The previous articles show how different methods of counting ranked ballots can produce different outcomes. The following elections are often cited by critics of RCV, and help illustrate how those outcomes can occur in practice. In many cases, these outcomes can be explained by how the rules of a particular system interact with voter behavior, rather than indicating a simple error or failure.

While many elections proceed without controversy, some have been cited by critics as examples of unexpected or confusing outcomes. A recurring issue is not simply that voters supported minor candidates, but that some ballots do not continue to the final round if they do not include a usable ranking for the remaining candidates. These outcomes can appear counterintuitive, especially when compared to first-round results.

The most common themes raised by critics include:

• Final outcomes that differ from first-round expectations, where candidates who were not leading in first-choice votes, sometimes despite low initial support, ultimately win.
• Voters who select only one candidate, whose ballots may not count in the final round if that candidate is eliminated.
• Ballots that mark multiple candidates at the same ranking level in systems not designed to handle indifference, which can cause those rankings to be unusable.
• Delays or errors in reporting results.
• Claims that the system may broadly favor one party over another.

This article reviews several commonly cited elections, focusing on what happened, what critics point to, and what happened mechanically in each case.

What happened
Pierce County used ranked choice voting for several county offices in 2008. In the County Executive race, Shawn Bunney led in the first round, but Pat McCarthy won after later rounds and vote transfers.

What critics point to
Critics point to 30,308 ballots that did not continue to the final round, out of 312,771 total ballots. They also note that the first-round leader did not win the final result.

What happened mechanically
In the first round, of 312,771 total ballots cast, Shawn Bunney received 108,527 votes (35 percent) and Pat McCarthy received 83,598 votes (27 percent), with the remaining 120,646 votes distributed among other candidates. In the final round, Bunney received 130,064 votes (46 percent of continuing ballots, 42 percent of total ballots) and McCarthy received 152,399 votes (54 percent of continuing ballots, 49 percent of total ballots), while 30,308 ballots did not continue because they did not include a usable ranking for either of the final two candidates.

What happened
A San Francisco Board of Supervisors race required multiple rounds before a winner was determined. The final result was decided after repeated eliminations and vote transfers.

What critics point to
Critics highlight that 8,200 ballots determined the winner, while 9,503 ballots did not continue to the final round and another 605 ballots were set aside for overvotes, out of roughly 17,000 total ballots. They compare these figures to argue that the final round involved fewer active ballots than the number of ballots that did not continue or were set aside.

What happened mechanically
In the first round, of 17,270 total ballots cast, Lynette Sweet received 2,655 votes (15 percent) and Malia Cohen received 1,988 votes (12 percent), with the remaining votes distributed among other candidates. In the final round, Cohen received 4,945 votes (56 percent of continuing ballots, 29 percent of total ballots) and Sweet received 3,822 votes (44 percent of continuing ballots, 22 percent of total ballots), while 9,503 ballots did not continue because they did not include a usable ranking for either of the final two candidates, and another 605 ballots were set aside for overvotes.

What happened
No candidate received a majority of first-choice votes. Bruce Poliquin led in the first round, but after additional rounds of counting and vote transfers, Jared Golden won the final result.

What critics point to
Critics point to 8,253 ballots that did not continue out of 289,624 total ballots. They note that those ballots did not continue into the final round. They also note that the candidate who received the most first-choice votes did not win.

What happened mechanically
In the first round, of 289,624 total ballots cast, Bruce Poliquin received 134,184 votes (46 percent) and Jared Golden received 132,013 votes (46 percent), with the remaining 23,427 votes distributed among other candidates. In the final round, Poliquin received 138,931 votes (49 percent of continuing ballots, 48 percent of total ballots) and Golden received 142,440 votes (51 percent of continuing ballots, 49 percent of total ballots), while 8,253 ballots did not continue because they did not include a usable ranking for either of the final two candidates.

What happened
A large field of candidates led to multiple rounds of counting. The process took time, and an early unofficial report included test ballots before the error was corrected.

What critics point to
Critics point to about 140,000 ballots, nearly 15 percent of roughly 940,000 total ballots, that did not continue into the final round. They also note that about 135,000 test ballots were included in early unofficial results.

