new simple legal strategy to get IRV
On 11 Oct 2015, at 00:10, robert bristow-johnson rbj@audioimagination.com wrote:
On 10/9/15 4:00 AM, Juho Laatu wrote:
I just note that there can be also simple ballots like in https://en.wikipedia.org/wiki/Open_list (see the picture). If one wants to expand that to ranked ballots with high number of candidates, one could simply allow the voter to write multiple numbers in the ballot instead of only one. One could thus cast a ranked vote by writing few numbers, e.g. "23 74 74 5 234 321".
and a machine is gonna scan that?
Manual scan by default. In Finland the open list ballots (bullet votes) are counted today by hand right after the polling station closes (within say 1 hour). I'd propose to do the same also with the (extended) ranked ballots. With ranked ballots you need to introduce also computers to store the ranked data, and the process takes a bit longer.
Machine scanning could be used too if it is reliable enough. Maybe so that machines use human help for ballots whose interpretation is not obvious.
Juho
On 10/9/15 6:43 AM, Kristofer Munsterhjelm wrote:
On 10/09/2015 10:08 AM, Clinton Mead wrote:
How did the 2009 Burlington Vermont Mayoral "fail"?
this sends me back to 2010 when i first joined this list. i will try to
itemize the basic items below.
It was one of the rare cases where it didn't elect the Condorcet winner,
well, it was the second time IRV was used. perhaps a fluke but one
might think that the method is problematic with a higher probability
than "fluke" if it breaks down on the 2nd try.
but
unlike plurality, at least it didn't elect the Condorcet loser out of
the three strongest candidates. IRV in this case gave a more
representative result than would have been the case with plurality.
but it still elected a candidate (with a 252 vote margin) when 587
more voters expressed on their ballots that they preferred someone
else. that is simply a failure to elect the majority voter preference.
And in almost all cases IRV does this.
I'd say it didn't go far enough. IRV works like a patch on Plurality
that lets it discard fringe parties that can't win anyway. But when you
get multiple large parties, the patch isn't sufficient and weird
behavior like center squeeze happens.
exactly. the Vermont Progressive Party was literally founded by Bernie
Sanders (ever hear of him on the east side of the pond?) even though
Bernie continued to identify himself as "Independent". it's funny now
that he's running against Hillary for the Democratic party nomination.
anyway, the Vermont Progs have at least a dozen people elected to the
State House and a half-dozen councilors elected to the Burlington city
council (besides having two Burlington mayors elected since Bernie).
IRV is more likely to fail when we get closer to the ideal (at least my
ideal) of multiple viable parties and of viable independent
candidates. that's what happened in Burlington in 2009.
But that's me :-)
Perhaps there's a more strategic explanation for the repeal itself. IRV
clearly wouldn't have pleased the Plurality supporters; they'd have
preferred seeing Wright win. And IRV wouldn't have pleased those who
found its outcome counterintuitive in the other direction; they'd have
preferred seeing Montroll win.
it's because 587 more voters marked their ballots that they preferred
Montroll over Kiss than there were voters that preferred the oppoisite.
and Montroll really beat Wright by 930 votes (but Kurt just does not get
it and still calls Andy a "weak candidate").
IRV might have held against either side,
but it couldn't hold against both at the same time.
I'd like to emphasize that I'm speculating, though. I don't live in
Burlington; someone who does would probably know the situation better
than I do.
it is true that many people in Burlington did not understand exactly
what went wrong in 2009, but they knew something was wrong.
here are my concise reasons for why IRV failed. first here are the
reasons we adopted IRV in the first place:
1.If a majority (not just a mere plurality) of voters agree that
candidate A is better than candidate B, then candidate B should not be
elected.
2.The relative merit of candidates A and B is not affected by the
presence of a third candidate C.If a majority (not just a mere
plurality) of voters agree that candidate A is better than B, whether
candidate C enters the race or not, indeed whether candidate C is better
(in the minds of voters) than either candidates A or B (or both or
neither), it does not reverse the preference of candidate A over
candidate B.If that relative preference of candidate is not affected
among voters, then the relative outcome of the election should not be
affected (candidate B winning over candidate A).In the converse, this
means that by removing anyloser from the race and from all ballots,
that this should not alter who the winner is.
3.Voters should not be called upon to do “strategic voting”.Voters
should feel free to simply vote their conscience and vote for the
candidates they like best, without worrying about whom that they think
is most electable.Voters should be able to vote for the candidate of
their choosing (e.g. Perot in 1992 or Nader in 2000) without risk of
contributing to the election of the candidate they leastprefer
(perhaps Clinton in 1992 or Bush in 2000).They should not have to
sacrifice their vote for their ideal choice because they are concerned
about “wasting” their vote and helping elect the candidate they dislike
the most.An ancillary concern; a candidate should not have to worry
about electing his/her most strident opponent by choosing to run against
another more acceptable opponent.
