Free State Project Forum

Please login or register.

Login with username, password and session length
Pages: [1] 2 3 4 ... 8   Go Down

Author Topic: getting rid of cumulative count  (Read 30537 times)

JasonPSorens

  • Administrator
  • *****
  • Offline Offline
  • Posts: 5705
  • Neohantonum liberissimum erit.
    • My Homepage
getting rid of cumulative count
« on: November 12, 2002, 12:16:53 am »

Well, I for one am finally persuaded that Cumulative Count needs to be junked in favor of an alternative.  The best alternative is Instant Runoff Voting, also called Preferential Vote or Alternative Vote or Single Transferable Vote.  It is one of the methods used in the "experiment" thread.  The way it works is simply that all voters rank the candidates, from first choice to last choice.  The candidate with the fewest first-place votes is eliminated, and the second-place votes of those voters who gave first-place votes to the eliminated candidate are then distributed as if they were first-place votes.  Then the process is repeated, until one candidate has a majority of votes.  It is pointless to vote dishonestly under this system, because there are no "wasted votes."  The system is easy to explain and the result is easy to compute.  Its advantage over rating voting is chiefly in computation, but Eddie Bradford has pointed out some paradoxes as well that could emerge in rating voting.

So my question is: Is there anyone here who would withdraw his Statement of Intent if the Participation Guidelines were changed to make Instant Runoff Voting the state vote method rather than Cumulative Count?  It seems as if most people support IRV rather than CC, but consider this an official testing of the waters.  If there's no strong opposition to replacing CC, then I'll try to get the Board to agree to do this.
« Last Edit: November 19, 2002, 10:35:45 pm by Jason P. Sorens »
Logged
"Educate your children, educate yourselves, in the love for the freedom of others, for only in this way will your own freedom not be a gratuitous gift from fate. You will be aware of its worth and will have the courage to defend it." --Joaquim Nabuco (1883), Abolitionism

PongGod

  • ****
  • Offline Offline
  • Posts: 64
  • Live free... and let the socialists die!
Re:getting rid of cumulative count
« Reply #1 on: November 12, 2002, 01:49:11 am »

The first question that came to my mind about the potential "cracks" in such a system was this:  isn't it conceivable that the candidate garnering the fewest first-place votes is actually a rather popular choice in which many (or even most) of the respondents picked it 2nd or 3rd?  In other words, it's no one's favorite, but everyone thinks its much better than average.  In many other voting systems, such a candidate could actually win, but we would be ruling it out from the start.

I realize that this scenario is rather unlikely, but I just threw out there as food for thought.
Logged
None are more hopelessly enslaved than those who falsely believe they are free. --Johann Wolfgang Von Goethe

JasonPSorens

  • Administrator
  • *****
  • Offline Offline
  • Posts: 5705
  • Neohantonum liberissimum erit.
    • My Homepage
Re:getting rid of cumulative count
« Reply #2 on: November 12, 2002, 08:43:07 am »

Yes, you're absolutely right: IRV isn't a perfect voting system (nothing is).  But its advantage is that it completely eliminates any incentive to vote strategically or dishonestly.
Logged
"Educate your children, educate yourselves, in the love for the freedom of others, for only in this way will your own freedom not be a gratuitous gift from fate. You will be aware of its worth and will have the courage to defend it." --Joaquim Nabuco (1883), Abolitionism

varrin

  • Former FSP President
  • FSP Participant
  • *****
  • Offline Offline
  • Posts: 999
  • THE air male
    • Varrin's FSP Info Page
Re:getting rid of cumulative count
« Reply #3 on: November 12, 2002, 12:37:24 pm »

I just have one question.  Will we be required to rank all 10 states or can we vote for fewer than 10?

If we can vote for fewer than 10, then strategic voting becomes more of an issue IIRC based on previous study of the system.  If we can vote for all 10, there still could theoretically be some strategic voting based on the presumed preferences of some people but it seems as though it would be somewhat reduced.

So the question remains, vote for all 10 or fewer than 10?

And what will we do with people who just insist on voting for fewer than 10?  And how many states do people who have opted out get to vote for?  That *could* raise some strategic issues (i.e. rank the leading excluded contender last to sway towards included states).

I know, nothing's perfect....

