[EM] Part 3 of Approval vs Condorcet

[Michael Garman]

>> If Approval had been the proposal over that period, it would be in use
in all 50 states by now.

I find it incredibly hard to believe that there wouldn’t be a partisan
backlash the first time approval cost one or the other — or both — the
election. The major parties famously love giving third parties a fair shot.
Not denying that it’s a fine system that ought to be used in more places
than it currently is, but this claim is absurd.

On Fri, Mar 22, 2024 at 8:07 AM Michael Ossipoff <email9648742 at gmail.com>
wrote:

> Part 3:
>
> …
>
> Part 2 didn’t send. I wrote it on my phone, & it failed to send. Hopefully
> a copy will be returned to me, so that I can try again to send it. But I’m
> writing this Part 3 on the desk-computer, so that I can save it in Word, so
> that, whatever happens with unsuccessful sending, I’ll have it to resend.
>
> …
>
> Approval-Advantages continued:
>
> …
>
> Let’s talk about enactment:
>
> …
>
> The incomparably easier implementation makes Approval much more enactable.
> Let me say more about easy implantation: Approval can be implemented with
> no new balloting-equipment, & also without even any software-modifications.
> Here’s a way:
>
> …
>
> The existing Plurality-count software is designed for collecting  a
> single vote in a race. So tell the count-program that there are many 2-way
> races (one for each candidate). The ballot has a line for each candidate’s
> name & voting-bubble. But, alternating with those lines, are blank lines,
> where the count-program expects votes for the nonexistent other candidate
> in the fictitious 2-way race.
>
> …
>
> So, when the counting is done, the count-program will report the results
> for all those fictitious 2-way races, including 0 for each of the
> vote-totals for the nonexistent opponents, & also the vote-totals for each
> actual candidate.
>
> …
>
> In that way, the existing Plurality count-software will give the totals
> for each actual candidate, summed over all the ballots.
>
> …
>
> So that’s why I said that Approval can be implemented at zero cost. No
> software modification needed.
>
> …
>
> Now, what does FairVote (under various names) have to show for its 35
> years of expensively promoting IRV (under various names)?  Two states &
> some cities.
>
> …
>
> Two states.
>
> …
>
> Sorry, but I don’t call that success. Sure, if the people of Oregon &
> Nevada have been well-enough deceived, there could be two more states this
> year. But 4 out of 50 sounds more like failure for a 35-year effort.
>
> …
>
> If Approval had been the proposal over that period, it would be in use in
> all 50 states by now.
>
> …
>
> Critreria:
>
> …
>
> Just as an added bonus:
>
> …
>
> Approval passes several criteria that Condorcet fails:
>
> …
>
> Participation, Consistency, & IIAC (all without loss of Pareto).
>
> …
>
> I don’t include FBC, because Condorcet’s FBC-failure that we discussed at
> EM a long time ago is so rare as to be strategically-irrelevant.
>
> …
>
> If burial is a problem, then FBC-failure comes back when we need
> favorite-burying drastic defenses to try to protect CW from burial. But the
> wv Condorcet-versions are so burial-deterrent, in 2 separate ways
> (Minimal-Defense & autodeterrence*), that burial can be ignored when voting.
>
> …
>
> *Even without any defensive-truncation, burial’s backfire is 10 times more
> likely than its success.
>
> …
>
> [Conclusion of this short Part 3]
>
> …
>
> If Part 2 hasn’t been sent yet, & doesn’t automatically get sent,
> hopefully a copy will be returned to me, & I’ll then send it along.
>
> …
>
> I have to say that, from now on, anything long will be written in word on
> the desktop computer instead of on the phone, for safekeeping of long
> messages in case they fail to send & are lost.
> ----
> 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/20240322/92335b3c/attachment.htm>