Open letter to STAR voting promoters
Robert,
What rhyme or reason is used to determine where the Approval cutoff is?
The idea is to give the voters as much freedom to rank among unapproved
candidates as among approved ones.
Also it is supposed to be analogous with grades children are given at
school, where A B C is a "pass" and D E F is a "fail". It shouldn't be
kept a secret from the voters that this is how their grading will be
interpreted.
I'm not necessarily a huge fan of this particular balloting rule. I find
it a bit too restrictive. If there are five candidates I should be free
to strictly rank all of them and approve all but one.
But I do very much like the idea of Condorcet methods using some sort of
unlimited ranking ballot with an explicit approval cutoff (and meeting
the Double Defeat criterion).
But you are apparently allergic to that idea. Do you think this
attitude of yours would be widespread in the US? Haven't a lot of
people got accustomed to giving out simple "Like/Dislike" binary
ratings on the internet?
Chris B.
On 6/06/2024 6:00 am, robert bristow-johnson wrote:
On 06/05/2024 8:55 AM EDT Chris Benham cbenhamau@yahoo.com.au wrote:
What do you think of Adam Tarr's old idea of using A-B-C--D-E-F
"grading" ballots with the idea of using them to infer ranking to elect
a Condorcet winner and if there isn't one then interpreting the A-B-C
grades as Approval?
Why not just A-B as Approval? Or why not A-B-C-D as Approval? Or why not just A as Approval (I know that would result in Condorcet-Plurality). What rhyme or reason is used to determine where the Approval cutoff is?
He liked Condorcet//Approval but also possible with these ballots is
Smith//Approval and Margins-Sorted Approval and a few others.
I don't see such a big problem with high-resolution score ballots as
long as it is made clear to the voters (and is true) that their
approximate ratings of the candidates will work just as well for them as
their exact ones.
Please take a look at these templates for RCV legislation: https://drive.google.com/file/d/1DGvs2F_YoKcbl2SXzCcfm3nEMkO0zCbR/view .
Even though the actual legislation is not verbatim with these templates, the IRV template (first one) has actually been used for the content of the ballot question for an initiative for IRV. The second template has BTR-IRV.
Now consider the last three templates ("Straight-ahead Condorcet"). So this is specifically about what to put in subdivision (4) to replace:
"(4) If no Condorcet winner exists in subdivision (3), then the candidate having the plurality of first preferences is elected."
What exactly will the language be to replace that line of code?
"(4) If no Condorcet winner exists in subdivision (3), then ..." what?
Perhaps,
"(4) If no Condorcet winner exists in subdivision (3), then the candidate receiving the most first and second preferences is elected."
I might call that Condorcet-Bucklin. Bucklin and Approval have something in common and that common property is that it's "counting marks" and not counting people. This may be perceived by some legal analysts as violating One-person-one-vote, I dunno.
Would that be a little more resistant to the burying strategy than Condorcet-Plurality? How about comparing it to Condorcet-TTR, which is nearly Condorcet-IRV?
"(4) If no Condorcet winner exists in subdivision (3), then the two candidates receiving the most first preferences shall runoff against each other and the candidate in that pair that is preferred over the other by the greater number of voters shall be elected."
How would MO's "wv-Condorcet" be worded? (Since he is apparently not plonking me as was threatened.)
It needs to be short and sweet. ("Sweet" means that it ostensibly elects the candidate with the "greatest" voter support, however that voter support is defined. But the definition of that "greatest" voter support needs to be simple and understandable and make sense to pedestrians.)
--
r b-j . _ . _ . _ . _ rbj@audioimagination.com
"Imagination is more important than knowledge."
.
.
.
Election-Methods mailing list - see https://electorama.com/em for list info
On 06/06/2024 12:55 AM EDT Chris Benham cbenhamau@yahoo.com.au wrote:
What rhyme or reason is used to determine where the Approval cutoff is?
The idea is to give the voters as much freedom to rank among unapproved
candidates as among approved ones.
Also it is supposed to be analogous with grades children are given at
school, where A B C is a "pass" and D E F is a "fail".
Gee, that's a good reason to base legislation for grown-ups on.
