Approval Voting on the Ballot in Fargo!


On July 30th:

Sprague said he expects the verification will take about two weeks though state law allows up to 30 days.

Is approval voting getting on the ballot in Fargo or not?


A lot of people here think that approval voting is just fine. I regret to have to say that approval voting simply will not perform. It’s undoubtedly not as awful as RCV/IRC, but I predict that it will alienate the voters. This is because of the “double bind quandary.”

See The Double Bind Quandary:

Unfortunately, approval voting cannot thoroughly disrupt the spoiler effect, and so if it is adopted, it will likely fail, and thus the two-party-system with elite rule will likely survive. This would bring severe disappointment for voters who supported an alternative to the current single selection system. I regard the terms “plurality voting” and “first past the post” as pathetic terminology. In reality, all systems tend to produce “plurality” results. There is no “post” in “first past the post,” even metaphorically.

Approval voting, whereby voters can grant just one (1) vote to as many candidates as they approve of suffers from a “double bind quandary”. Suppose the voters using the approval method are presented with three candidates: two of whom are ruling establishment supported (presumably a Republican and a Democrat), and one of which is truly desired. The voters will face a quandary. Should they vote for both the desired and the lesser-evil ruling establishment candidate, or just for one or the other of these two, so as to contribute to the optimal, or at least lesser evil outcome? I presume that for most voters this is simply undecidable. Perhaps this should be called a “triple bind quandary”, and it could result in strongly desired, but non-ruling establishment candidates almost never becoming elected. A simple score/range method could effectively thwart this quandary.

So I suspect that if approval is adopted it will very possibly fail to deliver what it promises. And that would be a major setback for the future adoption of score/range voting.

Add-On Multi-Winner Proportional Representation
Shortest Splitline Thread
  1. What preference is more important to them? Is it the preference between the Independent and the Democrat or the preference between the Democrat and the Republican? Which preference will give their ballot more bang for its buck? That is what will decide whether such a voter approves of the Democrat or not.

  2. Are you saying that because voters will be too afraid of the other party (Democrat or Republican) winning that they will always approve of at least one of the two and as a result, one of the two will always win? Because if that’s what you’re trying to say, more then 2/3rds of voters are dissatisfied with the two party system and want a third party and approval voting passes the favorite betrayal criterion and to top it all off, Congress has consistently had less then a 20% approval rating, which means that in a race with a Dem a Rep and a third party/Independent, I’d place my bets on the third party/Independent winning that seat, even in the unlikely scenario in which all voters distinguish a preference between the Dem and Rep by approving of one of the two. In the current system where voters always distinguish a preference between the Dem and the Rep, the winning dem or rep’s vote total is usually not too far from 50% (just look at the most recent presidential elections) and 50% is not an impossible bar for third parties/Independents to pass. In fact, politicians should definitely be appr

  3. Score/range voting reduces to approval voting when voters are maximally strategic so any problem approval voting has, score/range voting has as well.


The approval voting petition will be officially verified and approved by the City of Fargo, North Dakota Commission on Monday, August 27th. The measure to make Fargo, ND the first city in the nation to utilize approval voting for local elections will be on the ballot this November.


I agree that there is a flaw with Approval, specifically, the inability to give fractional approvals, which leads significant portions of the electorate to bullet vote (to preserve their honest favorite).

That said, Counted found that Approval results approximate Score results reasonably well.

Further, even if Approval is seen as being problematic, once people have used it once or twice, it should be trivial to get them on board with converting to Score; some academics studying approval have said that their study participants often ask for more options than just yes/no, without any prompting from the experimenters.


As I said in the “theory” forum, I began deconstructing RCV/IRV in the pre-election season of 2004. At that time, I promoted something called “Consecutive Runoff Approval Voting”, which method consisted of three ordinary consecutive approval voting elections (runoffs). Please see:

Consecutive Runoff Approval Voting

This was ignored, probably because people didn’t wish to endure three consecutive runoffs. So the project moved on to the strategic hedge simple score method.

While people at the Center for Election Science tend to use emotionally charged terms such as “honest voters,” “sincere voters,” “favorite betrayal,” etc., I eschew such terminology. I speak of “strategic voters,” “rectitudinous voters,” and so on. These terms have the great advantage of being relatively neutral.

