One of the ideas behind leading the presidential primary voting with the Iowa caucuses is that the state serves an important role in narrowing down a party’s field of candidates. Its folksy let’s-all-get-in-a-room-and-discuss methodology seems like a purer form of politics, one that rewards direct outreach from candidates and grass-roots organizing — which seems, in the abstract, like a powerful reason beyond tradition to continue to let Iowa go first.

That may all be the case. But it doesn’t mean that the caucuses are immune to the vagaries of political spending or to other thumbs on scales. It doesn’t resolve an issue of increasing importance in Democratic politics in particular: that Iowa is largely white but asked to evaluate candidates for an increasingly diverse party.

It also doesn’t mean that Iowa will necessarily do much to winnow the field. And in 2020, it seems increasingly likely that the picture after the caucuses will be as confused as it is beforehand.

Over the past several days, two major polls of Iowa have been released. At a glance, they seem to offer starkly different pictures of what’s likely to happen in next month’s caucuses. A poll from CNN, the Des Moines Register and Mediacom, conducted by Selzer & Co., has Sen. Bernie Sanders (I-Vt.) in the lead with 20 percent support. A poll from Monmouth University released Monday, though, has former vice president Joe Biden leading, with 24 percent of respondents identifying him as their first choice.

But both polls show a field with four candidates in the top tier. Biden’s lead in the Monmouth poll is a bit bigger than Sanders’s in the Selzer one, but neither is outside the margin of error.

Remember, though, that Iowa doesn’t simply cast votes in favor of a candidate. Instead, each caucus location haggles over who should win its delegates, meaning that supporters of candidates who do not generate enough support in a location are forced to either leave or support another candidate. If we add voters’ second choices to the mix, the picture shifts. Suddenly, former South Bend, Ind., mayor Pete Buttigieg steps into the lead in Monmouth’s poll, with 40 percent of respondents identifying him as their first or second choice. In Selzer’s poll, that position goes to Sen. Elizabeth Warren (D-Mass.).

If you think that’s confusing, just wait until Iowa Democrats release the results on the night of the caucuses. Not only will they release data showing total support after voters have been pushed to their second (and third) choices, but they’ll release data on the level of support each candidate got the first time around. On top of that, they’ll release the total number of delegates each candidate earned, which should largely correspond to the post-shift numbers but could differ in important ways.

The Monmouth poll suggests that Biden and Buttigieg will be declared the winners of the first round and overall voting. The Selzer poll suggests it will be Sanders and Warren. In other words: Any of the four leading candidates may be able to make the case that they “won” once caucusing is over.

It’s worth noting that there have been big shifts in each poll since November. Biden and Sanders improved in the Monmouth poll, for example, while the Selzer poll is most notable for Buttigieg’s decline.

But while Monmouth, too, shows Buttigieg slipping as voters’ first choice, its new poll also shows his position to have improved after second choices are factored in.

How likely is it that Buttigieg will need to rely on being voters’ second choice? It’s quite likely. While the percentage of respondents in both the Monmouth and Selzer polls who say they’re firmly decided on who they plan to support has increased since November, in neither case did a majority of respondents say that. In Monmouth’s poll, more than half said they were open to changing their minds, though 13 percent of the total said that was unlikely.

Monmouth’s summary of its poll notes that in November, only one candidate had a majority of his supporters say they were firmly committed to supporting him — Sanders — but now a majority of supporters of each of the top four candidates hold that position. But in no case do at least two-thirds of each candidate’s supporters say they won’t change their mind, meaning that between 35 percent (for Biden) and 47 percent (for Warren) of voters who’ve declared a first choice might change their minds before the caucuses.

Put another way, only 45 percent of respondents in Monmouth’s poll both 1) support one of the top four candidates and 2) say they won’t or are unlikely to change their minds. A lot can happen in that other 55 percent of caucusgoers.

So who’s going to win Iowa? Allow me to put on my stoned-out beatnik persona: What does win even mean, man? Polling shows it’s a toss-up, and the results that are made public will allow for a variety of interpretations.

Iowa gets to go first and has helped crown nominees in the past. If it’s going to do that this year, which seems uncertain, it’s hard to guess who that lucky candidate will be.