Poll on 4 voting-systems

I invite you to vote among these 4 voting systems:

Our current vote-for-1 Plurality system
Approval Voting
Score Voting
Majority-Disquaification (a ranking-method)

I'll define these voting-systems here:

You don't need a definition of Vote-for-1 Plurality.

1, Approval Voting:

By marking their names on the ballot, you approve 1 or more candidates, as many or as few as you want to.

The winner is the candidate approved by the most people.

[end of Approval definition]

2. Score Voting:

You rate each candidate within some ballot-specified range, such as from 0 to 10, or from 0 to 100, etc.

The winner is the candidate who gets, over the whole electorate, the highest points-total.

[end of Score definition]

3. Majority-Disqualification:

You can rank, in order of preference, the candidates whom you especially want to be the one who wins.

You're counted as "approving" all of the candidates whom you rank.

(but you have the option of indicating that you deny approval to any particular 1 or more candidates in your ranking.)

The winner is the most approved candidate who doesn't have another candidate ranked over him/her by more than half of the voters.

Voters are allowed to rank several candidates at the same rank-level if they want to--at any rank-level.

[end of Majority-Disqualification definition]
[
Voting-Instructions:

1. You're invited to approve as many of the voting-systems as you want to, in the click-spot ballot at the top of this page.

2. You're also invited to rank the candidates, in replies to this thread--as follows:

Rank, in order of preference as many of these 4 voting-systems as you want to--the ones that you approve.

The ballots, if any, will be counted by Majorilty-Disqualification.

As an example, here is my ballot:

1. Majority-Disqualification
2. Score
3. Approval

I didn't rank Vote-for-1 Plurality, because I don't approve it.

November29

 

Meta777--

Thanks for the favorable words. It's good to hear that there are people here who agree that we should have a better system than Plurality, which is the worst one possible.

I like Instant-Runoff (IRV) and Ranked-Pairs too.

Ranked-Pairs is the ideal best of the methods that look only at pairwise-defeats.

I want to say why I chose Majority-Disqualification as the ranking-method to include in the poll, but this is just a brief preliminary message. Another message will be posted shortly after this one.

But, briefly Majority-Disqualification is briefly-defined, and it meets the Favorite-Betrayal Criterion (never gives any incentive to vote someone over your favorite), doesn't have a chicken-dilemma problem, and gives solid protection to your ranked & approved candidates over the other candidates.

I'll be starting another posting in a few minutes. It might be fairly long, so I'd guess that it will be posted within about an hour or so.

More about my choice of ranked voting-system for the poll then.

November29

 

Now, about Instant-Runoff (IRV) & Ranked-Pairs:

Instant-Runoff:

First, let me define IRV here, for anyone who isn't familiar with it:

The voter votes a ranking of candidates. Only one candidate at any particular rank-level.

Repeatedly, delete from all of the rankings, the candidate who currently tops the fewest rankings.

When one candidate tops most of the rankings, then s/he wins.

[end of IRV definiition]


IRV is a good method, with an excellent combination of properties. But it has a problem. Not a prohibitive problem, but IRV isn't for every electorate.

I designated IRV as the count method for the post-election presidential-poll that recentlyI posted here. That poll only got two votes. The winner: It was a tie between Jill Stein & Donald Trump. But the poll is still open. It doesn't have a closing-date. The title of that poll's posting is something like "Instant Runoff ranked presidential poll".

If you haven't voted in that one, then I encourage you to do so. Break the tie.

Here's Instant-Runoff's problem:

I said that Majority-Disqualification gives solid protection to your ranked & approved candidates, against the other candidates. IRV doesn't give solid protection to your ranked candidates against anyone. Here's what can happen:

Say there are 3 candidates, A, B, & C. Say that B is the "Condorcet winner" (CW). That's the candidate who could beat each one of the others in separate pairwise elections. The A voters prefer B to C, and the C voters prefer B to A. B has a majority over each of the other candidates. The CW is the ideal compromise, and the most rightful winner.

But, in IRV, if B happens to be the favorite of fewest, then IRV immediately eliminates B. If the B voters have ranked A in 2nd place, then A will win, ending up at the top of most of the rankings. How do the B voters & the C voters feel about that? You have an angry majority. Not good.

