Splitting this topic off because I don't want it to derail the Wolf Comittee [sic] Results

Continuing the discussion from Wolf Comittee Results:

Two questions:

  1. Why the magic numbers 7 and 9? Is there not a reason that I might want to hedge some candidate as, say, a 4? Would it not depend on which election cycle we are in? (Example: in a close state the year after SSV is enacted, voters should probably max-min. But after many years the elites will mostly be getting scores of 2-5, if they get points at all.)
  2. If you are thinking strategically, how do you dispute the following argument for why strategic Score is equal to strategic Approval?
    • Suppose you could only give at most 1 point to each candidate. You would then choose a set S of candidates to give 1 point to, and abstain on the others.
    • Now suppose instead you can give at most 2 points. Since giving 1 point is unlikely to change the result, the candidates who should qualify for your second point should be the same set S.
    • And so on for your third, fourth, … and tenth point. The only time this argument would fail is if the election were so close that your set S changed after, say, the 6th point.
1 Like

I think splitting that discussion off was a good idea. I do believe there is far more here to be considered than is apparent at ‘first glance’. I don’t expect to persuade everyone that I’m right, but perhaps it is time for a few folks to look at election methods from new perspectives.

As a long-time political agitator I clearly perceive an absolute sociomechanical disaster in the making, where we are about to be plunged into a dark ‘RCV’/‘IRV’ electoral tunnel. I very strongly recommend that every voter forced to use a ranked ballot should be given the ability to mark on their ballot whether they want the ballots to be tabulated using ‘RCV’/‘IRV’ tabulation or by using simple ranked voting (SRV) tabulation. SRV tabulation grants 10 votes (or points) to the first rank-place candidate, 9 to the second rank-place candidate, and so on, all the way down to a ‘final zero’ rank place. And the votes are then simply added up, just as they would be with the (significantly superior) score/range method. But that’s not all:

In, for instance, the ‘single winner’ case, the tabulation should be totally completed in each individual ‘precinct’, and the winner in each individual ‘precinct’ should be granted ‘all of the votes’ from that precinct – ‘All of the votes’ comprising the total number of ballots that were cast there. This would effectively cause both ‘RCV’/‘IRV’ tabulation and SRV tabulation to effectively be ‘precinct summable’. In addition, the results of both ‘RCV’/‘IRV’ tabulation and SRV tabulation precincts could be combined (added together) seamlessly, so each precinct could decide on which method they would prefer to use. (Please note that a somewhat weakened version of the hedge strategy can be employed with SRV.) How could anyone take the stance that the voters aught to be denied this option?

=/ Why the magic numbers 7 and 9? /= – Said above

Well they are not ‘magic’ of course – They simply reflect my heuristic prediction of how people would vote after a few election cycles.

~~~~~~~~~~~~~~~~~~~~ //
[Said above]
= Now suppose instead you can give at most 2 points. Since giving 1 point is unlikely to change the result, the candidates who should qualify for your second point should be the same set S.

= And so on for your third, fourth, … and tenth point. The only time this argument would fail is if the election were so close that your set S changed after, say, the 6th point.
// ~~~~~~~~~~~~~~~~~~~~

This reasoning looks rather akin to mathematical proof by induction. My perspective is that such sociomechanical reasoning simply does not apply to systems wherein human beings are able to apply strategy, and that they will use strategy heuristically, and not in accord with rigid mathematical algorithms. This is an issue that is involved with something larger:

True democracy requires much more than some mere efficient electoral system – It demands certain cultural features – It requires a sufficiently high level of cultural sophistication. For example, no mere electoral system, however excellent, will be viable in a society wherein two factions are on the verge of engaging in a civil war.

It’s fairly simple to allow equal-ranking, in which case IRV is improved and SRV (which, confusingly, is the same abbreviation used for Score Runoff Voting/STAR) is equivalent to Score.

I think I have a good guess in expecting that FairVote and about a thousand other organizations and public figures are going to foist the ‘RCV’/‘IRV’ method on perhaps 20 or even more states in 2020. Now they are not going to ‘spring’ for any ‘equal ranking’, and I think that is very clear. So the strategy I suggested above seems to be the only one available at this time.

Since the ballot design for both methods is identical, how can any of the ‘RCV’/‘IRV’ promoters explain that the voters should be denied the simple choice of which tabulation method they prefer? I don’t think they could get away with it.

I am aware that STAR has been called ‘SRV’ in the past. But I prefer to snap it up at this point. There won’t be very much confusion.

“It’s too complex to be as viable as IRV, and anyways voters are harming themselves by supporting their candidates in this way.” It’s even quite easy for them to argue that it takes the public in the wrong direction (towards cardinal methods) or to pretend this is Borda (which most generally oppose.)

Wouldn’t this be like winner-take-all in the Electoral College, where most precincts wouldn’t matter because they tend to favor particular candidates, and only swing precincts would be contested?