It shouldn't be kept a secret from the voters that this is how
their grading will be interpreted.
Yeah, but it's not so much about communicating the rules to the voters, but imposing or defining the rules. Just because we can communicate the rule to the voters does not make it a sensible rule.
Here's a rule I like: FPTP ballots, but we elect the candidate with the second-greatest number of votes. Let's communicate that clearly to the voters, so they can well consider their well-informed vote. That transparency inoculates it from being a bad rule.
If multi-winner, we elect the candidates with most votes except the single candidate with the plurality of votes. That won't bring on any strategic or tactical voting. Voters can just vote sincerely and not worry about regretting their vote.
I'm not necessarily a huge fan of this particular balloting rule. I find
it a bit too restrictive. If there are five candidates I should be free
to strictly rank all of them and approve all but one.
I think you mentioned earlier the idea of counting any ranked candidate an approval. Only candidates completely unranked are not approved. This criteria makes a little sense because of some consistency in meaning.
But I would be at the other end, where I would count only the two highest-ranked candidates as "approved". That's the first non-trivial expansion of inclusion over just the top-ranked candidate as "approved", which is equivalent to Condorcet-Plurality. It's just that I think it's unlikely that you would approve anyone below your second-favorite, whom you might not really approve, anyway. It's that they're the lessor evil.
But I dunno that I would like Condorcet-Approval at all. It has to simple to explain in legislative language. Maybe it should be MinMax. But I still think that Condorcet-Plurality or Condorcet-TTR would be easiest to explain to legislators as to clear intent and function in the language that they are legislating. The law should say what it means and mean what it says.
But I do very much like the idea of Condorcet methods using some sort of
unlimited ranking ballot with an explicit approval cutoff (and meeting
the Double Defeat criterion).
I want it to be simple: "If more voters prefer A to B, let's not elect B if we can, at all, avoid it." If we cannot avoid it, some other measure of voter support must be used. The manner of that measure should make sense to the policy makers and the voters. This is so that when someone loses, that candidate and the voters for that candidate can understand and accept why they lost.
I'm starting to think that Condorcet-TTR will be the easiest to explain to people why the candidate elected was the candidate with the most consistent voter support.
But you are apparently allergic to that idea. Do you think this
attitude of yours would be widespread in the US?
I think this, elections in non-corrupt participatory democracies should have these principles:
- Equality of citizenship and wide enfranchisement to citizens
- Well-warned elections
- Equal and unhindered access to the polls for everyone
- The secret ballot
- One-person-one-vote, the equality of the value of our vote
- Majority rule
- Process transparency, deterministic outcome
- Peaceful transfer of power, rule of law.
Them's be principles.
Also...
A. Voters and policy makers understandably are skeptical about changing the rules to a system that has been in use for centuries. They need to understand why a change makes things better. And they may wonder if the change is a sneaky maneuver by some interest group to make new rules to help themselves out. Like let's change the rules to require a $40 poll tax. Who's that gonna help?
B. I am unabashedly in favor of strict equality of our franchise and vote and every angle of the notion "One-person-one-vote". That's why Condorcet. Whenever someone is elected that is not the Condorcet winner, then, for certain, some minority group of voters had their votes counting with more effect than a larger group of voters that preferred someone else for election. And then, for certain, the election was spoiled (there is a loser whose presence in the race materially changed the outcome).
C. I don't want voters to have to think that they gotta calculate how to vote to promote their own political interests. Who do you most want elected? Mark that candidate #1. Now, if that candidate wasn't around and you had to choose from the remaining pool, then who do you most want elected? Mark them #2. It shouldn't be any more complicated than that. And voters should be able to rest easy that they didn't shoot themselves in the foot by coming to the poll and voting. Voters should feel confident that if they cannot get their first choice elected, their full voting power will go toward their second choice. It shouldn't matter what chronological order candidates are "eliminated". There should be no chronological elimination.
D. For process transparency, we need a redundancy of tabulation of the vote in elections with the public, the media, competing campaigns able to also determine the winner of an election quickly (like election night) from publicly available data posted at the source (the precincts). At the precinct level, the ballot box (this would include the tabulator machine that the ballots go in) must be the only section in the data path that is opaque. It must be opaque to provide for the secrecy of our individual vote. (Can you imagine of candidate totals were displayed on a board and we could see them increment just after a ballot was inserted?).
