Maine IRV proponent doesn't understand IRV
33 views
Skip to first unread message
Clay Shentrup
Jul 13, 2016, 12:37:14 PM7/13/16
to The Center for Election Science
http://ivn.us/2016/07/05/yes-5-no-lesser-two-evils-maine/
Quote: You should never have to vote for the “lesser of two evils” when there is another candidate you like better.
This person is either ignorant or lying.
Abd ul-Rahman Lomax
Jul 13, 2016, 3:43:50 PM7/13/16
to The Center for Election Science
At 12:37 PM 7/13/2016, Clay Shentrup wrote:
><http://ivn.us/2016/07/05/yes-5-no-lesser-two-evils-maine/>http://ivn.us/2016/07/05/yes-5-no-lesser-two-evils-maine/
"I don't care as long as we win."
That piece is crafted to appeal to people with
little knowledge of voting systems. It sounds great.
>Question 5 would empower Maine voters to rank
>candidates for public office. “Yes on 5” would
>give Maine voters the freedom to vote for the
>candidates they like best without worrying that
>they will help to elect the candidates they like
>least. This simple reform gives more choice and
>more voice to voters in our democracy.
FairVote has developed and honed, with extensive
experience, arguments that work. Truth would
complicate them. They are on a mission to win,
and don't actually care about anything else. Example:
http://ivn.us/2016/06/01/taxpayers-waste-millions-two-elections/
Money "wasted" on elections (runoffs) with low
turnout. But these are the voters who care. These
elections improve the quality of results, but
FairVote is not interested in "quality of
results," and they have no concept of what that would even mean.
Most people, thinking of what this campaign
manager for FairVote wrote, will think of
elections where there are two major candidates,
both "evil" and then the voter's favorite, who
has no hope to win. So, yippee! They get to vote
for their favorite, then put a clothes pin on
their nose and vote for the lesser evil. But
wait, I thought they would have to vote for the
lesser evil. That's right, they don't have to,
but this returns them to the same condition as
with plurality, only they don't get to vote in a
runoff. In plurality with a majority requirement
in the first round, they really can vote for
their favorite and not worry, because if not
voting for a frontrunner causes majority failure,
there will be a runoff. And then they can
carefully consider if it matters to them if
Tweedle Dum or Tweedle Dee are elected, and if it
does, they can vote, and if it doesn't they can leave it to poeple who care.
I.e., they don't have to go to the work of
choosing between two evils in a runoff system,
unless they decide it's important. They can vote sincerely in the first round.
Except for one problem in systems that eliminate
all but two. If write-ins are allowed, this can
be fixed, but if not, there is center squeeze.
The lesser of two evils might not make it into
the runoff, unless you vote for this one in the primary.
When there are three or more viable candidates, IRV can break down badly.
But don't confuse people with facts.
>None of Maine’s Democratic, Republican, or
>Independent governors have been elected to their
>first terms by a majority since 1966. In fact,
>less than 40% of voters have picked Maine’s
>governors in half of all of elections in the last 40 years.
>
>“Yes on 5” would restore majority rule and
>ensure that candidates who are opposed by a majority of voters could never win.
It's just not so. First of all, they play fast
and loose with "opposed." What voters do is to
express preferences. The classic democratic
position is that results should actually be
supported by a majority. IRV does very poorly at
determining such. There are much better methods.
Runoff, with its flaws, still manages this. It
truly manages it if write-ins are allowed, and it
would do even better if an advanced voting system
were used for both rounds. The problem is the
elimination of the most broadly supported
candidate in IRV because of not being among the
top two in first choices (or something similar to that, in effect).
>A victory for “Yes on 5” means that Maine’s next
>governor would be elected to his or her first
>term by a majority for the first time in more than 50 years,
IRV creates a faux majority. It does not
magically cause majorities. In races between
three or more candidates, majority failure is
common with IRV, except that they not only
eliminate candidates with not enough votes in
early rounds of counting, they also eliminate
*ballots* that don't contain votes for a
remaining candidate. These voters voted against
the IRV winner. But IRV will proclaim that there
was a "majority in the last round," and then if
you point out the problem, they will say that it
was your fault, you should have voted. But, wait,
I should have voted for a candidate I detest? I
thought you told me I wouldn't have to do that!
