[EM] Poll on voting-systems, to inform voters in upcoming enactment-elections

Sat Apr 6 12:26:49 PDT 2024

On Sat, Apr 6, 2024 at 06:43 Chris Benham <cbenhamau at yahoo.com.au> wrote:

> Michael,
>
> Why do you think that  STAR is better than IRV (aka Hare) ?
>
As you know, I’ve been a lot more favorable to Hare than most. But:

1. I’ve realized that the choices, sometimes strategic, that Approval puts
on the voter are a lot better than the count-complexity &
computation-intensiveness—& consequent count-insecurity & count-fraud
vulnerability of the ranked methods such as Condorcet & Hare.

That count-fraud problem is the worst, but, overall, Approval is much
easier to introduce, describe, explain, propose—& much easier & inexpensive
to implement (can be implemented at zero cost, without even
software-modification), administer & audit against error &
count-fraud…compared to any other multicandidate-suitable method.
 …especially the ranked-methods.

I’d much rather trust voters to use Approval well than trust that the
count/m-staff won’t be bribed to do fraud.

Approval is the absolutely minimal method that allows & counts
multicandidate comparison…& therefore the unique completely unarbitrary
method.

Score & STAR? They have a strategy-problem that Approval doesn’t have, when
some people vote sincerely (merit-proportionate) & others vote
strategically (all-or-nothing).

But I nonetheless would rather have that than the problems of the
rank-methods.

Likewise, Score & STAR are less simple than the absolutely minimal
Approval, & so I much prefer Approval.  …but Score & STAR don’t even come
close to the computational complexity of Condorcet & Hare.

I prefer Score(0-5) to STAR because of the strategy-complications &
problems of the runoff, & the resulting FBC failure that you pointed out.
But, for public political elections, STAR still beats Condorcet & Hare, due
to their count-complexity.

2. Hare, while sharing the count-complexity of Condorcet (& additionally
requiring either one big central count, or lots of back-&-forth information
transfer between precincts & Central) doesn’t share Condorcet’s
strategy-freeness…requiring a favorite-burying top-end defensive-strategy
to protect acceptables from unacceptables.  …with the risk that timid
lesser-evil voters will do the same to protect their lesser evil.

Sure, Hare elects the favorite of the largest faction of the
mutual-majority when there is one,  but the support isn’t always mutual.

You can hope that, with Compromise as your 2nd-choice, if Favorite
eliminates Compromise, Compromise’s voters will transfer to Favorite
because they’re both progressives.  But, for one thing, if Compromise’s
voters transfer the other way, an unacceptable could be elected. That’s too
big a chance to take when there are completely unacceptable candidates.

For another thing, if they transfer to a perceived progressive, it might
not be Favorite, & might not even be very (or at-all) progressive.

I used to think that Hare was alright for polls, but Hare can have its
problem & do a funny trick, do a Silly, even in a poll. It demonstrated
that in a Free-&-Equal poll.

But I like Hare for choosing a pizza 🍕 topping or a 🍿 etc.

Easy handcount too.

I prefer Condorcet, especially RP(wv), for polls, & to propose to
jurisdictions where people insist on rankings. But not for public political
elections, for the abovestated reasons.