So the precinct posts vote tallies that can be publicly inspected and this essentially commits the government to those numbers except for a very few provisional ballots that are later adjudicated to be valid and counted. We all know what those numbers are and they won't change much. This prevents any corrupt high election official from "finding, uh, 11780 votes". There is no way that they can pad the numbers without someone else noticing. This is Precinct Summability and the decentralization of the tabulation of the vote.
Only the section of the data path where the ballots go into a box and, at the end of the day, a printout of tallies comes out of the box, only there is process opacity. And that can be scrutinized with courts ordering recounts and ballot bags can have their seals broken and be opened and ballots recounted to verify what was originally posted at any specific precinct. But after the precinct posting summable tallies, then no other part of the process should be opaque. It should all be transparent. If all precincts report tallies on election night, no one should be surprised about who is later certified the winner by authorities (unless it's extremely close and a recount including adjudicated provisional ballots flips the outcome).
E. The methodology should also be transparent. The law should say what it means and mean what is says. This is why I'm for the two-method Condorcet where the first method is simply applying the Condorcet criterion. And the contingency method (if no CW) should also be transparent that voters and policy makers understand and agree to the basis or measure of voter support used to determine which candidate has the most voter support. Then the application of the method should be mindless - just an algorithm.
Really, the reason I'm for Condorcet RCV is simple basic principles of democracy.
Haven't a lot of people got accustomed to giving out simple
"Like/Dislike" binary ratings on the internet?
But it's not about what (or who) we like on the internet. It's about a collective decision to choose who we give the keys to the government to.
--
r b-j . _ . _ . _ . _ rbj@audioimagination.com
"Imagination is more important than knowledge."
.
.
.
Robert,
Just because we can communicate the rule to the voters does not make it a sensible rule.
Of course, and I didn't say it did. The person who suggested
Condorcet//Approval using ABCDEF grading ballots didn't seem to think
this was necessary, but I would have the grades labelled "A-approve,
B-approve, C-approve, D-disapprove, E-disapprove, F-disapprove".
I want it to be simple: "If more voters prefer A to B, let's not elect B if we can, at all, avoid it." If we cannot avoid it, some other measure of voter support must be used. The manner of that measure should make sense to the policy makers and the voters. This is so that when someone loses, that candidate and the voters for that candidate can understand and accept why they lost.
Yes, that's great.
But I would be at the other end, where I would count only the two highest-ranked candidates as "approved".
I hope you mean candidates in the two highest rating-slots, meaning in
the case of these ABCDEF ballots candidates given an A or a B. That is
of course possible, but since Condorcet is incompatible with
Later-no-Help ranking among unapproved candidates is more suspicious
than among approved ones so I don't like forcing the voters to do most
of their ranking there (if they want to strictly rank a lot of
candidates). So I would be more open to limiting the Disapproval slots
to just two.
I don't want voters to have to think that they gotta calculate how to vote to promote their own political interests.
That's nice but it isn't fully possible. All Condorcet methods and
Hare fail Favorite Betrayal and so are vulnerable to Compromise
strategy. All Condorcet methods are vulnerable to Burial strategy.
But the problem is much greater in some methods than in others.
Chris B.
On 6/06/2024 6:18 pm, robert bristow-johnson wrote:
On 06/06/2024 12:55 AM EDT Chris Benham cbenhamau@yahoo.com.au wrote:
What rhyme or reason is used to determine where the Approval cutoff is?
The idea is to give the voters as much freedom to rank among unapproved
candidates as among approved ones.
Also it is supposed to be analogous with grades children are given at
school, where A B C is a "pass" and D E F is a "fail".
Gee, that's a good reason to base legislation for grown-ups on.
It shouldn't be kept a secret from the voters that this is how
their grading will be interpreted.
Yeah, but it's not so much about communicating the rules to the voters, but imposing or defining the rules. Just because we can communicate the rule to the voters does not make it a sensible rule.
Here's a rule I like: FPTP ballots, but we elect the candidate with the second-greatest number of votes. Let's communicate that clearly to the voters, so they can well consider their well-informed vote. That transparency inoculates it from being a bad rule.
