[EM] Legacy IRV limitations

[Richard, the VoteFair guy]

On 12/19/2023 8:08 AM, Toby Pereira wrote:
> Richard, quite a few of your "misrepresentations" are simply cases where 
> you disagree. I think Greg Dennis has already addressed a couple of them 
> but things like "Score voting would be a reasonable choice in elections" 
> and "Approval voting would be suitable for general elections" should not 
> be included in your list.

I don't claim that my list is authoritative.  It's just based on the 
evidence I've seen.  With that in mind, ...

Where is score voting used in a governmental election?

(Giving each candidate an upvote, or a downvote, or neither (neutral) 
doesn't count as score voting, even though it does qualify as a rating 
ballot rather than a ranking ballot.)

STAR voting does use score voting as the first step, but the second step 
of top-two runoff is needed to compensate for the many unfair results 
that would come from using score voting alone.  And so far the most 
significant adopters of STAR voting have arisen because of claims that 
it's better than "ranked choice voting" (RCV) without clarifying that 
those promotions regard FairVote's (legacy) version of RCV as the only 
way to count ranked choice ballots.

The fact that score voting isn't significantly used is, to me, a clear 
indication that it's not "a reasonable choice."  Especially because so 
many other better methods are gaining traction for use in governmental 
elections.

As for Approval voting, even the people who promote it acknowledge that 
it's primary advantage is ease of adoption regarding the ballot marking 
process and the counting process.  (They also claim it's easy for a 
voter to figure out how to choose which candidates to approve, but I 
don't see evidence that supports that claim.)  Now that ranked choice 
ballots are getting adopted relatively quickly (within the limits of 
governmental slowness), that simplicity advantage is disappearing.

Also I haven't heard fans of Approval voting claiming that the results 
are better than using ranked choice voting with a good counting method.

In particular, I've never heard anyone make a meaningful claim that 
Approval voting would be a good choice for U.S. presidential general 
elections (even if all 50 states suddenly switched to that method).

I admit that Approval voting is suitable for use in primary elections.

However it's even easier, less confusing, and just as fair, to nominate 
a second candidate (from the same party) when the candidate with the 
most votes fails to get majority support in a plurality/FPTP primary 
election.  That second nominee is the one who gets the second-most 
number of plurality votes.

The people who strongly support Approval voting seem to be promoting it 
as a stepping stone to score voting.  Yet, as indicated above, score 
voting doesn't have any significant adoption, just some people promoting 
its adoption.

Of course Approval voting, and possibly score voting, are better than 
plurality voting, but that's a very low standard for comparison.

Richard Fobes
The VoteFair guy