1. Ranked Choice Voting
2. Details about Ranked Choice Voting
3. Data on Ranked Choice Voting
4. Who Wins RCV Races?

Who Wins RCV Races?

This section explores how RCV works in practice in the United States, particularly the types of outcomes it produces. The section assesses RCV’s success at electing majority winners, evaluates whether incumbent candidates succeed in RCV elections, explores come-from-behind wins, and as well as  RCV’s impact on technical grounds, namely monotonicity and its tendency to elect Condorcet winners.

Majority Winners

Majority rule is a fundamental principle of our democracy. However, our current system often elects winners with less than majority support or even less than half of the votes. Between 1992 and 2019, 49 senators from 27 states were elected with less than 50 percent support. 

Primary elections in particular can lead to nominees with small pluralities. Because votes are often split between many candidates, it often takes a relatively small proportion of votes to win. In some highly-contested races, winners earn as little as 22% of the vote, meaning 78% of people voted against them, but this small plurality is enough to win the nomination in an election with many candidates. This chart shows plurality wins in primary elections for the U.S. House of Representatives in 2020. 

Ranked choice voting solves this problem by requiring that the winner receives over 50% of votes active in the final round, therefore electing nominees and candidates with the broadest support.

https://flo.uri.sh/visualisation/9228151/embedInteractive or visual contentflourish-embed-iframenowidth:100%;height:600px;allow-same-origin allow-forms allow-scripts allow-downloads allow-popups allow-popups-to-escape-sandbox allow-top-navigation-by-user-activation0

Majority winners in RCV elections

• There have been roughly 300 single-winner ranked choice elections in the United States that included at least three candidates (meaning no candidate can win a majority by default. When there are two candidates, one candidate must mathematically win over 50% of votes, except in the event of a tie). A majority winner was identified in the first round in about 40% of these races. The remaining 60% races were decided by instant runoff before declaring a winner.
• Sometimes, winners of RCV elections don’t earn a majority of total votes cast. A winner is declared when a candidate has a majority of votes active in a given round of counting (excluding ballots that have become inactive). This has occurred in roughly one-third of single-winner RCV elections in the United States. In these races, the winner was preferred by a majority of voters who expressed a preference between the finalists.
• Inactive ballots occur more often when jurisdictions limit the allowed number of rankings. Half of the elections in which the winner earned less than 50% of total ballots cast in the first round occurred in elections where voters were limited to only 3 choices.Learn more about the majority criterion with RCV in our FAQ.

Incumbency

• A 2021 report found that a city’s use of RCV is not a significant factor in the likelihood of incumbent candidates winning reelection. 
  • Ranked Choice Voting and Incumbent Success, by FairVote. (2021). 

• RCV has a neutral effect on incumbents, according to 2017 work by FairVote. Incumbents are re-elected at the same rate both with and without RCV.
  • RCV and Incumbency in the Bay Area, by FairVote California. (2017).

Come-From-Behind Winners

A “come-from-behind” winner is a candidate who did not have the most votes in the first round, but secured enough second, third, or other choice preferences to win in a later round.

A “come-from-behind” winner is not an “upset” or unfair result. It is a natural feature of RCV that means it is working how it is supposed to, that is, rewarding candidates with broad support over those who can only win by small pluralities. Nonetheless, the first-round winner is most often the overall winner. 

• Twenty RCV races were won by a candidate other than the first-round leader. That amounts to 4% of all RCV races in the United States since 2004, and 10% of all races that used multiple rounds of counting.

• Of the 20 come-from-behind wins in RCV elections, 18 were won by the candidate who began in second place.

• Two were won by the candidate who began in third place: San Francisco's 2010 election for the 10th District which elected Malia Cohen, and San Francisco's 2020 election for the 7th District which elected Myrna Melgar. Analysis of ballot data reveals that both would have won head-to-head match-ups against any other candidate in those elections, making them the "Condorcet Winner” or “beats-all” winner. These two cases are examples of RCV electing a Condorcet winner when a two-round runoff election would not have done so.