>
> And surely anyone here on this list can nominate any method they choose
> (and have it accepted/acknowledged) regardless of whether or not the
> method's supporters want it nominated.
>
>
>
> Chris Benham
>
>
> On 6/04/2024 10:46 pm, Michael Ossipoff wrote:
>
>
> This is to acknowledge the nominations of Smith//Default-Approval,
> Smith//Explicit-Approval, Margins-Sorted Approval, & Smith//DAC.
>
> I’d say include STAR, because that’s what its advocates would want.  …or
> would they? Its enactment is going to be voted-on in Eugene next month, &
> what if it finishes low here? That would be worse for the Eugene initiative
> than not including it.
>
> Of course showing voters about methods’ popularity here is my
> stated-purpose for the poll, & the fact that it’s about to be voted on for
> enactment would seem to suggest including it.
>
> But the advocates of STAR have been working hard, completely in good
> faith, & STAR is a lot better than IRV. Those are two good reasons to let
> EqualVote decide on STAR’s inclusion in the poll.
>
> I’ll ask the EqualVote group, & go by what they say.
>
> (In fact STAR, while more complicated than Approval, has nothing like the
> amount of count-complexity of Condorcet, or the consequent amount of
> count-insecurity & count-fraud vulnerability. I personally don’t propose
> STAR, because I regard it as an inbetween compromise between Approval & the
> ranked-methods, & I want the absolutely minimal. (I only propose Condorcet
> to jurisdictions where people insist on rankings.) …but, by my
> simplicity-standard, STAR scores high, even though I don’t propose it.)
>
> So the nominations list so-far is now (listed in order of nomination):
>
> Approval
> RP(wv)
> Schulze
> IRV
> Plurality
> MinMax(wv)
> Black
> Baldwin
> Benham
> Woodall
> Schwartz-Woodall
> Smith//Approval (of all ranked)
> Smith//Approval (of what is specified)
> Margin-Sorted Approval
> Smith//DAC
>
>
>
> On Sat, Apr 6, 2024 at 04:03 Chris Benham <cbenhamau at yahoo.com.au> wrote:
>
>>
>> I would like to nominate several methods.
>>
>> Smith//Approval (Ranking):
>>
>> Voters rank from the top only those candidates they "approve",
>> equal-ranking allowed,
>> the most approved member of the voted Smith set wins.
>>
>> Smith//Approval (specified cutoff):
>>
>> Voters rank from the top however many candidates they wish and can also
>> specify an approval
>> cutoff/threshold. Default approval is only for candidates ranked below no
>> others (i.e. ranked top
>> or equal-top).
>> The most approved member of the Smith set wins.
>>
>> Margins Sorted Approval (specified cutoff):
>>
>> Voters rank from the top however many candidates they wish and can also
>> specify an approval
>> cutoff/threshold. Default approval is only for candidates ranked below no
>> others (i.e. ranked top
>> or equal-top).
>>
>> A Forrest Simmons invention. Candidates are listed in approval score
>> order and if any adjacent pairs
>> are pairwise out of order then this is corrected by flipping the
>> out-of-order pair with the smallest
>> margin. If there is a tie for this we flip the less approved pair. Repeat
>> until there are no adjacent pairs
>> of candidates that are pairwise out of order, then elect the
>> highest-ordered candidate.
>>
>> Smith//:DAC
>>
>> Voters rank from the top however many candidates they wish, equal-ranking
>> allowed.
>> Eliminate candidates not in the Smith set and then apply
>> Woodall's Descending Acquiescing Coalitions method.
>>
>> There is a method I hate that is apparently contending in the real world:
>> "STAR". Given the stated purpose of
>> this poll, is there a case for including it?
>>
>> Chris Benham
>>
>>
>>
>> *Michael Ossipoff* email9648742 at gmail.com
>> <election-methods%40lists.electorama.com?Subject=Re%3A%20%5BEM%5D%20Poll%20on%20voting-systems%2C%0A%20to%20inform%20voters%20in%20upcoming%20enactment-elections&In-Reply-To=%3CCAOKDY5BkSGJkX%3D7zWXBr2t1SBNVMNj96wm-T8ubvr_wGM5h51w%40mail.gmail.com%3E>
>> *Wed Apr 3 22:13:28 PDT 2024*
>>
>>
>> ------------------------------
>>
>> EM used to do a lot of polls, but now never does. So I wouldn’t propose
>> one, if it weren’t for the fact that, this year, the voters of at least two
>> states are going to vote on whether to enact a certain voting-system.
>>
>> It seems to me—tell me if I’m wrong—that those people have a right to know
>> how people familiar with voting-systems feel about the relative merits of
>> some voting-systems.
>>
>> So, though I claim that polls are valuable for demonstrating the experience
>> of using the voting systems, & how they work, & what they’ll do—& are
>> therefore useful & worthwhile for their own sake—this poll that I now
>> propose isn’t a poll for its own sake.
>>
>> It is, as I said, proposed for the important practical purpose of letting
>> the voters in the upcoming enactment-elections know how we feel about the
>> relative merits of some voting-systems, including the one that they’re
>> about to vote on the enactment of.
>>
>> The voting-method for the poll:
>>
>> It seems to me that Schulze is the most popular ranked voting-system, among
>> the people at EM.
>>
>> …& it seems to me that the last time we voted on EM’s collective favorite
>> voting-system, Approval won.
>>
>> Those seem the top-two, in EM popularity.
>>
>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20240406/d5bf9cd1/attachment.htm>