[EM] Comments on a posting from Benjamin Grant

Michael Ossipoff email9648742 at gmail.com
Mon Jan 6 08:25:53 PST 2014


On Sat, Jan 4, 2014 at 4:47 PM, Jameson Quinn <jameson.quinn at gmail.com>wrote:

>
>
>
> 2014/1/4 Michael Ossipoff <email9648742 at gmail.com>
>
>> [quote]
>> Let's say that election after election people
>> see that more and more people are voting for Nader,although he is not
>> winning.  Thinking optimistically (as some people like to) that this might
>> be the year that Nader could take it all, they put all their money on
>> Nader
>> - they vote Nader, but *not* Gore. The result? Gore's numbers drop,
>> Nader's
>> numbers rise a little, but Bush still get's the most!
>> [/quote]
>>
>> That's the old standard criicism against Approval. The answer to it
>> hasn't changed, over all the thousands of times that objection has been
>> made. Approval isn't a pairwise-count methods. Approval doesn't aspire to,
>> or claim to achieve, the pairwise-cound ideal, whereby we can help
>> Compromise against Worst, while also helping Favorite against Compromise.
>>
>> For that, we need rank-balloting.
>>
>
> I believe that this is a mistake. Yes, it is true, only in a ranked method
> can the same ballot simultaneously count full-strength for Green over Dem
> and for Dem over Repub (or of course similarly for Libertarian>Repub>Dem).
>

The rank-balloting ideal: Fully help Compromise against Worst, while still
fully helping Favorite over Compromise. With the 3 Green scenario methods
that I recommend, that ideal is actually available to mutual majority
members.



> But that's not really needed.
>

That depends. If we're willing to wait several election-cycles before
getting better government, and if we don't mind strategy-need, then, sure,
we don't need better than Approval. Approval would be good enough. But we'd
like better. And maybe some feel that we need better government as soon as
possible, because every year a lot of harm is being done. Then, all the
extra years that Approval can take, to find the voter median, might be

Of course, in the Green scenario, conditions would already be pretty good,
and immediate improvement wouldn't be as necessary, so Approval's slowness
wouldn't be as problematic. But the Approval critics are right about the
fact that if some progressives try for too much, they could split the
progresive majority, and we could sink back to Republocrat government. Why
would anyone want that voting dilmma??

So it can't really be said that the rank-balloting ideal isn't needed. I
name IRV, Benham and Woodall so often, that I'm going to refer to them as
the Green scenario three (GS3). GS3 meet MMC, and have no chicken dilemma,
thereby achieving the rank-balloting ideal for muual majority members.
Benham and Woodall additionally meet the Condorcet criterion.

MJ doesn't meet MMC, but I admit that the MJ MMC failure that I found isn't
a very bad one. But no one claims that MJ doesn't have chicken dilemma. No
one claims that MJ meets the Condorcet Criiterion. Chicken dilemma can
easily lose all of MMC's benefit. With MJ, the rank-balloting ideal isn't
available to anyone.



> All you need is a method that will count your ballot full-strength for
> whichever you prefer out of the two actual voted frontrunners (which is
> critically different, of course, from the two presumed frontrunners going
> into the election).
>
>

If you could know in advance who those two frontrunners would be, then even
Plurality would be fine. The problem is when there are more than two
candidates, and you don't know who the frontrunners are. That's when the
rank-balloting ideal is really a good thing to have.


>
>
> And I believe that in practice, median systems (Majority Approval
> Voting/MAV; Majority Judgment; Graduated Majority Judgment; Majority Choice
> Approval; ER-Bucklin; etc.) will do that.
>
>

All of those have chicken dilemma, and none achieve the rank-balloting
ideal for anyone.



> Assume Green voters give the Green candidate an A, the Dem a D, and the
> Repub an F; and Dem voters all give the Repub an F; then as long as there
> are over 50% Green+Dem, Repub will safely lose; and the minute there are
> over 50% Green, the green will win. (Again, in a right-leaning area, the
> same argument applies for a generally rightist third party such as the
> Libertarians).
>
>

I'd be a liar if I claimed to find MJ simple. You seem to be saying that MJ
meets MMC, and doesn't have chicken dilemma, and that it acheves the
rank-balloting ideal for a mutual majority, as do GS3.  But, if Repub will
lose even if Dems defect against Greens, and the Dems can thereby make Dem
win instead of Green, that's a chicken dilemma.

You say the Greens wil in they have a majority. With GS3, the Greens would
win merely if they're bigger than Dem, and {Green, Dem} have a majority.

By the way, I feel that it's most pessimistic to think that the best that
the Greens, or progressives in general, can achieve is a majority that
includes them and the Democrats. I believe that we'll find that there's
actually a majority that _don't_ want sleaze and corruption.

I said that, unlike GS3, MJ has chicken dilemma and doesn't guarantee the
rank-balloting ideal for anyone, and you haven't contradicted that claim.




>
>
> Yes, you can construct artificial scenarios where a median system does not
> elect the CW. However, the "blurrier" you make the electorate (with
> different voters within a faction casting similar but not identical
> ballots), the harder it becomes to make such scenarios come out. In
> practice, I am certain that the electorate would be "blurry" enough that
> such scenarios would be vanishingly rare.
>
>

In all of the polls that we did at EM, Approval and Score elected the CW.
So did Plurality. But we weren't a strategy-inclined electorate. Chicken
dilemma would easily spoil MJ"s results.