A majority of the voters ranked B over A, and yet A won. That can't be called good protection of a majority's expressed wishes.

That's exactly what happened in Burlington, Vermont, in their 2nd Instant-Runoff election. As I said, there was an angry majority, and they immediately repealed IRV, and went back to using 1-Vote Plurality. That isn't what we want to happen in electoral reform.

I said that Majority-Disqualification meets the Favorite-Betrayal Criterion--It will never give anyone incentive to vote someone over his/her favorite. That can't be said of IRV:

In the situation I described above, the C voters could have avoided it. They could have protected B from elimination, by ranking B in 1st place, over their favorite (C).. Then B wins instead of A. the CW wins. That "favorite-burial strategy" by the C voters is the only thing that could save B from elimination, and enforce majority rule. The B voters themselves didn't have anything that they could do to protect their candidate.

That's why I say that iRV fails FBC. It fails FBC, and fails to elect the CW, in a way that got it repealed in Burlington's 2nd IRV election.

But I still like IRV, and I say that it's a good method. It just (as I said) isn't for every electorate.

If you know someone who wants to enact IRV, tell them that they've first got to understand and accept its problem. They've got to like and want IRV's unique offering and tradeoff, including its problem. "Otherwise don't enact it. And please don't enact it if you're going to rank a compromise or a perceived frontrunner over your favorite, to protect the compromise or frontrunner from elimination."

With that problem-disclosure and warning, I consider IRV to be a fine method. But, regrettably, no one gave that declaration to the people of Burlington. When the problem happened, they were rightly angry about having been deceived, and having IRV sold to them without a disclosure of its problem.

IRV has been heavily-promoted all around the country. To say that that promotion is "misleading" would be a big understatement. But because we're polite at this forum, I'll just say it that way.

But I like IRV. ...accompanied by the problem-disclosure & warning.

No voting-system is perfect. They all have some kind of strategy problem or considerations.

IRV has no chicken-dilemma. That always comes at some price. IRV's problem isn't really more than the going-rate for the price of that, and for IRV's other great properties and criterion-compliances.

And IRV's problem has mitigations:

1. If you're a C voter, and the B votes rank A in 2nd place (thereby giving the election to A when B is eliminated), well doesn't that mean that B is closer to A than to C? And, if so, then, with A & B being similar, then as a C voter, does it really make much difference whether the winner is A or B? Is that really a significant loss for you?

2. If C is big enough to eliminate B, then presumably C must be well-known. Now, if you, a C voter, like B, then B & C are similar enough that it's likely that the B voters like C too. ...and will rank C in 2nd place instead of A. So, if B doesn't win, then C, your favorite, will win.

...and, if you don't like B...if you don't like the CW, then you probably won't like the results of any voting-system.

----------------------------

So, with those mitigations, IRV's problem probably usually isn't as bad as it might sound. Maybe the Burlington Republicans should have been better sports about it. But it wasn't their fault that they weren't: They'd been seriously misled. IRV had been misprepresented to them. There was no problem-disclosure and warning. They didn't make an informed choice when they enacted IRV. They were suckered by promotion. They had a right to be angry.

But IRV has another problem too:

IRV isn't "precinct-summable":

A method is precinct-summable if each precinct only needs to send, to central count headquarters, some vote-totals for the various candidates, by which the count can be completed at central count headquarters.

If a method isn't precinct-summable, then it's necessary for each precinct to send its thousands of ballots in to central count headquarters, and the entire big count must be conducted there--away from the populations of the precincts.

That makes count-fraud a lot easier. Count fraud is much easier to perpetrate, and much more difficult to detect and prevent, if the voting-system isn't precinct-summable.

Now, if we had public ballot-imaging, then any voting-system would be secure from count-fraud. But we don't.

With the current situation, count-fraud detection is touch-I&-go. We pay 6 million dollars for a recount.

...by the same people who did the "count"? :^)

And what guarantee is there that the ballots have been securely-stored and transported, and haven't been modified, or some of them replaced?

...and that the recount will be any more trustworthy than the count?

No, under these questionable conditions, we don't want to make count-fraud even easier than it already is. A new voting-system must be precinct-summable.

