FairVote's nonsense blog against Score/Approval Voting

[Clay Shentrup]

> http://www.rangevoting.blogspot.com/

Has anyone ever written a thorough rebuttal to this pile of uninformed and deceptive propaganda?

[Clay Shentrup]

un·work·a·ble

adjective

not able to function or be carried out successfully; impractical.

Clearly Approval Voting isn't "unworkable". Maybe what FairVote was trying to say was something like "bad".

But, the dominant paradigm is Plurality Voting. Is FairVote saying that's workable? The insanity of this post boggles the mind.

[Toby Pereira]

On Wednesday, 10 February 2016 04:13:41 UTC, Clay Shentrup wrote:

> http://www.rangevoting.blogspot.com/

Has anyone ever written a thorough rebuttal to this pile of uninformed and deceptive propaganda?

I'm surprised you haven't. By the way, if you remove all instances of "later-no-harm", the word count would approximately halve.

[Clay Shentrup]

On Thursday, February 11, 2016 at 10:21:28 AM UTC-8, Toby Pereira wrote:

if you remove all instances of "later-no-harm", the word count would approximately halve.

Hah!

"Our voting system makes it safe to rank your 2nd candidate after your 1st candidate. Well, actually it doesn't. It makes it safe for your favorite candidate to have your 2nd favorite candidate ranked behind him. Although..it doesn't make it safe for you to rank your favorite candidate in 1st place. Oh and Approval Voting is bad because it'll cause everyone to vote for their favorite candidate. Which is not at all in contradiction with our huge grievance against Plurality Voting, that people will often not vote for their favorite candidate. Anyway, cardinal voting methods are unworkable."

[William Waugh]

On Tuesday, February 9, 2016 at 11:13:41 PM UTC-5, Clay Shentrup wrote:

> http://www.rangevoting.blogspot.com/

Has anyone ever written a thorough rebuttal to this pile of uninformed and deceptive propaganda?

A frustrating thing about their 'blog is that it doesn't seem to take comments. So, there's no way to be sure that their readers see a rebuttal if one of us publishes one.

Usually when I argue against IRV, the first thing I cite is its pathological response to small changes in the location of the center of public opinion as demonstrated under the heading of "Shattered" and under the headings farther down, in  http://zesty.ca/voting/sim/

Then I point out that IRV does not conform to Frohnmayer balance, which I think is key to preventing vote splitting. I suppose a variant could be made that would have balance, but it would have to allow a group of equal-graded candidates at the top as well as one at the bottom. Or require ranking all candidates. But in regard to requiring ranking all candidates, I see no justification for requiring a voter to rank candidates that the voter either genuinely grades as equivalent or chooses to for strategic purposes.

[William Waugh]

On Tuesday, February 9, 2016 at 11:13:41 PM UTC-5, Clay Shentrup wrote:

> http://www.rangevoting.blogspot.com/

Has anyone ever written a thorough rebuttal to this pile of uninformed and deceptive propaganda?

I'm wondering whether Fairvote is a deliberate plant by the 1% to maintain their ability to split votes. 

[Warren D Smith]

What's "Frohnmayer balance... the key to preventing vote splitting"?

the rangevoting.org criticisms page had tried to do some rebuttal-like stuff...
I don't know if you consider it adequate or on point for this...:

http://rangevoting.org/RVcrit.html


--
Warren D. Smith
http://RangeVoting.org <-- add your endorsement (by clicking
"endorse" as 1st step)

[William Waugh]

On Friday, February 12, 2016 at 12:28:22 AM UTC-5, Warren D. Smith (CRV cofounder, http://RangeVoting.org) wrote:

What's "Frohnmayer balance... the key to preventing vote splitting"?

It's a constraint, that is met by just those voting systems in which for any vote that can be cast, there is a balancing vote that will cancel the effect of the first vote on the electoral outcome.

For example, in Approval Voting, with candidates A, B, C, and D, where I vote for A and B and oppose C and D, you can cancel my vote by voting for C and D and opposing A and B.

"Electoral outcome" includes not only the determination of the winner, but also a measure of the relative popularity of the other candidates.

