Well, here's my thing, if someone did vote Debra > RON > Far > Tsu, why should they vote that way? Simply, it's Debra > RON if you do that. I don't understand why we should treat RON as a "candidate" and eliminate it early on. In fact, that option should exist even down to the final two candidates to determine if a majority agree that we should reopen nominations.

So, this makes sense, but legislatively it would take someone far smarter than me to implement.

Anyhow, the problem here is that say:

7 ballots say Far, RON, Debra, tsu
6 ballots say tsu, RON, Debra, Far
2 ballots say RON, Debra

Under this situation, more people would prefer RON rather than either candidate, but because RON isn't the first choice of enough people, it would be eliminated.

Is there a way that we can write a law/do the math that once a ballot hits "RON" that is logged, even if we count the candidates below? Does that make sense?

-tsunamy
[forum admin]

Reply

Well, here's my thing, if someone did vote Debra > RON > Far > Tsu, why should they vote that way? Simply, it's Debra > RON if you do that. I don't understand why we should treat RON as a "candidate" and eliminate it early on. In fact, that option should exist even down to the final two candidates to determine if a majority agree that we should reopen nominations.

So, this makes sense, but legislatively it would take someone far smarter than me to implement.

Anyhow, the problem here is that say:

7 ballots say Far, RON, Debra, tsu
6 ballots say tsu, RON, Debra, Far
2 ballots say RON, Debra

Under this situation, more people would prefer RON rather than either candidate, but because RON isn't the first choice of enough people, it would be eliminated.

Is there a way that we can write a law/do the math that once a ballot hits "RON" that is logged, even if we count the candidates below? Does that make sense?

You seem to be identifying one of the main issues with IRV, yes, but there's not a great way to solve that. I guess are you saying that once we've done all the runoffs, if the winner is ranked below RON in a majority of ballots as originally cast, RON should nevertheless win? Maybe that could work... but it's definitely manipulating IRV to be something more Condorcet-like.

~ Aumeltopia ~

Raccoons are bandits! First they steal your food . . . and then your heart/identity!

Reply

Well, here's my thing, if someone did vote Debra > RON > Far > Tsu, why should they vote that way? Simply, it's Debra > RON if you do that. I don't understand why we should treat RON as a "candidate" and eliminate it early on. In fact, that option should exist even down to the final two candidates to determine if a majority agree that we should reopen nominations.

So, this makes sense, but legislatively it would take someone far smarter than me to implement.

Anyhow, the problem here is that say:

7 ballots say Far, RON, Debra, tsu
6 ballots say tsu, RON, Debra, Far
2 ballots say RON, Debra

Under this situation, more people would prefer RON rather than either candidate, but because RON isn't the first choice of enough people, it would be eliminated.

Is there a way that we can write a law/do the math that once a ballot hits "RON" that is logged, even if we count the candidates below? Does that make sense?

You seem to be identifying one of the main issues with IRV, yes, but there's not a great way to solve that. I guess are you saying that once we've done all the runoffs, if the winner is ranked below RON in a majority of ballots as originally cast, RON should nevertheless win? Maybe that could work... but it's definitely manipulating IRV to be something more Condorcet-like.

Yeah, at that point you might as well just switch us back to Condorcet voting.

Reply

7 ballots say Far, RON, Debra, tsu
6 ballots say tsu, RON, Debra, Far
2 ballots say RON, Debra

Out of interest, I looked up a bunch of voting systems and found this interesting one called Modified Borda Count that gives:

4 candidates, so it's 4 points for being ranked first, 3 for second, and so on, giving us:

7 ballots is 7x4 for Far, 7x3 for RON, 7x2 for Debra and 7x1 for tsu
6 ballots is 6x4 for tsu, 6x3 for RON, 6x2 for Debra and 6x1 for far
2 ballots is 2x2 for RON and 2x1 for Debra

Note that because the last two ballots were not prepared to add the other two candidates, their first place votes have less power. This doesn't actually make much of a difference in this case because adding another to the end would just give all of their rankings extra points, but it does stop the vote becoming into a slugfest where everyone just votes for their favourite candidate - if people are willing to compromise (and let's be honest no voting system can save an electorate who is completely unwilling to compromise) then they're rewarded with more voting power with each compromise candidate they give.

Far has 34
tsu has 31
RON has 43
Debra has 28

...Giving RON the victory, followed by Far in 2nd place, tsu in third and Debra in fourth.

I'm going to give you guys the chance to comment on this, but I thought it would be interesting to share. The system seems to favour the most consensual candidate - that is, a candidate that the most people can agree with, rather than a candidate who has the most absolute support.

(This post was last modified: 10-31-2020, 09:02 PM by Jebediah.)

Reply

No Borda count.

No no no no no no no.

~ Aumeltopia ~

Raccoons are bandits! First they steal your food . . . and then your heart/identity!

Reply

No Borda count.

No no no no no no no.

Why not?

Reply

I mean, my interest is in making this is easy for everyone to understand and easy for the EC to count.

I actually don't think this is terrible, but it might be a tad difficult to count?

-tsunamy
[forum admin]

Reply

I mean, my interest is in making this is easy for everyone to understand and easy for the EC to count.

I actually don't think this is terrible, but it might be a tad difficult to count?

I'd have to admit it is a bit difficult - it's relatively easy to wrap your head around (to figure out the score given to each vote you can just go 'bottom ranking on the ballot is one point, next one up is 2, and so on" which makes life significantly easier) but it is a bit tedious to count.

That being said, once you simplify it to that rule it's not much more tedious than counting for approval voting except with a bit of a twist.

Reply