IRV's lack of precinct-summabilty disqualifies it for our official public political elections (at least until such time as we have public ballot-imaging).
-------------------------------------
Ranked-Pairs:

I like Ranked-Pairs (RP) too, as I said in my previous post.

One problem that it has is that its definition is longer than Majority-Disqualification's definition. That's because RP's definition requires supporting definitions of several terms:

Supporting definitions:

X "beats" Y if more voters rank X over Y than vice-versa.

If X beats Y, then Y has a "defeat" by X.

A cycle is a cyclic sequence of defeats, such as X beats Y beats Z beats X.

If X beats Y, then the "strength' of that defeat is measured by the number of people who ranked X over Y.

The RP count:

1. A defeat is "affirmed" if it isn't the weakest defeat in a cycle whose other defeats are affirmed.

2. A candidate wins if s/he doesn't have any affirmed defeats.

[end of Ranked-Pairs definition]

Now, line #1 of the count-rule, above, defines the count, but it doesn't specify a procedure.

So here's a procedure:

a) Write the defeats in a vertical list, ordered with stronger defeats higher.

b).Starting at the top of the list, and working down, "consider" each defeat as follows:

...The defeat is affirmed if it isn't in a cycle with defeats that have already been affirmed.

...when the entire list has been thus considered, there will be some affirmed defeats.

--------------------------------------

That's the procedure for determining which defeats are affirmed.

When that's been determined then as I said, a candidate wins if s/he has no affirmed defeats.

Among the methods that look only at the strength of pairwise-defeats, RP is the only one that never unnecessarily disregards a defeat.

----------------------------

So why didn't I choose RP as the rank method to include in the poll?

For one thing, its definition, with the necessary supporting definitions of the terms that are used, is too long.

Also. I feel that, if you're going to use a rank-balloting method, then it might as well have the best properties. RP, and Condorcet in general, doesn't meet FBC, the Favorite-Betrayal Criterion that I defined earlier in this post.

Additionally, Condorcet, including RP, has the chicken-dilemma, and doesn't have any easy & reliable way to avoid it.

And, as I mentioned in my previous post, Majority-Disqualification has a briefer definition, and meets FBC, and doesn't have a chicken-dilemma problem. Those things make Majoriity-Disqualification the best choice for the rank-balloting method to include in the poll.

Rank-balloting methods are relatively arbitrary, and wordily-defined. It wouldn't be a good idea to include more than 1 rank-balloting method in the poll, because I want people to particicipate, to vote.

I really want to find out how people feel about those 3 methods.

I thank you, and the other poll-participant, for your input. ...for answering my question.

That question is the sort of thing that we who discuss voting-systems often argue about: Which methods would people like? Which would they like the most? I'm grateful for your answers to my question.

To anyone who hasn't voted yet: I invite you to vote in this poll.

November29th

 

I left a word out of this procedure-description. Here's what it was intended to say:

So here's a procedure:

a) Write the defeats in a vertical list, ordered with stronger defeats higher.

b).Starting at the top of the list, and working down, "consider" each defeat as follows:

...The defeat is affirmed if it isn't the weakest defeat in a cycle with defeats that have already been affirmed.

...when the entire list has been thus considered, there will be some affirmed defeats.

November29

 

When I said that rank-balloting voting-systems are arbitrary, that wasn't intended to disparage rank-balloting voting-systems.

I meant that rank-balloting voting systems are arbitrary in the sense that there many ways to count the rankings. Whatever rank-count you propose, someone could, and probably will, say, "Why that rank-count instead of a different one?" or "I claim that this other rank-count is better than the one that you propose."

That could easily result in no agreement being arrived at, regarding how to count the rankings.

And it's true that rank-count rules tend to be wordy.

I like rank-balloting voting-systems, because, with them, it's possible to avoid chicken-dilemma problems, and sometimes a few other strategy improvements.

But rating-methods, like Approval, 0 to 10, -10 to +10, etc, are perfectly good enough too, and would be a tremendous improvement if they were adopted.

November19

 

All are really bad ideas for too many reasons to list. Beyond too complex for most people, elections are not about who you "approve of" or how you "grade" candidates. The OP does not even hint why either of those are even a meritorious dictated selection process.

 

Anything other than one vote for one candidate would be too overwhelming for 75% of the population.

KISS

keep it simple stupid.

