Thanks for your answer to the question about proportionately and the example of an election in which one candidate secures a strong majority of the vote but receives only one seat on the Council -- and then ends up under-represented (and potentially outvoted) 2-1 by candidates who receive lesser votes.
That is more than a hypothetical. Most of the top vote getters in recent Council elections have garnered about 40-55% of the vote in their large primary fields. So, outcomes of this kind will be a distinct possibility in the first round of vote tallies – and possible, although less likely, thereafter. Even though the system will preclude crediting candidates with more than 25%+1 of the vote, there will be a calculation of votes received before they are capped and their surplus transferred to others. So, this aspect of the process, although complicated, will still be transparent enough to show the actual level of support for winning candidates. And that’s when the disparities in representation will be clear.
You are right to point out that most proportional representation systems rely on multiple parties (and, I'd add, parliamentary systems) to achieve their purpose. But that's not how it will work in Portland's nonpartisan elections. The outcomes of the new system will rarely be strictly proportionate and sometimes quite disproportionate when measured by the voting power granted to winners in relation to the levels of support they receive from their electorates. And where outcomes are disproportionate, their effect will be contrary to the goal of better reflecting the will of the voters.
Proponents of this system contend it is designed to give voice to voters whose preferred candidates might never be able to pass a 50%+1 threshold. That’s fine, when its outcomes reflect, at least roughly, the will of the voters. But in cases where candidates receive majorities or near majorities and end up in the minority in the representation of their districts, that will clearly not be the case.
Proportionality refers to the selection of candidates, not to the amount of power each elected official has over legislation. Some want to claim there is a problem by only observing first ranked choices and ignoring all the other ranked choices on a person’s vote. This is ridiculous given that the name is literally “Ranked Choice Voting.” Moreover, not having first ranks alone determine the outcome isn’t something that’s going to be resolved in Mapps’ proposal. The person with the most first ranked choices would have no guarantee of winning.
With respect to gaming RCV (both single-seat IRV and multi-seat STV), I think it's commonly asserted that votes can't be strategized because it's not practically possible. The behavior of the runoffs makes results so chaotic that it's not really feasible to improve the odds of getting the result you want by ranking strategically instead of honestly. You'd have to have extraordinarily accurate insight into what other voters are going to do. For example, in the recent Alaska House election, it's likely that many Palin voters would have ranked Begich first if they knew in advance that ranking Palin first was going to cause Peltola to ultimately win. But there's just not good enough information about other voters' sentiment to be able to strategize that way in most elections. Palin was a realistic contender. All three of the candidates were, so it was infeasible to figure out in advance if there was a way to strategize a ballot.
With our current FPTP elections, it's usually much easier to discern which candidates have a shot and which ones are spoilers. Voters in our gubernatorial race know that Johnson has no real chance of winning, for example. They may like her best, but choose to strategically vote for Kotek or Drazen because they want their vote to have a meaningful impact and are willing to betray their true favorite to do so. But if our gov race were IRV, it wouldn't necessarily be so clear. It would be much more difficult, probably impossible to know if you should vote strategically or honestly. So it makes sense to simply vote honestly. It's not that RCV can't be strategized, it's more that voters rarely have sufficient information to know in advance whether strategic voting is to their benefit.
Regarding "gaming" the voting system, it's not entirely correct to say that the voters ultimately determine the outcome. Some theoretical political scientists will say: tell me the outcome you want, and I'll tell you the voting rules that will ensure it, given what we know about the preferences of those voting. In practice, we don't change voting rules regularly in municipal elections; we don't always know voter preferences; and voter preferences can change over time. So the theoretical point is not conclusive. The implication is that regardless of whatever rules you create, someone will be able to figure out a way to game it, at least to dramatically increase the probability of the outcome that someone prefers, so long as the person or people understand the rules and the preferences voters will bring into the process. One need only observe the machinations in Oregon's governor's race, such as following the trail of contributions to candidates, to see this in action. Or watch an astute parliamentarian at work, such as U.S. Senate and House leadership to find examples. How long this manipulation can be sustained is another matter. If voters learn about it, they might indeed have the ultimate say but that's not guaranteed.
