Ok, your example demonstrates a case where Condorcet elects someone
who is not liked very much (but majority would change e.g. the
Democrat candidate to PW if they were allowed to do that).
I note that the name "Professional Wrestler" is not very natural name
in the example since people that are called "PW"s are generally
considered not to be good politicians. Therefore PWs would probably
be ranked more often at the last position (e.g. 10 D 2 R 0 PW). So,
the end result is not as bad as it seems. The PW must be a very good
PW. (Same with OMRLP.)
As you said, Condorcet and IRV don't see the ratings. I give another
example of IRV. C=Centrist, ER=ExtremeRight, EL=ExtremeLeft. (I use
the slightly exaggerating word "extremist" (although I just
complained about use of "PW" above) to make it clear that there is no
exact match to the main parties of current two-party systems.)
34% - 10 ER 8 C 0 EL
34% - 10 EL 8 C 0 ER
32% - 10 C 5 EL 0 ER
IRV first drops C and then elects EL. The centrist candidate would
have been a quite good end result, also from sincere ratings point of
view.
One more example on Range (that takes the ratings into account but is
vulnerable to some strategies). The example is as above but the
numbers are a bit different to make it clear who the "font runners" are.
Sincere opinions:
40% - 10 ER 8 C 0 EL
40% - 10 EL 8 C 0 ER
20% - 10 C 5 EL 0 ER
Actual ballots when voters exaggerate their votes and apply Approval
style strategy after observing that EL and ER are the "major
candidates" / "front runners" (and after assuming that the final
decision will be between these two):
40% - 10 ER 0 C 0 EL
40% - 10 EL 0 C 0 ER
20% - 10 C 10 EL 0 ER
Range picks EL as the winner. The sincere opinions would have given C
the best score but strategies changed the situation.
In summary, all methods have some problems. One needs to estimate
which problems are lesser and which worse. Use of ratings would maybe
be nice, but in competitive elections (where giving sincere ratings
means losing power) methods like Condorcet seem to perform better.
Condorcet allows the majority to rule and make decisions where the
sum of utilities is actually smaller that what electing the proposal
of the minority would have given. Out of the ranking based methods
Condorcet seems to me to perform better than IRV. (But if you need
IRV to make the case easy to explain, that is maybe not a
catastrophic move. :-)
Juho
On Apr 25, 2007, at 3:32 , Tim Hull wrote:
I know the Condorcet winner is preferred to every other candidate -
however, I'm in particular assuming ballots like this:
48% - 10 D 2 PW 0 R
47% - 10 R 2 PW 0 D
5% - 10 PW 5 D 0 R
(the numbers being the sincere range rating for the candidate)
Under Condorcet, PW would win despite the fact that he or she is
barely liked by anyone.
Under range and IRV, D would win. I know that Condorcet and IRV
don't use ratings, but you need to take into account
the fact that #2 is not always a strong #2 or is some eccentric
joke candidate. For instance, imagine a similar election in the UK
with Labour, the Conservatives, and the Official Monster Raving
Loony Party (assume no Liberal Democrat ran)... Would an OMRLP MP
really be a quality result? It may be entertaining, though...
On 4/24/07, Juho <[EMAIL PROTECTED]> wrote:
On Apr 25, 2007, at 0:40 , Tim Hull wrote:
The partyless method is seen as a plus - our current parties as
somewhat diverse in their composition, and people generally don't
like the "vote counts for candidate and party" when you can have
wildly diverging ideologies on the same ticket. It also
encourages party discipline and "voting in bloc" at the Assembly
level, something no one likes the idea of...
As far as Condorcet for single-winner, it's yet another complex
explanation and has the issue of failing "later-no-harm", which I
feel would cause massive amounts of strategic and bullet voting,
no matter how low the real risk of LNH failure.
That's called "uneducated and mistaken voters" ;-). The cases where
there would be some real reason to vote that way are extremely
rare. Note that also IRV is not free of strategic voting related
problems. I don't think Condorcet performs poorly here. (Negative
propaganda can be made though on any method.)
It also can elect centrists with very weak support along the
lines of my "pro wrestler" example (assuming that he'd get a 2 or
1 our of 10 in Range).
Note that a Condorcet winner, even if coming from a small party, is
a candidate that majority of voters would prefer in comparison to
any other candidate. I'd say that is strong support, although the
number of first place rankings in the ballots may not be as high as
with some other candidates.
Juho
Also, dominance by two major parties would be a significant
improvement over the status quo - as of now we have dominance by
*1* major party.
On 4/24/07, Juho <[EMAIL PROTECTED] > wrote:
On Apr 24, 2007, at 6:26 , Tim Hull wrote:
> In this case, the only *tested* method which is fully candidate
> based (i.e. no party lists, open or closed) - and does not use
> anything other than votes cast for candidates to determine winners
> - is STV.
(There are also other interesting methods like http://
en.wikipedia.org/wiki/Proportional_approval_voting and http://
en.wikipedia.org/wiki/CPO-STV. STV is however more established and
closer to real life, so I don't recommend any more complex or
experimental systems to be promoted in your case.)
(I have also written about MultiGroup that is a method that could,
despite of seeing candidates as members of various groupings, be
fully based on individual candidate decisions on what kind of
groupings/ideologies the want to promote and benefit of ( i.e. not
"party lists" but "candidate lists of groups he/she likes"). This one
is also experimental, so not for you.)
> In the case of voting, it seems like a good idea for the method
> of voting to be consistent for everyone. Hence, it only seems
> logical to use IRV. Doing anything else would only make the
> explanation of how voting works twice as long, and make said effort
> more likely to fail.
(You didn't say if you want the method to be consisted to the voters
or also to the ones who will decide what method will be taken into
use. If it is enough to provide a consistent voting experience to the
votes, any ranked ballot based method would do. But I guess you refer
also to the latter case.)
> Until these is a good, *proven* single-winner/multi-winner
> combination that works well, I don't see this type of situation
> changing.
(Does the "combination" mean combination of multi-seat and single-
seat "districts" (within a multi-winner election) or combination of
"government" and "chairman" elections? I guess the latter is the
case. Also other combinations would work technically, but maybe would
be more difficult to explain to the decision makers (= not work
well).)
> In my push to implement a better voting system than our truncated
> Borda/FPTP combo, I see IRV and STV as the best chance to actually
> make a change. I don't see myself trying to push two separate and
> complicated systems (one alone is hard enough), or trying to sell a
> system that has not been widely used anywhere.
Ok, you know best what is possible and what not. Note however that
with IRV you'll choose a direction where the major parties will be
favoured (centrist compromise candidates from smaller parties
probably won't be elected). Maybe that is ok in the environment in
question.
> In short - I would say that the lack of any good, tested multi-
> winner system with a better-than-IRV single-winner version is part
> of why IRV is so popular...
(I guess this you mean that this is the reason "why IRV is so
popular" to you in your current case (not in general).)
My summary of the STV-IRV combination is that
- IRV favours big parties (Condorcet would not, and also it would be
ranked ballot based)
- explaining STV and IRV to the decision makers at one go is a bonus
- you have decided to use a partyless method, which is ok, but I'm
still wondering if the existing major groupings will agree with this
- STV-IRV would surely be a significant improvement to your current
voting practices
Juho
___________________________________________________________
Copy addresses and emails from any email account to Yahoo! Mail -
quick, easy and free. http://uk.docs.yahoo.com/trueswitch2.html
----
election-methods mailing list - see http://electorama.com/em for
list info
----
election-methods mailing list - see http://electorama.com/em for
list info
----
election-methods mailing list - see http://electorama.com/em for list info