 

Well, you listed 2 claimed reasons, so let's start with those:

How complex is it to rate each candidate as "Approved" or "Unapproved"?

How complex is it to rate each candidate from 0 to 10, or from -10 to +10?

Want a complex problem? Ok, how about the problem of choosing whether to abandon your favorite, or to abandon the frontrunner-compromise who might be the only candidate who can beat the worst candidate. If you like that problem, then you'll love 1-Vote-Plurality.

Approval & Score don't have that problem.

But yes, a lot of people agree with you that the definitions and count-rule of the rank-balloting methods are unacceptably wordy. I hear that and take that objection seriously. If someone is going to propose one new voting system, it shouldn't be a rank-method.

Approval & Score are much better proposals.

I'm grateful for the input about that.

Yes, our current 1-Vote-Plurality elections aren't about that. They're about being allowed to support only one candidate, even when that means voting a sleazy, despised compromise over your favorite. Could that be the problem?

An election is about choosing a winner, from among a set of candidates.

There's nothing that says that can't or shouldn't be done by letting voters rate the candidates (as in Approval or Score), and electing the overall highest-rated candidate. (In Approval that would be the most approved candidate).

No, an election isn't inherently or necessarily about making that choice by 1-Vote-Plurality.

1-Vote-Plurality requires, forces, you to support only one candidate, and to designate all of the other candidates as "Un-Supported". If you want to support a disliked compromise (as so many people do), then 1-Vote-Plurality forces you to vote that s/he's better than your favorite. That's a lie, but 1-Vote-Pluralitly requires it, if you want to support a compromise.

Forced falsification of preferences is so common, familiar & well-known in our Plurality voting, that you think that's just a fact of life. Wrong. It's an unnecessary problem, quite absent from Approval-Voting or Score-Voting.


Voting freedom would consist of letting the voter himself/herself be the one to decide whom s/he wants rate favorably or unfavorably.

The question is, "What reason can you give, why that freedom shouldn't be permitted?"

If you can't give a reason for denying a freedom, then your advocacy for continuing to deny it isn't convincing.

So, what's your best objection to allowing voters to rate the candidates, and electing overall best-rated candidate?

Approval would elect the candidate who is rated "Approved", "Acceptable" by the most voters. How would that be bad?

If you don't have a reason, that's ok.

November29

 

I applaud the thought, but I'm not sure that these systems will work. For the first one, we'll have more problems with popular vote vs actual votes than we have now. Because let's say you vote for two people, the one you want most and the other one. And let's say there's another guy who votes for the runner up candidate you votes for as his first choice. Who has precedence? From your system design, it would seem like each person is awarded only one vote, so in reality, the runner up could win even though the majority of the country would rather not have him.

Second one is basically the same as the first one. The US is a republic, not a democracy. If we do things like this, both the republican and democrat portion of our governmental system will start falling apart. This system essentially makes the radical have more of a say in the election. Because a staunch, radical Trump supporter for example would mark 10 whereas a moderate may mark 6. The radical gets more votes.

Third one is basically the first one but with the extra clause so.


Fixing our system is simple, actually. The main issue is that the winner takes all. So in a state like Florida, where something like 52% was Republican but 48% was Democrat, the Republicans get all the votes. But that's not representative of the population. We, of course, have to remember that we live in a Republic, so it makes sense to consolidate like that, we aren't going off the popular vote. But I think that, if 52% is Republican, 52% of the electoral votes should go Republican, like in Maine. I propose we do it like this.

We do the votes by congressional district, like we do now. Only, if the district votes a certain way, that elector is the one who choses the candidate even if the rest of the state votes the other way. The party with the majority also gets the two senator votes from the state, and if its a tie, each gets one senator vote. This makes every state a swing state, in that, you have to win swing districts now too. So California, which is very blue, would still award some of its votes to the Republicans because of those little congressional districts. This would also make gerrymandering unfavorable once each district is somewhat politically homogenous.

This system has a historical precedence too. Basically what happened was that in the 1800s, states had about the same population as congressional districts do today. So voting by state was actually representative of the population but still part of the republican form of government. This started becoming a problem in the late 1800s, 1876 to be exact, because populations started rising. So you could win 51% of a state and get all the electoral votes, but the majority of the population actually didn't vote for you. But if we shrink the scope to only congressional districts, you would win both the popular vote and the electoral vote if you won the election.

