I don’t know quite how to render an awkward silence in print, so imagine the beginning of this column starts with a long pause, followed by a clearing of the throat, the sense of mental gears turning and then, eventually, a “Let me get right back to you on that,” and the click of a phone call ending.
“Before we started talking, I thought I understood this,” one campaign strategist said as I pressed him on how the vote-counting would actually work under the radically new ranked-choice-voting system in the all-important Democratic primary of New York’s mayoral race.
These are phone calls with the most well-informed, well-wired practitioners of New York City politics. They are operatives, pollsters, field organizers and campaign managers, and the candidates themselves, and even they could not describe what exactly is going to happen once votes start getting counted on June 22.
Here is what we know: Voters will go to the polls and select not just their favorite candidate but their five favorite candidates, ranked in order of preference. There are currently 40 people who have declared that they are running for mayor of New York, from front-runners like Scott Stringer and Eric Adams to Paperboy Prince, a rapper who wants to legalize psychedelics in the city and, at a recent mayoral forum, tossed a pie into his own face to raise awareness about food insecurity. Only about ten of them, however, are likely to make the ballot, although that could change as Mayor de Blasio and Governor Cuomo face pressure to suspend the usual signature-gathering requirements in the wake of the pandemic.
While there is concern that some voters will be confused by any new kind of ballot, and that few people actually have a fourth or fifth preference for mayor — in a recent special City Council election in Queens, the first time RCV was tried in the city, there were stories of many voters leaving the bottom four slots on their ballots blank, and some insiders predict that the majority of voters will do the same in the mayor’s race — the big point of confusion is what happens after all the vote is in. To win, a candidate needs 50 percent of the vote. RCV is often described as creating rounds of voting (while requiring voters to go to the polls only once), with candidates falling away after each new count until a winner is declared.
At a basic level, here’s how it works: Let’s say there are ten candidates, who, by coincidence, are named Candidate One, Candidate Two, and so on, through Candidate Ten, and who by further coincidence got the number of first-place votes in that same order. (Candidate One had the most, Candidate Two the second most, and so on.) Because Candidate Ten received the fewest first-place votes, he is eliminated first. Ten’s voters are now “free agents,” as it were, and will go to the person listed second on their ballots. Let’s say all of Ten’s voters listed Candidate Nine as their second-place vote. They will then be counted as first-place votes for Candidate Nine. It’s still not enough for Nine (Ten didn’t have many votes to go around), and so after Candidate Nine is eliminated, all of the people who voted for Candidate Ten as their first choice and Candidate Nine as their second choice will now go to their third choice — Candidate Eight, or Candidate One, or whomever. All of the people who voted for Candidate Nine as their first choice, meanwhile, will also be free agents, and will go to their second choice — presuming their second choice wasn’t Candidate Ten — as first-place votes, until someone gets above 50 percent.
If this seems needlessly complicated, the previous way of doing things wasn’t exactly ideal either. Under the old system, if no one got above 40 percent, there was a runoff election a few weeks after the primary between the top-two vote-getters — and that second round tended to have extremely low turnout. In that situation, though, the candidates at the end at least knew whom they were competing against and which voters they needed to target. Ranked choice is supposed to force candidates to campaign in every part of the city, since they will need to be the second or third choice of voters they could have written off before. It gives voters a reason to stop strategically voting: New Yorkers can vote as first preference for a cause candidate who favors, say, LSD legalization but be secure in the knowledge that their vote will not be “wasted.” Hence, as we have seen, a lot more candidates enter in ranked-choice fields.
People working on this race say they all have paid close attention to the recent election in Queens (where RCV was rendered moot after the eventual winner won with over 50 percent of first-choice votes before any reallocations) and in cities around the country when it was first implemented. In Oakland, California, the first time RCV was tried, the candidate who “won” with the most first-place votes on Election Day ended up losing once second- and third-place votes were tabulated, something that happens in roughly one-third of all RCV elections, according to one strategist working on the mayor’s race. In San Francisco, meanwhile, London Breed, the pick of the city’s political Establishment, prevailed over two progressives who were running together because Breed picked up most of the second-place votes of a handful of minor candidates who were also on the ballot.