They would say, then, that you'd face the same
choice in a runoff. Yes. and then one can really
consider it, see the new arguments in a narrowed
race, it is far more democratic. And if I can
write in a candidate, I can still vote for my
favorite, and if an advanced method is used in
the runoff *I could vote for my favorite and for
a frontrunner.* Or not, as I choose.
FairVote is selling garbage, calling it caviar.
And when there is a failure, as happens with IRV
from time to time, where voters realize they
would have had a better result if they hadn't
voted, they will toss out the method and there
goes election reform for a long time. They won't
mention that it's expensive to canvass IRV. They
won't mention the high ballot spoilage rates. Etc.
There is a far simpler reform that is practically
cost free which would improve the system.
Approval Voting. There is the older ranked
system, Bucklin voting, which, combined with
runoffs with majority failure in the primary,
would make far better choices and is easy to vote and canvass. Etc.
What is needd in states is a process to study
voting systems and generate recommendations, and
that process should be carefully managed to
generate the best advice. The League of Women
Voters ran something like that, but without care,
and obviously manipulated by FairVote. And
supposedly LWV is nonpartisan, but not on this issue, obviously!
><http://ivn.us/2016/07/05/yes-5-no-lesser-two-evils-maine/>http://ivn.us/2016/07/05/yes-5-no-lesser-two-evils-maine/
>
>
>Quote: You should never have to vote for the
>“lesser of two evils†when there is another candidate you like better.
>
>Quote: You should never have to vote for the
>
>This person is either ignorant or lying.
I vote for ignorant combined with a heavy dose of
>This person is either ignorant or lying.
"I don't care as long as we win."
That piece is crafted to appeal to people with
little knowledge of voting systems. It sounds great.
>Question 5 would empower Maine voters to rank
>candidates for public office. “Yes on 5” would
>give Maine voters the freedom to vote for the
>candidates they like best without worrying that
>they will help to elect the candidates they like
>least. This simple reform gives more choice and
>more voice to voters in our democracy.
FairVote has developed and honed, with extensive
experience, arguments that work. Truth would
complicate them. They are on a mission to win,
and don't actually care about anything else. Example:
http://ivn.us/2016/06/01/taxpayers-waste-millions-two-elections/
Money "wasted" on elections (runoffs) with low
turnout. But these are the voters who care. These
elections improve the quality of results, but
FairVote is not interested in "quality of
results," and they have no concept of what that would even mean.
Most people, thinking of what this campaign
manager for FairVote wrote, will think of
elections where there are two major candidates,
both "evil" and then the voter's favorite, who
has no hope to win. So, yippee! They get to vote
for their favorite, then put a clothes pin on
their nose and vote for the lesser evil. But
wait, I thought they would have to vote for the
lesser evil. That's right, they don't have to,
but this returns them to the same condition as
with plurality, only they don't get to vote in a
runoff. In plurality with a majority requirement
in the first round, they really can vote for
their favorite and not worry, because if not
voting for a frontrunner causes majority failure,
there will be a runoff. And then they can
carefully consider if it matters to them if
Tweedle Dum or Tweedle Dee are elected, and if it
does, they can vote, and if it doesn't they can leave it to poeple who care.
I.e., they don't have to go to the work of
choosing between two evils in a runoff system,
unless they decide it's important. They can vote sincerely in the first round.
Except for one problem in systems that eliminate
all but two. If write-ins are allowed, this can
be fixed, but if not, there is center squeeze.
The lesser of two evils might not make it into
the runoff, unless you vote for this one in the primary.
When there are three or more viable candidates, IRV can break down badly.
But don't confuse people with facts.
