[EM] Unranked ballot election challenge

[Tom Ruen]

Thanks Forest,

Reply inserted...

----- Original Message -----
From: "Forest Simmons" <fsimmons at pcc.edu>
To: <election-methods-list at eskimo.com>; "Tom Ruen" <tomruen at itascacg.com>
Sent: Wednesday, March 28, 2001 8:23 PM
Subject: Re: [EM] Unranked ballot election challenge


> Tom, in your example below you keep switching between 546 and 526 for the
> size of the A + C faction. (It doesn't make any difference in the winners
> by the two methods.)
>

Thanks for catching the 546/526 - the first should have been 526 and
otherwise the tallies seem correct.

> Here's my two cent's worth:
>
> I don't think that the same population of voters would have voted the same
> on their ballots under the two different methods.
>
> If the Approval voters had known that their ballots were to be counted
> Unranked IRV style, and also had fairly accurate information about the
> size of the various factions, then (most of) the B+C and A+C factions
> would have dumped C to have more influence on the outcome between the
> perceived frontrunners A and B.
>
> Many of these would have been dumping their favorite.

In this case polls would show a near 3-way first place tie. In this place
some voters will bullet vote for a favorite, others will double vote against
their least favorite. I actually think the A+B vote is fairly reasonable -
5% voting for both to guarantee one against C can win. The A+C and B+C
voters are strange ones if we imagine a simple political spectrum.

I agree that voters might have voted different knowing if they get full or
half votes, but in such a close 3 way election, I don't think there is any
clear strategy under either system. About 92% did bullet vote.

>
> Unranked IRV fails FBC.

*** FBC= "Favorite Betrayal Criterion - By voting another candidate over his
favorite, a voter should never get a result that he considers preferable to
every result he could get without doing so."

I'm not sure what "voting over one's favorite" means with unranked ballots.
Surely leaving a favorite unranked CAN NOT help one's favorite. Offering a
tie with another to help defeat a least liked choice seems an identical
compromise under Unranked IRV as Approval.

>
> Unranked IRV might work as well as Approval with zero information, but in
> the real world it would still suffer from the spoiler / lesser evil
> problem.

Approval and Unranked IRV both demand polling information to allow voters to
consider how far to compromise. I see both methods as suffering from
spoilers in the sense that voters are encouraged to vote ALSO for a stronger
second favorite. However I see no spoilers in the sense of voter being
encouraged to NOT voting for one's favorite at all.

Under what conditions in Unranked IRV would a voter benefit from bullet
voting for a second favorite? I'm sorry, but I can't see this.

Tom

>
> Forest
>
> -----------------------------
> Tom Ruen wrote ...

> Comparing Unranked-IRV with Approval, these methods have divergent winners
> in less than 2% of random elections with 3 candidates and many voters.
This
> is a very small variation, but worthy to be considered.
>
> Here's one (random) election I picked to demonstrate:
>
> Approval ballots:
> C=7886 (32.0%)
> B=7411 (30.1%)
> A=7298 (29.6%)
> A+B=1103 (4.5%)
> A+C=526 (2.1%)
> B+C=405 (1.6%)
>
> Looking at these ballots we can see a very close 3 way race among 3
> candidates. We can see C has the largest united coalition and A and B have
> the largest compromise coalition.
>
> If we measure by approval:
> A=7298+1103+526=8927  (36.2%) (First)
> B=7411+1103+405=8919  (36.2%) (Second)
> C=7886+405+526=8817    (35.8%) (Third)
> A wins by a hair over B.
>
> IRV gives a different result:
> Round 1:
> A=7886+(1103+546)/2=8112.5 (32.9%) (Third)
> B=7411+(1103+405)/2=8165.0 (33.2%) (Second)
> C=7886+(405+526)/2=8351.5 (33.9%) (First)
> Eliminate A
> Round 2:
> B=7411+1103+405/2=8716.5 (35.4%) (First)
> C=7886+405/2+526=8614.5 (35.0%) (Second)
> NOTA=7886 (29.6%) (Third)
> B wins
>
> Well, so here we have a case to consider. B has more core supporters than
A,
> and A has more compromise supporters from C. Splitting votes compared to
> full votes makes a difference.
>
> Which result more accurately represents voter preference?
>
> Tom Ruen
>