[EM] reply to venzke - range "random skewing" effect is not a problem

[Abd ul-Rahman Lomax]

At 03:00 AM 3/13/2007, Juho wrote:
>I guess you, as a Range expert, pretty well know what the anticipated
>problematic scenarios are. Problems may arise e.g. when opinion polls
>tell that Democrats would get only 49% of the votes (against 51% of
>the Republicans) and therefore their supporters decide to put some
>additional weight in their votes and vote strategically in Approval
>style. This would make the Democrats win.

Why? I really think this should be realized: I expect, at least 
initially, major party supporters to vote under Approval exactly the 
same as they currently vote under Plurality. Almost all will bullet vote.

In the election scenario, though, suppose it is Approval (the 
arguments are quite the same with Range). If the Dems have 49 percent 
of the votes, then we could surmise that they will all vote only for 
the Dem. And we would likewise surmise that the Repubs will vote only 
for the Repub. Why would any of them approve both the Dem and the Repub?

(Likewise, in an election with strong top-two candidates, say Dem and 
Repub, we can expect that most voters will vote 0 for one and 100 for 
another. It is third party supporters and independents who will vote 
intermediate votes. And not room was allowed for them in the 49-51 scenario.)

>It is possible that Republicans would counter by applying the same
>strategy and the situation gets balanced again.

Possible? *Why would one expect anything else*?

Party partisans will, presumably, bullet-vote. They may, some of 
them, add additional approvals or range ratings, but not where it 
could shift the election, and these extra votes are moot, they have 
no effect on the outcome. It is what happens with *other voters* that 
is interesting. Range and Approval allow third party voters to 
participate fully in the pairwise election between the top two, if 
they want to do so. If they really believe in "Tweedledum and 
Tweedledee," they aren't forced to do so.... Participating fully 
means voting Approval style. In Range as well as in Approval.

I'm claiming that these scenarios of Range allegedly producing bad 
results because of "strategic voting" are, essentially, nonsense. The 
example here doesn't even look at the actual behavior, it just 
asserts that it is a problem. And it is not.

Other examples depend on the idea that Approval style voting is 
somehow inferior or undesirable. It is an expression of strong 
preference. If you have a strong preference, express it! If you don't 
care that much, why, then, you can express a weak preference, but 
don't come crying if the election method discounts your weak 
preference. That is exactly what it is designed to do. Consider it, 
but not as much as a strong preference. This is the whole point of 
considering preference strength!

>  But as a result of
>this race on "whose voters are more strategic"

You can't get more strategic than Approval. If you care, vote that 
way. What is imagined here is that a political party has the voters 
in its pocket, so that they automatically will vote as the party 
tells them to vote. But this is not the real world of elections.

>  Range elections may
>become in essence Approval elections.

I don't know how many times this nonsense has been repeated. "Range 
becomes Approval." No, Range will *never* become Approval unless you 
can somehow get all the voters to not express intermediate ratings. 
Range *allows* voters the freedom to express intermediate rankings 
and, we could argue, it is strategically optimal for them to do so. 
The idea that it is not depends upon a contradiction in terms. That 
is, a weak preference that is strong enough that I'll exaggerate it. 
The limit of this insanity would be "I don't care if A or B wins the 
election, they are both equally good, but I prefer A to B so I'll 
bullet vote for A."

>The achieved results of
>Approval voting are not very bad in terms of achieved social utility.
>The worst scenarios are ones where some parties/groupings vote in
>Approval style while others do not. In these cases it seems obvious
>that the social utility would not be good.

It is not obvious at all.