What existing voting system is this most similar to?

So the system goes as thus.
On any given ballot you will be given 3 votes. There could be more or less, but 3 seemed like a nice number. So on the ballot you will mark who you want Most, Next Most, and Next, Next Most. A vote of Most will equal 3 points, a vote of Next Most will equal 2 points, and a vote of Next, Next Most will equal 1 point. There will also be a slot where a person can fill in their Next Most and Next, Next Most vote so it won’t be counted, but those ‘dropped’ votes will still be counted for other reasons.
So all the ‘points’ are tallied, if one wanted to get an estimate of how many ‘people’ voted for who, the total points could be divided by six. Percentages for federal funding could still remain the same, and ‘dropped’ votes don’t count towards those percentages. Otherwise, it’s the same majority rules as our current system.
Now, my question is, what currently existing system is this most similar to? The point of this system is to try and get rid of the two-party mentality, and I think it would work. What would some weakness of this system be? And lastly, how likely is it that we would ever be able to get our voting system changed while the Senate and House are still ruled by the two dominating parties?

Scroll down on this page to the paragraph on Majority Preference Voting, a form of proportional representation. As the explanation states:

Ok, now I know what system it’s most similar to, can someone answer my other questions?

Thanks for your response Sternvogel!

Just a WAG, but I’d guess this could cause a candidate with a small yet enthusiastic crowd of supporters win over a candidate a majority would consider acceptable, although not ther #1 choice.

Consider a situation with five candidates and your 3-2-1 point system. 100 people are entitled to vote. Candidate A is a blockhead, but 20 of the 100 voters absolutely love him and give him their 3 points; they don’t give out the 2 and 1 points because they don’t want to support anyone but their darling. All the other 90 voters think A is a blockhead and would rather die than give him any points. A yields 60 points altogether.
Candidate B is nobody’s absolute favorite, but everybody (except for the 20 supporters of #1) thinks he’s acceptable. On everybody’s priority list (except for the 10 supporters of A), he’s ranking higher than A, even among voters who don’t give him any points (because there are five candidates and they can only give out three votes - B could be their number 4 and A number 5). I suppose that, based on this, you could draw up a situation in which A does not only defeat B but also all the other three candidates, although the majority of voters would prefer B (or some other guy).

*): No references to any recent political events.

I believe Kenneth Arrow did a number of studies that showed that there is no voting system that meets the basic requirements you would think a voting system should have. Unfortunately, I have never been able to find a laymen’s writeup of his work anywhere, but it’s great stuff.

So, there is a weakness to yours, because all voting systems have anti-democratic weaknesses, but I do not have the materials to track down exactly what that weakness is.

Well, Google might help…


…and some of these seem to think I am overstating the case. Nevertheless, very enlightening.

Chapter 3 onwards of the above link is really good:


The system you describe in the OP is pretty much what is known as Instant Runoff Voting, or IRV. It’s the system that’s used in Australia.

One thing that no-one has mentioned so far is that you can change the effectiveness of this voting system based on whether or not you allow blank spaces on the ballot.

For example, if you do allow blank spaces, it means that people can choose to vote only for their main, preferred candidate, without giving any weight to the other candidates. A completed ballot might look like this:

Candidate A [ ]
Candidate B [1]
Candidate C [ ]
Candidate D [ ]

This would give first preference to Candidate B, but if Candidate B were eliminated, then this vote would be of no use at all to any of the other candidates. A system allowing blanks might, depending on the rules, even produce a ballot that looks like this:

Candidate A [ ]
Candidate B [1]
Candidate C [2]
Candidate D [ ]

In this case, the voter is happy to support Candidate C if B gets knocked out, but under no circumstances wants to support candidates A or D.

On the other hand, if you do not allow blank spaces, the ballot would look like this:

Candidate A [3]
Candidate B [1]
Candidate C [2]
Candidate D [4]

In this scenario, the voter may end up helping candidate A, whether he wants to or not.

It’s not quite clear to me which of these scenarios is more likely to result in a real challenge to the two-party system. If you allow blanks, you might get some third-party voters who give no preferences to the major parties, thus helping small parties at the expense of large ones. But you might also get major party voters who give no preferences to small parties, thus effectively making the election a regular, first-past-the-post affair, like the current US Presidential election.

On the other hand, if you don’t allow blanks, you’re effectively forcing people to cast a vote for one of the two major parties somewhere along the line, as the smaller parties get eliminated.

To work this out, you’d need a good knowledge of statistics, combined with information about how different groups would be likely to vote if they were given a choice to leave some squares blank.