If multi-winner, we elect the candidates with most votes except the single candidate with the plurality of votes. That won't bring on any strategic or tactical voting. Voters can just vote sincerely and not worry about regretting their vote.
I'm not necessarily a huge fan of this particular balloting rule. I find
it a bit too restrictive. If there are five candidates I should be free
to strictly rank all of them and approve all but one.
I think you mentioned earlier the idea of counting any ranked candidate an approval. Only candidates completely unranked are not approved. This criteria makes a little sense because of some consistency in meaning.
But I would be at the other end, where I would count only the two highest-ranked candidates as "approved". That's the first non-trivial expansion of inclusion over just the top-ranked candidate as "approved", which is equivalent to Condorcet-Plurality. It's just that I think it's unlikely that you would approve anyone below your second-favorite, whom you might not really approve, anyway. It's that they're the lessor evil.
But I dunno that I would like Condorcet-Approval at all. It has to simple to explain in legislative language. Maybe it should be MinMax. But I still think that Condorcet-Plurality or Condorcet-TTR would be easiest to explain to legislators as to clear intent and function in the language that they are legislating. The law should say what it means and mean what it says.
But I do very much like the idea of Condorcet methods using some sort of
unlimited ranking ballot with an explicit approval cutoff (and meeting
the Double Defeat criterion).
I want it to be simple: "If more voters prefer A to B, let's not elect B if we can, at all, avoid it." If we cannot avoid it, some other measure of voter support must be used. The manner of that measure should make sense to the policy makers and the voters. This is so that when someone loses, that candidate and the voters for that candidate can understand and accept why they lost.
I'm starting to think that Condorcet-TTR will be the easiest to explain to people why the candidate elected was the candidate with the most consistent voter support.
But you are apparently allergic to that idea. Do you think this
attitude of yours would be widespread in the US?
I think this, elections in non-corrupt participatory democracies should have these principles:
- Equality of citizenship and wide enfranchisement to citizens
- Well-warned elections
- Equal and unhindered access to the polls for everyone
- The secret ballot
- One-person-one-vote, the equality of the value of our vote
- Majority rule
- Process transparency, deterministic outcome
- Peaceful transfer of power, rule of law.
Them's be principles.
Also...
A. Voters and policy makers understandably are skeptical about changing the rules to a system that has been in use for centuries. They need to understand why a change makes things better. And they may wonder if the change is a sneaky maneuver by some interest group to make new rules to help themselves out. Like let's change the rules to require a $40 poll tax. Who's that gonna help?
B. I am unabashedly in favor of strict equality of our franchise and vote and every angle of the notion "One-person-one-vote". That's why Condorcet. Whenever someone is elected that is not the Condorcet winner, then, for certain, some minority group of voters had their votes counting with more effect than a larger group of voters that preferred someone else for election. And then, for certain, the election was spoiled (there is a loser whose presence in the race materially changed the outcome).
C. I don't want voters to have to think that they gotta calculate how to vote to promote their own political interests. Who do you most want elected? Mark that candidate #1. Now, if that candidate wasn't around and you had to choose from the remaining pool, then who do you most want elected? Mark them #2. It shouldn't be any more complicated than that. And voters should be able to rest easy that they didn't shoot themselves in the foot by coming to the poll and voting. Voters should feel confident that if they cannot get their first choice elected, their full voting power will go toward their second choice. It shouldn't matter what chronological order candidates are "eliminated". There should be no chronological elimination.
D. For process transparency, we need a redundancy of tabulation of the vote in elections with the public, the media, competing campaigns able to also determine the winner of an election quickly (like election night) from publicly available data posted at the source (the precincts). At the precinct level, the ballot box (this would include the tabulator machine that the ballots go in) must be the only section in the data path that is opaque. It must be opaque to provide for the secrecy of our individual vote. (Can you imagine of candidate totals were displayed on a board and we could see them increment just after a ballot was inserted?).