>
>
> Why is MAV better than Ossipoff's proposals (IRV, Benham, and Woodall)?
> For one thing, none of the 3 are summable. In practice, that is not an
> insurmountable barrier, but it does mean that software is many times harder
> to certify, and that fraud-prevention checks are similarly trickier and/or
> less reliable.
>
>

On the contrary, any voting system can be completely free of count-fraud.

All that's needed is public imaging of the ballots:

Voting is by simple paper ballots, marked with pencil, pen, punch, etc.; or
by voting machine, such as touchscreen. If it's by machine the machine
ouputs a paper ballot with the voter's ranking. The voter examines it it to
assure hirself that the ranking is as s/he intended, and then drops it in a
slot in the ballot-box.

The following takes place in front of a public crowd, and a set of
observers from a set of parties across the politica spectrum:


One at a time, each ballot is laid face-up on a table. Above the table is a
frame containing digital cameras belonging to and operated by all of the
observer parties, and one belonging to and operated by the elections
department. The ballot is stamped with a sequential order number, and then
imaged by all the digital cameras.

When all ballots have been thus imaged, they're securely moved to secure
storage, protected by locks, cameras, and alarms belonging to and
controlled by the observer parties.

Any party, and the elections department, can and must share its digital
ballot set with whoever requests it. Anyone can scan the images digitally,
and do a computer count of the ballots. If there is a dispute about one or
more ballots, then the original ballots are retrieved from storate, and
re-examined and re-imaged. Disputes are easily resolved, because the
ballots are sequentially numbered.

This system is entirely invulnerable to count-fraud, regardless of what
rank-count is used.

Motivated to find ways to criticize IRV, many have criticized its
"non-summability", making it into a problem that it isn't.

An IRV count with N candidates is, at worst, like N Plurality counts. Yes,
there'd need to be communication between precinct and Central.
Communication is likewise needed now, with Plurality. If we can do one
secure Plurality count, then we could do N secure Plurality counts.  ..or
one secure IRV count.

Some like to imply that the communication must be by insecure lines, or by
trusted courier :-)  No, it can be by publication--broadcast and website
posting.   ...in the same way that Plurality count results are published.



>
>
> For another, MAV is easier to understand.
>

If MAV is as easy to understand as MJ, the MAV isn't easy to understand. I
don't know what MAV is.

IRV is easy to understand, and is very popular. I've posted my very brief
IRV definition. Of the at least five U.S. political parties that offer a
new voting system in their platform, all of those offer IRV as that voting
system. As I said, Benham is a minimal change from IRV, additionally
offering the Condorcet Criterion.

Therefore, we already have a head-start with GS3, for the Green scenario.


>
>
> For a third, I am confident that MAV results would give significantly
> better social utility than IRV
>

...until you consider strategy. Without any strategy, with just sincere
rating, Score gives the highest social utility. But that's a utopian hope,
not realistic under any soon-forseeable conditions.


> , and slightly better than Benham and Woodall.
>

No. Not with chicken dilemma, and not when GS3 guarantee the rank-balloting
ideal for mutual majorities.

The CW is typically the SU maximizer. Benham and Woodall are a lot more
likely to elect the CW than MJ is (or MAV, if MAV ls like MJ or Approval).

At EM, in the earliest years of this millennium, or mabe even a little
before, I demonstrated that, with distances measured by city-block distance
(as opposed to Euclidean distance), the CW is particularly likely to be the
SU maximizer. I don't remember exactly what it was that I demonstrated.
Maybe I demonstrated that, with city-block distances the CW is aways the SU
maximizer.

And I claim that city-block distance is the relevant distance-measure in
issue-space.

Euclidean disance is for things which, like spatial distance components,
aren't each individually what their aggregation is, and therefore aren't
summable to get that aggregation. A candidate's issue-undesirabilities are
not in that category.

Benham and Woodall, free of chicken dilemma, and meeting MMC and the
Condorcet Criterion (CC), are particularly likely to elect the CW when
there is one--and there usually will be one.



> That is to say, MAV clearly does better with the same amount of voter
> honesty
>

Certainly not. I don't know what MAV is, but if it's like MJ or Approval,
then it can't match Benham or Woodall, or even IRV. When IRV fails to elect
the CW, it will still always choose from the innermost mutual majority set,
if that mutual majority rank sincerely.




> , and I see no reason to believe that voters would be any more honest
> under the latter systems
>

Howabout because there's no chicken dilemma, and because the full benefit
of MMC is available to members of a mutual majority.




> , and certainly not by enough to offset MAV's advantage.
>
>

A strong statement. What's MAV, and what is its advantage?



>
>
> ...
>
> Like Ossipoff, I am convinced that Approval is the first step to election
> reform
>

It probably would be, if we're talking about electoral reform under current
conditions. I've discussed why that's very unlikely to happen., Any kind of
reform, including electoral reform, can happen only _after_ we elect a
better government,  by Plurality.  ...and only after we all demand, and
get, verifiable count-results (in other words, legitimate elections, and,
thereby, genuine legitimate democracy).

After the election of a progressive government...I call that the Green
scenario (even though the progressive party needn't be the Greens
(GPUS)--It could be the Justice Party, the Pirate Party, G/GPUSA, SPUSA, or
any of the communist parties).

The first step toward electoral reform, or any reform, is electing a
progressive government, a progressive party, via Pluraity. And the obvious
default first improved voting system will be IRV, because it's what those
(at least) five political parties offer in their platforms. As I said,
Benham is such a minimal change from IRV, that Benham will be the obvious
and natural improvement on IRV.

Michael Ossipoff






>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20140106/f71303a1/attachment-0004.htm>


More information about the Election-Methods mailing list