Resolution criteria
This market will resolve based on the number of rounds required to determine a winner in the 2025 New York City Democratic mayoral primary using ranked-choice voting (RCV). The options are:
2 or Fewer: Resolves "Yes" if a candidate secures a majority (over 50%) of first-choice votes, eliminating the need for additional rounds, or if only one elimination round is needed.
3: Resolves "Yes" if exactly three rounds are needed to declare a winner.
4 or More: Resolves "Yes" if four or more rounds are required to determine the winner.
Official results from the New York City Board of Elections will serve as the source for resolution. (vote.nyc)
Background
In New York City's RCV system, voters rank up to five candidates in order of preference. If no candidate receives more than 50% of first-choice votes, the candidate with the fewest votes is eliminated, and votes are redistributed to the next choice on those ballots. This process repeats in successive rounds until a candidate achieves a majority. (vote.nyc)
Considerations
The number of RCV rounds depends on the distribution of first-choice votes and the total number of candidates. A larger field or a more evenly distributed vote can lead to more rounds before a majority is reached. For instance, in the 2021 Democratic mayoral primary, eight rounds were necessary to declare a winner. (en.wikipedia.org)
@JedRakin Funny thing is, the way RCV is usually taught it would be 4+ rounds. It seems the misconception is fairly widespread https://polymarket.com/event/nyc-mayoral-dem-primary-of-rcv-rounds?tid=1751046041681
@JedRakin I've now spent an hour trying to figure this out and it's much more complicated than I thought. It looks like the way the results ended up for this election will be without precedent in NYC, so I honestly have no clue how they'll tabulate it.
vote.nyc doesn't mention their batch elimination method that I can see, but according to Fair Vote there would have to be several rounds:
"In RCV elections which allow batch elimination, multiple candidates in last place(s) will be eliminated in a single round if their combined vote total in that round is less than the vote total of any other candidate."
So while the totals for candidates 4-12 combined do not exceed the votes for Lander (nor do the votes of 3-12 exceed the vote total for Cuomo) they can't all be eliminated at once because the combined totals of Bartholomew, Prince, and the Write-ins are greater than the total for Ramos.
In 2021 the first round only removed the write-ins, but there are currently more total write ins than votes for Bartholomew.
@MarkFrost This market's resolution will speak for itself, but I think you've misinterpreted that line. It specifically says any other candidate, not the next candidate above them. Since the candidates in 3rd-12th combined are less than any other candidate (both Zohran and Cuomo), they can be batch eliminated.
Further supporting that is Section 1057-g of the New York City Charter, which is more specific with "Election is mathematically impossible. The term "election is mathematically impossible" applies to a candidate who cannot be elected because such candidate's vote total in a round, plus all votes that could possibly be transferred to such candidate in future rounds from candidates who received a fewer or an equal number of votes, would not be enough to surpass that of the candidate with the next highest vote total in such round." and then "Batch elimination. The term "batch elimination" means the simultaneous elimination of multiple candidates whose election is mathematically impossible." and then "If no candidate receives a majority of highest rank votes, tabulation shall proceed in rounds. In each round, the number of votes for each continuing candidate shall be counted; each continuing ballot shall count as one vote for its highest ranked continuing candidate for that round; and exhausted ballots shall not be counted for any continuing candidate. A round ends with one of the following outcomes:
(a) If there are two continuing candidates, the candidate with the most votes shall be declared the nominee of his or her party for a primary election, or elected winner for an election for which nominations were made by independent nominating petitions.
(b) If there are more than two continuing candidates, the last place candidate shall be eliminated and a new round shall begin; provided, however, that batch elimination shall occur at the same time as such elimination of the last place candidate, unless such batch elimination would result in only one continuing candidate, in which case no such batch elimination shall occur."
Since 3-12 all have fewer combined votes than Zohran or Cuomo, all of them have mathematically impossible elections, so they will all simultaneously be eliminated in batch elimination.
The question I was referring to in my comments was write-ins. Traditionally, due to the way BOE's code is designed, they eliminate write-ins first in their own round and then proceed to the process illustrated above. That would make it 3 rounds. But there's a separate clause in their manual canvass instructions that says if write-in votes exceed that of the last place registered candidate, the write-ins must be sorted into every candidate listed on write-in ballots. In the provisional results, write-ins currently exceed the vote total of last place candidate Selma Bartholomew. In my research, this has never happened in NYC since ranked choice was implemented in 2021.
That brings up the question, will the BOE simply verify that none of the write-in candidates' vote total exceeds that of the last place candidate and then proceed as normal (would result in 3 rounds) or will the BOE divide up each write in candidate and then run the RCV tabulations (which would exclude the write-in elimination step altogether and bring it down to 2 rounds).
@HenryRodgers Are you referring to write-ins exceeding a listed candidate or some other element of the situation when you say that?
@andreiklepatch Yes the former, I haven’t been able to find another example where that was the case in NYC, but let me know if you have.
@HenryRodgers I also looked through all previous examples now and am pretty sure this never happened before. I'd say that 2 is probably a bit undervalued at these prices but really no way of knowing how it will go.