WS--
Let me word my answer better, now that I've had a more thorough look at the
example.
Your objection was that if the Blues order-reverse against Red, and if the
Reds do defensive truncation, that will elect Hitler. In my most recent
posting, I correctly pointed out that Hitler doesn't
WS, in his paper,mentioned a strategy that Rob LeGrand calls Strategy A, and
which I call BF-Certain.
BF stands for Best Frontrunner.
I'll get to this more, soon, but, for voting in an Approval election, that
strategy is only any good if it's fairly certain which of the two expected
Juho--
Warren Schudy's statement that, in Range, it's never optimal to give a
rating other than top or bottom is correct in public elections. Other than
the stipulation of public elections (with so many voters that your own
ballot won't significantly affect the pair-tie probabilities),
[Note to list members: Lomax isnt noted for his brevity, being one of the
few people who post longer messages than I do. So you could say that this
reply is a lot to be sending to your inbox. Im aware of that, and so Im
assuring you that I wont do it often. Not daily or weekly. If there
WS's example shows the quirkiness of even the best rank methods. They'll do
unintended things that we'd rather they didn't do. Approval has a stability
and behavioral simplicity that the rank methods don't have.
But the best rank methods, such as SSD, MDDA and MAMPO, offer strategic
ws wrote:
We show that approval voting strategic equilibria are closely related to
honest
Condorcet Winners. There exists an approval equilibrium with a clear
font-runner F and runner-up R if and only if the F is the clear Condorcet
Winner and R the Condorcet runner-up. In contrast, we show
WS:
You said that Hitler wins in your example of order-reversal and truncation.
Would that it were so. Actually Blue wins instead of Hitler :=(
Blue is the only candidate with no pairwise defeats. Condorcet doesn't look
at pairwise opposition unless it's in a defeat.
It would be better for
Forest--
As I was saying before, if everyone honestly and very accurately indicates
the most likely winner, then the method is a lot like DSV, without the
incentilve to misrepresent one's preference ordering.
As you said, the strategy incentive is in the indication of the most likely
winner.
Forest--
The use of those checkmarks to inform automatic strategy is a shortcut to
DSV's way of determining each voter's best strategy.
It was you who first suggested that it's in a voter's best interest to
approve every candidate who is better than that voter's expectation in the
Forest--
The use of those checkmarks to inform automatic strategy is a shortcut to
DSV's way of determining each voter's best strategy.
It was you who first suggested that it's in a voter's best interest to
approve every candidate who is better than that voter's expectation in the
I said that I suspect that no nonprobabilistic method can be more
strategy-free than wv Condorcet and its close relatives that trade one of
its criterion-compliances for another. I don' have a proof. It's only a
suspicion. It was suggested by my brief look at DSV and Nash Equilibrium
With the better rank methods, ordinary truncation is powerful enough to
thwart and penalize offensive order-reversal. What more is needed for
truncation?
I dropped my proposal of power-truncation for that reason, and because it
would make strategic or no strategic truncation into a big
Between two evils, I always pick the one I never tried before. - Mae West
If only voters were willing to try something different from what they've
been unsuccessfully trying for decades.
Mike Ossipoff
election-methods mailing list - see http://electorama.com/em for list info
Id said:
What an elaborate counterfactual story. Its amazing what lengths to which
some people will go, to make Plurality fail Condorcets Criterion without
mentioning preference.
I've already answered about that. It's based on a privileged balloting
system. My criteria make no
Chris wrote:
If the balloting rules don't allow the voters to fully express their
intended ranking, then we assume that the voters
vote to express as much of it as the balloting rules allow, giving
priority to expressing as many of their intended
strict pairwise preferences as possible
I
A votes-only FARCS approach would completely obscure the purpose and benefit
of the majority defensive strategy criteria, thereby defeating their
purpose.
A criterion would be of little use if its wording didnt express its
important guarantees.
That was the purpose of this posting, but I
Basically its this:
Everything mentioned in my criteria is an obvious element of the situation.
FARCSs intended rankings (which, even in principle, couldnt be intended)
have no real-world interpretation.
The use of those fictitious rankings is very cumbersome and awkward,
roundabout
Chris had said:
I'd be interested in seeing an example of MD failure that you agree (or
are
content) with.