4.Election policy that decreases convenience for voters will decrease
voter participation.Having to vote once for your preferred candidate,
and then being called on to return to the polls at a later date and vote
again for your preferred candidate (if he/she makes it to the run-off)
is decidedly less convenient and we must expect that significantly fewer
voters will show up for the run-off.Electing candidates with decreased
legitimate voter participation cannot be considered as democratic or as
indicative of the will of the people, as electing candidates with higher
voter participation.This is in keeping with the same motivation as
“Motor Voter” and voter registration efforts; increasing voter
participation and giving more citizens a stake in who is elected to office.
IRV in Burlington 2009 failed criteria #1, #2, and #3. because there
was no runoff, i guess it satisfied criterion #4.
even though this was not listed, IRV in Burlington 2009 failed
Monotonicity and IRV never satisfies "Precinct Summability" which i
consider important for transparency and election integrity. as
elections get larger (like statewide or nationwide), this Precinct
Summability becomes even more important.
--
r b-j rbj@audioimagination.com
"Imagination is more important than knowledge."
On 10/10/15 5:30 PM, Juho Laatu wrote:
On 11 Oct 2015, at 00:10, robert bristow-johnsonrbj@audioimagination.com wrote:
On 10/9/15 4:00 AM, Juho Laatu wrote:
I just note that there can be also simple ballots like in https://en.wikipedia.org/wiki/Open_list (see the picture). If one wants to expand that to ranked ballots with high number of candidates, one could simply allow the voter to write multiple numbers in the ballot instead of only one. One could thus cast a ranked vote by writing few numbers, e.g. "23 74 74 5 234 321".
and a machine is gonna scan that?
Manual scan by default. In Finland the open list ballots (bullet votes) are counted today by hand right after the polling station closes (within say 1 hour). I'd propose to do the same also with the (extended) ranked ballots. With ranked ballots you need to introduce also computers to store the ranked data, and the process takes a bit longer.
Machine scanning could be used too if it is reliable enough. Maybe so that machines use human help for ballots whose interpretation is not obvious.
listen, here in Vermont, i have worked on a few different election
recounts where we had to examine thousands of ballots by hand. many
people mark even bubble or bullet ballots so poorly that even that fails
machine scanning. i know there is Optical Character Recognition (OCR),
but i would not trust that to recognize numbers written by voters on
ballots. numbers like "3" and "8" get confused. and other pairs, like
"1" and "7" and "5" and "8". or "5" and "6".
--
r b-j rbj@audioimagination.com
"Imagination is more important than knowledge."
On 11 Oct 2015, at 00:43, robert bristow-johnson rbj@audioimagination.com wrote:
On 10/10/15 5:30 PM, Juho Laatu wrote:
On 11 Oct 2015, at 00:10, robert bristow-johnsonrbj@audioimagination.com wrote:
On 10/9/15 4:00 AM, Juho Laatu wrote:
I just note that there can be also simple ballots like in https://en.wikipedia.org/wiki/Open_list (see the picture). If one wants to expand that to ranked ballots with high number of candidates, one could simply allow the voter to write multiple numbers in the ballot instead of only one. One could thus cast a ranked vote by writing few numbers, e.g. "23 74 74 5 234 321".
and a machine is gonna scan that?
Manual scan by default. In Finland the open list ballots (bullet votes) are counted today by hand right after the polling station closes (within say 1 hour). I'd propose to do the same also with the (extended) ranked ballots. With ranked ballots you need to introduce also computers to store the ranked data, and the process takes a bit longer.
Machine scanning could be used too if it is reliable enough. Maybe so that machines use human help for ballots whose interpretation is not obvious.
listen, here in Vermont, i have worked on a few different election recounts where we had to examine thousands of ballots by hand. many people mark even bubble or bullet ballots so poorly that even that fails machine scanning. i know there is Optical Character Recognition (OCR), but i would not trust that to recognize numbers written by voters on ballots. numbers like "3" and "8" get confused. and other pairs, like "1" and "7" and "5" and "8". or "5" and "6".
I'd trust humans to be the final judges. Some votes will be rejected because of unclear numbers (or because of additional markings in the ballots), but that has not caused any problems. It is possible that one or two seats will change in the second final check (within next one or two days), but there are never any complaints about the process or agreed outcome. The votes are counted locally right after the election (and checked second time and final results announced later) in the presence of representatives of multiple parties, so there will not be any systematic bias in any direction.