You said – “more then 2/3rds of voters are dissatisfied with the two party system and want a third party…” Well that “third party” will likely become the new “second party” due to double bind quandary. As I have said about approval voting, when voters are faced with a “strongly undesirable”, a “moderately undesirable” and a “desirable” candidate, they will find it very difficult to decide whether to vote for both the “moderately undesirable” and the “desirable” candidate, or just the “moderately undesirable” or just the “desirable” candidate (the double bind quandary). So, the chance of electing the “desirable” one will be very questionable, and we may seldom or never get a “desirable” one at all.

With strategic hedge simple score voting, they could give 10 votes to the “desirable” ones, and perhaps 7, 8, or 9 to the “moderately undesirable” ones, while simply abstaining from bestowing any votes at all to the undesirable ones. This method is spoiler effect suppressive, and it would fatally vitiate the double bind quandary.

You said – “Score/range voting reduces to approval voting when voters are maximally strategic so any problem approval voting has, score/range voting has as well.” Oh my lord I so wish you folks would pull yourselves out of this nasty rut! In real life, disciplines such as quantitative algebra might be surmised to “reduce” to algebraic logic because they depend on Boolean “relations” such as “equals” and “greater than.” But other than in the most rarefied sense this sort of thinking is completely pointless. In practice, approval and score/range voting are very different methods. And your assertion that “any problem approval voting has, score/range voting has as well” is simply wrong (e.g. you cannot use strategic hedging with approval voting).

I suspect that the adoption of approval voting will poison the well for strategic hedge voting, and the public will not keep adopting new voting methods continually.

As far as I can see, psephology (the study of political elections) is the new alchemy, and needs to be brought to a merciful end. Strategic hedge is the only solution that will yield positive results.


It’s official:


That exists with the current system as well, though, so I doubt it will alienate voters any more than they already are.


I see no evidence for this. Note that in the 2014 Maine exit poll, the finish order literally reversed, causing the independent to take the lead. That is how you escape two-party duopoly.


Thank you, ClayShentrup, for providing an explainable question. In case there is still any preliminary question, please let me revisit a point made above by parker_friedland:

"Score/range voting reduces to approval voting when voters are maximally strategic so any problem approval voting has, score/range voting has as well."

My position is that this is dead wrong – NO, score/range voting should NOT reduce to approval voting, provided that strategic voters are reasonably competent. (And rectitudinous voters who attempt to vote “honestly” are simply perpetrators of electoral malpractice.)

To drive this point home, we should clear up something else. The “classic” example of the consequences of the spoiler effect seems to be the 2000 presidential election involving (among other candidates) George W. Bush, Al Gore, and Ralph Nader. This event is usually analyzed from the perspective of “liberals” or “progressives.” But we know that concepts pertaining to election methods ought to be viewed from a “neutral” perspective. Maybe we could just call them Josh shrub, Hal Bore, and Nalf Raider. Or better Nastycrats, Sneakycrats, and Vegecrats. Or even better, (individual-voter-assessed) strongly undesired, moderately undesired, and more-or-less desired candidates. Now let’s look at the differences between approval voting and strategic hedge simple score voting:

Let’s pretend that the available candidates are a Nastycrat, a Sneakycrat, and a Vegecrat. With (my preferred version of) approval voting, you could only bestow to each desired candidate one vote, and bestow no vote to (abstain from voting for) each undesired candidate. Suppose you desperately want prevent the election of the Nastycrat, but would much prefer the election of the Vegecrat. What can you do? There are just two options here: You can bestow a vote on only the Vegecrat, or you could bestow a vote on both the Vegecrat and the Sneakycrat, hoping in the latter case to avoid election of the Nastycrat. Now, this choice is very likely to be undecidable, and also frustrating. The outcome will presumably be random, and we could end up never, or seldom ever, obtaining a Vegecrat.

