Where The People Are Without Strategy There Is Tyranny


The ruling Western securitary / corporate pirate complex uses the ruse of false national democracy / sovereignty to conceal the tyranny of its masters and to justify its wars. The population is controlled by means of an illusion of self-government that is sustained by two-party / few-party locked-in systems.

The population is mesmerized by "liberalism", "conservatism", "Republican / Democrat party" circuses, and an endless number of petty factional disputes. Meanwhile the rulers impose austerity while expending vast resources in their continual quest for world domination. Real democracy would end all of that, so therefore, the pirates will insist that it must never be allowed to come into existence.

In the 21st century, a sudden surge of very heavily financed campaigns have emerged to promote what may be termed "democracy reform". For now, the great majority of these "reforms" are based on arcane theory-based election methods, which are debated among election method theory analysts. Only recently have reforms, based upon practical methods, begun to be promoted by practical analysts who are overtly concerned for the welfare of the masses.

The famous investigative historian Eric Zuesse has this to say about the current U.S. tyranny:

:~~~~~~~~~~~~~~~~~~~~ //
Strategic Culture Foundation -- Why Calling US a 'Democracy' Is Both False & Dangerous to Do -- by Eric Zuesse -- 3/15/19

It's false because it is definitely untrue, and that's not merely because America has a higher percentage of its residents in prison than does any other nation on this planet, but also because the only scientific studies that have been done of the matter show -- they prove -- scientifically -- that the US is a dictatorship by its very wealthiest residents, against all the rest of the population. Traditionally, that's called an "aristocracy," not a democracy, but ever since Mussolini in the 1920s, it came to be called "fascism," which is the successor to "feudalism" and thus is merely the modernized form of feudalism. What used to be called by such terms as "monarchy" or "aristocracy" is thus now called "fascism" but the leopard is the same regardless of what it is labeled, and what it really is [is] a dictatorship. Mussolini sometimes instead called fascism "corporationism" and it certainly is today's United States Government, even if some people choose to call it 'democracy'. It's what the US Government has been scientifically proven to be: dictatorship, by the richest few (the controlling owners of the international corporations), against the entire public.
// ~~~~~~~~~~~~~~~~~~~~:

The common people do of course know that voting, as it is done in the present day ("single-select" or "choose-one" voting), is generally useless, and about half of them do not bother. Even if it was useful, they would only do it because people have a powerful instinct to do things that are personally sacrificial but socially necessary. Voting as it is designed today is useless simply because of the two-party / few-party lock-in effect.

Of course some people in Western culture, due to learned mass assimilatism, will voluntarily flock toward predominant institutions. However, everybody knows that if a Republican, a Democrat, and a "Nader" are available, only the Republican or the Democrat will win, so it is pointless to vote for the "Nader", since their vote would then merely be sacrificed -- it would count for nothing due to the two-party lock-in effect which the single-select system ensures -- and they would thereby be deprived of the opportunity to vote for the lesser-evil candidate.

The only way that the voters can overcome two-party lock-in is by utilizing some system of "graded voting" that allows them to give a maximum number of votes to a "Nader", somewhat fewer votes -- or "hedge votes" -- to a "lesser evil" predominant party candidate, and no votes at all to a "greater evil" candidate. This is "hedge voting", and it is the only type of method that allows preference for the "Nader" while requiring only a small sacrifice of the ability to deprive opportunity for the "greater evil" candidate. Since election officials who design ballots and subsequently tabulate votes are never to be trusted too greatly, all aspects of election systems must be designed to operate as simply as possible.

The best system is the strategy-tolerant simple score method which allows voters to grant from (1) to (10) votes (the "score") to each candidate. The voter who grants (10) votes to a candidate knows that she or he is allotting a 100% portion of assertable support. (8) votes would grant an 80% portion, and so on. As an example of hedge voting, (8) or (9) votes might be granted to a lesser-evil candidate. Obviously, an abstentious "non-grant" of votes would allot (0) votes, and thus no support. This would result in only ten possible vote grants for each candidate. When the election is finished all of the votes are simply added up. This system is "pure-summative", and its entire tally procedure can be completed locally. The results represent a low quantity of data, which only needs to be passed to larger tabulations once. This would quickly disrupt the two-party lock-in, and thus enable voters to elect officials whom they truly prefer.