Thanks for your answer to the question about proportionately and the example of an election in which one candidate secures a strong majority of the vote but receives only one seat on the Council -- and then ends up under-represented (and potentially outvoted) 2-1 by candidates who receive lesser votes.
That is more than a hypothetical. Most of the top vote getters in recent Council elections have garnered about 40-55% of the vote in their large primary fields. So, outcomes of this kind will be a distinct possibility in the first round of vote tallies – and possible, although less likely, thereafter. Even though the system will preclude crediting candidates with more than 25%+1 of the vote, there will be a calculation of votes received before they are capped and their surplus transferred to others. So, this aspect of the process, although complicated, will still be transparent enough to show the actual level of support for winning candidates. And that’s when the disparities in representation will be clear.
You are right to point out that most proportional representation systems rely on multiple parties (and, I'd add, parliamentary systems) to achieve their purpose. But that's not how it will work in Portland's nonpartisan elections. The outcomes of the new system will rarely be strictly proportionate and sometimes quite disproportionate when measured by the voting power granted to winners in relation to the levels of support they receive from their electorates. And where outcomes are disproportionate, their effect will be contrary to the goal of better reflecting the will of the voters.
Proponents of this system contend it is designed to give voice to voters whose preferred candidates might never be able to pass a 50%+1 threshold. That’s fine, when its outcomes reflect, at least roughly, the will of the voters. But in cases where candidates receive majorities or near majorities and end up in the minority in the representation of their districts, that will clearly not be the case.
Proportionality refers to the selection of candidates, not to the amount of power each elected official has over legislation. Some want to claim there is a problem by only observing first ranked choices and ignoring all the other ranked choices on a person’s vote. This is ridiculous given that the name is literally “Ranked Choice Voting.” Moreover, not having first ranks alone determine the outcome isn’t something that’s going to be resolved in Mapps’ proposal. The person with the most first ranked choices would have no guarantee of winning.
With respect to gaming RCV (both single-seat IRV and multi-seat STV), I think it's commonly asserted that votes can't be strategized because it's not practically possible. The behavior of the runoffs makes results so chaotic that it's not really feasible to improve the odds of getting the result you want by ranking strategically instead of honestly. You'd have to have extraordinarily accurate insight into what other voters are going to do. For example, in the recent Alaska House election, it's likely that many Palin voters would have ranked Begich first if they knew in advance that ranking Palin first was going to cause Peltola to ultimately win. But there's just not good enough information about other voters' sentiment to be able to strategize that way in most elections. Palin was a realistic contender. All three of the candidates were, so it was infeasible to figure out in advance if there was a way to strategize a ballot.
With our current FPTP elections, it's usually much easier to discern which candidates have a shot and which ones are spoilers. Voters in our gubernatorial race know that Johnson has no real chance of winning, for example. They may like her best, but choose to strategically vote for Kotek or Drazen because they want their vote to have a meaningful impact and are willing to betray their true favorite to do so. But if our gov race were IRV, it wouldn't necessarily be so clear. It would be much more difficult, probably impossible to know if you should vote strategically or honestly. So it makes sense to simply vote honestly. It's not that RCV can't be strategized, it's more that voters rarely have sufficient information to know in advance whether strategic voting is to their benefit.
Regarding "gaming" the voting system, it's not entirely correct to say that the voters ultimately determine the outcome. Some theoretical political scientists will say: tell me the outcome you want, and I'll tell you the voting rules that will ensure it, given what we know about the preferences of those voting. In practice, we don't change voting rules regularly in municipal elections; we don't always know voter preferences; and voter preferences can change over time. So the theoretical point is not conclusive. The implication is that regardless of whatever rules you create, someone will be able to figure out a way to game it, at least to dramatically increase the probability of the outcome that someone prefers, so long as the person or people understand the rules and the preferences voters will bring into the process. One need only observe the machinations in Oregon's governor's race, such as following the trail of contributions to candidates, to see this in action. Or watch an astute parliamentarian at work, such as U.S. Senate and House leadership to find examples. How long this manipulation can be sustained is another matter. If voters learn about it, they might indeed have the ultimate say but that's not guaranteed.