 

Approval is used in the election of the Secretary General of the U.N. Centuries ago, it was used for a long time in the election of the Pope.

Not work? How could it work less well than our 1-Vote-Plurality, the worst possible voting-system?

With Approval, the winner will always be the candidate approved by the most people. The candidate whom the most people consider approvable. The candidate the most people consider satisfactory.

How is that worse than what we get when 1-vote-Plurality gives many or most people a strong incentive to vote for someone they dislike, as a "lesser-of-2-evils"?

With Approval no one will be voting someone they don't like over their favorite. For that reason alone, Approval will work incomparably better than 1-Vote-Plurality.

If I approve A & B, and someone else just approves B, then B gets more approvals from us two voters than A does.

I consider A & B to be alright. The other person only likes B, and doesn't consider A satisfactory.

Maybe I approve B, in addition to A, because I know that A has little chance of winning. I want at least one of A or B to win.

...or maybe I feel that B is nearly as good as A, or that they're both excellent. In any case, as I said, Approval always elects the candidate considered by the most people to deserve approval. The candidate that the most people consider satisfactory.

Each person can rate as "Approved" as many candidates as s/he wants to.

The candidate who is satisfactory to most people is the most justified winner. S/he's a lot less likely to be disliked by a majority...compared to the winner in 1-vote-Plurality, where many or most people vote for someone they dislike, as a lesser-of-2-evils.

That could happen a lot more easily, and happens often, with 1-Vote Plurality.

If a majority dislike someone & want him to not win, then they won't approve him.

Yes, Score is like Approval, with the flexible option to give fractional approvals (ratings less than the maximum).

Maybe the government would start being our government for a change.

Why would the government start falling apart if we start electing the candidate who is considered satisfactory to the most people?

The result would be that we'd have government that most consider satisfactory. How would that be a bad thing?

Everyone can, would, & should give a 10 to their favorite. ...and probably to all of the very best candidates. Maybe give a lower rating to a candidate who isn't quite deserving of a 10, but whom you want to give some help against the worst candidates.

You expressed concern about majority-rule. Then what's wrong with disqualifying candidates over whom a majority rank someone else?

Weed-out candidates against whom someone else has a majority. Why not?

...and then elect the remaining candidate who is satisfactory to the most people.

November29

 

I disagree. The government here is neither a republic nor a democracy.

"Democracy" means government by the people.

"Republic" means a thing that is of the people.

"Our" government is not ours, because it is neither of, by, nor for the people.

"Our" government is of, by, & for the rich.

November29

 

Amen!

Make that AMEN!

Any of these methods would probably put the entire state of Florida out of the voting business.

- - - Updated - - -

We ARE a republic...whether some want to think jadedly about it or not.

 

The voting system I would prefer for presidential elections would be what we pretty much have with some changes. All states would have a winner take all electoral votes, but only if a candidate receives 50% plus one vote. In our last election, there were 14 states where neither Trump nor Clinton received 50% plus one vote. In the 36 states plus D.C. this would have left Trump ahead in the electoral count 198-182. The remaining 14 states since no candidate received the required 50% plus one would go to awarding their electoral votes via who won the congressional districts plus awarding the final two to which candidate won the state with a plurality of the vote. Clinton won seven of these states and Trump seven. Being I have no idea of which candidate won the congressional districts within these 14 states, I can’t post what would have happened or what the final electoral vote count would have been.

Either that or do it the way we do in Georgia, if no candidate receives 50% plus one, there is a runoff between the top two vote getters. I also like the Louisiana jungle primary system where all candidates are listed on the ballot regardless of party and there are at times four, five or more candidates from the same party. The top two finishers if no one gets 50% plus one vote face each other in a runoff like the recent decided Louisiana senate seat decided yesterday. But I think in order to gather all the electoral votes from any states, a candidate should receive a majority, 50% plus one vote.

 

Marking the candidates who are satisfactory to you is a lot easier than having to decide whether to abandon your favorite, or abandon the chance of affecting the election by voting for the best frontrunner, the candidate who is the only one who can beat the worst candidate (the Democrat, at this forum--Well, for me, the Democrat is one of the two worst).