The voting system that allows a voter to choose plurality or antiplurality (and no other alternatives) meets the constraint, but still is a poor voting system due to lack of expressiveness.

William Waugh

[Clay Shentrup]

On Thursday, February 11, 2016 at 8:17:01 PM UTC-8, William Waugh wrote:

I'm wondering whether Fairvote is a deliberate plant by the 1% to maintain their ability to split votes.

I've wondered this myself, but I think it's a little too far fetched. It would just be too brilliant. Most conspiracists make dumb mistakes.

[Warren D Smith]

Aha. This "Frohnmayer Balance"
is exactly the condition needed to make Rivest-Smith
"VAV antifraud voting protocol" work.

[William Waugh]

There has to be a mistake somewhere, because in http://rangevoting.org/RivSmiTBadd.html you remark that "As Rivest & Smith's paper shows, VAV handles plurality elections securely." But vote-for-one Plurality clearly does not satisfy the balance constraint. For example, if Bush, Nader, and Gandhi are running, there is no antivote to a vote for Gandhi.  It would have to be a vote for Bush and Nader, and that isn't allowed under vote-for-one Plurality; it's disqualified as an "overvote".

[Warren D Smith]

On 2/12/16, William Waugh <2knuw...@snkmail.com> wrote:
> There has to be a mistake somewhere, because
> in http://rangevoting.org/RivSmiTBadd.html you remark that "As Rivest &
> Smith's paper shows, VAV handles plurality elections securely." But
> vote-for-one Plurality clearly does not satisfy the balance constraint. For
>
> example, if Bush, Nader, and Gandhi are running, there is no antivote to a
> vote for Gandhi. It would have to be a vote for Bush and Nader, and that
> isn't allowed under vote-for-one Plurality; it's disqualified as an
> "overvote".

--yes, you are correct; however, VAV when used for plurality
just added a special anti-plurality ballot to the mix, overcoming this obstacle.
But with, say, IRV, you could not do that, there is no such thing as
an IRV anti-ballot
and no reasonable way to make one, far as I can see.

Anyhow, Rivest-Smith no longer seems to be the greatest way to
fraudproof elections;
there now are rival schemes which seem more attractive.

[Eric Sanders]

And how did they get the bassist from Nirvana? He doesn't seem like a 1% conspiracist... unless Kurt was too! :-)

[Clay Shentrup]

On Tuesday, February 16, 2016 at 9:40:50 AM UTC-8, Eric Sanders wrote:

And how did they get the bassist from Nirvana? He doesn't seem like a 1% conspiracist... unless Kurt was too! :-)

Yeah, Krist is just uninformed on the details. I saw him speak in Seattle in 2006 and asked him about IRV's failure to elect a Condorcet winner. He confusedly asked, "Is this the..non-monotonic anomaly?" Complete lack of clarity.

[William Waugh]

On Friday, February 12, 2016 at 6:06:42 PM UTC-5, Warren D. Smith wrote https://groups.google.com/d/msg/electionscience/3ONjarOCEsg/EOm3sSLrFwAJ

...

Anyhow, Rivest-Smith no longer seems to be the greatest way to
fraudproof elections;

there now are rival schemes [that] seem more attractive.

What are they?

[William Waugh]

Is there a better term than "Frohnmayer balance" for Frohnmayer balance? I haven't seen where anyone brings it up earlier than he does.

Is it necessary to choose a voting system that meets the balance constraint, if we want to prevent vote splitting?

The vote-for-one system fails the balance constraint and specifically the way it fails it is by according more political power to voters who approve fewer candidates than to those who prefer more count of candidates. The votes for the partisans of the larger group of candidates get split among those candidates and so their power gets diluted. This results in two-party dominance (2PD).

If a system failed balance in the opposite way, by giving more power to approvers of more count of candidates and less to approvers of fewer, it would encourage every faction to bring in as many clone candidates as they could. I don't see that such a system would particularly help democracy.

I don't know offhand of a third way a system could fail the balance constraint, assuming its only input consists of the votes.

[Clay Shentrup]

I would say those are some of the weakest arguments you could make against IRV.