I replied:
:
Id agree and be content with an example that doesnt violate SFC or
SDSC.
Kevin then said:
Here is such an example. Suppose these are the sincere
Maybe there should be criteria for evaluating criteria. For instance, FARCS
doesnt pass the laugh test.
I consulted my JoAnn Q. Citizen consultant.
I said, Im going to tell you two criterion definitions. Theyre supposed
to be very similar, but they dont sound at all alike. Tell me
Chris said:
In a way Approval is worse. In my example, the five AB compromisers
might correctly believe that A has at least as good a chance of winning as B
and that C has the least chance to win. They don't need to be
convinced that their favourite isn't viable, just (given their abhorrence
Id said:
Ive never denied that rank methods can add advantages not available in
Approval. Ive even said that I myself would prefer a good rank method for
our public elections, though I myself, as a voter, would be content with
Approval. It would be a nice luxury to rank the best
First of all, if your criteria system doesnt apply to Approval and Majority
Favorite, how can you call it acceptable?
My criteria, as Ive said, apply seamlessly and universally to all methods
(certainly to all proposable methods).
You showed how CR can fail a votes-only FARCS SFC.
Chris said that people arrive at the polls intending to vote a certain way
in a rank method, and then find out that it's (say) Plurality or Approval.
Say it's Plurality. Their ranking that they arrive with would reasonably
have their favorite in 1st place (Yes, I know it's a no-no to speak of
I think Douglas Woodall just considers that in his preferential
election rule (PER) universe all ballots are ranked
ballots with truncation allowed and above-bottom equal ranking not,
Maybe Douglas Woodal lives in a universe of his own. Criteria are more
useful if they're about the universe in
I'd said:
Obviously majority rule is violated by an outcome that is contrary to what
a majority have voted that they want. For instance, if a majority vote B
over A, then we can assume that, if A or B wins, they vote that it be B.
Chris says:
That is reasonable, and granted for the sake of
I'd said:
Obviously majority rule is violated by an outcome that is contrary to what
a majority have voted that they want. For instance, if a majority vote B
over A, then we can assume that, if A or B wins, they vote that it be B.
Chris says:
That is reasonable, and granted for the sake of
Two anagrams of Hillary Rodham Clinton:
1. Chill liar, do not harm N.Y.
2. My liar troll had no chin.
Mike Ossipoff
election-methods mailing list - see http://electorama.com/em for list info
Shoo, special miff.
-- David Cary
Translation?
Here's another anagram of the same name:
Ah, droll CNN military ho
Mike Ossipoff
election-methods mailing list - see http://electorama.com/em for list info
Chris Benham had said:
Regarding the above example, I can't see any justification in the actual
votes for suggesting that majority rule is violated by electing A.
All three candidates have a majority-strength defeat.
I replied:
Correct--they do. But electing A violates
When I said that, with 2-level balloting there's always a beats-all
candidate, who always wins in Approval, someone could say that the same is
true of Plurality. Sure, but Plurality doesn't have free 2-level voting, so
none of this means much for Plurality.
Likewise, when I defined emphatic
The FBC that Ive been using is the one that requires that there be no way
of improving ones outcome by voting someone over ones favorite. That
strictest of FBCs is met by Approval, -1,0,1, MDDA, and MAMPO.
It appears that DAMC doesnt meet that FBC. But to know that voting some
Kevin's FBC/SF is much more concise in its meaning (though not its wording),
and clearer and more useable than the expectation FBC that I named in my
previous posting. Limiting it to that set of equal-ranked candidates does
make it much more useable than my expectation FBC. That's probably
Chris--
You wrote:
Mike,
Does this compromising one C voter have to unapprove C?
I reply:
No.
Referring to this example,
52: AC (offensive order-reversal)
100: BA
50: C/B
You continued:
ACBA. Approvals: A152, C102, B100. AC 152-50, CB 102-100, BA 150-52
DMC and ASM
Is it the version of FBC that is being used in San Francisco?
Mike Ossipoff
election-methods mailing list - see http://electorama.com/em for list info
One way that DAMC _won't_ fails FBC: If a majority rank Favorite over
Compromise, then, by voting Compromise over Favorite, you can't change the
fact that a majority have ranked Favorite over Compromise, and so you can't
get rid of that majority defeat of Compromise. That made it seem to me
Id like to improve my three examples that I posted the other day, so that
theyre genuine FBC failure examples for DMC.