The campaigns seem to be largely of the belief that the best way to prevail under RCV is to mostly ignore it. The winner of the race is likely to be the winner of the race, goes the logic, or the person who received the most first-choice ballots at the start, and so the key is to gobble up support where you can. But there are some exceptions: Shaun Donovan, an Obama Cabinet official, has not polled higher than fifth place and is making a concerted effort to land on as many ballots as possible, if not in first place, then at least somewhere, figuring that the winner will be the person who ends up chosen by the greatest number of voters somewhere on their ballot.
A predominant school of thought holds that ranked choice is going to benefit Andrew Yang simply based on his name identification — that most voters won’t be familiar enough with the other candidates to know whom to put further down on their ballots. Because of this, Yang has been getting the bulk of the incoming flak from the rest of the field in an attempt to disqualify him among the swath of the voters who aren’t certified members of the Yang Gang. “I think you are going to see people start making the case that Andrew Yang is uniquely unqualified to hold the job and would be a disaster at it, so that even if he gets all of his first-place votes, he doesn’t pick up any more along the way,” said one strategist.
A main argument for RCV is that it’s supposed to make campaigns more civil, since votes are no longer a zero-sum game. To some extent, strategists say, that has been true. Besides the attacks on Yang, most of the negative campaigning has been through “no fingerprints” oppo research dumped in the press, such as revelations that former Citi executive Ray McGuire served as a banker to the Koch brothers, or that Eric Adams went on junkets to Azerbaijan, or that Yang once told an Ohio woman during his presidential campaign that he was trying to escape New York too. The prevailing levels of congeniality will probably keep falling though. For one thing, it is unlikely that there are many voters who would take offense if another candidate attacked their first choice. For another, as one operative working on the race told me, as we get closer to Election Day, candidates are going to be searching for ways to get attention and negative and personal attacks are often the best way to attract the attention of the press.
It is not far-fetched to consider a scenario where second-place votes play a determinative role. Most campaigns believe that any tipping point like that would come not from the candidates who are eliminated first but from candidates in the middle of the pack — those who are bopping along with 10 or 15 percent of the vote. Being the second-place choice of a disproportionate share of mid-tier ballots could be enough to nudge a first-tier candidate over the top in a close race. If the race looks like it will come down to two people, those two are likely to attack each other as viciously as in the campaigns of old — after all, neither will be eligible for the other’s second-place votes.
Last month, Gustavo Rivera, a state senator from the Bronx, became the first person in the history of New York City politics to endorse two candidates in one race, announcing that Stringer was his first choice and Diane Morales, a nonprofit executive, his second. Some wondered if this meant that Stringer and Morales were now going to run as slate, with Stringer telling his voters to support Morales and Morales telling her supporters to vote Stringer. So far, that hasn’t happened, but it touches on a key strategy at work: The best way to win a ranked-choice election is to go after the voters who are already committed to a rival candidate and explain why you are best as their second choice. “The key is you don’t tell people how to vote. You ask them for their vote knowing that they like someone else better,” said one pollster working on the race.
Rivera’s dual endorsement was read by many as a way for him to appeal to two of his constituencies at once — white progressives and lefty Latinos — as a way to create favorable news for both campaigns, but it points to the way in which institutional players could be more important than usual, according to David N. Schleicher, a professor at Yale Law School and an expert on RCV and municipal elections.
Few voters will come to the polls with a preference beyond their first or second choice. Will unions and democratic clubs, rabbis and reverends, announce not just that this is my candidate, but here is how the whole rest of how your ballot should look? That kind of influence on voters could be crucial, Schleicher says.
“Journalists hate ranked choice because it is a hard story to tell,” he said. “But this is an election in which most voters will have no clue what is going on. It is the summer during a pandemic, a lot of people are going to be out of town. The usual political strategy gets more complicated, and institutional players become even more important because no one knows what is going on. But it is going to be wild to watch.”