Could FairVote be pushed to advocate Equal-Rank IRV?

This is essentially IRV but you can equally rank multiple candidates, and they each get one vote. If everyone only uses the first rank, this is essentially Approval, so wouldn’t it actually be better? It’s also not a big shift from IRV. The biggest issues are that it still requires centralized counting (but so does PR) and that you can’t equally rank in most STV implementations, so that might confuse voters (having equal ranking allowed and not on the same ballot.)

The ideal system probably would allow scoring candidates in each rank, but that’s not practical.

Here is an example I found against ER-IRV:

45:Right=Left>CentreRight
35:CentreRight>Right>Left
20:Left>CentreRight>Right
First-preference tallies
Right:45 CentreRight:35 Left:65
CentreRight has the lowest tally, and so is eliminated then Right wins. This time no coordination was needed. As long as the Right suporters knew that Right had more first-prefernces than CentreRight, and a pairwise win against Left, then each individual Right supporter got an increased expectation by insincerely upranking Left from last to equal-first with no risk. This would also work if the numbers 45/35/20 were replaced with 49/48/3. I suggest the right numbers in your “paradoxical” row should be IRV1,ER-IRV(fractional)2, ER-IRV(whole) 5!

but just like in Approval or Score, the Left voters can stop this by equally ranking Center-Right with Left.

I have begun to think that ‘centralized tabulation’ is not the way to frame the complication that it presently refers to. Consider: simple score only requires one-time ballot inspection tabulation, and STAR requires two-time ballot inspection, but RCV potentially requires multiple-time ballot inspection. However, neither simple score nor STAR requires ballot editing, while RCV always has the potential requirement for (multiple-time) ballot editing. This editing may be done on paper, or in some other, likely more abstract way. Multiple-time ballot inspection and ballot editing both create excessive ‘information trafficking’ within tabulation processes, but the editing is far, far worse.

Ultimately, some central tabulation facility must be empowered to declare a winner, regardless of whatever election system happens to be in place. Weakest contestant elimination can only be performed in one central tabulation facility, and with RCV, there always exists the potential for this elimination procedure to be required to be performed multiple times. This is clearly the worst form of excessive information trafficking.

The ‘equal-rank IRV’ certainly sounds like a plan that would entail quite a great amount of information trafficking during tabulation. Also, why do people worry about what FairVote thinks? (Probably because they have the Big Bucks!) Anyway, they are not in the business of listening to people, and they have an amazing track record of being ‘less than truthful’.

I’m wondering if any local or state RCV chapters could be convinced to do something like this, especially if cardinal advocates promised greater support/donations/publicity in exchange. Ultimately, RCV is inevitably going to take a few states and many cities, so it’d be worth it to mitigate the damage and potentially uplift the method into something as good or better than Approval in terms of the quality of the winner and evaluation of support for 3rd parties.

By the way NIST (National Institute of Standards and Technology, U.S. Department of Commerce) has invented its own terminology:

NIST Election Terms Glossary - Draft

There is no good reason to trust this bureaucratic ‘entity’. We don’t have to obey the ‘standards’ invented by this industry-dominated agency.

Couldn’t you have put that in a separate post?

I think that Equal-Rank IRV is better than IRV, so it is acceptable as a last resort. When do you advance to the next rank? When all rank-1 candidates are out? Or is each rank weighted by the proportion of eliminated candidates on the rank higher?

1 Like

When all rank-1 candidates are out, I’m guessing. So the benefit is that a voter can choose to compromise, ensuring consensus candidates don’t get eliminated, or they can rank them as a lower preference so that their vote transfers to them as a last resort.

IRV already has the flaw that if you rank medium-strength candidates first, you can have your second and third eliminated prematurely. Top-ranking several candidates would only make that worse…

…which degenerates the ballot into an Approval ballot, which Isn’t So Bad.


1 Like