Since around the start of the 21st century the ruling Western securitary / corporate pirate complex has been propping up NGOs that tout an "RCV"/"IRV" method (or perhaps more accurately, methods). The "RCV"/"IRV" method is not voting in the ordinary sense because instead of being pure-summative, it is summative-eliminative. The ballot design requires voters to assign one candidate to one "place" -- from first to last in a sequence of places -- and this is called a "ranked ballot". It should be obvious that some places should be allowed to be "empty", with no assigned candidate. In the typically prescribed "RCV"/IRV" method the 1st place votes are summed, and if no one immediately has more than 50% of the total votes, then the candidate with the fewest votes is "eliminated". And then -- on each ballot -- the candidate (if any) in the place below the eliminated one is reallocated upward. Presumably, all candidates below the eliminated one are reallocated upward. This process is reiterated until some candidate "wins" over 50% of the votes. Of course if someone does have over 50% of the votes in the first "round" they win with an autochthonous majority -- but if elimination rounds are required, they only will "win" with an artificial "majority". The procedure just prescribed is not the one ordinarily utilized in practice; it usually must be modified or constrained in some severe fashion. It is completely unrealistic to suppose that any "RCV"/"IRV" method not drastically constrained will ever be reasonably amenable to hand-counted paper ballots in modern times.

"RCV"/"IRV" methods are not strategy-tolerant in any positive sense; but they are in strategy-tolerant for various negative purposes. The previously mentioned "score" method has often been criticized for violating the "later-no-harm" principle -- that is, it is possible for a lesser-evil candidate to win due to voters granting her or him any votes, even if less votes are granted to them than to other most-preferred candidates. But this is really not a significant concern. Also, "bullet voting", where a voter grants only one vote to only one candidate is an an extremely poor strategy for the score method -- hedge voting is the actual very good strategy. However, bullet voting is a very good strategy for "RCV"/"IRV" voting. This is because "RCV"/"IRV" always violates the "sooner-no-harm" principle. With that method, any candidate granted 1st place automatically diminishes the chance to win for all candidates granted lower places in the sequence of "ranks" -- even though those candidates may be virtually as much preferred as the 1st placed one (these are sometimes called "clone" candidates). This is an invitation to chaos since a great number of "clone" candidates can mutually eliminate each other, and then some almost universally unfavored candidate can win, due to the "bullet" votes of some very small minority. "RCV"/"IRV" voting is truly full of unexpectable quagmires.

Ranked voting ballots are by no means exclusively applicable to only the "RCV"/"IRV" vote tallying method. The very same ballot design could be utilized for "simple ranked voting". Simple ranked voting is very simple. Voters cast votes on a ranked place ballot in accordance with their intentions: =/ 1st > 2nd > 3rd > 4th… /=. Only one candidate (or no candidate at all) may be chosen for each place. The 1st place candidate is granted 10 votes, 2nd place gets 9 votes, 3rd place gets 8 votes, and so on. The 10th place candidate is granted 1 vote, and any further places are granted no votes. All of the votes are simply added up, and the candidate who was granted the most votes is the winner. Absolutely nothing is done to interfere with the voters' use of the hedge strategy, so this tallying method would be very disruptive of two-party lock-in, even though it would be less responsive than the simple score system.

All that the voters need is the ability to cast one vote to determine whether the "RCV"/"IRV" vs, the "simple ranked" tallying method will be utilized. Of course they will eventually decide to avoid the many pitfalls of the former method.

There exists an "approval" voting method whereby voters may choose to "approve" or refrain from "approving" any number of candidates, and all of the "approve" votes are simply summed up. However, this method does not enable use of the hedge strategy, so it will not ensure abolition of two party lock-in.