V-

Logged
Departed Fresno, PRC (Peoples Republic of California): October 18, 2004
Arrived Keene, FS (Free State!): October 25, 2004!
To contact me, please use email, not PM here.

zeeder

  • ****
  • Offline Offline
  • Posts: 24
  • I'm a chicken!
Re:getting rid of cumulative count
« Reply #4 on: November 12, 2002, 01:46:52 pm »

Quote
Personally, I'm getting the impression that we're trying to complicate the piss outta something that should be dirt simple.

   lol that was exactly what i was thinking.

   I am not knowledgable about all of the voting processes out there but I would reccommend this.
   Everyone gets one vote for what they believe would be the easiest state to free. The state that comes in last is dropped. Vote again. The state(s) that come in last are dropped. vote again. Repeat until one state is left.  This would assure the best state IMO and be alot of fun to bet on:)


Zeeder
Logged
For the average American freedom of speech is simply the freedom to repeat what everyone else is saying and no more

Sylvain Poirier

  • ****
  • Offline Offline
  • Posts: 35
    • Liberal theory of power
Re:getting rid of cumulative count
« Reply #5 on: November 12, 2002, 02:17:15 pm »

I just imagined another vote, I do not know if it is famous or not.
It is just another way of computing from the rankings, so from the same poll we could produce this other result.

Say a state A is better than a state B if a majority of people have ranked A above B.
Then define the elected state to be the state which is better than all the others.
It is unique because between two states, one is better than the other, so they cannot both be the best.

But it may not exist : among the bests there might be A > B > C > A but it is very unlikely. If it ever happened we could solve the problem by giving each one 2 points per first ranking among the three and 1 point per second ranking, and see which one has more points.

Another remark: I think voters should be allowed to give equal rank to two states, as a way to do abstention of choice between them. In this case among A,B and C one would give only 1 point to the better(s).

If we assume that there will exist two states A and B which will be better than the others, then people are not interested to cheat in their vote to influence the result between A and B because the winner only depends on how many times A is ranked higher than B.

But all these details may be pointless, as the last poll (whose method was also very good and therefore should not give a significantly different result), seemed to clearly present one state as much better than the others.
Logged

Michelle

  • FSP Participant
  • *****
  • Offline Offline
  • Posts: 748
Re:getting rid of cumulative count
« Reply #6 on: November 12, 2002, 02:21:33 pm »

I haven't really been following all the discussions about different methods of voting.

Rather than going back and trying to make sense out of members' opinions, is there some source that I could go to and read about the pros/cons of various methods?
Logged
Please join NHLA today! http://www.nhliberty.org With every new member we gain political weight to support liberty-friendly candidates and promote liberty throughout NH.
Support the Liberty Scholarship Fund. Please make a donation today! http://www.lsfund.org

JasonPSorens

  • Administrator
  • *****
  • Offline Offline
  • Posts: 5705
  • Neohantonum liberissimum erit.
    • My Homepage
Re:getting rid of cumulative count
« Reply #7 on: November 12, 2002, 02:25:17 pm »

I'm not too sure about the concept of "dishonest" or "strategic" voting really. If its the individuals vote, then its theirs to do with as they please... even to not cast it if thats what they wish.

Well, the problem is that if some people vote strategically and others do not, the strategic voters have an advantage and could cause a state to win that does not have the broadest support.

Quote
If there are restrictions on how or where one can place (or not place) said points, then its certainly not freedom of choice. It then negates the entire purpose of a free vote in the first place.

No one's talking about putting restrictions on how to cast your vote, just about moving to a new voting system that "naturally" eliminates all incentive to vote dishonestly.
Logged
"Educate your children, educate yourselves, in the love for the freedom of others, for only in this way will your own freedom not be a gratuitous gift from fate. You will be aware of its worth and will have the courage to defend it." --Joaquim Nabuco (1883), Abolitionism

JasonPSorens

  • Administrator
  • *****
  • Offline Offline
  • Posts: 5705
  • Neohantonum liberissimum erit.
    • My Homepage
Re:getting rid of cumulative count
« Reply #8 on: November 12, 2002, 02:26:40 pm »

  Everyone gets one vote for what they believe would be the easiest state to free. The state that comes in last is dropped. Vote again. The state(s) that come in last are dropped. vote again. Repeat until one state is left.  This would assure the best state IMO and be alot of fun to bet on:)

Instant Runoff Voting is *exactly* the same thing!  The only difference is that you don't have multiple rounds: by ranking the states you indicate your preference for the next "round" should your first choice be eliminated.
Logged
"Educate your children, educate yourselves, in the love for the freedom of others, for only in this way will your own freedom not be a gratuitous gift from fate. You will be aware of its worth and will have the courage to defend it." --Joaquim Nabuco (1883), Abolitionism

JasonPSorens

  • Administrator
  • *****
  • Offline Offline
  • Posts: 5705
  • Neohantonum liberissimum erit.
    • My Homepage
Re:getting rid of cumulative count
« Reply #9 on: November 12, 2002, 02:28:18 pm »

I haven't really been following all the discussions about different methods of voting.

Rather than going back and trying to make sense out of members' opinions, is there some source that I could go to and read about the pros/cons of various methods?

This thread should provide a good start:
http://forum.freestateproject.org/index.php?board=5;action=display;threadid=644
Logged
"Educate your children, educate yourselves, in the love for the freedom of others, for only in this way will your own freedom not be a gratuitous gift from fate. You will be aware of its worth and will have the courage to defend it." --Joaquim Nabuco (1883), Abolitionism

ZuG

  • FSP Participant
  • ****
  • Offline Offline
  • Posts: 313
Re:getting rid of cumulative count
« Reply #10 on: November 12, 2002, 02:55:08 pm »

Well, I for one am finally persuaded that Cumulative Count needs to be junked in favor of an alternative.  

I for one am interested in your reasoning. What finally persuaded you, and what are the potential problems?

Logged
Before you post that comment, think. The people you are talking to will soon be your neighbors. Would you talk to your neighbor in the way you are talking to those on the forum?

JasonPSorens

  • Administrator
  • *****
  • Offline Offline
  • Posts: 5705
  • Neohantonum liberissimum erit.
    • My Homepage
Re:getting rid of cumulative count
« Reply #11 on: November 12, 2002, 03:07:29 pm »

Well, what really did it for me was the demonstration that under Cumulative Count the only rational way to cast your ballot is to give all of your votes to your favorite candidate.  The problem is that if you give some to lower-ranked candidates, you're reducing the probability that your first choice will win, which you'll never want to do.  This makes Cumulative Count even worse than simple plurality rule, which is just about the worst possible system for us.  Under plurality rule, everyone gets 1 vote and the candidate with the most votes wins.  W/ CC many, perhaps most, people are giving all of their votes to a single candidate - so far, just like plurality rule.  But some people would be voting honestly and distributing their votes among all their preferred candidates.  These people would be exploited for the benefit of the dishonest voters under CC.  (I write "dishonest," but perhaps a lot of the people casting strategic ballots are not even aware of the fact that they are misrepresenting their preferences.)  So CC is like a skewed, irrational variation of plurality voting.  :(
Logged
"Educate your children, educate yourselves, in the love for the freedom of others, for only in this way will your own freedom not be a gratuitous gift from fate. You will be aware of its worth and will have the courage to defend it." --Joaquim Nabuco (1883), Abolitionism

ZuG

  • FSP Participant
  • ****
  • Offline Offline
  • Posts: 313
Re:getting rid of cumulative count
« Reply #12 on: November 12, 2002, 03:16:22 pm »

That does make sense. If you explain it that way when (if) you change the voting system, I doubt that you'll have many defectors, if any at all. Nobody wants to be fair-minded about the ranking and then get screwed by those who aren't. That the same thing that happens in current politics, we don't need any more of that.
Logged
Before you post that comment, think. The people you are talking to will soon be your neighbors. Would you talk to your neighbor in the way you are talking to those on the forum?

5pectre

  • FSP Participant
  • ****
  • Offline Offline
  • Posts: 322
  • terrorism is the symptom not the disease -1.5/-7.7
    • home
Re:getting rid of cumulative count
« Reply #13 on: November 12, 2002, 03:19:11 pm »

Well, I for one am finally persuaded that Cumulative Count needs to be junked in favor of an alternative.

I like the idea of approval voting combined with some kind of rating system.

I posted this before, but i think it is relevant to this thread:

http://www.sciencenews.org/20021102/bob8.asp
Logged

ZionCurtain

  • ****
  • Offline Offline
  • Posts: 404
  • I'm a llama!
Re:getting rid of cumulative count
« Reply #14 on: November 12, 2002, 05:23:40 pm »

Are we going to see another practice poll with this new method? I am interested in how it will affect the eastern bias. That will be the true test.
Logged
Pages: [1] 2 3 4 ... 8   Go Up