[EM] inaccurate Fargo approval voting results

Sun Jun 9 11:24:44 PDT 2024

Maine House 2018 with Jared Golden? You can’t just pick and choose “high
profile” to mean “outlier I want to focus on.”

On Sun, Jun 9, 2024 at 2:23 PM Closed Limelike Curves <
closed.limelike.curves at gmail.com> wrote:

> Michael—that seems a bit uncalled for.
>
> I did forget about the NYC race, you're right. Like I said, IRV only has
> pathologies about 10-15% of the time, so out of the first 4
> competitive+high-profile IRV elections (NYC, the two Alaska House races,
> and the Alaska Senate race) only 2 have had gross pathologies (Alaska
> Senate was a decapitation pathology that avoided a center-squeeze).
>
> Although, with that said, I don't think the New York primary was a good
> showing for IRV either. Every method had Eric Adams as the winner
> (including FPP), but 17.5% of the ballots were exhausted by the end of the
> race, and NYC almost declared the wrong winner because of errors in
> vote-counting.
>
> On Sun, Jun 9, 2024 at 10:51 AM Michael Garman <
> michael.garman at rankthevote.us> wrote:
>
>>
>>
>> Oh, I’m sure the highly-funded, endorsed, and publicized candidates in
>> the NYC mayoral election not named Adams would love being called
>> insignificant! Brilliant analysis from the politics understander.
>> On Sun, Jun 9, 2024 at 1:49 PM Closed Limelike Curves <
>> closed.limelike.curves at gmail.com> wrote:
>>
>>> Michael—the issue is it fails much more than 1% of the time, it's just
>>> hidden from sight. We can see the same mistake in FPP. If it looks like 97%
>>> of FPP elections "got the right result", there's clearly *something*
>>> fishy going on in that number.
>>>
>>> And there is! The issue is that those 97% and 99% figures are based on
>>> the incorrect assumption that which candidates choose to run isn't affected
>>> by the electoral system. If everyone who wanted to run for office *did*
>>> run, pathologies would be much more common. We saw this in Alaska, where
>>> the top-4 format made the pathologies that usually get swept under the rug
>>> by the nomination stage visible. In the *very first* major election
>>> with multiple significant candidates, we saw a simultaneous
>>> monotonicity/participation/majority failure. This lines up much more
>>> closely with what models predicted: in competitive 3-candidate elections,
>>> all of these pathologies should be quite common, happening around 5-15% of
>>> the time.
>>>
>>> The reason IRV looks like it "basically works" is because its
>>> pathologies mean moderate and third-party candidates know they have no hope
>>> of winning, so they never run in the first place. At that point, in a
>>> 2-party system, all voting systems will return the same results (because
>>> it's just a simple majority vote).
>>>
>>> I think this bears repeating: *a low rate of **empirical** pathologies
>>> is often a negative, not positive, indicator*. If your dataset has no
>>> examples of center-squeeze, that means your system is so bad at electing
>>> Condorcet winners that moderate candidates are refusing to run in the first
>>> place. Similarly, we'll know Condorcet methods are working if (sincere)
>>> Condorcet cycles start popping up all the time. That's how we'll know we've
>>> successfully depolarized our politics and broken free of the old
>>> one-dimensional political spectrum (where the median voter theorem protects
>>> us from cycles).
>>>
>>> On Sat, Jun 8, 2024 at 2:44 PM Michael Garman <
>>> michael.garman at rankthevote.us> wrote:
>>>
>>>> I'm happy to make tweaks! I don't have all the answers, Bob. I've never
>>>> pretended to.
>>>>
>>>> My issue is with the people -- many of whom are on this list -- who
>>>> view something that fails less than 1% of the time as insufficiently
>>>> preferable to the status quo and would encourage voters to reject it in
>>>> favor of preserving FPP.
>>>>
>>>> Advocate for the best currently viable option != accept it as the best
>>>> we'll ever have.
>>>>
>>>> On Sat, Jun 8, 2024 at 5:40 PM robert bristow-johnson <