It is entirely possible to transform any pure-summative election method into a multi-winner proportional representation election method. That is, there is a method for implementing a proportional representation (multiple winner) election (e.g. for a legislature) without the involvement of parties using any such method:

“Tranches” correspond to seats in a legislature, but also, approximately, to non-majority groups or interests. Here are the fundamental parameters and variables (through utilization of the concept of "tranches -- which provide a kind of "curved score"):

S = The total number of seats to be filled, which will equal the number of "tranches" or "layers".

N = A tranche Number (these tranche numbers run from 0 to (S - 1) The “strongest winner” is in the topmost tranche, and the tranches form “layers,” with tranche #0 occupied by the strongest winner, tranche #1 below tranche #0, tranche #2 below tranche #1, and so on down to #(S - 1).

W = The strongest Winner’s total number of votes.

C = The “tranche ceiling”, or end point at the top of each tranche (to be determined for each given N (tranche number), by the equation below).

C = W*(1 - (N/S)^2)


Strongest winner’s total = 310

Total number of tranches = 8 (For a total number of seats = 8. And the bottomost tranche = 7)

For each tranche number (N) there is a calculated ceiling number (C):

C = 310 * (1 - (N/8)^2)

(C must be rounded off -- banker's rounding is recommended.)

As described above, this proportional method would lead to serious problems due to effectively random outcomes. However, these problems effectively vanish if a "paladin preservation" technique is employed. With this, any incumbent candidate who receives enough votes to whichever tranche they already occupy (presumably this would be a "paladin" with an well-liked established track record) will win that tranche once again, even if some other candidates would otherwise win it. This might appear to "fly in the face" of the common notion of "toss the rascals out", but improved election methods would remove the "rascals" automatically in any case.

Election method theory analysts have observed that this proportional method violates all sorts of (seemingly) important principles; but practical analysts will observe that it will nonetheless "just work", and will assure proportional representation strong enough to obviate the need for redistricting in many cases.

In the end, the true Silver Bullet that will abolish faulty election methods will be the ability of voters to vote for the method that is to be employed in the following election. Even the old "select one" method would likely be sufficient for this.

A National Citizens' Assembly on Electoral Reform


First, you sound like Just Another Conspiracy Theorist… although I have said several times here that I also feel like a conspiracy theorist, so…

Simple ranked voting will not work if there are several ideal candidates and you want to give them all 10 points. Forget that system because (1) FairlybadVote will never support it (as it fails LNH) and (2) Score Voting is better in basically every way.

Tranche voting does not make any sense – why not just use simple score voting and independent districting? (That, or asset voting… but given the dramatic race condition that has developed, Asset has virtually no chance against IRV, Score, and STAR.)


I sincerely thank you, NoIRV, for your willingness to offer criticism.

My responses here will not be in the order you presented them.

<> Recent studies indicate that 83% of Americans bear traits that confirm that they may be presumed to be “conspiracy theorists”. (Not really. I’m just joking. But who knows?) (Maybe Them?)

<> You claimed: =/ “Simple ranked voting will not work if there are several ideal candidates and you want to give them all 10 points.” /= That is mostly true; it will work rather poorly in that case.

<> You claimed: =/ “Score Voting is better in basically every way.” /= This is certainly correct. Simple ranked voting is far less “voter friendly” than score voting. It is, by far, less responsive (some would say “less expressive”). It is relatively pretty bad, but it is still vastly less dreadful than “RCV”/“IRV”.

<> You claimed: =/ “FairlybadVote will never support it (as it fails LNH)…” /= Now, here is where you stumble into the pitfall! Consider the implications of one of my favorite words: Strategy! Remember I said:

=/ "Ranked voting ballots are by no means exclusively applicable to only the “RCV”/“IRV” vote tallying method. The very same ballot design could be utilized for “simple ranked voting”. /= This means that the very same ballots (with their ranked votes) can be tallied by both of these methods. I also went on to say:

=/ All that the voters need is the ability to cast one vote to determine whether the “RCV”/“IRV” vs, the “simple ranked” tallying method will be utilized. Of course they will eventually decide to avoid the many pitfalls of the former method. /= This is strategy! We do not need to even ask FairlybadVote to support simple ranked voting at all. No – Not. At. All.