>None of Maine’s Democratic, Republican, or
>Independent governors have been elected to their
>first terms by a majority since 1966. In fact,
>less than 40% of voters have picked Maine’s
>governors in half of all of elections in the last 40 years.
>
>“Yes on 5” would restore majority rule and
>ensure that candidates who are opposed by a majority of voters could never win.
It's just not so. First of all, they play fast
and loose with "opposed." What voters do is to
express preferences. The classic democratic
position is that results should actually be
supported by a majority. IRV does very poorly at
determining such. There are much better methods.
Runoff, with its flaws, still manages this. It
truly manages it if write-ins are allowed, and it
would do even better if an advanced voting system
were used for both rounds. The problem is the
elimination of the most broadly supported
candidate in IRV because of not being among the
top two in first choices (or something similar to that, in effect).
>A victory for “Yes on 5” means that Maine’s next
>governor would be elected to his or her first
>term by a majority for the first time in more than 50 years,
IRV creates a faux majority. It does not
magically cause majorities. In races between
three or more candidates, majority failure is
common with IRV, except that they not only
eliminate candidates with not enough votes in
early rounds of counting, they also eliminate
*ballots* that don't contain votes for a
remaining candidate. These voters voted against
the IRV winner. But IRV will proclaim that there
was a "majority in the last round," and then if
you point out the problem, they will say that it
was your fault, you should have voted. But, wait,
I should have voted for a candidate I detest? I
thought you told me I wouldn't have to do that!
They would say, then, that you'd face the same
choice in a runoff. Yes. and then one can really
consider it, see the new arguments in a narrowed
race, it is far more democratic. And if I can
write in a candidate, I can still vote for my
favorite, and if an advanced method is used in
the runoff *I could vote for my favorite and for
a frontrunner.* Or not, as I choose.
FairVote is selling garbage, calling it caviar.
And when there is a failure, as happens with IRV
from time to time, where voters realize they
would have had a better result if they hadn't
voted, they will toss out the method and there
goes election reform for a long time. They won't
mention that it's expensive to canvass IRV. They
won't mention the high ballot spoilage rates. Etc.
There is a far simpler reform that is practically
cost free which would improve the system.
Approval Voting. There is the older ranked
system, Bucklin voting, which, combined with
runoffs with majority failure in the primary,
would make far better choices and is easy to vote and canvass. Etc.
What is needd in states is a process to study
voting systems and generate recommendations, and
that process should be carefully managed to
generate the best advice. The League of Women
Voters ran something like that, but without care,
and obviously manipulated by FairVote. And
supposedly LWV is nonpartisan, but not on this issue, obviously!
Kevin Baas
Jul 13, 2016, 5:21:46 PM7/13/16
to The Center for Election Science
The advantage of transferable vote systems such as STV (or IRV in the single winner case), is that if you vote for a candidate that can't win, your vote can still get transfered to e.g. a second or third choice, and this still count for something.
This allows you to list your favorite candidate first and the lesser of two evils second, without worrying about your vote not counting.
This is not to say that STV or IRV passes the favorite betrayal criterion. There are still cases where voting for the lesser of two evils is the better strategy. But because, unlike score-voting and , well, every non-transferable vote method, it passes later-no-harm, it is possible to vote for a second choice without harming your first choice.
So while a change in expression of preference can effect the winner in that direction (As it should, to be pareto efficient), the ability to transfer votes dramatically reduces the likilihood of your vote being wasted, etc.
he may have been off by a subtle mistake, but the jist of what he's saying is absolutely true and very important for selecting a tallying method.
Brian Olson
Jul 14, 2016, 2:59:18 PM7/14/16
to electio...@googlegroups.com
"""You should never have to vote for the “lesser of two evils” when there is another candidate you like better."""
This is straight up true.
Trying to achieve it with IRV is severely suboptimal.
--
You received this message because you are subscribed to the Google Groups "The Center for Election Science" group.
To unsubscribe from this group and stop receiving emails from it, send an email to electionscien...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Reply all
Reply to author
Forward
0 new messages