Feb. 29.

Iowa and New Hampshire have held a privileged place in the calendar since 1972, when nominating power first moved from party bosses in the proverbial smoke-filled rooms to primary voters. Many criticize these two states’ prominence, arguing that the two small, rural, predominantly white states don’t accurately represent Americans or Democrats. To add diversity to the calendar, in 2008, the Democratic Party allowed South Carolina and Nevada to vote next before the rest of country.

While these four states don’t bring lots of delegates, they matter immensely as early signals of candidate strength. Together they have 155 delegates, only 3.9 percent of the total pledged Democratic delegates, but they can profoundly affect a candidate’s momentum. That’s because the media overemphasizes strong showings in these early states and because candidates can often translate such attention into a boost in polling and fundraising, which helps them in the later states.

What happens in February will affect the race’s trajectory

If a single candidate wins all of the early carve-out states, or even three of the four, that candidate will become a strong front-runner, heading into Super Tuesday poised to win a significant portion of delegates.

After all, a majority of Democrats are reporting that it is more important that the nominee can beat President Trump in November than that the nominee agree with them on the issues. With that outlook, they may be willing to quickly coalesce around a candidate.

That won’t be the case if the candidates split the early states. If there’s no clear front-runner, the diversity of Super Tuesday states combined with the proportional-representation rules may intensify the chaos rather than defuse it.

Thirty-one contests will be held in the first seven weeks. The remaining 26 states vote over the next 11 weeks, with some lulls when few or no states vote. If Super Tuesday delivers no obvious nominee, the race will probably stretch on for quite a while, with candidates continuing to divide delegates proportionally.

Correction: An earlier version of this post inaccurately stated that candidates must win at least 15 percent of the nationwide vote to seat delegates at the national convention. In fact they must win at least 15 percent of the vote in a district or state. The editors regret the error.

Caitlin Jewitt is an assistant professor of political science at Virginia Tech and author of “The Primary Rules: Parties, Voters, and Presidential Nominations” (University of Michigan Press, 2019).