The Democrat would never get an approval from me.

The media claims that the Democrat & the Republican are the "frontrunners". No, they're the bottom-runners.

You might be surprised by who the frontrunners would really be, if we had honest, verifiable vote-counting, and if we had honest, open, participatory, agenda-free media (without which there can be no such thing as an informed vote).

"The real voting-power belongs to him who counts the votes."


- - - Updated - - -

Strictly speaking, we aren't a republic unless our government is of the people. It's of-the-rich-people, so I guess you could call it a republic-of-the-rich.

November29

 

Yea...ok. Well, lemme ask you this, is there any foreseeable way rich people cannot run the government? Has there every been a time when government was not the elite? Even under your proposed systems, people have to campaign and such, so they need to be able to afford not to work for a while, they need to fund their campaigns, they need all sorts of things, that only money can buy.

And furthermore, elitism is something that is clearly necessary for human development. People never have a problem with the elite if they have a chance at becoming that elite. The only reason that you have a problem with it is because its becoming increasingly hard to get rich. If you were rich, you wouldn't care.

And if you don't think the government gives the people a say, move to North Korea and see what that really means.

 

But it all comes back to the issue of population. First off, the Pope centuries ago....way to bring up my corrupt Church. And the secretary general of the UN is not voted on by the people, now is it. He's voted on by the general assembly, which is like over a hundred people. Contrary to popular belief, things that work for small populations like popular vote do not work if you have a population over 300 million. Which is why we can't compare ourselves to countries like Sweden or Canada. Approval would be such a difficult and daunting task that we may as well stick with our current system, which is broken by population already.

 

Fair enough, it wasn't a good comparison, because the Secretary-General & the Pope are elected by smaller electorates than a U.S. president.

So you can say that Approval hasn't been tried for an electorate of 300 million. But 1-Vote-Plurality has been tried, and it's been giving us notoriously bad results, a cynical, disaffected population, half of whom don't bother to vote, because the media tells them that (because of 1-Vote-Plurality's ridiculous problem), they have to hold their nose & vote for a "lesser-evil".

How bad could it be to elect the candidate who is satisfactory to the most people? It's like finding a guy who keeps sticking an ice-pick into his foot. You try to suggest that he not keep doing that, and he replies, "No way! I haven't tried not doing this! I'm going to stay what what I've tried."

That depends on what you mean by "work".

Approval would work by electing the candidate who is satisfactory to the most voters.

1-Vote-Plurality "works" by forcing people to choose between abandoning their favorite, or abandoning their chance to affect the election-result by voting for the best (alleged) frontrunner. You call that "working"?

The electoral college "works" by giving a Rhode-Islander many times the voting power, compared to a Texan. How democratic is that?

Which would be a difficult & daunting task?

1. To mark the candidates who are satisfactory to you?

(instead of having to decide whether to abandon your favorite, or abandon your chance of affecting the election-result by voting for the best (alleged) frontrunner, the only candidate who can beat the worst candidate?)

(I'll add that that decision is even more difficult due to a media system that is consistently and systematically disinformational and agenda-driven.)

or

2. To count Approval ballots? They're like Plurality votes. The existing voting equipment and ballots would work fine. The existing software would work fine too. All that would be needed would be to turn off the provision that detects & rejects "overvotes". Cost of changing so Approval? Zero.

If we can count Plurality votes, then we can count Approval votes.

Like the guy who keeps sticking the ice-pick into his foot, because it's all he knows, and at least he's tried it.

November29



, which is broken by population already.[/QUOTE]

 

we are a republic...even if some want to deny that we are.

 

I prefer different voting systems based on the scenario. Ideally all government would be small enough to have a single caucus in a stadium wherein people raise their hand to indicate voting. A modern caucus if you will. This would be inherently FPTP.

Given we do have a large, imperialistic nation state I like STV. Works well here in Australia.