All we have to do is demand that the voters be allowed to choose between these two tallying methods by simply voting for one, or the other. Are they going to insist that the voters do not deserve this choice?

This is strategy, my friend. And unlike approval voting, simple ranked voting is a reasonably usable method that would serve as a viable stepping stone toward score voting. Strategy, my friend!

<> You claimed: =/ “Tranche voting does not make any sense – why not just use simple score voting and independent districting?” /= It is true that people living within various districts sometimes comprise minorities, simply because they may tend to have shared values. But let’s consider minorities in a perhaps simpler form.

Suppose “bikers” (people who ride motorcycles a lot) come to agree that they are suffering some sort of oppression, and that they need some representatives in an elected body to give voice to their concerns. Well, by right of assembly, they can agree to support one or more biker-friendly candidates who, while they might not garner enough votes to win in the upper tranches, could still gain occupancy in some of the lower tranches, and thus voice the biker’s concerns (they would still have just the same powers as the higher-tranche representatives). If this is not pure proportional representation, then what is? And the entire procedure is very simple and easy to safely implement.

Thanks again, and I hope this clarifies what these potential reforms are intended to accomplish.


This advocacy topic article is about how the issues most important to me relate to the upcoming very well financed “Unrig Summit” gatherings. I try very hard to work as an “investigator”, and this means that I must look beyond superficial notions such as “left wing”, “right wing”, “communist”, “heresy”, “conspiracy”, “labels”, and so on. I look to see how seemingly unrelated things potentially interact.

The entire “Unrig movement” is remarkably odd. The Nashville “Unrig Conference” is (somehow) “hosted” by the “RCV” promoting “RepresentUS”, since at the bottom of the =/ https://unrigsummit.com/ /= website it says: =/ © 2018 RepresentUs. All Rights Reserved. /=.


:~~~~~~~~~~~~~~~~~~~~ //
#Unrig Core Team

Robert David Steele, Chief Enabling Officer (CeO) of Earth Intelligence Network (EIN), is the founder of #UNRIG. A former Marine Corps infantry officer and former spy for the Central Intelligence Agency, his last government position was as the senior civilian responsible for creating the Marine Corps Intelligence Activity. He served as Chief Executive Officer (CEO) of Open Source Solutions, Inc. from 1993-2007. Establishing EIN in 2006, with Internal Revenue Service (IRS) accreditation in January 2017, he has dedicated himself to teaching individuals and organizations about the value of holistic analytics, true cost economics, and Open Source Everything Engineering (OSEE). He was recommended for the Nobel Peace Prize in January 2017. #UNRIG is an educational project intended to last 90-120 days (now extended to 365 days)

Connect with Robert via [email protected]. His personal web page is http://robertdavidsteele.com. He curates the blog http://phibetaiota.net.
// ~~~~~~~~~~~~~~~~~~~~:

More about “Unrig” and its founder Robert David Steele:

:~~~~~~~~~~~~~~~~~~~~ //
The Center for Media and Democracy (CMD) publishes SourceWatch – Page: Robert David Steele

Robert David Steele (born July 16, 1952) is a conspiracy theorist who was a candidate for the Reform Party’s nomination for President of the United States in the 2012 presidential election until February 23, 2012.


In 2017 he founded #UNRIG. In an interview with Jeff Rense (on 20 July 2017) [1] Steele thanked Rense for putting him in contact with David Duke. He said that on July 21 he would be beginning a 30 day tour of all fifty states on which he said he was taking “Ron Paul, Dennis Kucinich, Jesse Ventura and Ralph Nader and Patrick Buchanan and Vichell Good, Robert Steele and Cynthia McKinney with me.” This was not true as he was talking about an Election Reform Bill that he mentioned the previous month in this context:

=/ “Sponsor an Election Reform Summit immediately, followed by the introduction of the Election Reform Act of 2017 by Rand Paul in the Senate and Tulsi Gabbard in the House, and commit to barnstorming the country with Ron Paul, Dennis Kucinich, Jesse Ventura, Ralph Nader, Patrick Buchanan, Cynthia McKinney, and others – #UNRIG is yours to command.” /=
// ~~~~~~~~~~~~~~~~~~~~:

These folks surely do get around!