So the precinct posts vote tallies that can be publicly inspected and this essentially commits the government to those numbers except for a very few provisional ballots that are later adjudicated to be valid and counted. We all know what those numbers are and they won't change much. This prevents any corrupt high election official from "finding, uh, 11780 votes". There is no way that they can pad the numbers without someone else noticing. This is Precinct Summability and the decentralization of the tabulation of the vote.
Only the section of the data path where the ballots go into a box and, at the end of the day, a printout of tallies comes out of the box, only there is process opacity. And that can be scrutinized with courts ordering recounts and ballot bags can have their seals broken and be opened and ballots recounted to verify what was originally posted at any specific precinct. But after the precinct posting summable tallies, then no other part of the process should be opaque. It should all be transparent. If all precincts report tallies on election night, no one should be surprised about who is later certified the winner by authorities (unless it's extremely close and a recount including adjudicated provisional ballots flips the outcome).
E. The methodology should also be transparent. The law should say what it means and mean what is says. This is why I'm for the two-method Condorcet where the first method is simply applying the Condorcet criterion. And the contingency method (if no CW) should also be transparent that voters and policy makers understand and agree to the basis or measure of voter support used to determine which candidate has the most voter support. Then the application of the method should be mindless - just an algorithm.
Really, the reason I'm for Condorcet RCV is simple basic principles of democracy.
Haven't a lot of people got accustomed to giving out simple
"Like/Dislike" binary ratings on the internet?
But it's not about what (or who) we like on the internet. It's about a collective decision to choose who we give the keys to the government to.
--
r b-j . _ . _ . _ . _ rbj@audioimagination.com
"Imagination is more important than knowledge."
.
.
.
Election-Methods mailing list - see https://electorama.com/em for list info
In fact, some of us suggested to Richie allowing equal-ranking,
each vote counted whole, & he immediately refused it.
Although I don't like to agree with what Rob Richie says, he's correct
in this case. In other words, I agree this would not be a good change.
Maybe. Probably. …but it was suggested because it’s better garbage
than ordinary IRV. ... I wouldn’t advocate garbage, but I
suggested that garbage to Richie because it’s a
“Lesser-Of-2-Garbages”. ... At least it avoids IRV’s worst.
IRV has two major weaknesses:
-
dismissing equal-rank markings
-
not eliminating a candidate who is clearly less popular than the
candidate with the fewest transferred votes.
Both are easy to remedy.
I dislike IRV, yet I find myself supporting it as a stepping-stone to
better election methods. That's because a ranked choice ballot is
better than a STAR ballot, and better than a Score ballot, and better
than an Approval ballot.
Richard Fobes
the VoteFair guy
On 6/5/2024 7:52 AM, Michael Ossipoff wrote:
On Wed, Jun 5, 2024 at 01:18 Chris Benham <cbenhamau@yahoo.com.au
mailto:cbenhamau@yahoo.com.au> wrote:
__
Mike,
In fact, some of us suggested to Richie allowing equal-ranking,
each vote counted whole, & he immediately refused it.
And he was very correct to do so, because the resulting ER-IRV
(whole) method is garbage.
Maybe. Probably. …but it was suggested because it’s better garbage than
ordinary IRV.
I wouldn’t advocate garbage, but I suggested that garbage to Richie
because it’s a “Lesser-Of-2-Garbages”.
At least it avoids IRV’s worst.
[I don’t have any comments farther down in the text. It’s easier to say
that than to delete the unreplied text.]
On 6/4/2024 8:19 PM, Closed Limelike Curves wrote:
< ...
So, I suppose my real question is: is there any voting system
that satisfies both properties? Sincere favorite and
later-no-help are enough for voters who follow good polls to
work out the Condorcet winner, by making min-maxing the
optimal strategy. On the other hand, I don't know
if they're necessary or if they're compatible with honesty in the
zero-information case.
Since you seem to have asked me this question, and no one else has
answered it, I'll say ...
I believe that attempting to "satisfy both properties" is likely to
increase the failure rates of yet other fairness criteria.
To me, a zero failure rate and a one-in-a-hundred-thousand failure rate
are not significantly different. Both are close enough to zero to not
be a concern in elections.
The place where this is bad is the zero-information setting. True
zero-info almost never exists, but low-information settings crop up if
you have a sincere Condorcet cycle, several clones, or a nonpartisan
local election. In these settings, it would be very nice to have
voters be perfectly honest.
