Molly Whitevery cool illustration of ranked choice voting and how it worked in Maine’s gubernatorial Democratic primary
via Reddit (https://www.reddit.com/r/Maine/comments/1ua4e7n/sankey_diagram_of_the_democratic_gubernatorial/)
acm@molly0xfff very cool! I'd love for Philadelphia to do this for mayoral primaries, as we have a tendency to have a winner with like 32% of the vote, and the outcomes might involve more buy-in this other way!
Bill Zaumen@acm_redfox @molly0xfff Don't use ranked-choice voting. To see why, read up on Arrow's impossibility theorem. We almost had a failure in California where there was a chance that all the Democrats would be eliminated in the election for Governor because there were two Republicans and a lot more Democrats in the race. Democrats far outnumber Republicans in California.
https://en.wikipedia.org/wiki/Arrow%27s_impossibility_theorem
Todd Knarr@bzdev @acm_redfox @molly0xfff Californiaca primaries don't use RCV. It's "top two go to a runoff". With RCV the outcome you fear can't happen because to win one of the Republicans would have to end up with an outright majority.
RCV does have a flaw but it's one that can affect the outcome only in exceptionally weird and rare circumstances
JTW@tknarr @bzdev @acm_redfox @molly0xfff the problem is that the flaw of instant runoff voting (which is the system being marketed as RCV) arises exactly when the "spoiler" candidate becomes viable. It happened in Alaska in 2022: Begich (center-right) was preferred over Peltola (center-left) by a majority of voters and over Palin (right wing) by a (different) majority. But because he had fewer first choice votes he was eliminated and Peltola was elected over Palin (who was the spoiler).
@tknarr @bzdev @acm_redfox @molly0xfff a candidate who beats every other candidate head to head is called the Condorcet winner, and failing to elect such a candidate is what has got instant runoff repealed in the past, setting back electoral reform. OTOH, FairVote, which is lobbying for IRV under the name RCV, is also working to replace single-member gerrymandered districts with multi-member districts which could be filled proportionally, which is the real solution.
@jtwcornell91 @bzdev @acm_redfox @molly0xfff I think you're falling prey to taking polls as indicative of voting, which they aren't. If he was the first choice of more voters than either Peltola or Palin, why did he get the fewest first-choice votes of the three?
@tknarr @jtwcornell91 @acm_redfox @molly0xfff In a separate toot, I provided an example of voting on a restaurant: in that example, if the vote was Italian versus Mexican, Italian would win. If it was Italian versus Chinese, Italian would win. With all three and an instant run off election, Italian was eliminated. If you come up with some fix, but still meet the conditions where Arrow's theorem applies, you will have a different case where it fails.
@bzdev @jtwcornell91 @acm_redfox @molly0xfff Thing is, neither of your two represent the actual election. Try this: have everyone rank the three choices 1st, 2nd and 3rd. Then, for each ballot, give their 1st choice 3 points, their 2nd 2 and their 3rd 1 point. If a choice isn't ranked, it gets 0 points. Winner is the choice with the most points.
@tknarr @jtwcornell91 @acm_redfox @molly0xfff Recently, we had a three-way race in a primary after our congress person retired. The system nearly failed - a tie between 2nd and 3rd place occurred, which was resolved in a recount. The tie involved 30,249 votes each for two of the candidates. During the recount, they found a 12-vote error in which those votes were tallied but not counted due to an operator error.
So don't tell me it works!
.https://www.latimes.com/california/story/2024-05-01/ca-16-results-recount-tk-tk
@bzdev @jtwcornell91 @acm_redfox @molly0xfff Sounds like it worked as designed. The recount would've happened even if it weren't a tie due to the slim margin. The recount found the error, it was corrected and the correct outcome resulted.
Checks like recounts are part of the system and can't be ignored.
@tknarr @jtwcornell91 @acm_redfox @molly0xfff LOL - it nearly *failed* as designed: the design did not consider the possibility of a tie and how to resolve that. Finding an error merely meant that they were lucky.
@bzdev @jtwcornell91 @acm_redfox @molly0xfff IIRC if you look deeper into the laws around the primaries there is a provision for handling ties (coin toss, as is typical for elections). And I notice you don't note that _all_ election systems have the same problem with ties, and require similar provisions for resolving them. So again you're complaining that systems aren't perfect but offering no alternatives that aren't worse.
@tknarr @jtwcornell91 @acm_redfox @molly0xfff Is Todd serious? I suggested an alternative that is (a) understandable, (b) easy to implement, and (c) is one for which there is no spoiler effect. The winning candidate apparently funded the recount, apparently because the rules for a tie meant that all three would be in the general election & the guy with the most votes preferred to have one opponent. That is not good for public confidence in the fairness of an election.
@bzdev @jtwcornell91 @acm_redfox @molly0xfff Things like multi-seat districts have their own problems which have been documented. The biggest one is that since the elections are for _parties_, not representatives, it further disconnects the representatives from the voters. This is the last thing we want because the biggest problems come when reps hew to the party line over their constituents' views.
@bzdev @jtwcornell91 @acm_redfox @molly0xfff You can't solve problems stemming from parties by messing around with election mechanics. You need to push parties out of the picture.
RCV's focus is on making the mechanics of the election produce a result closer to the overall support level expressed by the votes. It also reduces the power of parties by removing the cost of supporting candidates outside the party-approved ones.
@bzdev @jtwcornell91 @acm_redfox @molly0xfff There's also the small matter of those changes not being legal. You'd need Constitutional amendments changing how the legislatures are set up first, and even at the state level that's unlikely. At the federal level it's virtually impossible.
@tknarr @jtwcornell91 @acm_redfox @molly0xfff For Todd's edification, in California constitutional amendments are proposed quite often as part of the initiative process. That's why we have provisions such as "The people shall have the right to fish upon and from the public lands of the State and in the waters thereof,..." or "The right to marry is a fundamental right" (added to repeal one that said otherwise for a specific minority).
@bzdev @jtwcornell91 @acm_redfox @molly0xfff Getting an amendment on the ballot in CA is probably doable. Passing it, though, is going to be virtually impossible given the political dynamics of urban vs. rural areas of the state. You'd also have to successfully navigate all the legal challenges the rural areas would raise.
@tknarr @jtwcornell91 @acm_redfox @molly0xfff Todd, are you serious? Initiatives changing the California Constitution appear on the ballot all the time. Some pass, some don't. Here's an example:
PERMITS LEGISLATURE TO EXCLUDE NEWLY CONSTRUCTED RAIN-CAPTURE SYSTEMS FROM PROPERTY-TAX REASSESSMENT REQUIREMENT. LEGISLATIVE CONSTITUTIONAL AMENDMENT.
Permits Legislature to allow construction of rain-capture systems, completed on or after January 1, 2019, without requiring property-tax reassessment. Fiscal Impact: Probably minor reduction in annual property tax revenues to local governments.
@tknarr @jtwcornell91 @acm_redfox @molly0xfff Todd, I did read what you wrote, which was incorrect. I'm going to mute you.