[EM] Justifying my MMPO suggestion

[Michael Ossipoff]

I called MMPO the best method for current-conditions electorates in this
country, if they want or need ranking.

That might call for a better answer to some of the objections to MMPO.

"Current conditions" refers to dishonest, intentionally disinformational,
agenda-driven mass media, and a public who believe them.

I'm only talking about voting here, & I make no claims about what voters
elsewhere need.

Referring to Kevin's MMPO bad-example, I asked who is wronged in that
example.

Chris gave the best answer. He said that, because the winner should be A or
B (& I agree that would make more sense),...

...then the A voters are wronged, because they have a right for A to win
with 50% probability.

For now, let's assume that's so.

The word "outrage" or "outrageous" was used. How outrageously are they
wronged?

Well, all of the A voters said that C is at least as good as the other
candidate whom they'd like to win with 50% probability.

That doesn't make it very outrageous if C wins.

Yes, C might be a write-in, and might be Donald Duck, or someone widely
disliked, like Hillary Clinton.

But, if write-in are allowed, then the A voters, by not voting a 2nd
choice, are saying that _any_ write-in would be as good as A.

Astonishingly irresponsible voting? Sure.

The method's fault? No.

Another thing:

Surely the badness of a bad-example depends on its plausibility. This
bad-example is highly implausible.

Yes, MMPO can fail to elect a CW.

But, if there is no indifference, a CWv (voted CW) will win.

If the CWv loses because of large-scale indifference, well that just
doesn't sound very complainable.

When MMPO avoids chicken dilemma, it does so by electing A. Some complain
that that violates the Plurality Criterion.

But A is the most favorite candidate who doesn't have majority pairwise
opposition.

C has about twice the pairwise opposition.

As I said, if Plurality says A shouldn't win, then Plurality is mistaken.

Yes, examples can be contrived in which MMPO does terribly by the
positional standard, of the pairwise defeats standard.

But how surprising is it if a method that doesn't use a standard does
pooorly by that standard in an example contrived to achieve that?

It's _because_ MMPO doesn't use those standards that it has its remarkable
combination of properties.

FBC, Weak CD, MAM-like strategy.

I suggest that a voting-system proposal should offer a list of methods.
...ordered by how much you recommend them.

The public &/or the initiative proposal committee could then make their own
choice.

Of course, if you include Plain MMPO in the list, then you must include
pages to tell & answer the objections that have been expressed against it.

That could greatly multiply the length of your proposal.

But, if feasible, I suggest that Plain MMPO be included because of its
properties.

Michael Ossipoff
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20160928/8f9b5f07/attachment.htm>