First Past The Post is the cause of many of the political ills in the United States, and I used to think RCV was the answer. But it is a disaster:
1. The rate of spoiled ballots goes way up, and disproportionately so in lower-income areas (https://rangevoting.org/SPRates.html#minn). Suppressing 3 to 5% of the vote in poor districts would be devastating; the winning margin is smaller than that in lots of elections.
3. RCV doesn't break us out of two-party dominance. Once a third-party candidate gains more support, they can still be a spoiler under RCV, forcing you to choose between the lesser of two evils.
4. Audits are essential to ensure an accurate count, but RCV makes them vastly more difficult because you can't add up the votes from smaller districts to get the overall result.
5. Let's have voting machine companies write more unreviewed proprietary software and make it more complicated, that sounds like a great idea!
There's a much better method that actually does help us escape two-party rule. It's simple, cheap, well-tested, accessible, easy to understand, and doesn't even require changing any ballots or software. That method is Approval Voting (https://electionscience.org/library/approval-voting-versus-i...). Simply:
• Use the same ballots as we do today
• Let people vote for all the candidates they like
• Count all the votes
Done. No new ballots, no new procedures, no new software, and extremely simple to explain to voters.
Here's the intuition for why this works. When you vote using the current system, expressing support for one candidate requires you to withdraw your support from another candidate. RCV is the same in that you can only have one first choice: if you want to rank one candidate higher, you must rank another candidate lower. That's what keeps you from voting for your true favourite. With Approval Voting, your decisions about each individual candidate are fully independent: how you vote on one candidate has no effect on any other candidate. So you can express your full support for your favourite without harming the "lesser evil" candidate.
RCV is frequently criticized by mathematicians and economists, who are familiar with its many problems. This is one of those uncommon situations where the theorists and pragmatists agree: Approval is better from both perspectives.
1. The rate of spoiled ballots goes way up, and disproportionately so in lower-income areas (https://rangevoting.org/SPRates.html#minn). Suppressing 3 to 5% of the vote in poor districts would be devastating; the winning margin is smaller than that in lots of elections.
2. With RCV, sometimes voting for someone you like will make them lose (https://rangevoting.org/ClayIrv2.html).
3. RCV doesn't break us out of two-party dominance. Once a third-party candidate gains more support, they can still be a spoiler under RCV, forcing you to choose between the lesser of two evils.
4. Audits are essential to ensure an accurate count, but RCV makes them vastly more difficult because you can't add up the votes from smaller districts to get the overall result.
5. Let's have voting machine companies write more unreviewed proprietary software and make it more complicated, that sounds like a great idea!
There's a much better method that actually does help us escape two-party rule. It's simple, cheap, well-tested, accessible, easy to understand, and doesn't even require changing any ballots or software. That method is Approval Voting (https://electionscience.org/library/approval-voting-versus-i...). Simply:
• Use the same ballots as we do today
• Let people vote for all the candidates they like
• Count all the votes
Done. No new ballots, no new procedures, no new software, and extremely simple to explain to voters.
Here's the intuition for why this works. When you vote using the current system, expressing support for one candidate requires you to withdraw your support from another candidate. RCV is the same in that you can only have one first choice: if you want to rank one candidate higher, you must rank another candidate lower. That's what keeps you from voting for your true favourite. With Approval Voting, your decisions about each individual candidate are fully independent: how you vote on one candidate has no effect on any other candidate. So you can express your full support for your favourite without harming the "lesser evil" candidate.
RCV is frequently criticized by mathematicians and economists, who are familiar with its many problems. This is one of those uncommon situations where the theorists and pragmatists agree: Approval is better from both perspectives.