In these examples, with DMC, either the B /or C voters need to use informed
Approval strategy to save the CW, enforce majority rule, and defeat the
greater evil, or else
Because DAMC recognizes majority defeats, and because River is similar to
Ranked Pairs, a Condorcet version, DAMC lilkely meets SFC SDSC. Because of
having Ranked-Pairs as a relative, DAMC might also meet GSFC.
Because it recognizes only majority defeats, that might confer FBC
compliance:
Matthew Welland wrote:
there is an academic
path and a pragmatic path.
Dont fall into the Democrat trap of believing that pragmatic has to be
less than really good. MDDA is extremely simple, briefly-defined. And yet it
has a very impressive list of criterion-compliances. MAMPO is even
My best reason to withdraw my suggestion for power-truncation is that it
amounts to encouraging and facilitating order-reversal on a grand scale.
That's exactly what would take away the benefit of SFC compliance. For me,
SFC and SDSC are the whole reason to have a rank method.
Mike Ossipoff
MDDA and MAMPO (if modified to allow voting an Approval line instead of
saying that all ranked candidates are approved) share UncAAOs advantage of
a majority being able to make a candidate lose while still ranking him. But,
since its necessary to know to not approve him, its not as if no
It seems to me that I once found out that options like ARLO spoil FBC
compliance. So if one wants FBC, one shouldn't include ARLO.
Of course Condorcet, including SSD, doesn't meet FBC anyway, so the above
doesn't apply to them.
But, with MDDA or MAMPO, that would be an additional reason to
SFC describes conditions under which a voter doesnt need any strategy at
all. Are there ways of achieving that under broader conditions?
Yes, thats what ARLO does.
But are there other ways? To get more strategy-freedom under more
conditions?
Summer 04, I asked if a method could
Are you claiming C wins with certainty???
Yes, of course. Doesn't that follow from the definition of the method?
The set of options approved on both ballots is {A,C} of which C is the
most approved member.
That seems really lame. Such a method would certainly fail to elect A
even if
Chris--
I guess you're right: I wanted a way for someone to _really_ trash the
candidates they don't like. But, as you point out, it seems to give
strategists a way to defeat compliance wilth my own criteria.
I withdraw my recommendation of power-truncation.
Mike
election-methods
I'd said:
But if we're going to have the added definition-wordiness of Condorcet,
we should get what Condorcet can offer, including compliance with the
Plurality Criterion, SFC, GSFC, and SDSC, and URNEC.
Juho replied:
And maybe [will you provide] pointers to the definitions of these
My definitions can be found at the following website:
http://www.barnsdle.demon.co.uk/vote/sing.html
URNEC probably isn't there, though I intend to add it there.
SFC and SDSC are there.
Mike Ossipoff
election-methods mailing list - see http://electorama.com/em for list info
When I shot the margins monster with URNEC, a few years ago, margins
disappeared from EM for a while. But it came out of its grave again, like
some kind of Romero cannibalistic zombie. But now it's been shot with the
Plurality criterion, and this time it has really been finished off for good.
When I shot the margins monster with URNEC, a few years ago, margins
disappeared from EM for a while. But it came out of its grave again, like
some kind of Romero cannibalistic zombie. But now it's been shot with the
Plurality criterion, and this time it has really been finished off for good.
Juho--
In a posting to a different mailing list, Markus pointed out that margins
fails the Plurality Criterion, and that wv Condorcet passes the Plurality
Criterion.
For me, Plurality isn't essential. For instance, I consider MDDA a good
proposal.
But if we're going to have the added
Juho said:
I very much support evaluating also the performance with sincere votes /
the utility function that a methods tries to implement (in addition to
evaluating its strategy resistance).
I reply:
Often, when people speak of methods in terms of strategy, they speak of
Forest--
Alright, but the C voters are still truncating their approval, arent they?
They still need that strategy in order to put the choice to the A voters
about accepting the Nash equilibrium or else. True, the C voters dont have
to abandon A to the degree that theyd have to in wv. So
Lomax says:
Okay, that rarity of rarities, an original post by me. A great deal of
criticism of voting methods is based on how the methods behave with
strategic voters, those who vote insincerely to gain some presumably
favorable outcome
I reply:
Mr. Lomax is missing the point. It isnt
Forest had correctly said:
Under winning votes the C faction can take defensive action and
truncate to 20 C. The resulting position is a Nash Equilibrium.
