[EM] No evidence that IRV doesn't fail. Reasons why it must.

[Paul Kislanko]

>Paul said:
>> Where I disagree a bit with James  is that I think it is too
>> much to ask the voters to quantify their subjective rankings
>> by coming up with their own weights.
>
>I never suggested we should ask the voters what their weightings were.
What I did say was that our
>knowledge that, for most voters, their first preferences were more
important than their subsequent
>preferences, could not justify the activities of those who were
constructing voting systems that
>depended on assigning specific values to successive preferences when they
had not asked the voters
>what those values might be.  There are two possible logical ways forward:
1. if you must assign
>values, ask the voters. OR  2. if that is not possible, or is unreasonable,
don't play with voting
>systems that assign arbitrary values based on some set of assumptions.

Then we're in agreement. May be the first time I've posted something that
wasn't obviously wrong...

I got the gist of "There are two possible logical ways forward:
>1. if you must assign
>values, ask the voters. OR  2. if that is not possible, or is unreasonable,
don't play with voting
>systems that assign arbitrary values based on some set of assumptions."

There are some who support item 1, and as a voter I do not like that one for
the reasons I spelled out in the previous post. It is less clear to me that
item 2 is something I'd agree with.

I would suggest that the vote-counting method be the same for all voters,
and it doesn't really matter that any part of the method is arbitrary, since
from the voter's perspective the entire method is arbitrary. Half of the
elections I get to participate in are Plurality-based, and the other half
are Plurality-based but if no one gets a majority there has to be a run-off
between the top 2. From the voter's perspective ALL rules are arbitrary.

Systems that count votes based upon "arbitrary" values might be perfectly
acceptable, if the "some set of assumptions" upon which they are based meet
reasonable and generally-acceptable criteria. Now, what THAT might be is a
subject of (considerable) debate, but the mere fact that the weights
assigned to my vote by the election method are arbitrary is not in itself a
problem. As a voter, all I need to know is what they are.