[EM] Easy voting & handcount. Approval vs RV.

Michael Ossipoff mikeo2106 at msn.com
Fri Feb 16 05:55:10 PST 2007



Before saying more, I want to say that, in spite of anything that I say 
below here, SSD is what I want for public elections.

If you've ever participated in an EM poll, the ones where we use all sorts 
of balloting, then you know that ranking is _lots_ easier than rating. And 
you know that, especially when there are lots of candidates, Approval is 
easier than ranking.

In fact, Approval's very easy voting and handcount, in EM polls, will give 
you a new appreciation for Approval.

In general, I much prefer Approval to RV (in its versions other than 
Approval). Approval has a simple elegance and beauty that RV doesn't even 
come close to. In Approval, a sincere vote is essentially the same as a 
strategic vote.

And, though Approval doesn’t give you the kind of expressiveness that 
Condorcet does, it gives you another kind of expressiveness that 
rank-balloting can’t give: Expressiveness about _how_ good or bad a 
candidate is. You express that when you vote for him/her or not. You can 
rank all the candidates in order of preference, but that doesn’t tell which 
ones are disgusting. With Approval you say something that you don’t say in a 
ranking.

Yes, I’d rather have SSD, but if it’s to be a handcount, or there are lots 
of candidates,  and you might not want to ask people to rank that many 
candidates, then the much easier Approval offers a very good substitute for 
Condorcet’s expressiveness, Approval’s own kind of expressiveness. Approval 
would be my 2nd choice then.

RV lets you express fine gradation of preference? Illusory. RV’s strategy 
problem stifles expressiveness more than it provides it.

Now, having said all that about Approval vs. RV, I have to also say that, 
due to the special situation in our elections, RV would probably give a 
better result than Approval in our public elections, in spite of Approval’s 
general superiority to RV, and in spite of Approval’s great aesthetic 
superiority to RV.

I’ve also talked about RV’s much greater familiarity and popularity, and 
people’s widespread misunderstanding about Approval and “1-person-1-vote”.  
So RV is probably a good public proposal, if SSD turned out to be 
un-gettable.

By the way, I highly recommend another EM poll. In fact I recommend that we 
do them regularly. You don’t know how you feel about the voting systems and 
balloting systems until you actually vote them and count them. Take my word 
for it. In EM polls I substitute Cloneproof SSD for SSD. CSSD is equivalent 
to BeatpathWinner, the easier, but less obviously-motivated and justified,  
implementation of CSSD.

When we do a poll, I suggest using rank-balloting, Approval balloting, and 
ratings. I suggest asking voters to vote sincerely, not strategically in the 
ratings, since the Approval ballot takes care of strategic voting.

When a definite answer isn’t desired, it’s enough to just get answers from 
the various methods used. When a definite answer is desired from a 
multi-mode poll, then I suggest Voter’s Choice, which we’ve used in past 
multi-mode polls.

Mike Ossipoff





More information about the Election-Methods mailing list