Juho
--
r b-j rbj@audioimagination.com
"Imagination is more important than knowledge."
Election-Methods mailing list - see http://electorama.com/em for list info
On 10/10/2015 11:43 PM, robert bristow-johnson wrote:
On 10/10/15 5:30 PM, Juho Laatu wrote:
On 11 Oct 2015, at 00:10, robert
bristow-johnsonrbj@audioimagination.com wrote:
On 10/9/15 4:00 AM, Juho Laatu wrote:
I just note that there can be also simple ballots like in
https://en.wikipedia.org/wiki/Open_list (see the picture). If one
wants to expand that to ranked ballots with high number of
candidates, one could simply allow the voter to write multiple
numbers in the ballot instead of only one. One could thus cast a
ranked vote by writing few numbers, e.g. "23 74 74 5 234 321".
and a machine is gonna scan that?
Manual scan by default. In Finland the open list ballots (bullet
votes) are counted today by hand right after the polling station
closes (within say 1 hour). I'd propose to do the same also with the
(extended) ranked ballots. With ranked ballots you need to introduce
also computers to store the ranked data, and the process takes a bit
longer.
Machine scanning could be used too if it is reliable enough. Maybe so
that machines use human help for ballots whose interpretation is not
obvious.
listen, here in Vermont, i have worked on a few different election
recounts where we had to examine thousands of ballots by hand. many
people mark even bubble or bullet ballots so poorly that even that fails
machine scanning. i know there is Optical Character Recognition (OCR),
but i would not trust that to recognize numbers written by voters on
ballots. numbers like "3" and "8" get confused. and other pairs, like
"1" and "7" and "5" and "8". or "5" and "6".
If you have check digits or checksums, that gets rid of most silent
failures; how many they get rid of depends on what kind of errors you
want to correct and how much redundancy you're willing to add to the ID
- e.g using five-digit codes for ten candidates can handle more errors
than using four-digit ones.
The machine could then signal an error if the ID doesn't read as valid
to its software, and then have a manual counter (or multiple) verify the
ID by hand.
I don't know how good handwriting recognition is at the moment, though;
perhaps it would have to be all manual. But there are at least ways to
avoid silent errors like one candidate being mistaken for another. I
also imagine one could not use digits that look too alike, for instance
using only the digits {1, 2, 4, 6, 8, 9}.
On 10/10/15 5:57 PM, Juho Laatu wrote:
On 11 Oct 2015, at 00:43, robert bristow-johnsonrbj@audioimagination.com wrote:
On 10/10/15 5:30 PM, Juho Laatu wrote:
On 11 Oct 2015, at 00:10, robert bristow-johnsonrbj@audioimagination.com wrote:
On 10/9/15 4:00 AM, Juho Laatu wrote:
I just note that there can be also simple ballots like in https://en.wikipedia.org/wiki/Open_list (see the picture). If one wants to expand that to ranked ballots with high number of candidates, one could simply allow the voter to write multiple numbers in the ballot instead of only one. One could thus cast a ranked vote by writing few numbers, e.g. "23 74 74 5 234 321".
and a machine is gonna scan that?
Manual scan by default. In Finland the open list ballots (bullet votes) are counted today by hand right after the polling station closes (within say 1 hour). I'd propose to do the same also with the (extended) ranked ballots. With ranked ballots you need to introduce also computers to store the ranked data, and the process takes a bit longer.
Machine scanning could be used too if it is reliable enough. Maybe so that machines use human help for ballots whose interpretation is not obvious.
listen, here in Vermont, i have worked on a few different election recounts where we had to examine thousands of ballots by hand. many people mark even bubble or bullet ballots so poorly that even that fails machine scanning. i know there is Optical Character Recognition (OCR), but i would not trust that to recognize numbers written by voters on ballots. numbers like "3" and "8" get confused. and other pairs, like "1" and "7" and "5" and "8". or "5" and "6".
I'd trust humans to be the final judges.
in a recount, yes. but ballots should not be designed to preclude
machine tabulation in normal use. and i don't think that OCR technology
is good enough presently to rely on for governmental elections where the
voting population is large. when elections come out so close that we
wonder who the real winner is, then yes, there should be careful manual
recounts mandated by law with procedures established in advance and put
into law.
--
r b-j rbj@audioimagination.com
"Imagination is more important than knowledge."
On 10/10/2015 11:14 PM, robert bristow-johnson wrote:
On 10/9/15 6:28 AM, Kristofer Munsterhjelm wrote:
"Place a mark on the line corresponding to the rating you want to give.
Further to the right is better."
