Vote Counting Amendment

[Jon Seattle]

(The following is a draft of an amendment I am posting as an individual for discussion. It is not endorsed by any party. See the following for an example: http://forums.slcds.info/viewtopic.php? ... t=45#p9791 .)

Vote Counting Amendment

This amendment is intended to correct three problems with our current voting system:

1. Zero-based Borda counts are easier to compute and understand.
2. Our current approach to counting eliminated factions has the consequence of making each successive elimination stronger for every eliminated faction, so that eliminating a faction A, has less of an impact on A’s count, than eliminating A and B.
3. The language in the constitution could be considered ambiguous about eliminating multiple factions, and even allowing elimination of all factions.

In order to correct this, the following will replace the first paragraph of article 1 section 2 of the constitution:

Representative seats are chosen by means of the Sainte-Laguë method using scores generated by zero-based Borda-counts for ranked factions. Voters will have the option to not rank any faction except for one. k eliminated factions subtracts [(3 - k) / 2] from the score of each eliminated faction.

Citizens will rank the list of candidates from their first pick faction to serve on the RA. Candidates will be ranked using Borda counts without the possibility of eliminating a candidate. The candidates with the highest Borda count within the allocation provided by the CDS-wide election will serve in the RA.

[Jon Seattle]

Several points:

The formula in this bill may seem more complicated, but how it works is very easy to understand: when you eliminate a faction you take one point away from its final Borda score. (For other numbers of points, just adjust the constant 2 in the formula.)

If you you like the idea of greater expressiveness, the ability to cross-off factions with a well defined meaning, this is the bill to support.

On the other hand if you think there should be a strong incentive to eliminate all but one faction, and that people should really vote for one faction only, you would support the current system.

if you think the system should be hard to understand and unintuitive so people will make more mistakes, then you would support the current system. Under the current system crossing off A, B, and C will take more points away from A, than just crossing off faction A.

[Jon Seattle]

There was some debate on the General forum about wether a law should be changed to deal with moral issues. This law does not do that. We could consider laws that reward, are indifferent to, or punish selfish behavior.

1. The current system gives a bonus for eliminating factions from your ballot. The more factions you eliminate, the more points are deducted from each eliminated faction.

2. There are standard variations such as Modified Borda counts (see http://en.wikipedia.org/wiki/Borda_coun ... orda_count ) that take away points if you eliminate factions from your ballot.

3. The proposed change has the benefit of neither rewarding nor punishing elimination. Every time you eliminate a faction you take away the same number of points from that faction. Its the only approach that leaves a level playing field.

[cleopatraxigalia]

Am I correct then that this assumes that any faction running for a seat is an acceptable option for the community and should have a chance? The criteria now are only that three people say they are a faction, is that correct? And is this okay with everyone?

How do we make sure a faction that the commmunity does not desire to be in office does not gain a seat with this system?

[Jon Seattle]

A really interesting question! I don’t try to address that issue with this bill. The effect of this bill will be to spread out the Borda counts for factions so it should make the seat allocation a bit less even, but I doubt it will change seat allocation by a huge amount.

The real issue here is not the Borda counts, but the Sainte-Laguë method. The Sainte-Laguë method is a system of proportional allocation (in fact it was used in the US Senate for a while, known there as Webster’s method). Its a heck of a strong leveler. With a large number of parties, it tends to give just about all parties some voice. I suspect it will work that way no matter what we do with the Borda points.

Another way of approaching this may be to require a new faction to have a certain number of voters sign a petition before a they can get on the ballot, and maintain a certain small percentage of Borda points in each election in order to stay in.