The 2020 Democratic field may be the largest presidential primary field in modern history. With Sen. Michael Bennet entering the 2020 presidential race on Thursday, there are now 20 “major” Democratic candidates, according to FiveThirtyEight’s standards, and additional candidates might enter the race or reach major status. But it’s unlikely all these candidates run well into the 2020 primary season. Some Democratic candidates will likely drop out even before the Iowa caucuses, which are scheduled to kick off the voting process on Feb. 3, 2020. And large candidate fields have historically winnowed pretty quickly a month or so after Iowa, though there are reasons to wonder if 2020 could be different.
To get a sense of how many candidates might drop out this cycle and when that might happen, we gathered data for each party’s nomination process in all the presidential primary cycles from 1980 to 2016 where the party’s incumbent wasn’t running for re-election. (We started with 1980 because in the ’70s, some major candidates announced after the Iowa caucuses,1 and before that the nomination process was very different from today’s primary system.) To figure out who was running each year, we used a list of candidates put together by FiveThirtyEight editor-in-chief Nate Silver, though those candidates may not all be “major” under our current definition. Dropouts included anyone who ended a run in a major-party primary by conceding, suspending a campaign, endorsing an opponent or withdrawing, even if they continued campaigning as an independent or member of a third party.
The 14 nomination processes we looked at varied a great deal in terms of the size of the field, ranging from two candidates to 17, but on average, about one of every 4 candidates have dropped out before the Iowa caucuses.
A few candidates usually drop out before the Iowa caucuses
The total number of notable presidential candidates in a primary field and the number who dropped out before the Iowa caucuses, 1980-2016
Cycle | Party | Total candidates | Dropped out before Iowa | Share of Total |
---|---|---|---|---|
1980 | R | 9 | 2 | 22% |
1984 | D | 8 | 0 | 0 |
1988 | D | 11 | 2 | 18 |
1988 | R | 7 | 1 | 14 |
1992 | D | 8 | 1 | 13 |
1996 | R | 12 | 3 | 25 |
2000 | D | 2 | 0 | 0 |
2000 | R | 12 | 6 | 50 |
2004 | D | 10 | 2 | 20 |
2008 | D | 10 | 2 | 20 |
2008 | R | 12 | 4 | 33 |
2012 | R | 12 | 4 | 33 |
2016 | D | 5 | 2 | 40 |
2016 | R | 17 | 5 | 29 |
Average | 9.6 | 2.4 | 23 |
In some cycles, an obvious front-runner limited the number of candidates who ran, but in some cases, a bunch of people threw their hats into the ring even if someone else was clearly leading in the polls. In the 2000 Republican primary, for example, Texas Gov. George W. Bush was polling at close to 50 percent in the first half of 1999, but 11 other Republicans were still in the race. In the second half of 1999, however, Bush was polling at about 60 percent, and half of the GOP field had dropped out by the time of the Iowa caucuses in January 2000. And in other cycles with no clear front-runner, like the 2004 Democratic primary or 2008 Republican primary, the field attracted a large number of candidates.
But there’s a key difference between a presidential primary with a huge field — like the 17 candidates who vied for the GOP nomination in 2016 — and a primary where just two people are running, like the 2000 Democratic primary: More candidates drop out sooner. After all, there is only so much support — voters, money, campaign volunteers, and so on — to spread around, which can make sticking it out until Iowa more difficult in a larger primary field.
Average share of candidates who dropped out before the Iowa caucuses, by size of field, 1980-2016
Field size | total cycles | Avg. share who drop out | |
---|---|---|---|
≤10 | 8 | 16% | |
>10 | 6 | 32 |
If we split the primary data into two groups, fields with 10 or fewer candidates and fields with more than 10 candidates, we see that the share of candidates who drop out before Iowa is twice as high in bigger fields.
So for the 2020 Democratic field, it wouldn’t be that surprising if around a third of the candidates dropped out before Iowa. If there end up being 21 major candidates, for example, we might expect around seven to leave the race before voting even starts.
The candidate fields tend to winnow even faster shortly after the Iowa caucuses. For example, during the 2016 Republican primary, the 17-candidate field had dropped to 12 in the year before Iowans cast their votes, and seven more candidates dropped out in the four weeks after Iowa, leaving just five people still in the race. That four-week threshold — which has usually included most or all of the other early contests (in New Hampshire, South Carolina and, in more recent times, Nevada) — gives us a sense of what might happen to the Democratic candidate field in the next 10 months. On average, about two-fifths of the field was still in the race four weeks after Iowa; usually, that’s roughly five candidates, though if these proportions hold up in this cycle, we could be looking at eight or nine Democrats still in the race next March.2
Not many candidates are left a month after Iowa
Number of candidates who announced a run vs. number who were still running during Iowa caucuses and four weeks after Iowa, for fields with more than 10 major candidates, 1980-2016
Number of candidates | |||||
---|---|---|---|---|---|
Cycle | Party | Original field | during Iowa caucuses | 28 days after Iowa | Share of field left 28 days after Iowa |
2016 | R | 17 | 12 | 5 | 29% |
1996 | R | 12 | 9 | 5 | 42 |
2000 | R | 12 | 6 | 3 | 25 |
2008 | R | 12 | 8 | 5 | 42 |
2012 | R | 12 | 8 | 5 | 42 |
1988 | D | 11 | 9 | 8 | 73 |
Average | 12.7 | 8.7 | 5.2 | 42 |
We may see a number of candidates dropping out right after this four-week post-Iowa period in 2020, as this cycle’s Super Tuesday — when 13 states are scheduled to vote on the same day — is currently slated to be at the beginning of the fifth week after the Iowa caucuses.3
However, history might be an imperfect guide for 2020, as there haven’t been many Democratic races with a big field. In fact, just one recent Democratic primary has featured more than 10 candidates: the 1988 cycle, in which eight of the 11 major candidates were still going four weeks after the Iowa caucuses, so the dropout rate wasn’t very high.4
And there’s some evidence that Democratic candidate fields get winnowed more slowly than Republican ones, which could be due in part to the Democrats’ proportional delegate-allocation rules. Republican contests use more winner-take-all or winner-take-most rules to divvy up delegates, so it’s possible that more Democrats keep hanging around because they have a shot at earning delegates even in states where they know they won’t finish first. Plus, candidates like Sen. Kamala Harris and former Rep. Beto O’Rourke have a strong incentive to make it to Super Tuesday on March 3 — their delegate-rich home states of California and Texas are scheduled to vote that day, which could give their campaigns a big boost.
Moreover, it now seems easier to circumvent traditional channels of political power to attract support. Just look at technology businessman Andrew Yang, or author and motivational speaker Marianne Williamson, who have each attracted a sizable number of donors in an effort to qualify for the upcoming 2020 primary debates. Or consider Sen. Bernie Sanders, who mounted a strong challenge to his party’s front-runner in 2016 thanks in part to an unprecedented reliance on small-dollar donors. All three have focused more on appealing directly to voters than on winning over the important players within the party itself. So in 2020, we may discover that the incentives encouraging people to run have changed in a way that leads to fewer candidates dropping out than we might expect based on past contests.
And remember, a few more candidates may still get into the Democratic race before any drop out, so the field may not yet have reached its maximum size. But if modern presidential primary data is to be trusted, at least a handful of the candidates we’re hearing a lot about this year won’t be running anymore by the new year.
From ABC News: