Maine Democratic voters go to the polls Tuesday to select delegates to the national convention.

Even after former Vice President Joe Biden’s resounding victory in Saturday’s South Carolina primary, most prognosticators believe Vermont Sen. Bernie Sanders will win easily in Maine, as he will in his home state. He might even poll the most votes in Massachusetts, home of his liberal rival, Sen. Elizabeth Warren.

But how should those victories – or any primary victories in a crowded primary field – be understood? Maine voters have twice opted to run our elections using ranked-choice voting, electoral rules that allow voters to indicate which candidate they would favor if their first choice does not fare well in the voting. While the Legislature has opted to use ranked-choice voting in presidential primaries, that law will not take effect this year – and is being challenged once again by those favoring systems producing plurality winners.

What would the results in Maine look like if Maine used ranked-choice voting in Tuesday’s primary? While we can never know for sure, because candidates would campaign differently under different rules and because voters might make different decisions, we can draw some interesting conclusions from recently conducted polls.

Last week YouGov conducted a poll of 1,507 registered California voters for Stanford University’s Hoover Institution and the Bill Lane Center for the American West. The poll has a sampling error of plus or minus 3.1 percent; respondents answered the survey over a three-day period, after the Nevada caucuses and the South Carolina debate but before Biden’s victory in South Carolina.

On that survey, we asked respondents about their choice for presidential nominee under the vote-for-one-candidate system, and then we asked them how they would have ranked candidates had they been given that choice. In analyzing the results, we accepted the Democratic Party rules holding that only candidates receiving at least 15 percent of the vote would receive any delegates. Thus, we eliminated the candidate with fewest votes, reallocating that candidate’s voters to their second-choice candidate, and continued to do so (allocating to the highest remaining candidate) until all remaining candidates had at least 15 percent of the vote.

As is expected in Maine, Sanders led the preference poll. His margin over Biden and Warren was just under 10 percent. Only those three candidates polled 15 percent statewide; thus, they were the only candidates who would receive an allotment of the delegates awarded statewide. Former New York City Mayor Michael Bloomberg was at 13 percent; he and the other candidates listed – former South Bend, Indiana, Mayor Pete Buttigieg, Rep. Tulsi Gabbard, Sen. Amy Klobuchar and businessman Tom Steyer – would all be shut out of the statewide allotment of delegates, though one or more might have qualified for delegates apportioned by congressional district. The press would report these results as a big victory for Sanders, with Biden and Warren acknowledged for winning some delegates and the others as losers.

However, that is not the message that reflects California voters’ real preferences. If Californians had ranked candidates, Sanders would still have led, but his margin over Biden would have been about 2 percent, not 10 percent. Warren would have trailed Biden by 5 percent, further behind him than she was with single preferences noted. Bloomberg would have surpassed the 15 percent threshold. Biden and Bloomberg, the two candidates perceived as most moderate, would have benefited far more than the more liberal candidates. As one example, take the supporters of Steyer, who in fact dropped out of the race after South Carolina (as have Buttigieg and Klobuchar). His voters, when reallocated according to our poll, would have gone largely to Biden and Bloomberg, with only a few to Warren and even fewer to Sanders.

Would the Maine results be different under ranked-choice voting? We do not know for sure, but interpreting voter true preferences when many candidates are running is more complex than simply declaring whoever gains a plurality of the votes as the winner. These results from our California poll show clearly that the Democratic Party is more evenly split than declaring a plurality leader as the winner suggests.

Other YouGov polling, administered on a national sample for The Economist throughout the primary season, suggests that self-described “very liberal” Democrats are enthusiastically for Sanders, that “liberal” Democrats are split among the candidates and that “moderate” Democrats favor Biden or Bloomberg over the others. The party is split – and when we look at the results in Maine, we should appreciate that a plurality for one candidate does not imply support from all party voters.

Comments are not available on this story.