What happened mechanically
In the first round, of 942,031 total ballots cast, Eric Adams received 289,403 votes (31 percent) and Kathryn Garcia received 184,463 votes (20 percent), with the remaining 468,165 votes distributed among other candidates. In the final round, Adams received 404,513 votes (50.4 percent of continuing ballots, 43 percent of total ballots) and Garcia received 397,316 votes (49.6 percent of continuing ballots, 42 percent of total ballots), while 140,202 ballots did not continue because they did not include a usable ranking for either of the final two candidates.

What happened
The city used ranked choice voting for a multi-winner Charter Commission race. After transfers, some candidates who began near the top of the first-choice count did not win seats, while a candidate who began with fewer first-choice votes ultimately won.

What critics point to
Critics point out that one candidate began with about 4 percent of the vote, roughly 1 in 25 voters, and still won a seat. They also note that a candidate with more than 1,800 votes lost to a candidate who initially had 367 votes.

What happened mechanically
The race began with 11 candidates competing for 4 at-large seats. In the first round, 8,757 active first-choice votes were cast. Steven DiMillo received 1,851 first-place votes (the 2nd most), while Patricia Washburn received 378 first-place votes (the 7th most, about 4 percent). As lower-ranked candidates were eliminated, their votes were transferred based on later rankings. Over successive rounds, Washburn accumulated votes from eliminated candidates and ultimately reached 3,744 votes, securing a seat. DiMillo, despite starting with a higher first-choice total, was eliminated in a later round with 2,563 votes and did not win a seat. By the final round, 2,089 ballots were inactive and did not continue because they did not include a usable ranking for the remaining candidates.

What happened
A crowded mayoral race required multiple rounds to determine a winner. The final result was very close.

What critics point to
Critics highlight that more than 4,000 ballots, over 18 percent of roughly 21,000 total ballots, did not continue to the final round. The final margin was only 21 votes.

What happened mechanically
In the first round, of 21,246 total ballots cast, Monica Zoltanski received 4,382 votes (21 percent) and Jim Bennett received 4,130 votes (20 percent), with the remaining votes distributed among other candidates. In the final round, Zoltanski received 8,620 votes (50.06 percent of continuing ballots, 40.57 percent of total ballots) and Bennett received 8,599 votes (49.94 percent of continuing ballots, 40.47 percent of total ballots), while 4,027 ballots were non-transferable because they did not include a usable ranking for either of the final two candidates.

What happened
The election required multiple rounds of counting. Democrat Mary Peltola won after transfers while Republican candidates received a combined majority of first-choice votes.

What critics point to
Critics highlight that nearly 15,000 ballots did not continue into the final round, out of roughly 188,000 total ballots. More than 11,000 came from voters who selected only one Republican candidate. They also point to the combined first-choice vote for Republican candidates.

What happened mechanically
In the first round, of 188,666 active first-choice votes, Mary Peltola received 74,817 votes (40 percent) and Sarah Palin received 58,339 votes (31 percent), with the remaining votes distributed among other candidates and write-ins. In the final round, Peltola received 91,266 votes (51.48 percent of continuing ballots, 47.46 percent of final tabulation ballots) and Palin received 86,026 votes (48.52 percent of continuing ballots, 44.74 percent of final tabulation ballots), while 14,997 ballots did not continue because they did not include a usable ranking for either of the final two candidates.

What happened
An Oakland school board race was initially reported with the wrong winner due to a ranked-choice tabulation error. The result was later corrected.

What critics point to
Critics note that the candidate originally treated as the third-place finisher was later found to have actually won. The race involved about 26,000 total ballots, and the error was discovered about two months after Election Day.

What happened mechanically
The Alameda County Registrar of Voters stated that the tabulation issue was due to a configuration error in the ranked-choice voting system. The same configuration was used across multiple contests, but only one race was affected in a way that changed the reported outcome. After the issue was identified, the Registrar conducted a review, worked with the system vendor, and retabulated the ballots using the correct settings.

If there are other RCV elections that have been cited or contested, please mention them in the comments and they can be added here.