If we use “qualified for at least one debate” to define “serious candidate for a party’s nomination,” then the race in the Democratic presidential primary winnowed down to 16 when Rep. Tim Ryan ended his campaign last week.  To put this in perspective, the oversized Republican field of 2016 reached its maximum at 16 candidates.  One might question whether it is fair to consider John Delaney a major candidate simply because he qualified for a debate, but we are also considering Jim Gilmore a major candidate for Republicans in 2016 under our definition.

This has led some to call for more aggressive winnowing of the Democratic field.  Take this Slate article, which urges multiple candidates to drop out since they have almost no chance of becoming the nominee, including reasonably strong candidates such as Pete Buttigieg, Amy Klobuchar and Cory Booker. The Democratic Party itself seemingly agrees, having substantially raised the polling and fundraising requirements to qualify for the December debate.

This discussion is getting ahead of things.  Here is why:

1. We don’t know a lot about how primaries work, much less mega-primaries. Since the presidential nomination process was turned over fully to the voters in 1972, we’ve had a total of 24 presidential primary campaigns. Several of these were effectively uncontested – the 1984 Republican race, the 1996 Democratic race, the 2004 Republican race and the 2012 Democratic race -- lowering the number of cases to 20.  Still more saw only weak opposition arise, such as the 1972 Republican race. 

Even with the remaining primaries, we don’t see much that resembles the current contest.  Consider the Democratic side. In 2016, at the peak, six major candidates (and that is being generous to Jim Webb, Lincoln Chafee and Lawrence Lessig) ran for president. In 2008 it was eight.  The 2004 primaries saw 10 candidates run, while 2000 was a two-man race.  In 1992, Democrats had a six-candidate field, 1988 saw 11, and 1984 had eight.  On the Republican side, it is a similar story.

How does this history translate to the current “mega-field”? We don’t know, because we only have one other such field in recent times: the Republican’ in 2016.  We really can’t base much off of this. A careful analyst must also consider that the factors driving these large fields – the rise of online fundraising and super PACs, the availability of the Internet to bypass media and party gatekeepers, and so forth – also should alter our definitions of “viability.”  Put differently, the old adage that there are three tickets out of Iowa and two out of New Hampshire may no longer hold true.

Think of it this way: In the 1996 Iowa Republican caucuses, eighth place was Morry Taylor (you will be forgiven if you have to Google him).  In 2016 it was John Kasich, who eventually finished in third place in the delegate count.

2. Undecideds and “the one percent” matter. A lot. One other side effect of the large field is that an unusually large number of votes are being held by the “undecided” category, and by “one-percenters,” my term for candidates that have only minuscule support. To see what I mean, consider the national RCP average. “Undecided” along with candidates receiving 2% of the vote or less currently account for about 20% of the total.  This means that about 20% of the field is either given over to candidates who really are long shots for the nomination (but see below) or are undecided.  It’s not an exaggeration to say that undecided is currently in third place.

We see similar effects in the early primary states. Undecideds/minor candidates total 19% of respondents in Iowa (good for second place) and are leading Nevada with 27%. In South Carolina 18% of voters are undecided or for minor candidates (second place).  Only in New Hampshire does a relatively large segment of the electorate seem to have settled on a major candidate; still, the 9% total for undecideds/minor candidates is still good enough for fourth place.

The field will continue to winnow, and those voters will go somewhere.  Joe Sestak’s 0.5% and Beto O’Rourke’s 1% average may seem inconsequential, but combined they could be the difference between Tulsi Gabbard finishing in eighth place and fifth.

3. There isn’t much separating top from bottom here.

One of the problems that Republicans had in 2016 was that the incentives for candidates to drop out simply weren’t there, and this was a direct consequence of the size of the field.  Most of the 2020 minor candidates have been involved in races where they have come from behind or had a substantial surge in the polls.  Right now none of them are more than three points out of fifth place in Iowa, none is more than nine points out of fourth place in New Hampshire, none is more than five points out of fourth place in Nevada, and none is more than seven points out of fourth place in South Carolina.

To be clear, I’m saying that finishing fifth, fourth, fourth and fourth would get a candidate the nomination.  But because the Democratic race is sequential, scoring at this level in a few early races probably extends a candidate’s lifespan, and all of these candidates are within striking distance of such a showing.

Moreover, since the eighth- and ninth-place candidates will likely be dropping out, that will free up a significant number of voters.  Again, where those voters go is sort of up for grabs; it seems like they should gravitate to the major candidates, but their status as voters for minor candidates suggests that there are things about the major candidates they already dislike.

4. Big things happen late in the primary season.

If you’re tuned in enough to be reading this, the 2020 presidential election probably already seems interminable.  For most voters, however, it is only getting started.  This means that lots of minds will be changed, and big movements for candidates can occur.  At this point in 2015, Donald Trump’s main challenger in the polling was Ben Carson, who would actually eclipse Trump briefly in the national polls in early November. Ted Cruz and John Kasich, who would be the last two candidates standing, were in fourth and ninth place, respectively.  In 2011, on the Republican side, we had just witnessed Rick Perry’s poll collapse, and we were in the midst of Herman Cain’s rise.  Newt Gingrich was at 9% in the polls; by mid-December he would have a 13-point lead and have around a third of the Republican electorate in his camp. Rick Santorum was at 2% in the polls, and would not begin his surge until January.

What about 2007? The Republican poll leaders were Rudy Giuliani and Fred Thompson, neither of whom would win a primary. Eventual nominee John McCain was in third place with about 15% of the vote, while Mike Huckabee was at just 7%. In Iowa, which Huckabee would eventually win, the former Arkansas governor was only just starting his surge; at the beginning of the month he was in fifth place. Barack Obama trailed Hillary Clinton by over 20 points nationally at this point in 2007. In 2003, the race looked like a two-man race between Dick Gephardt and Howard Dean.  Eventual nominee John Kerry was in third place, albeit with only 9% of the vote.

To be sure, none of these candidates were at 2% of the vote.  But none of these candidates were in 16-person fields either.  It is perfectly reasonable for candidates to wait and see if they can catch fire; someone usually does at this point in the primaries.

5. Joe Biden is the X-factor.  Looming above all of this is the figure of Joe Biden.  Biden’s vote share has been remarkably stable, but he’s slipping into second place in the early states and is toying with third place in New Hampshire.  If this turns into a panic among “establishment” Democratic donors, his candidacy could collapse rather quickly.  That would be a lot of votes, in addition to “undecided” and “one/two-percenters” potentially up for grabs.  Against this backdrop, it is understandable why candidates like Klobuchar stick around, at least for now.