This mammoth movement constitutes an unprecedented coup in the United States. This surely must involve millions, tens of millions, perhaps hundreds of millions of dollars.

And what will we all get out of this? We will get dreadful “RCV”/“IRV” voting. The “RCV”/“IRV” sponsoring RepresentUS organization is based in the very same town where I live! And they have just launched an all-out campaign to mandate that voting method right here in Massachusetts.

Here again is what I suggested above:

:~~~~~~~~~~~~~~~~~~~~ //
All that the voters need is the ability to cast one vote to determine whether the “RCV”/“IRV” vs, the “simple ranked” tallying method will be utilized. Of course they will eventually decide to avoid the many pitfalls of the former method. /= This is strategy! We do not need to even ask FairlybadVote to support simple ranked voting at all. No – Not. At. All.

All we have to do is demand that the voters be allowed to choose between these two tallying methods by simply voting for one, or the other. Are they going to insist that the voters do not deserve this choice?
// ~~~~~~~~~~~~~~~~~~~~:

So why don’t we just do that? Let the voters decide, probably by just using the old “choose-one” method, which way they want their votes to be tabulated in each Massachusetts election.

I myself will continue to march in front of the local District Court.


Ok, I see now. However…

  1. The meaning of your ballot is fundamentally different. (In IRV, your second choice is basically a backup, only counted if your first is eliminated earlier. In SRV, your second choice is an “almost-equal”, getting 90% of the support that your first gets.)
  2. SRV can be counted in precincts by simply sending each candidate’s total score from that precinct. IRV requires either sending the entire ballots (much more data, has bribery issues) or several back-and-forth communications.
  3. IRV has been propagandized to the moon, and much of that propaganda also concerns the elimination system.
  4. If the main advocacy/propaganda groups “FairVote” and “RepresentUs” do not support SSV or SRV, the chances of either of those even making the ballot is slim.

The resemblance between the two systems is highly superficial. It would be better to ask the people to choose between IRV and SSV, but both referenda have the same low chance of being implemented.


Let us imagine an alternate world in which Simple Score Voting is the dominant system that people are talking about. You lead the movement that is going to put SSV on a ballot in, say, Maine. And then some random guy, who sounds a bit like a conspiracy theorist, comes up and says “Hey wait! I have this other system (describes some obscure system… say, Bucklin) and I think it works better! How about we let the people decide between SSV and Bucklin in the referendum?”

Would you (A) listen to the person and change the question, or (B) just ignore them and do the referendum on the single question?

(If you chose B, then that is probably exactly what “fairvote” is thinking about CES.)


Thanks again, NoIRV, your interpretation of what I propose is probably nearly exactly correct. I want to emphasize that it seems to be most convenient to just assume that “ballot design” and “tabulation method” are two very distinct components of election systems.

So what I have just proposed is a “simple ranked voting system” (SRV) that uses exactly the same ballot design as the “ranked choice voting system” (RCV). However, the SRV tabulation method is radically different, resembles simple score voting as much as possible, and most importantly, allows for a reasonable, albeit far from optimal, degree of hedge voting that stands a far better chance of disrupting two-party (or few-party) lock-in than, for example, approval voting (AV). I think it would be a much better steppingstone to SSV than, say, AV, and hopefully not just another pitfall.

The plan is that, in addition filling out the heavily promoted ranked ballot, there will also be a “little” choose-one question on the ballot: “Do you prefer the votes of this election to be counted by means of RCV tabulation, or by means of SRV tabulation?”

So, depending on which tabulation method wins by dint of having been granted the most votes, the entire election will be tabulated in one of these two very different ways.

Please observe that this will cause no problems whatsoever for (some folk’s) beloved rectitudinous (“honest”) voters! They can still vote just as “honestly” as they want! Until this question of which tabulation method will be used in each venue becomes the settled standard in a few elections, things will be a bit tricky for strategic voters. But who cares about them?

