[EM] Score Chain Climbing and TACC

[Kevin Venzke]

Hi Forest,

Le mardi 25 janvier 2022, 01:30:04 UTC−6, Forest Simmons <forest.simmons21 at gmail.com> a écrit : 
> Score Chain Climbing generally disappoints both burial and Burr dilemma defectors.
> 
> That's why it is becoming my favorite method.
> 
> SCC
> 
> While more than one candidate remains eliminate the highest score candidate that
> does not pairwise defeat the lowest score remaining candidate.
> 
> The Burr defector, like the burial culprit is typically a fairly strong
> candidate that sees a chance to bury or truncate an opponent that he does not
> defeat pairwise, but might well come out ahead of if the opponent's score is
> lowered.
> 
> SCC is practically tailor made to disappoint this kind of manipulation ... the
> lowered score candidate still defeats her detractor pairwise, and her lowered
> score makes her the pairwise eliminator at some early stage ... the lower her
> score, the earlier her chance for revenge!

I'm not sure I'm following your logic. In a three-candidate cycle, the chain
climbing methods (like TACC) elect whichever candidate defeats the weakest
candidate, according to the strength metric being used (implicit approval or
Score etc.). Right?

At first glance this doesn't sound promising from a burial perspective because
if we have a non-CW candidate (the "rival") with strategizers, we expect them to
defeat the "pawn" pairwise and we would often expect that pawn to be the weakest
of the three candidates. That suggests that the rival will succeed in their
strategy most of the time (unless they accidentally turn the pawn into the CW).

Yet experimentally I have found TACC (with implicit approval) to have relatively
low burial incentive. Better than C//A(implicit) and C//FPP, but worse than
fpA-fpC or C//IRV or BPW.

How can this be? I believe it's because with TACC, the act of giving the pawn
new, falsified preferences is quite likely to cause the pawn to *no longer* be
the weakest candidate by the metric of implicit approval. That is, the failure
situation for TACC doesn't occur as predictably or as often as we might be
inclined to guess.

Indeed running simulations on the TACC algorithm but using first preferences
instead of implicit approval, I find some of the highest burial incentive
anywhere.

This makes me believe that SCC is not going to be as good as TACC, since it
does less to ensure that the weakest candidate (by Score) won't be the pawn in
the strategy.

Kevin