Primaries are about winning. Frequently, a candidate winning or losing a single state by a few points completely changes the trajectory of the contest. In 2016, Bernie Sanders put a good scare into Hillary Clinton. He’d have struggled to win no matter what, but losing Iowa 49.9% to 49.6% and finishing a few points behind in Nevada made his task infinitely harder. They entered South Carolina with Hillary winning narrowly twice and Bernie by a large margin (NH) once. Imagine if he was on a three state winning streak.
Four years prior, Rick Santorum beat Mitt Romney in Iowa by a few votes. Literally. The problem was Romney was declared the winner on caucus night. It took three weeks to decide Santorum actually had a few more votes. By then, the narrative was already established, and while Santorum stuck around and won a few states, Romney had the nomination in order after winning Florida.
John McCain won the 2008 GOP nomination on the strength of beating Romney by 5 points in New Hampshire, Mike Huckabee by 3 points in New Hampshire, and Romney by 5 again in Florida. In 2008 Romney barely lost the important contests and fell short. Four years later he looked like he won Iowa when he didn’t and got the nomination.
A few point win in a primary or caucus is a much narrower thing than the same result in a general election. Polling often moves by 10 or 20 points in a single week as voters take previous results into account and start actually focusing on their final primary decision. Sometimes even when it doesn’t look close, it winds up close.
Right now, each of the first three states look very, very, very close. The most recent Iowa poll from CBS/YouGov had Biden, Buttigieg, and Bernie all tied at 23%. Elizabeth Warren trails at 16%, but that’s well within range. She’s thought to have the best ground team in Iowa. And remember, any candidate who falls short of 15% in a given caucus precinct loses their supporters to one of those who cleared the barrier. Who Amy Klobuchar, Tom Steyer, Andrew Yang, or Cory Booker voters decide to move to will likely make the difference.
That’s just one survey, but FiveThirtyEight’s new projection system is similarly confused. Biden is given a 1 in 3 chance of winning, Bernie 1 in 4. The favorite is estimated to wind up with 26.8% of the vote, even after the sub-15% candidates are redistributed. It’s very possible the top three or four finishers will wind up within five percent of each other. Normally, there’s a huge perception difference between first and fourth. But what if that’s only 5,000 voters?
New Hampshire is close too. Again, FiveThirtyEight is conflicted. They have Sanders with a 31% chance of victory, Biden 27%. That leaves plenty of room for Buttigieg and Warren too. A new survey from Monmouth had Buttigieg at 20%, Biden 19%, Sanders 18%, and Warren 15%. Other polls have them in a different order, and sometimes further apart, but you get the idea.
Then there’s Nevada. Biden is leading, but Sanders is close. If one of them wins the first two, does it change who finishes first here? Let’s say Bernie gets 25% in Iowa, Biden 23%. And then similar things happen in New Hampshire. If we assume record turnout in both states, about 14,000 total votes would separate them. Is it reasonable for this to change our perception and give one the patina of a winner and the other a loser with 48 states left to go?
Historically it matters. In the 1976 Republican primary, Gerald Ford beat Ronald Reagan by 2.8% in Iowa and 1.4% in New Hampshire and then went on a winning streak. Reagan rallied, but never managed to catch up. There are some important differences this year. Republicans tend to make most of their primaries winner-take-all. So it’s not just a matter of perception, it’s delegates too. Democrats allocate delegates proportionately. Some statewide, others by congressional district.
If someone barely wins, they’ll have barely more delegates. There are even situations where a candidate could finish a little behind in overall votes and due to distribution, have an extra delegate or two than the winner. From 1984 to 2016, Democrats gave super delegates–DNC officials, senators, representatives, etc. enough to make up 20-25% of the delegates needed for nomination.
So when Barack Obama and Hillary Clinton ended the primary season with a very similar amount of delegates (more for Obama, but very narrowly), the super delegates were able to put Obama over the top. Their support for him began with his narrow Iowa victory, so another instance of some close wins having a disproportionate effect. For 2020, super delegates won’t weigh in on the first ballot. It’s unlikely they’ll declare who they are supporting as far ahead of time as in 2008 for Obama and 2016 for Clinton.
If the delegate count won’t be much impacted, that reduces the chance of small differences mattering. Another normal reason is money. Candidates frequently count on early wins to create fundraising momentum to keep them in the game. Each of the top four had a strong fourth quarter, and when we see the final FEC report for Q4, we’re likely to see they’re in decent cash positions too.
Yang is raising money and his supporters don’t seem easily discouraged, so even if he isn’t a likely nominee, he won’t need to leave because he’s broke like most mid/lower-tier candidates do. Steyer can spend all day. And Steyer’s wealth is a rounding error for Mike Bloomberg, who isn’t even playing in the first four states, and doesn’t care about qualifying for debates.
The only candidate who absolutely needs an early W to keep going is Klobuchar. Minnesota borders Iowa, she’s raised the least of any candidate with a chance, and doesn’t have the same base of small contributors as Yang, Sanders, Buttigieg, or Warren.
Plus, as we know, Iowa and New Hampshire are not demographically representative of the country. Even less representative of the Democratic Party. With so much commentary on how the leading candidates are all white, the idea of closing up shop before primary voters of color can weigh in seems like a horrible idea.
In a previous year, Joe Biden would have to win one of the first three states, and it would help him tremendously to win one of the first two. But I think it might not matter much this time. As long as it’s close. Even if Sanders wins the first three, or Buttigieg or Warren the first two, I wouldn’t count Biden out.
Winning helps for sure. This time, unless the first contests are blowouts, I think voters will see this as a long primary season, not a short sprint.