~~~~~~~~~~~~~~~~~~~~ //
[Said above]
“It’s too complex to be as viable as IRV, and anyways voters are harming themselves by supporting their candidates in this way.” It’s even quite easy for them to argue that it takes the public in the wrong direction (towards cardinal methods) or to pretend this is Borda (which most generally oppose.)
// ~~~~~~~~~~~~~~~~~~~~

They can argue that SRV is no good all they want. Still, some precincts will choose it, and come to like it much better.

~~~~~~~~~~~~~~~~~~~~ //
[Said above]
Wouldn’t this be like winner-take-all in the Electoral College, where most precincts wouldn’t matter because they tend to favor particular candidates, and only swing precincts would be contested?
// ~~~~~~~~~~~~~~~~~~~~

If the voters used strategic hedging in SRV districts, it would render that meaningless. So-called ‘winner-take-all’ is a meaningless slogan.

All that is asked is that the voters be given a simple choice. Nothing more.

How will they choose it if it isn’t implemented in the ballot measure enacting IRV, the campaigning and control of which will be in FairVote’s hands?

Voters’ ballots would be thrown out randomly depending on whether they’re in the minority of their precinct, so they might have to strategically choose which precincts to cast their votes in, which then makes it possible to identify which precincts are the ones that supporters of Candidate Y will vote at, making it possible to target, intimidate, and identify those voters with high probability. Also, winner-take-all precincts probably don’t preserve vote equality enough to be constitutional when the precincts have all casted ballots for the same office(s).

=/ How will they choose it if it isn’t implemented in the ballot measure enacting IRV, the campaigning and control of which will be in FairVote’s hands? /=

I am talking about a strategy for dealing with the likes of FairVote. I do not expect some sort of ‘total victory’.

=/ Voters’ ballots would be thrown out randomly depending on whether they’re in the minority of their precinct, so they might have to strategically choose which precincts to cast their votes in, which then… /=

I’ve never heard of any situation in which voters get to choose which ‘precinct’ they may cast their votes in. This makes no sense that I can discern.

But doesn’t that imply they are not being strategic? In every sense I have heard, strategy in any exact game involves algorithms.

Because SRV “fails” the “later-no-harm” criterion that the IRV propagandists have been using to shun any system that is not choose-1, two-round runoff, and IRV. The propagandists will argue that the SRV question creates extra complexity for what they will call no real benefit. (Besides, how do you combine SRV votes into IRV votes within the same race?)

I think what AssetVotingAdvocacy is talking about is this: Assume there are 3 precincts: North, East, and West.

North:
51 A
49 B

East:
26 A
24 B

West:
50 B
10 A

With your model, it seems like A wins 150-60, despite B having 123 total votes over A’s 87.

If you’re in a city split 50/50 liberal/conservative, and you’re a liberal voted who lives in a predominantly conservative part of the city, you’d go to a more liberal part of the city to cast your ballot, because that makes it more likely that the majority casting ballots there are liberal, and therefore your vote will get counted. Party leaders would have an incentive to indicate which precincts are safe for their voters to vote in.

Here, B voters from the West part of the city should probably all cast their votes in the North or East part of the city, to ensure they can swing those precincts in their favor and capture all of their votes.

IRV has substantial opposition from people who would rather have FPTP. In the US, IRV has trouble attracting support from anyone who isn’t a Democratic partisan, or part of the Sanders faction. However, IRV doesn’t have universal support even among those places. So it’s only able to succeed in very blue communities like Takoma Park, or places that lean Democratic where recent events have given IRV momentum among Democrat-leaning voters (like Maine, where Paul LePage won due to vote splitting).

This could lead to precinct gerrymandering. It also turns voting into a form of gambling. If I vote at my precinct, I’m betting that my preferred competitive candidate will win my precinct. If they don’t, my vote may work to defeat them and I should’ve stayed at home. Also, if I vote absentee, how would my vote be counted?

You think FairVote will successfully get IRV enacted for use in general elections of offices elected statewide in 20 different states by the end of 2020? $250 says they don’t.

1 Like

They also have RepresentUs working on their side, although that group suddenly went inactive (at least on their YouTube channel) after that guy did that big walk.

IRV has momentum, sure, but it’s primarily been at the local level. So far the only time it’s managed to reach state level elections is in Maine. As I’ve pointed out, the Republicans don’t like it. The Democrats only have trifectas in 15 states. They also could override the governor’s veto in Massachusetts, Vermont (with the help of the Progressive Party), and Maryland, but that still only makes 18. Maryland won’t do it: the legislature wouldn’t even pass a bill this year to allow Montgomery County to use RCV (or Approval, which got little attention) for county elections. It would be a stunning reversal if those legislators, who previously sought to prevent another locality from using it, decided next year that they want it used in their own district! The reason that bill died in Maryland was a general fear of change, a problem that isn’t unique to Maryland. This will be a problem elsewhere. So I don’t see a path to enactment in 20 states in one year.

There is also a lot of talk about a potential Democratic trifecta takeover at the federal level, but that technically won’t be in 2020.