Electoral College Reform starting with one or a few states

[Mike Ossipoff]

donald at mich.com writes:
> 
> 
> Mike writes:
> >When you speak of transferring the electoral votes of dropped
> >unwinnable candidates, that implies that you've given them
> >electoral votes. There's no need to give electoral votes to anyone
> >, even temporarily, before the choice process is completed.
> >
> 
> Dear Mike,
> It is necessary to divide each state's electoral votes according to the
> state's popular vote on each candidate. This is the way I am able to

No that isn't necessary. The reform states could simply give all
their electoral votes to their favorite winnable candidate, as
determined by the count of voter rankings in those states. That
was Steve's proposal.

> preserve the electoral vote ratio to population for each state.
> 
> I know that DEMOREP1 is in favor of doing away with the U.S.Senate and the
> two electoral votes that each state receives - but I am not going into that
> dispute in my presentation of the topic Electoral College Reform in one or
> a few states. The plan that I have presented fits into existing conditions.
> 
> The balance of your post deals with which is the best single-winner method
> to use - another dispute that I have not gone into in my presentation. I

No, I merely said that your procedure implies commitment to a particular
count method (Instant Runoff), even though you say you haven't gone
into that issue. If not Instant Runoff, then some hybridization of
it, since vote transfers are part of your procedure.

> made it clear that it is up to the Reformed States to decide which
> single-winner method to use.

Of course. Who else would decide that. But surely it will be ok to
suggest a method to them.
 
> 
> What I have presented is the order of a number of steps which need to be
> taken before using some single-winner method. You are free to advocate

Wrong. The steps that you've presented don't need to be taken. For instance
the initial proposal in this discussion doesn't take the steps you
list. In fact, some of the steps you advocate can't really be justified
in terms of the simple goal of electing the best president (as determined
by their sw count) that the reform states can elect. The task is so simple,
but you insist on adding procedure that's irrrelevant to, & contrary to,
that simple goal.

> whichever.



> - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
> Now to something else - Combinations and Vote-Sums
> 
> I never said nor intended that we must compute combinations that have no
> votes. If a combination has no votes it does not have a Vote-Sum. Only
> combinations with Vote-Sums are to be considered.
> 
> Having said that - I want to spell out two mathmatical limits on the number
> of Vote-Sums.
> 
> One: There cannot be more Vote-Sums than people that voted in an election.
> Two: There cannot be more Vote-Sums than the number of possible combinations.

Which would be astronomical in an election with, say, 15 candidates,
as discussed earlier.

> 
>      There cannot be more but there can be less - maybe a lot less.
> 
> Less - because thousands of people will happen, for any reason, to vote the
> same combinations.
> 
> Have no fear - we will not be facing Vote-Sums in the trillions and trillions.
> 
> This discussion started when I complained that it was unreal for the voters
> to only cast votes for three of fifteen possible combinations. It becomes

My point was that, if you insist on all 9 (not 15) effective possible
rankings (not counting equally-ranked candidates) in the 3-candidate
race, then how can you not also insist on a 15-candidate example not
also including all of its possible rankings--all 3.5546 trillion of
them (again, not counting equal ranking).


> more unreal when we realize that in order for this to happen the voters for
> the same candidate on the first pick would all vote lock-step for the same
> candidate on the second pick and then again vote lock-step for the same
> candidate on the third pick - very unbelivable. And this was supposed to
> happen with all three blocs of voters. Your example is reproduced below:
> 
>                       40 ABC
>                       25 BAC
>                       35 CBA
> 
> This example is not a valid example - it is suspect.

No lock-step needed. As I said, but you apparently missed it,
every progressive I talk to agrees that Clinton is probably at
least marginally better than Dole. Even I agree to that, on a
few token, cheap-talk issues. If the Democrats intentionally 
position themselves a tiny bit closer to the progressives than
the Republicans are, then no lockstep is needed to make the Democrat
the 2nd choice of the progressives. The same would hold true
in the rankings of the Republican voters, to whom the Democrat
would seem less-Republican than the Republican, but not as
un-Republican as the progressive. 

Have I succeeded in explaining that to you yet??

As for my examples in which all the Clinton voters vote the same,
I've made it clear that that's only for simplicity. I've pointed
out that in my Dole-voter truncation examples, Clinton wins by
Condorcet's method no matter whom the Clinton voters rank 2nd, or
if they don't rank anyone 2nd.

And I've pointed out that, if the Clinton voters divide their vote
equally between Dole & Nader, for 2nd choice, then the effect in
Instant-Runoff, is sthe same as if they didn't list a 2nd choice,
and that justifies the simplification of not having them list
a 2nd choice in the example.

If Dole is bigger than Nader, and if we can't say that the Democrat
voters would favor Nader over Clinton, then it's fair enough to
have them, in the example, divide their votes equally between Dole
& Nader. Actually, they'd be more likely list Dole 2nd, because
Dole is many times closer to Clinton, morally, ethically, & 
in his policies, than Nader is. I'm doing you & Instant Runoff a
favor when I divide the Clinton voters' 2nd choice vote equally
between Dole & Nader (or, equivalently, don't list a 2nd choice for
the Clinton voters). And simplfying the example, without making it
unrealistic, except in the sense that simplicity itself is unrealistic,
since there could be lots of candidates. But as I said, your Instant
Runoff, for instance, can & will easily fail even when there are just
3 candidates, so there's no need for more in the example. Feel free to
write a bigger example if you want to. No one's stopping you.


> 
> Yours,
> 
> Donald of New Democracy http://www.mich.com/~donald
> 
> 
> .-

Mike


> 


--