At this point I am convinced that the disastrous RCV is poised to be forced down our throats. The fat cats will throw their money bags to the RCV movement, and we will not, effectively, be given any real choice between RCV and score voting. Certainly not if the moneybag people have anything to say about it! So what can we do? Here is the key:

We simply demand that this little question about how the votes will be counted must be included on the ballot. Are they prepared to deny the voters this simple choice? What are they afraid of?

I say: Go. For. It.


Sure, but (1) how do we plan to convince those in charge to actually do this, and (2) what about the fact that the vote fundamentally means something different under IRV than SRV? (You cast your hedge-strategy SRV vote hoping that SRV wins… and then in fact the majority go for IRV? Or someone else fills out a second choice they barely care for only to find that second choice got 9 points?)

What are they afraid of?

They could be afraid of hedge strategies and the idea that the pirate party might not dominate after all.


Approval is equivalent to finer-grained Score because eventually the voters will learn to use probability to obtain the fine grain, in large elections. Decide how you would vote in a finer Score system, normalize to a range from zero to one, and take that number as the probability with which you should approve the candidate.

In support of your arguments about the benefits of Score for the single-winner case (if it can’t be eliminated), I refer to Warren D. Smith’s https://rangevoting.org/Cash3.html , which is nice and short and to the point.


I’ve seen you make this argument before. Say I were to use this tactic and decided to approve candidate A with probability 1; B: 0.9; C: 0.3; and D: 0. Are these probabilities be independent, or would I approve A only 10% of the time, A and B only 60% of the time, and A, B, and C 30% of the time (i.e. picking a random threshold between 0 and 1)?


This gives me visions of voters flipping coins to decide how to vote. Will stray coins be rolling out of voting booths?

You could question it from the perspective of cognitive load. Will people simultaneously make political decisions and figure out probabilities? I don’t think so.

I say this will not just NOT happen; this will just not even BEGIN to happen. Common sense should tell you people will never do this sort of thing.

Much easier to simply just let the voters grant candidates from (1) to (10) votes (or to abstain from granting some candidates any votes whatsoever). This imposes minimal cognitive load. For the important strategic hedge technique to disrupt few-party lock-in they probably only need from (7) to (10) votes. At one point I suggested (5) to (10) votes, since this would be easier to hand-count and generally less troublesome. But it has to -sell-, and it would be hard to explain the seemingly arbitrary restrictions.

Aside from proportionality issues – few-party lock-in, and unnecessary complexity are ultimately the real problems.


Independent probabilities would definitely work. Think in terms of relative frequencies if many people followed the strategy. However, I think your threshold method could also work, with the appropriate calculations.


Against what common sense says, we have what experience says, with one data point, FPtP, in which the public has picked up on the appropriate strategy, which is to tell people of moral character not to run lest they spoil the election for the less evil of the two money-supported “front runners”. So I’m suggesting the reasoning that if people have learned the strategy of one voting system to which they have been actually exposed for deciding contentious choices, they are likely to adapt similarly to a new system should they be exposed to it with similar import on the outcomes. And for Approval, that strategy is to use probability.

At the same time, because strategy is easier to teach for finer-grained Score than it is for mere Approval (i. e. two-valued coarsest-grained Score), I favor efforts to put finer-grained Score into effect for single-winner elections if we can’t convince our compatriots to eliminate single-winner elections. I suggest that values near the extremes are more strategically valuable than values near the middle, so the optimal range would not use equal spacing. I advocate {100, 99, 90, 50, 10, 1, 0}.


So… abolish the Senate and the Presidency? (Or leave them to legislatures and Congress to select?)

(If you advocate liquid democracy, that would become yes/no voting, and there would be no Score, IRV, or any of that stuff.)

100,99,90,50,10,1,0 is bad because (1) those are magic numbers, (2) it is not as simple, and (3) voters (thinking of myself here) may want to use 70s and 30s.