Candidate 1 worst |-------------------------------------------| best
Candidate 2 worst |-------------------------------------------| best
Candidate 3 worst |-------------------------------------------| best
Candidate 4 worst |-------------------------------------------| best
...
Candidate n worst |-------------------------------------------| best
Score Voting (a.k.a. Range Voting). Warren Smith will like it. it's
what judges at the winter Olympics use for figure skating (and some
other events). but my question continues to be: "how highly do you
score your second choice?" and third choice?
(Condorcet is better.)
That's part of the reason I prefer median ratings, because there's much
less incentive to exaggerate.
But if you want Range and don't like strategy, there's always this
DSV variant Warren has referred to in the past.
- Rearrange the ballots in random order and create a new Range election.
- Go down the list from first to last ballot.
2.1. Making a transformed version of the current ballot according to the
best Range strategy given the ballots that have already been entered
into the Range election.
2.2. Entering the transformed copy into the election. - Note who wins this Range election, or alternately note the outcome
(total score for each). - Repeat from 1 a very large number of times.
- Designate the candidate who won most often as the winner. Or
alternately, perform an ordinary Range election with each iteration's
outcome (from 3) as Range ballots.
The ballot format above could be used for rankings as well (if you're
willing to disregard equal rank) in the same way that you can turn any
rated ballot into a ranked ballot by sorting the candidates by their rating.
I believe computer recognition might work well enough to at least provide assistance to the humans by helping them recording the numbers. For example I might read the numbers myself first, and then check if the computer has the same opinion. If yes, then I press enter to proceed to the next ballot and accept computer's interpretation of the ballot. This would be "computer assisted counting".
I note that some ballots might contain numbers 1 and 7, and another ballot might contain numbers 1 and 7 as well, but the first 1 and the latter 7 might look exactly the same. One would need to make the interpretation based on seeing also the other numbers on the ballot. A different looking 7 would cause the 1 to be interpreted as 1 in the first ballot. The machines should be smart enough to take also this into account. Or alternatively they would just give up and leave the interpretation of these two ballots to humans.
Juho
On 11 Oct 2015, at 01:00, Kristofer Munsterhjelm km_elmet@t-online.de wrote:
On 10/10/2015 11:43 PM, robert bristow-johnson wrote:
On 10/10/15 5:30 PM, Juho Laatu wrote:
On 11 Oct 2015, at 00:10, robert
bristow-johnsonrbj@audioimagination.com wrote:
On 10/9/15 4:00 AM, Juho Laatu wrote:
I just note that there can be also simple ballots like in
https://en.wikipedia.org/wiki/Open_list (see the picture). If one
wants to expand that to ranked ballots with high number of
candidates, one could simply allow the voter to write multiple
numbers in the ballot instead of only one. One could thus cast a
ranked vote by writing few numbers, e.g. "23 74 74 5 234 321".
and a machine is gonna scan that?
Manual scan by default. In Finland the open list ballots (bullet
votes) are counted today by hand right after the polling station
closes (within say 1 hour). I'd propose to do the same also with the
(extended) ranked ballots. With ranked ballots you need to introduce
also computers to store the ranked data, and the process takes a bit
longer.
Machine scanning could be used too if it is reliable enough. Maybe so
that machines use human help for ballots whose interpretation is not
obvious.
listen, here in Vermont, i have worked on a few different election
recounts where we had to examine thousands of ballots by hand. many
people mark even bubble or bullet ballots so poorly that even that fails
machine scanning. i know there is Optical Character Recognition (OCR),
but i would not trust that to recognize numbers written by voters on
ballots. numbers like "3" and "8" get confused. and other pairs, like
"1" and "7" and "5" and "8". or "5" and "6".
If you have check digits or checksums, that gets rid of most silent
failures; how many they get rid of depends on what kind of errors you
want to correct and how much redundancy you're willing to add to the ID
- e.g using five-digit codes for ten candidates can handle more errors
than using four-digit ones.
The machine could then signal an error if the ID doesn't read as valid
to its software, and then have a manual counter (or multiple) verify the
ID by hand.
I don't know how good handwriting recognition is at the moment, though;
perhaps it would have to be all manual. But there are at least ways to
avoid silent errors like one candidate being mistaken for another. I
also imagine one could not use digits that look too alike, for instance
using only the digits {1, 2, 4, 6, 8, 9}.
Election-Methods mailing list - see http://electorama.com/em for list info
As already obvious from my previous mail (reply to Kristofer Munsterhjelm), I think cooperation of humans and computers could be used to make the process fast and also accurate (if we trust humans to be accurate when computers give up trying to interpret some of the tricky ballots).