• This chart shows all come-from-behind victories in the United States since 2004:

https://flo.uri.sh/visualisation/8395207/embedInteractive or visual contentflourish-embed-iframenowidth:100%;height:600px;allow-same-origin allow-forms allow-scripts allow-downloads allow-popups allow-popups-to-escape-sandbox allow-top-navigation-by-user-activation0

 

 

"Condorcet" winners

The Condorcet criterion states that the candidate who would win a one-on-one matchup against every other candidate should win the election.  RCV does not guarantee that the “Condorcet winner” will win, but it does make it extremely likely, and certainly out-performs traditional elections. Nonetheless, the Condorcet criterion is only one of many criteria we could use to evaluate an electoral system.

• It’s unclear how often single-choice plurality (our current voting method) or two-round runoff elections elect Condorcet winners because voters' back-choices are not  publicly disclosed.
  

• Of the 440 single-winner RCV elections in the United States since 2004 in which we have sufficient ballot data to assess whether the Condorcet winner won the election, all but one — the 2009 mayoral election in Burlington, Vermont —  were won by the Condorcet winner.

• In the Burlington election, the three strongest candidates (after all others were eliminated) had 37%, 34%, and 29% support. The third-place candidate, Andy Montroll, was eliminated. After the election, ballot image data showed that supporters of the top two candidates overwhelmingly ranked Montroll ahead of the other finalist, making Montroll the Condorcet candidate.
• It may be disputed whether it would have been better for Montroll to win the election despite attracting so little core support. However, it is certain that Montroll would have also lost under a two-round runoff election or a single-choice plurality election.

• In the rare instance where RCV would fail to elect the Condorcet winner, he or she must have attracted too little core support to place first or second in the final round. 

• Two RCV races elected the Condorcet winner when a two-round runoff would not have done so. In these races, both winners began in third place and finished in first place after the RCV tally. They are Connie Chan from San Francisco's 10th district in 2010 and Myrna Melgar from San Francisco's 7th district in 2020. In each case, a two-round runoff election would not have allowed the Condorcet winner to advance to the final round. 

 

The "monotonicity criterion"

Monotonicity means that ranking candidates lower doesn’t help them and ranking them higher doesn’t hurt them. Any voting method in which votes are counted in rounds — including RCV and two-round runoff elections — can have nonmonotonic outcomes but the realities of RCV in practice make it extremely unlikely.

To our knowledge, no group of voters in an RCV election has ever attempted to exploit the possibility of nonmonotonicity for strategic purposes. Doing so successfully requires highly unusual circumstances and a detailed and accurate prediction of how the electorate will rank the candidates. Such a degree of hindsight or voter control does not exist. As such, monotonicity under RCV is a largely academic question: it has never impacted an RCV campaign and is unlikely to impact a future one. Learn more in our FAQ.

Of some 500 RCV elections in the United States, only one had a possibly non-monotonic outcome: the 2009 mayoral race in Burlington, Vermont. Below are results from that election. 

Whether this election constitutes a non-monotonic outcome depends on how strictly the criterion is defined. No candidate could have won if voters merely ranked them lower. However, if some Wright voters (between 367 and 589 voters) had instead ranked Kiss first (and changed no other ballots), Wright would have been eliminated instead of Montroll, and Montroll would have beaten Kiss. In other words, a group of Wright voters could have caused Kiss to lose by ranking him higher. However, this would be Wright voters helping to elect Montroll. No group of voters could have elected their own preferred candidate by ranking them lower.

Evaluating electoral systems on criteria like Condorcet and monotonicity are certainly interesting (though often hypothetical) exercises. However, as per Arrow’s Impossibility Theorem, no system can meet every criteria in every circumstance. Therefore, we caution not to get too “stuck” on niche scenarios like the above when overall, RCV has proven success at electing broadly popular consensus candidates (as shown on this page).