The title of this thread, " Where The People Are Without Strategy There Is Tyranny", resonates with me. I hear arguments to the effect that one or another voting system resists strategy in more cases than another does, or that one compared to another incents or does not incent a specific category of strategy. They use this to argue for ranking systems and against evaluative systems. But the Gibbard theorem says all voting systems give an incentive for strategy, in that to use them to maximum power, a voter or a party or a faction has to take into account some estimate of how their opponents are likely to vote or where they stand toward the candidates. If there is a strategy that the enemies of the people can use, they will use it. So it makes a certain amount of sense to me that we should design systems that provide a strategy for the people, so the money class will not have all the advantages.

I described in another forum a single-winner system that I thought might provide the people a better strategy than even Score Voting provides. I could refer over there, but I want to change maybe some of the terms and symbols I used in describing the system.

A person on the other forum did post an argument against my system. I will refer to the original post and that person’s response if anyone so requests.

This time I will start with an example ballot, then explain what it means.

Nader, Gore: COMPL; Nader

My ballot consists of two parts. The semicolon delimits those. In my original post on this system, I called those parts “stanzas”. Which term to use is a social and psychological question. Part of the objection I received said, in paraphrase, a voter is going to be puzzled as to what the f— are “stanzas”. But I advocate that people argue first about the power relations that different voting systems would establish, then about the marketing considerations of how to name components.

My ballot consists of stanzas because the tally will run in multiple rounds, like IRV (and it is not precinct summable). The first stanza in my example says that as long as anyone other than Nader and Gore is still in the running, I am supporting Nader and Gore. The “COMPL” notation specifies that the stanza is triggered according to the complement of the set of candidates named in the stanza. So in this example, that’s why the interpretation says “anyone other than”.

My last stanza simply names Nader. Since it does not have the COMPL notation, that means that this stanza triggers if Nader is still in the running. But an implicit triggering rule is that a stanza only has effect in a round if none of the stanzas preceding it on its ballot triggered. Only the first one that triggers counts.

Moving from focusing on the example to laying out the general rule:

A ballot consists of so many stanzas as the voter chooses. Order matters among them. Each stanza can name any number of candidates. Any stanza can carry the optional mark COMPL.

The tally proceeds in rounds. The candidates in the running for the first round are the candidates named on any of the ballots. Each round begins with zeroed-out accumulators for the candidates who are still in the running. The round then examines the ballots. For each ballot, the first stanza is selected that applies, and only this stanza affects this round of the tally. The way it affects it is that each candidate named in the stanza gets a point. The determination of whether a stanza applies depends on whether it is marked COMPL. If it is so marked, the stanza applies if and only if any of the original candidates from the first round that is not named in the stanza is still in the running for the current round. So for example, if the stanza does not name Bush but Bush is still in the running, and the stanza is marked COMPL, it applies because of Bush. A stanza that is not marked COMPL applies if and only if any of the candidates named in the stanza is still in the round.

Once the round has examined all the ballots and accumulated the totals according to the above rules, the round eliminates from further consideration in the subsequent rounds, about half the candidates still in the running. This will be subject to some rounding rule, but if there are only three or two candidates remaining, only one candidate is to be eliminated. If only one candidate remains, that candidate wins.

So I’m guessing that this system would not require of ordinary voters, as accurate a guess about how the other voters will vote, as Score would require, for choosing the best strategic vote.

And I’m guessing that the IRV pathologies would not be duplicated in this system. But I could be wrong.


So basically, it lets people vote as either Approval (just vote something like “A,B,C; A,B,C Compl”?) or IRV or top-two-ish.


Each stage operates as Approval, but a voter can enter several contingencies to try to program her vote to respond well to the order of eliminations of candidates. Strategy can be IRV-like or Coombs-like (in the later case I am not totally sure how “like” it is) according to the voter’s bet on how the other voters feel. How to link that bet to the strategy, I am not sure yet. But my interest in the system comes from the strict equality it provides to the voters’ power, and from the thought that it doesn’t require so finely-tuned a strategy as Score does, and accordingly might appeal to IRV advocates, if I and/or others can find the right marketing phrases to use when talking to them.


I will remember your system (and its parent) when I establish my own computer-based company full of nerds and we need to vote on things that are more than 2 candidates.

(But I probably will not win out – Approval is just too simple for meetings!)