 

[/QUOTE]

Ok, again though, we aren't a democracy, we are a republic. That's number one. Number two, this is the failing of left-wing thinking that's so destructive. "if we don't know if it works, it is still better than a system that doesn't work" vs "we know what doesn't work but we are sticking to it" (that's right wing idiocy for you ) I'm not saying stick to this system, I know it doesn't work. I'm saying that these systems simply won't work for a population as large as the US. The statistical margin of error is already so large that adding in a vote by Approval would make it a lot worse. With a population so large as the US, you need to consolidate the votes into smaller elections, as a republic would work. What I'm saying is that this consolidation used to be done on the state level because the US population allowed it so, but nowadays, that level has to be shrunk even further to account for our larger population. It also shrinks the margin of error, because the less amount of people you have to count for, the more accurate you'll be. Doing the vote by congressional district also gives people more of a say, balancing out the democracy and the republic within the US government.

The voting system worked....up until 1876 because the US population rose to a point where you could win the popular vote but still lose the election. We already know what works, we just have to fix our broken system within the bounds of what works to make sure we keep up with the time.

As for our voting software, do you know how much the right criticizes the votes because of the "illegals"? switching to this system would get you more pushback from them than ever before. And they kinda do have a point, it is going to be a lot harder to account for voter fraud if a larger amount of votes can be counted.

 

What about having districts where candidates are popular, then have a decision based on the amount of votes from each district the candidate receives; I guess you can call the district County. Then after this have each district candidate up for decision from the multiple Counties and then the winner of these series of elections, is the State representative. Maybe this is what is already done? This, of course would be for Governor or State Executive.

 

You'd have to be a lot more specific about what problem you're referring to there. Counting all of the Satisfactory ratings wouldn't create or worsen any statistical problem. It isn't clear what statistical problem you're referring to.

The problem of proving identity & citizenship would be unchanged.

The assurance that only citizens are voting, and only voting once, is independent of whether 1-Vote-Plurality or Approval is used.

The problem of detecting & preventing count-fraud, likewise, isn't any more difficult with Approval than with 1-Vote-Plurality.

Count-fraud is the real problem. Currently the vote-counting is unverifiable. About 1/4 of the states don't even have a paper ballot to save as evidence. But even when there are paper ballots, the transport, storage, & counting of those ballots needs to be demonstrated to be secure.

How would we get security against count-fraud? By public ballot-imaging.

Traditionally, we had big public hand-count events, with the population watching, and with party-representatives participating. We need something like that again. "Make America democratic again."

One at a time, each ballot is laid on a table, and stamped with a sequential number. Above the table is a frame, on which are mounted digital cameras belonging to, & operated by, a number of political parties, all across the political spectrum. I'll call them the "observer-parties". Additionally, there's a camera owned & operated by the precinct.

Each ballot is imaged by each observer-party camera, & by the precinct camera.

Each party can scan its ballot-images, into a program that will do a count. Each party can do its own count, as can the official electoral system as well.

If any of these count results differ, then the disputed ballot (they're all numbered) could be retrieved from secure storage, and re-examined, and re-imaged.

Thus any disagreement could be resolved reliably.

Representatives of all the observer-parties would be observing the overall imaging event, and would ride with the ballots in the truck that transports them to and from the secure storage.

The secure storage should consist of a reinforced concrete building, windowless, each of which would be protected by cameras, locks, and alarms belonging to & operated by one of the observer parties. Each observer party could have its own locks, cameras & alarms protecting the stored ballots.

Attempted count-fraud or ballot-tampering would result in a mandatory sentence of life imprisonment, without possibility of parole, in a maximum-security prison.

Do we want count-results that are secure against count-fraud, or not?

November29

 

Current. The others are pretty nuts, and the data analysis required would make it extremely hard to implement at anything but a local level.

 

Translation: "I'm a conservative, and anything that's new to me must be nutty, by virtue of the fact that it's new to me."

Nonsense. "Extremely hard analysis" to add up Approval votes?

As for Score & MDDA, I want to break the news to you that we now have something called "computers". Even an ordinary household personal computer can do an enormous count, by a pairwise-count method such as MDDA, in a few minutes.

But of course, until we have good verifiable vote-counting, by public ballot-imaging, Approval would be a much better idea than MDDA, or Instant-Runoff, or even Score.

The "difficult implementation" objection to Approval is entirely ridiculous.

Here's what would be needed to change over to Approval:

1. Turn off the feature that detects & rejects "overvotes".

2. Add two new words to the ballot: Where it now says "Vote for 1", change that to "Vote for 1 or more".

How "extremely hard" would that be?