Juho
On 11 Oct 2015, at 01:06, robert bristow-johnson rbj@audioimagination.com wrote:
On 10/10/15 5:57 PM, Juho Laatu wrote:
On 11 Oct 2015, at 00:43, robert bristow-johnsonrbj@audioimagination.com wrote:
On 10/10/15 5:30 PM, Juho Laatu wrote:
On 11 Oct 2015, at 00:10, robert bristow-johnsonrbj@audioimagination.com wrote:
On 10/9/15 4:00 AM, Juho Laatu wrote:
I just note that there can be also simple ballots like in https://en.wikipedia.org/wiki/Open_list (see the picture). If one wants to expand that to ranked ballots with high number of candidates, one could simply allow the voter to write multiple numbers in the ballot instead of only one. One could thus cast a ranked vote by writing few numbers, e.g. "23 74 74 5 234 321".
and a machine is gonna scan that?
Manual scan by default. In Finland the open list ballots (bullet votes) are counted today by hand right after the polling station closes (within say 1 hour). I'd propose to do the same also with the (extended) ranked ballots. With ranked ballots you need to introduce also computers to store the ranked data, and the process takes a bit longer.
Machine scanning could be used too if it is reliable enough. Maybe so that machines use human help for ballots whose interpretation is not obvious.
listen, here in Vermont, i have worked on a few different election recounts where we had to examine thousands of ballots by hand. many people mark even bubble or bullet ballots so poorly that even that fails machine scanning. i know there is Optical Character Recognition (OCR), but i would not trust that to recognize numbers written by voters on ballots. numbers like "3" and "8" get confused. and other pairs, like "1" and "7" and "5" and "8". or "5" and "6".
I'd trust humans to be the final judges.
in a recount, yes. but ballots should not be designed to preclude machine tabulation in normal use. and i don't think that OCR technology is good enough presently to rely on for governmental elections where the voting population is large. when elections come out so close that we wonder who the real winner is, then yes, there should be careful manual recounts mandated by law with procedures established in advance and put into law.
--
r b-j rbj@audioimagination.com
"Imagination is more important than knowledge."
Election-Methods mailing list - see http://electorama.com/em for list info
On 10/10/15 6:11 PM, Kristofer Munsterhjelm wrote:
On 10/10/2015 11:14 PM, robert bristow-johnson wrote:
On 10/9/15 6:28 AM, Kristofer Munsterhjelm wrote:
"Place a mark on the line corresponding to the rating you want to give.
Further to the right is better."
Candidate 1 worst |-------------------------------------------| best
Candidate 2 worst |-------------------------------------------| best
Candidate 3 worst |-------------------------------------------| best
Candidate 4 worst |-------------------------------------------| best
...
Candidate n worst |-------------------------------------------| best
Score Voting (a.k.a. Range Voting). Warren Smith will like it. it's
what judges at the winter Olympics use for figure skating (and some
other events). but my question continues to be: "how highly do you
score your second choice?" and third choice?
(Condorcet is better.)
That's part of the reason I prefer median ratings, because there's much
less incentive to exaggerate.
But if you want Range and don't like strategy, there's always this
DSV variant Warren has referred to in the past.
- Rearrange the ballots in random order and create a new Range election.
- Go down the list from first to last ballot.
2.1. Making a transformed version of the current ballot according to the
best Range strategy given the ballots that have already been entered
into the Range election.
2.2. Entering the transformed copy into the election. - Note who wins this Range election, or alternately note the outcome
(total score for each). - Repeat from 1 a very large number of times.
- Designate the candidate who won most often as the winner. Or
alternately, perform an ordinary Range election with each iteration's
outcome (from 3) as Range ballots.
and what legislative body are we going to convince to adopt that?
The ballot format above could be used for rankings as well (if you're
willing to disregard equal rank) in the same way that you can turn any
rated ballot into a ranked ballot by sorting the candidates by their rating.
i still don't think it answers my question. for the voter motivated to
see his/her first choice getting elected, but wants to cover his/her ass
with a contingency vote, the question remains: "How highly does one
score their second choice? And third choice?"
it's directly answered with a Ranked-Order Ballot, but not with a Score
Ballot nor with the Approval Ballot. this is why i just can't jump on
Warren's bandwagon. with ranked-choice voting, the voter knows
directly what to do if the voter trusts that the tabulation method
does not do something goofy with the marked ballots. if the voter finds
out later that ranking or rating his/her second choice too high hurt
his/her first choice, that voter is still left with tactical thinking in
the voting booth.
--
r b-j rbj@audioimagination.com
"Imagination is more important than knowledge."