Ballots are prepared to be tabulated for Maine's Second Congressional District's House election Monday, Nov. 12, 2018, in Augusta, Maine. The election is the first congressional race in American history to be decided by the ranked-choice voting method that allows second choices. Credit: Robert F. Bukaty | AP

Ranked-choice voting faced no formalized opposition when it first appeared on Mainers’ ballots in 2016, allowing its proponents to make many bold claims about how the system would improve our elections.

In the Bangor Daily News and elsewhere, Mainers were specifically told that RCV would produce majority winners, reduce negative campaigning, and bolster the influence of independent and third-party candidates.

To test these claims, our analysts at The Maine Heritage Policy Center examined the data and outcomes of 96 RCV elections held throughout the country, including Maine, and determined these claims simply do not hold water.

The first and most important claim to address is the issue of majority winners. Through the elimination and reallocation of votes, we were told this system would restore majority rule over our elections – no longer could they be decided by a mere plurality.

The problem, however, is that RCV is so convoluted that few voters know how to maximize their influence over the process. Instead of filling out their ballot to the fullest extent, some voters rank one or just a few candidates they prefer (or make ballot marking mistakes) which have a significant impact on how a majority is determined.

When a voter makes ballot marking mistakes or only ranks candidates that are eventually eliminated from contention, their ballot becomes exhausted and no longer counts toward the final denominator used to determine a majority winner. It’s almost as if these voters never showed up on Election Day.

The prevalence of exhausted ballots is significant. Of the races we examined, an average of 11 percent of ballots became exhausted in each election and the eventual winner won with a “fake majority” 61 percent of the time. Because exhausted votes are excluded from the final denominator used to determine a majority, most winners of RCV elections do not win a true majority of the votes cast on Election Day; they receive a majority after the system throws out the votes of more than 10 percent of the electorate.

RCV also does nothing to increase the influence of non-major party candidates, and in some ways it actually makes things worse. Ranked-choice voting supposedly empowers voters to rank candidates by their true preference rather than wasting their vote on a candidate who does not have a legitimate chance of winning.

From the League of Women Voters Maine’s website: “Ranked choice voting allows voters to support their favorite candidate without worrying that they might ‘throw their vote away,’ or worse, split their votes with like‐minded voters and unintentionally help elect the candidate they like the least.”

The data tell a different story. In the 2018 2nd Congressional District election, two rounds of tabulation were necessary to declare Rep. Jared Golden the winner and he prevailed with “fake majority” because more than 8,000 ballots were exhausted. There were 289,624 valid ballots cast in the first round of the election, meaning a majority would be achieved when a candidate received 144,813 votes. However, 8,253 ballots were exhausted in the second round of tabulation and Golden was declared the winner with just 142,440 votes. This equates to 49.1 percent of the overall votes cast on Election Day; not a true majority.

These are primarily instances where a voter preferred only independent candidates Tiffany Bond or William Hoar, and did not rank either Golden or Bruce Poliquin. What was their reward for supporting their favorite candidate(s)? Because these voters exhausted their choices, they essentially, if unintentionally, threw their votes away in subsequent rounds of tabulation.

Another supposed selling point of RCV was that it would make our politics more civil and reduce negative campaigning. Little evidence from Maine exists to support this assertion.

According to Maine campaign finance data, independent expenditures (expenditures made by groups unaffiliated with a specific candidate) to support candidates declined by 40 percent in Maine’s 2018 gubernatorial primary compared to the 2014 cycle, while expenditures to oppose increased by 100 percent – from $0 combined in the 2006, 2010 and 2014 cycles to $207,500 in 2018.

This data, which indicates an increase in negative outside ads, calls into question the validity of the claim that the system improves the tone and civility of our politics. Any way you slice it, ranked-choice voting has not delivered on its promises to Maine people.

Fortunately, there’s still time to turn back.

The Maine Senate on Monday gave final approval to a bill that expands RCV to presidential elections. Gov. Janet Mills should take these findings into account and veto the legislation.

Jacob Posik is the director of communications for the Maine Heritage Policy Center.