I believe we should expect that in the future there will be better
opinion polls that provide more information for governmental elections.
In other words, zero information elections will become rare -- if the
outcome is important, as it is in governmental elections.
I think honesty would be very nice in every setting. But getting
honesty requires using rules that don't motivate voters to vote dishonestly.
In other words, if there's a way to cheat, cheating will occur.
Richard Fobes
the VoteFair guy
On 6/4/2024 8:19 PM, Closed Limelike Curves wrote:
* Every cardinal/rating method I know of is vulnerable to what used to
be called a "drama queen" meaning someone who exaggerates importance.
Expressed another way ...
I'm very interested in fleshing this out a bit more. In some situations,
this is the main drawback of score voting; min-maxing means your ballot
expresses less information than a full ranking and produces worse
results than most Condorcet methods. But in other situations, this is a
huge strength of rated voting that I think we really need to keep.
The place where this is bad is the zero-information setting. True
zero-info almost never exists, but low-information settings crop up if
you have a sincere Condorcet cycle, several clones, or a nonpartisan
local election. In these settings, it would be very nice to have voters
be perfectly honest.
The place where this is good is in the high-information setting where
voters follow the polls, or know . Then, every "sensible" voting method
will encourage exaggeration: if you rank or rate the best frontrunner at
the top, this maximizes your support for them, and therefore your impact
on the election. Similarly, ranking or rating a candidate last minimizes
your support for them.
The only difference in the case of cardinal/rating methods is you can do
this without being forced to engage in order-reversal. This is a big
f*cking deal. Lots of voters will refuse to engage in decapitation
to protect a Condorcet winner (e.g. Begich in Alaska): it's
counterintuitive, makes voters feel icky, and reduces their favorite
candidate's shot at winning. Even if they /were/ willing, this kind of
dishonest order-reversal strategy can make it impossible to identify the
sincere Condorcet winner. ("Hmm, is RFK Jr. trailing because of
the brain worm or the major parties burying him?")
In high-information situations, this kind of approval-thresholding is
the least-damaging kind of strategy:
- It's simple and obvious, so parties can't manipulate voters by lying.
- It has a single stable equilibrium point, so you don't need money or
party endorsements to prove you're a viable candidate. - It's sincere, so it lets you identify the Condorcet-winner and the
Condorcet runner-up. - It guarantees that if voters follow the polls, they'll choose the
Condorcet winner.
So, I suppose my real question is: is there any voting system that
satisfies both properties? Sincere favorite and later-no-help are enough
for voters who follow good polls to work out the Condorcet winner, by
making min-maxing the optimal strategy. On the other hand, I don't know
if they're necessary or if they're compatible with honesty in the
zero-information case.
I believe we should expect that in the future there will be better
opinion polls that provide more information for governmental elections.
In other words, zero information elections will become rare -- if the
outcome is important, as it is in governmental elections.
I mostly agree, which is why I think the perfect information case is the
most important. However, in the perfect information case I think cardinal
voting (specifically approval) is the best you can do. Anything else is
going to create an incentive to be insincere, which can obscure the
Condorcet winner or the runner-up (the most important pieces of information
to have).
Say you know who the Condorcet winner and runner-up are, and you're working
in a monotonic system. Now, the best strategy is to rank the better one of
these candidates at the top, which minimizes the probability that they'll
be involved in a cycle. The other candidate should be ranked at the bottom,
to maximize the chances that someone else will pairwise-beat them: the more
candidates you rank above them, the more likely it is one will beat them.
Now we have the question of what happens to the other candidates. Ideally,
we want to minimize the amount of dishonesty. Perfect honesty is impossible
now: unless the top frontrunner is your sincere favorite, there's no way to
give a completely sincere ranking anymore. But it* is *possible to get
*semi-sincere
*rankings out of voters.
Here’s what I want to know: can we keep this property, but still get strict
sincerity for the zero-information case?
On Thu, Jun 6, 2024 at 9:22 PM Richard, the VoteFair guy <
electionmethods@votefair.org> wrote:
On 6/4/2024 8:19 PM, Closed Limelike Curves wrote:
< ...