But suppose we have strategic hedge simple score voting. We can bestow between five to ten votes (bestow a 50% to 100% portion of assertable support) to each or all of the candidates, or else abstain from bestowing any votes to each (or all) of the candidates. Bullet voting, that is, either bestowing no votes (by abstaining) or bestowing ten votes to each (or all) of the candidates would be a dreadfully poor strategy. We would give zero votes to the Nastycrat, and give ten votes to the Vegecrat, but we would be foolish to give either zero or ten votes to the Sneakycrat. If we did that we would simply be imposing the double bind quandary upon ourselves. That choice could easily turn out to be a completely random decision. But, relying upon our situational awareness, we can still give, say, seven, eight, or nine votes to the Sneakycrat, thus providing the Vegecrat with the maximal advantage, while in effect still voting against the Nastycrat by giving some significant assertable support to the Sneakycrat.

So therefor, approval voting can only provide unpredictable, random immunity to the spoiler effect, but strategic hedge simple score voting can provide fairly reliable immunity from it.


OK prove me wrong.

And are you going to explain why it’s a poor strategy?

Why would this be foolish?

So? All you proved was that giving the most strategically possible vote (giving only max and min scores) involves making a tough strategic decision. That doesn’t mean that it’s an inferior way of voting. Do you know what else involves making a difficult strategic decision? Voting in the first place. But if you abstain from voting, you don’t have to make such a tough decision so abstaining is a superior way of voting, right? Wrong.

So if you can’t prove me wrong, I guess I’ll have to prove myself wrong for you.

Suppose that in a 3 voter election you prefer A to B to C. If the other two voters vote, they are going to give 4 stars to B and 5 stars to C but there is a 50% chance that they will show up.

If you vote A = 5 stars, B = 0 stars, C = 0 stars, there is a 50% chance the winner is A and a 50% chance it is C.
If you vote A = 5 stars, B = 5 stars, C = 0 stars, there is a 100% chance the winner is B.

But if you vote A = 5 stars, B = [anything other than 5 and 0] stars, C = 0 stars, there is a 50% chance the winner is A and a 50% chance that it is B.

However, while it is technically true that there are some examples where the most strategic way of voting does not involve casting an approval style ballot, those examples are only common when the number of voters is extremely small.

As the number of voters approaches infinity, this becomes the most strategic way to vote in score voting becomes this:


This is incorrect. Although there is a caveat in very tiny elections.


About your last point, not all voters will act strategically. Remember that (1) voting is irrational and (2) people really do vote for Greens and Libertarians, so we have living proof that some voters behave unstrategically. So yeah. If you want to have people not vote Approval style, then back STAR or (STAR + vanilla score)/2.


I thought we were talking about the case that voters are strategic. And the burden of proof would of been on rkjoyce because he/she was claiming that I was wrong, and when you are claiming that someone else is wrong, the burden of proof is on you.



And it won!!!


Score voting just let’s you water down support so it’s just watered down approval. Not sure why that’s a hard concept to grasp.


Wow. Usually when people talk about “watered down” they mean that you simplify it and reduce its functionality, especially in a negative way, but usually for some other gain. Like a “watered down” software that has half of the features of the original but runs on older systems or uses less memory or is free. From that perspective, Approval is watered down Score because the former is a proper subset of the latter that takes less sophistication on the part of the counting process.


Unfortunately, I am getting this sense that UnfairVote’s “bullet voting” proposition will turn out to come true.

But for exactly the opposite reason that UV would hypothesize. They say that voters bullet vote as a strategy when they actually have a stake in the outcome. However, to me, it seems like in a city election, many people would be unlikely to research the candidates as well as if it were presidential or congressional, and they would perhaps have either one guy they like or one guy they hate, and vote accordingly.

Besides, on the first year of a system, there will still be plenty of 2PD leftover effects, and combined with lack of knowledge will produce the exact outcomes that UV will pounce on.


What has happened here is that parker_friedland quoted me as presented as:

And then he continues:

Also ClayShentrup said I am incorrect in my claim that bullet voting is a dreadful strategy:

I have explained why strategic hedge voting is a great strategy, and bullet voting is a dreadful strategy with score voting (since it effectively constrains score so as to make it equivalent to approval):

This bullet voting would only be attempted by novice voters, who would soon learn that they must use the hedge strategy. “Honest” (really goody-goody rectitudinous) voters are guilty of electoral malpractice.

When will all these folks unlearn this giant “bullet voting” fallacy?

Daily Kos post could use some expertise