Chris writes:
Taking such defensive action causes B to win, so why would they want to do
that when they
prefer A to B? And I don't see
I said:
The B voters, by truncating, make the would-be reversers accept the Nash
equilibrium or suffer the consequences.
I meant The C voters instead of The B voters
The C voters, by truncating, make the would be reversers accept the Nash
equilibrium or suffer the consequences.
Mike
I said:
The least fortunate voter either rates 1 candidate 1 and N-1 candidates 0,
or rates N candidates 1 and rates N-1 candidatres 0.
I meant to say:
The least fortunate voter either rates 1 candidate 1 and N-1 candidates 0,
or rates N-1 candidates 1 and rates 1 candidate 0.
Mike Ossipoff
Juho says:
This kind of differences could also explain some unexplainable differences
in the attitudes on strategic voting on this list :-).
I reply:
No, it can explain them if theyre unexplainable. But it might explain them
if theyre _otherwise_ unexplainable.
I dont know what you
I distinguish strategy criteria from embarrassment criteria. SFC, SDSC and
FBC are strategy criteria, and are all-important for the best voting
systems, in my opinion.
The Plurality Criterion is an embarrassment criterion only, and therefore
isn't important to me.
Yes, MMPO's Plurality
Instead of Approval balloting that says to vote for as many as you want to,
say, instead:
The -1,1 method:
Each voter may give to any candidate -1 points or 1 point. The candidate
with the highest vote total wins.
[end of -1,1 definiltion]
Then you emphasize something that people miss:
Jan Kok wrote:
The statistical evidence at http://rangevoting.org/TTRvIRVstats.html
seems pretty good that IRV leads to two party domination in IRV
elections, while (delayed) top two runoff tends to lead to a strong
multiparty system.
Why do those two methods, which seem
Juho said (about margins poor properties with regard to unreversed Nash
equilibria):
This one did not change my feelings much. If you'd say something similar
about sincere votes
I reply:
Here I believe that youre saying you want something said about complete
sincerity rather than just
I'll start with order-reversal, because that's what Juho's example was
about:
Order-Reversal:
In the example below, the A voters prefer B to C, but are using offensive
order-reversal in order to take victory from B. The B voters could be
regarded as not having a preference among A and C, or
Juho--
You said:
Here's my example. It is in principle the same one I already used but now
presented as a bit more realistic scenario.
I reply:
Ok, if its effectively the same as your first example, then doesnt
everything that I said about your first example apply to this one too?
The first things that stands out about Juhos example is the fact that there
is no majority preferring B to C. So one could ask in what way it is a
problem for C to win instead of B.
C is the favorite of only one voter? We dont claim to be choosing the
winner by Plurality. Condorcet
Id said:
WV is much more strategy-free. The difference is unidirectional.
Juho replies:
I doubt the unidirectionality. I think the example I gave (Sincere votes:
49:A, 49:BC, 1:CB. Strategic votes: 49:A, 49:BC, 1:CA.) was an example of a
situation where WV is vulnerable to a strategy and
Id said:
WV is much more strategy-free. The difference is unidirectional.
Juho replies:
I doubt the unidirectionality. I think the example I gave (Sincere votes:
49:A, 49:BC, 1:CB. Strategic votes: 49:A, 49:BC, 1:CA.) was an example of a
situation where WV is vulnerable to a strategy and
Kevin said:
My problem with using sincere preference criteria all the time is that I
find it awkward to always word things as though I have no idea what kind of
method I'm talking about.
I reply:
But isn't that necessary if a criterion is to apply seamlessly to all
methods?
Mike Ossipoff
Yes, MAMPO looks like it belongs at the top of the simple alternative rank
methods, just below SSD, in my list of methods to propose. So now this is my
method proposing-order:
1. SSD
2. MAMPO
3. MDDA
4. MDDB
5. MMPO
6. PC
7. Smith//PC
8. SR
Of course its no surprise that, being better
Yes, MAMPO looks like it belongs at the top of the simple alternative rank
methods, just below SSD, in my list of methods to propose. So now this is my
method proposing-order:
1. SSD
2. MAMPO
3. MDDA
4. MDDB
5. MMPO
6. PC
7. Smith//PC
8. SR
Of course its no surprise that, being better
That's because MAMPO starts out about an Approval majority and then goes to
a pairwise method; while MDDA starts out about a pairwise majority, then
goes to Approval.