So, I suppose my real question is: is there any voting system
that satisfies both properties? Sincere favorite and
later-no-help are enough for voters who follow good polls to
work out the Condorcet winner, by making min-maxing the
optimal strategy. On the other hand, I don't know
if they're necessary or if they're compatible with honesty in the
zero-information case.
Since you seem to have asked me this question, and no one else has
answered it, I'll say ...
I believe that attempting to "satisfy both properties" is likely to
increase the failure rates of yet other fairness criteria.
To me, a zero failure rate and a one-in-a-hundred-thousand failure rate
are not significantly different. Both are close enough to zero to not
be a concern in elections.
The place where this is bad is the zero-information setting. True
zero-info almost never exists, but low-information settings crop up if
you have a sincere Condorcet cycle, several clones, or a nonpartisan
local election. In these settings, it would be very nice to have
voters be perfectly honest.
I believe we should expect that in the future there will be better
opinion polls that provide more information for governmental elections.
In other words, zero information elections will become rare -- if the
outcome is important, as it is in governmental elections.
I think honesty would be very nice in every setting. But getting
honesty requires using rules that don't motivate voters to vote
dishonestly.
In other words, if there's a way to cheat, cheating will occur.
Richard Fobes
the VoteFair guy
On 6/4/2024 8:19 PM, Closed Limelike Curves wrote:
* Every cardinal/rating method I know of is vulnerable to what used
to
be called a "drama queen" meaning someone who exaggerates importance.
Expressed another way ...
I'm very interested in fleshing this out a bit more. In some situations,
this is the main drawback of score voting; min-maxing means your ballot
expresses less information than a full ranking and produces worse
results than most Condorcet methods. But in other situations, this is a
huge strength of rated voting that I think we really need to keep.
The place where this is bad is the zero-information setting. True
zero-info almost never exists, but low-information settings crop up if
you have a sincere Condorcet cycle, several clones, or a nonpartisan
local election. In these settings, it would be very nice to have voters
be perfectly honest.
The place where this is good is in the high-information setting where
voters follow the polls, or know . Then, every "sensible" voting method
will encourage exaggeration: if you rank or rate the best frontrunner at
the top, this maximizes your support for them, and therefore your impact
on the election. Similarly, ranking or rating a candidate last minimizes
your support for them.
The only difference in the case of cardinal/rating methods is you can do
this without being forced to engage in order-reversal. This is a big
f*cking deal. Lots of voters will refuse to engage in decapitation
to protect a Condorcet winner (e.g. Begich in Alaska): it's
counterintuitive, makes voters feel icky, and reduces their favorite
candidate's shot at winning. Even if they /were/ willing, this kind of
dishonest order-reversal strategy can make it impossible to identify the
sincere Condorcet winner. ("Hmm, is RFK Jr. trailing because of
the brain worm or the major parties burying him?")
In high-information situations, this kind of approval-thresholding is
the least-damaging kind of strategy:
- It's simple and obvious, so parties can't manipulate voters by lying.
- It has a single stable equilibrium point, so you don't need money or
party endorsements to prove you're a viable candidate. - It's sincere, so it lets you identify the Condorcet-winner and the
Condorcet runner-up. - It guarantees that if voters follow the polls, they'll choose the
Condorcet winner.
So, I suppose my real question is: is there any voting system that
satisfies both properties? Sincere favorite and later-no-help are enough
for voters who follow good polls to work out the Condorcet winner, by
making min-maxing the optimal strategy. On the other hand, I don't know
if they're necessary or if they're compatible with honesty in the
zero-information case.
Election-Methods mailing list - see https://electorama.com/em for list
info
wv Condorcet isn’t a method. It’s a class of Condorcet methods that resolve
circular-ties based on pairwise-defeat strengths, where the strength of a
pairwise defeat is measured by the number of people ranking the defeater
over the defeated.
It isn’t a general definition of a class that specifies requirements for
defining as-yet unspecified new members. Instead it’s a collective name for
a few particularmethods that measure defeat-strength in that manner.
In fact, by “wv Condorcet”, I mean either RP(wv), Schulze, SSD, MinMax(wv),
& Smith//MinMax.