>>>> rbj at audioimagination.com> wrote:
>>>>
>>>>>
>>>>>
>>>>> > On 06/08/2024 4:20 PM EDT Michael Garman <
>>>>> michael.garman at rankthevote.us> wrote:
>>>>> >
>>>>> >
>>>>> > Oh no…it only works 99.2% of the time! The horror!
>>>>> >
>>>>>
>>>>> Could work 99.6% of the time.  Or we could revert to FPTP and it works
>>>>> 97% of the time (i.e. in less than 3% of RCV elections, a difference occurs
>>>>> in who is elected).  The failure rate of FPTP is not all that bad, yet
>>>>> there is your organization, FairVote, RCVRC, dozens of state/local
>>>>> organizations, with scores of employees committed to correcting that 3%.
>>>>>
>>>>> But not interested in correcting their own flaws.  So much so that
>>>>> they lie about the performance of their own product and they lie about the
>>>>> objective failures when they do occur.
>>>>>
>>>>> A bridge or some other construction that failed, unnecessarily, 0.4%
>>>>> of the time that it is used would not be considered acceptable.  But
>>>>> elections are less important than bridges.
>>>>>
>>>>> You see, Michael, when the stakes are high, even low failure rates
>>>>> become important.  The Electoral College hasn't failed all that often, but
>>>>> because in recent times it has failed twice, 16 years apart, that's
>>>>> bringing attention to the systemic failure (because this failure is
>>>>> favoring a particular side).
>>>>>
>>>>> As RCV is used more and more, these other failures will happen more
>>>>> often.  Each time this failure occurs, especially when the failure is
>>>>> unnecessary and avoidable, bad shit happens.  Like repeal efforts that get
>>>>> on the ballot.
>>>>>
>>>>> So the question is, do RCV advocates learn from these glitches?  Or
>>>>> continue to deny the failures or minimize the significance of the failure?
>>>>>
>>>>> You are clearly in the latter group.  You haven't learned yet that
>>>>> course corrections early in the voyage are less costly than later in the
>>>>> voyage.  We're quite early in the voyage of RCV reform, there is solid
>>>>> evidence of not heading in the intended direction, yet you continue to deny
>>>>> the indication of a course correction and insist that we dig in deeper and
>>>>> commit to exactly the same heading that has already demonstrated the need
>>>>> for adjustment.
>>>>>
>>>>> The need for adjustment is multi-faceted (another thing you ignore).
>>>>> Especially with the nut-cases challenging the legitimacy of election
>>>>> results *along with* T**** and sycophants actually trying to simply fudge
>>>>> the election results, any reform that causes a reversion or roll-back of
>>>>> process transparency *hurts* the election reform movement.  Right now, with
>>>>> FPTP, we have redundancy in tabulation and we can tell who wins *directly*
>>>>> from the tallies posted at the source (the polling places) on the evening
>>>>> of the election.  This is what protects us from some corrupt official
>>>>> simply fudging the numbers and "finding, uh, 11780, uh, votes."
>>>>>
>>>>> Secretary of State Raffenberger was not (and is not) corrupt, but
>>>>> Jeffery Clark (a different high official) obviously is.  We need process
>>>>> transparency to prevent a corrupt election official from fudging the
>>>>> numbers from the tallies.  We need decentralization and process
>>>>> transparency to prevent the QAnon assholes from making any credible claims
>>>>> that accurate tallies were fudged.  We lose all that with Hare RCV.  And
>>>>> like the 0.4% of the Hare RCV that failed, this loss is unnecessary.
>>>>>
>>>>> But you're still in denial.
>>>>>
>>>>> --
>>>>>
>>>>> r b-j . _ . _ . _ . _ rbj at audioimagination.com
>>>>>
>>>>> "Imagination is more important than knowledge."
>>>>>
>>>>> .
>>>>> .
>>>>> .
>>>>> ----
>>>>> Election-Methods mailing list - see https://electorama.com/em for
>>>>> list info
>>>>>
>>>> ----
>>>> Election-Methods mailing list - see https://electorama.com/em for list
>>>> info
>>>>
>>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20240609/0afc7e5e/attachment-0001.htm>