Criticize my Euler diagram


#1

First draft

  1. It probably has too many systems in it?
  2. I mostly wanted to clarify the never-ending confusion about STV/IRV/ranked systems and range/ranked.
  3. Also I am not a graphic designer.
  4. Also I maybe put some things in the wrong place.

I tried to include systems used in the real world, most common reform proposals, and then some to fill in the empty spaces

I was thinking I could draw a dotted line between IRV and STV? Since they are marketed using the same name and have the same procedure. But are they really more similar to each other than other systems are to each other, or is that just marketing?

and the unlabeled circle around Approval and SAV is for the 2-level rated variants.


#2

This is excellent!

  1. Nope, not too many systems. Looks great. I’d like if you could have an index to expand what PAV, SAV, SPAV are for someone who may only be learning of those through this chart.
  2. LGTM
  3. LGTM
  4. :man_shrugging:

Is there an order to the X/Y axis? For example, “Independence between candidates”


#3

This should go up here:

I’d like to edit this page to break it into “Single Winner” and “Multi Winner” electoral systems.
Thoughts?


#4

Is there an order to the X/Y axis?

No. How would you rate them by “Independence between candidates”?

One criticism on reddit was that FPTP is a single-winner system, and shouldn’t be in the middle, even though it’s used for things like US congress.
Borda and IRV are also used to elect Australian representatives and Nauru’s parliament, even though I have them listed as single-winner.
Bucklin’s WP article literally says single- or multi-winner though…


#5

I’d like to edit this page to break it into “Single Winner” and “Multi Winner” electoral systems.
Thoughts?

How would you split it? Things like ranked vs rated and multi vs single are orthogonal to each other, and you don’t want to duplicate things.


#6

Asset Voting.
Could be single mark. However, I at this moment prefer letting each voter mark up to W candidates (W = number of winners).
Designed for multi winner but still works in 1-member districts (like states with only 1 rep in congress),


#7

Yeah, I’m not sure. I just feel like the philosophy of how we teach voting science is very disorganized. I’d like there to be a curriculum that first focuses on Single Winner elections, then moves on to Multi Winner elections.


#8

Yeah those should have different names.


#9

Apportioned Approval, Apportioned Range, Apportioned STAR, Apportioned 3-2-1 are all (technically) a thing…


#10

add my method :smile: ( https://blog.opavote.com/2016/06/guest-post-rethinking-stv-fundamentals.html?m=1 ) it would be ranked, span both single winner and multi-winner, in the single winner it equates to borda, and its condorcet in both single and ranked.


#11

you imply borda isn’t condorcet, but I’m not aware of any examples of condorcet failures in borda. and I dont see any way there could be.


#12

#13

hmmm… yep. surprised I missed that case. I stand corrected. thanks.


#14

…that implies my method isn’t condorcet :frowning:


#15

Don’t feel too bad about that; Condorcet, No Favorite Betrayal, and Later No Harm all appear to be mutually exclusive. If your method met Condorcet, that would mean that it failed No Favorite Betrayal, which I consider to be the far more important criterion.