So, as another two-part method, I wouldn't say that MAMPO is more
complicated than MDDA or MDDB. So, if it adds Plurality
Ok, sure, its obvious (when my claim is challenged) that, though a
candidate at the voter median is the CW, its plausible that the CW neednt
be at the voter median: Suppose no candidate is at the voter median. Might
not a candidate who is much closer (in some sense) to the voter median
On Feb 20, 2007, at 15:39 , Michael Ossipoff wrote: Juho wrote: My
sympathies towards minmax(margins) come primarily from the way it handles
sincere votes. I reply: But there wont be sincere votes for it to
handle, to the extent that it doesnt allow sincere votes. Thats why
On Feb 20, 2007, at 15:39 , Michael Ossipoff wrote: Juho wrote: My
sympathies towards minmax(margins) come primarily from the way it handles
sincere votes. I reply: But there wont be sincere votes for it to
handle, to the extent that it doesnt allow sincere votes. Thats why
Because of the universal angry opposition that Approval seems to meet, with
members of the public, and because RV can be tarred with the same brush,
its a good idea to consider rank methods instead.
As I was saying, SSD is my favorite. My best girlfriend said that SSD wasnt
too
I was trying to post my message about alternative rank methods, but somehow
copied the Juho reply into that posting instead. I then posted the
alternative rank methods message with the subject line Rank methods for if
SSD is rejected.
Mike Ossipoff
election-methods mailing list - see
Juho wrote:
My sympathies towards minmax(margins) come primarily from the way it handles
sincere votes.
I reply:
But there wont be sincere votes for it to handle, to the extent that it
doesnt allow sincere votes. Thats why the defensive strategy criteria, and
the wv Condorcet methods
It calls for a little comment. A number of times before, I've proposed
polls, and usually a number of people voted. Enough to do a meaningful,
interesting count. This is the first one in which not even one person (other
than myself) voted.
Obviously that's always a possibility. It was a
Mike,
I guess a wiki would be the best way to do the poll then... The votes
could be posted in a single place, changed at will, and discussed on a
talk page.
I reply:
...changed at will is the part that I don't quite like. I've never
understood the appeal of a wiki, where what you send might
To comment on the last part first:
Kevin says:
(Although, incidentally, Mike's criticism that my scheme can't be used to
show that Approval and Range fail SFC is quite strange. SFC is defined on
sincere preferences. If Mike means what I use in place of SFC
I reply:
Yes, I thought Id
Dave said:
There can be debate as to remembering A=B. For each 2 such votes I would
count as if there was 1 each of AB and BA.
I reply:
AB and BA would make sense if youd power truncated A and B. But saying
that theyre equal is not the same as saying that each is better than the
other.
Juho wrote:
(There are good methods also on the other side of the fence,
like minmax(margins).)
I reply:
But, when saying that minmax(margins) is good, you've got to say what it's
good for. I've told what the wv methods are good for: The best ones meet
SFC, GSFC, and SDSC.
Mike Ossipoff
Anyone voting in this poll is urged to rank and rate as many candidates as
are of interest. Don't bother ranking or rating them all--only the ones who
have significantly more merit than the others I list all of these only to
accomodate all political persuasions.
If you choose to vote, post
I should add that the ratings should be from 9 to 100.
I didn't include Ron Paul or Mike Gavel, because if I included
little-heard-of candidates of the main parties because they're antiwar, I
could be accused of partisan candidate-inclusion. I included the Green and
Libertarian that were
(Ratings are difficult. Sometimes sincere ratings look strategic).
Approval RankRating
Michael Badnarik (Lib.) [x] [3] [80]
Peter Camejo [x] [1][100]
Hillary Clinton
Kevin says:
We could define Condorcet on sincere preferences also.
I reply:
We could and do. I define Condorcets Criterion based on preference. By
preference I mean what you mean by sincere preference. Its really the
only CC definition that makes any sense.