(…all of which have been amply defined here.)
If someone wanted a broader definition, that could be:
A method that uses defeat-strengths, which are defined as I described, &
has the defensive-strategy properties that I’ve specified for the
above-listed methods.
But the previous definition, as one of a certain few particular methods, is
what I mean by the term.
Yes, I blocked you, & you remain blocked.
But I’ve now blocked the most frequent posters (who never seem to stop
posting daily or nearly daily). … & nearly all of those who post
frequently by any standard.
Sometimes there’s a topic that I want to comment on, &, when I check for
them, some blocked sender has said something that I might not have already
answered enough times yet.
:-)
Sometimes I’m too lazy, or it’s too late in the day, to do something
worthwhile, like sending a proposal or trying to organize an
initiative-committee, & so, with so little EM mail now, due to
garbage-filtering, I might check to find out if any garbage-spewers have
said something that I’ve only answered 99 times.
It’s easy enough if I’m checking for their posts in a current topic. But
any less recent posts will be interspersed among thousands of unwanted
advertisements. …& they’re automatically deleted after a month anyway.
So it’s mostly recent garbage-posts that I ever take the time to check for.
But the important thing about this is that I’m not willing to have the
garbage-spewers’ **** dumped on my screen every day. So the blocking is
definitely serving it’s purpose.
On Wed, Jun 5, 2024 at 13:41 robert bristow-johnson <
rbj@audioimagination.com> wrote:
On 06/05/2024 8:55 AM EDT Chris Benham cbenhamau@yahoo.com.au wrote:
What do you think of Adam Tarr's old idea of using A-B-C--D-E-F
"grading" ballots with the idea of using them to infer ranking to elect
a Condorcet winner and if there isn't one then interpreting the A-B-C
grades as Approval?
Why not just A-B as Approval? Or why not A-B-C-D as Approval? Or why not
just A as Approval (I know that would result in Condorcet-Plurality). What
rhyme or reason is used to determine where the Approval cutoff is?
He liked Condorcet//Approval but also possible with these ballots is
Smith//Approval and Margins-Sorted Approval and a few others.
I don't see such a big problem with high-resolution score ballots as
long as it is made clear to the voters (and is true) that their
approximate ratings of the candidates will work just as well for them as
their exact ones.
Please take a look at these templates for RCV legislation:
https://drive.google.com/file/d/1DGvs2F_YoKcbl2SXzCcfm3nEMkO0zCbR/view .
Even though the actual legislation is not verbatim with these templates,
the IRV template (first one) has actually been used for the content of the
ballot question for an initiative for IRV. The second template has BTR-IRV.
Now consider the last three templates ("Straight-ahead Condorcet"). So
this is specifically about what to put in subdivision (4) to replace:
"(4) If no Condorcet winner exists in subdivision (3), then the candidate
having the plurality of first preferences is elected."
What exactly will the language be to replace that line of code?
"(4) If no Condorcet winner exists in subdivision (3), then ..." what?
Perhaps,
"(4) If no Condorcet winner exists in subdivision (3), then the candidate
receiving the most first and second preferences is elected."
I might call that Condorcet-Bucklin. Bucklin and Approval have something
in common and that common property is that it's "counting marks" and not
counting people. This may be perceived by some legal analysts as violating
One-person-one-vote, I dunno.
Would that be a little more resistant to the burying strategy than
Condorcet-Plurality? How about comparing it to Condorcet-TTR, which is
nearly Condorcet-IRV?
"(4) If no Condorcet winner exists in subdivision (3), then the two
candidates receiving the most first preferences shall runoff against each
other and the candidate in that pair that is preferred over the other by
the greater number of voters shall be elected."
How would MO's "wv-Condorcet" be worded? (Since he is apparently not
plonking me as was threatened.)
It needs to be short and sweet. ("Sweet" means that it ostensibly elects
the candidate with the "greatest" voter support, however that voter support
is defined. But the definition of that "greatest" voter support needs to
be simple and understandable and make sense to pedestrians.)
--
r b-j . _ . _ . _ . _ rbj@audioimagination.com
"Imagination is more important than knowledge."
.
.
.
Election-Methods mailing list - see https://electorama.com/em for list
info