[EM] IRV vs. plurality

[James Green-Armytage]

Dear Eric, and other voting methods fans,
>
You write:
>
>By every measure I consider important, I see no difference between 
>the two
 [plurality and IRV]
>. They should both be absolutely avoided.

I reply:
	What are the measures that you consider important? Clearly monotonicity
is one, but what are the others?
	Even if third parties are unlikely to win, I do not consider it
completely irrelevant that people are enabled to vote for them without
investing all their voting power in doing so. This should improve the
political standing of third parties, allow them access to debates,
matching grants, allows them to gain a foothold, etc. IRV has better
Condorcet efficiency than plurality... I listed a lot more benefits to IRV
over plurality. Are you going to tell me that you read them all and found
every one of them to be completely irrelevant and worthless?
	Spoiling can still occur in IRV, but it is much, much easier for it to
happen in plurality.
	I am not arguing that IRV is radically better than plurality. I am
arguing that it is at least marginally better, all other factors aside. As
you don't seem to be fond of plurality, I don't see why you are unwilling
to concede that point.

You write:
>As for IRV, this case clearly shows a tremendous problem with the system.

I reply:
	I am well aware of the flaw in IRV that you have presented here, and I
agree that it is quite a serious flaw. I think that my posting makes it
clear that I am aware of such problems.

I wrote:	
>>I am fairly sure that IRV, Condorcet,
>>and Approval people should try to treat each other as allies rather than
>>competitors, and that they should not try to strawman or exclude each
>>others' systems.

You replied:
>
>While there are at least be some questions regarding Approval vs. 
>certain versions of Condorcet, I don't there is any argument left in 
>the support of IRV over these two obviously superior methods.

I reply:
	Is this a non-sequitur? I can't entirely see how your paragraph is a
response to the one that you quoted.

	Anyway, I was not making that argument. I was arguing that IRV is
superior to plurality.
	However, I am admittedly not totally convinced that Approval is superior
to IRV. (It may well be, but nobody has convinced me of it yet.)
	Also, I think that a version of IRV that allows for equal rankings would
be better than plain IRV, and would incorporate many of the virtues of
Approval, with the added freedom of voters to differentiate between levels
of support. Aside from monotonicity, I would like to know what other
advantages Approval has over 'equal rankings' IRV. 

	I will say that I find it rather hard to imagine a normative argument
that Approval is superior to ranked pairs or beatpath Condorcet. I can
certainly accept the idea that Approval has a strong practical advantage
in being easily adoptable (without new equipment, with a relatively easy
voter education effort), but it seems clear that Condorcet allows for much
more thorough communication, and requires much less strategic guesswork on
the part of the voters.

You write:
>(However, I definitely prefer RP Condorcet (Deterministic #1-wv)...or 
>whatever Mike O. was calling it...don't have it in front of me at the 
>moment....over all other single-winner methods I am aware of.)

I reply:
	I don't necessarily disagree, but I am curious: what advantages do you
think ranked pairs has over beatpath / CSSD?