Lomax says:
The
I specifically invite Juho Laatu, Chris Benham, David Carey, and Markus
Schulze, and any other non-Americans on EM, to vote in the February 2008 EM
presidential poll.
I will do four counts:
1. EM-everyone
2. EM-Americans
3. EM-non-Americans
4. EM-everyone and their spouses, partners, kids,
First, I want to express my agreement with the statement that were here
because public elections are important. Were not here because of campus
elections or organizational elections, etc. I suggest that campus
governments and organizations be urged to use methods that are suitable for,
and
Anyone offering criteria should welcome criticism of them, without anger,
just as Ive been doing. But sometimes something that you werent expecting
takes you by surprise, and you lose your temper.
BC is a criterion, more than its Schulzes method dressed-up as a
criterion. Thats
Are apportionment academics related to voting-system academics?
You have probably noticed a certain cluelessness about voting-system
academics. Ive been checking out some apportionment writing on the
Internet, and apparently academics who write about apportionment share that
cluelessness.
be different than actual voting.SFC: If no one
falsifies a preference, and there's a CW, and a majority of all the voters
prefer the CW to candidate Y, and vote sincerely, then Y shouldn't win.
[end of definition] Michael Ossipoff wrote: Kevin and Chris posted
their criteria
Lomax said:
Just to be explicit about the application of this to equal ranking. At 10:28
AM 2/15/2007, Chris Benham wrote: Pasting from Mike's page: Some
definitions useful in subsequent criteria definitions: A voter votes X
over Y if he votes in a way such that if we count only his
As I said, ranking is lots easier than rating, as you know if youve
actually done both in EM polls.
Thats why Borda is useful for personal decision making.
And maybe, in that Utopian society that Lomax and Smith want to believe in,
the difficulty of rating might make Borda a better idea
RVs strategy problem in comparison to Condorcet is a subject that weve
already discussed. RVs strategy problem in comparison to Approval is that
sincere voters can be had by strategizers.
You might say, truly, that a person can make the choice about whether they
want sincerity or
that meets BC also meets those 5
criteria.
Michael Ossipoff wrote:
So I prefer my own preference-based wordings of my defensive strategy
criteria. However, I myself have used a votes-only, rank-methods-only test
for compliance with my criteria: Steve Epplelys Beatpath Criterion. Any
rank
the voters prefer the CW to candidate Y, and vote sincerely, then Y
shouldn't win.
[end of definition]
Chris quotes me:
Michael Ossipoff wrote:
Kevin and Chris posted their criteria that they incorrectly claimed
equivalent to SFC.
These same alternative SFCs have been posted to EM before
Kevin and Chris posted their criteria that they incorrectly claimed
equivalent to SFC.
These same alternative SFCs have been posted to EM before and thoroughly
discussed before.
In fact, we've been all over this subject before.
Though Chris's and Kevin's criteria clearly are not
Kevin had said:
Hi, By the way, you can also ensure a method satisfies SFC by having it
comply with this votes-only criterion: If more than half of the voters
rank A over B, but there is no majority of the voters ranking some
third candidate over A, then B doesn't win.
I replied
Regarding Kevin's puzzling statement:
, and if you did know that you qualify for it, you'd also know that you
don't need to use it, since your candidate should win no matter what you do
with your lower rankings.
I'd like to add that SFC is about not needing to use anything, not needing
to
Warren says:
ok, sorry for my dimness here.
I reply:
Thats ok, Warren; its not like its the first time.
Warren continues:
Thanks to Benham and Venzke we now have two readable definitions of SFC
I reply:
We agree on one thing: Warren was incapable of reading the definition of
SFC, and
When discussing Chris's SFC, I said There can be no majority voting X
over Y. What I meant to say was that, in Pluralitly, if a majority vote X
over Y, then there can be no majority voting anyone over X.
Mike Ossipoff
election-methods mailing list - see http://electorama.com/em for list
Ossipoff: But if Warren has a quibble about what it means to fully vote X
over Y, then
I refer him to my criteria SFC, GSFC, and SDSC.
--WDS: here is the definition of SFC given by Ossipoff (CW=Condorcet
winner):
SFC: If no one falsifies a preference, and there's a CW, and a majority
of
1 - 100 of 154 matches
Mail list logo
