[Lobowolf]

jdonn, on Nov 14 2008, 12:31 PM, said:

I strongly disagree with voting your preference out of who has a chance to win if it's not your first choice. That gives people crazy ideas like that 50 million people want George Bush to be president from time to time.

I agree (!!)

I don't see how any given vote for Nader is any less meaningful than a vote for Obama or McCain would have been.

[Al_U_Card]

Actually, it counts more! It takes one vote away from one of the candidates and that makes the other one one vote closer....

[mycroft]

It's better than a "no vote".

In the warped world that is U.S. politics, it does tend to work out that voting for a third-party candidate is a vote for the person you least want to win (as its -1 for the one you'd vote for *if* you were forced to pick).

In Canada, voting for a party gives them (if they win enough, but enough is rather small) $1.75/year until the next election for campaign funds. Therefore, there is a reason.

[mikeh]

JoAnneM, on Nov 14 2008, 12:13 PM, said:

Voting for a 3rd party candidate is doing nothing more than taking a vote away from either the Democrats or Republicans, and if you want your vote to mean absolutely nothing then go right ahead.  I would rather make a positive statement and vote for someone who is running with a chance of winning.

That depends on your timeline.

Here in Canada, the Green Party is slowly gaining popular support. Its share of the national vote is still small, and the party really didn't come close to winning a seat, altho a number of its candidates did quite well in their electoral areas.

I think that the party will continue to grow... and that it may, just may, someday attain a critical mass where it will finally be seen as having a real potential to affect the balance of power.

Our method is fundamentally different from the US, in that we have multiple parties (4 different parties have seats in parliament) and the Prime Minister is the leader of the party that is able to command a majority of votes in the legislature...which doesn't mean having 50%+1 of the seats or, as happens elsewhere, forming a coalition government: a minority government can function by pandering to another party on specific issues, by staying away from controversial topics, or by taking advantage of the reluctance of the opposition parties to trigger another election... our elections happen whenever the government decides to call an election or after the government is defeated in a vote of no-confidence (altho the election cannot be delayed more than 5 years)

Thus a party can gain some effective power even while a long way from having any chance of becoming the government. Indeed, the Bloc Quebecois, is a one-province party, with neither any desire nor any chance of success outside of Quebec, but it has influence in parliament because it has enough seats that, in combination with the other opposition parties, it can force an election.

All of this tends to grant a new party a slightly better chance here than a similar party in the US.. but here and there, the new party needs to attract voters in steadily increasing numbers even when, or especially when, the votes will be, short term, wasted.

[helene_t]

jdonn, on Nov 14 2008, 06:43 PM, said:

I actually think the best system theoretically would be to deduce the true opinion of each voter about each candidate on a scale of say 0 to 10 and then add up the total points. Of course it would never work in practice, mostly because people would rate their favorite 10 and everyone else 0.

A lot of research has been done on optimal election procedures.

It can be proven mathematically that every scheme with more than two alternatives is subject to "manipulation", i.e. it may sometimes be in the interest of voters to lye about there pov, e.g. by voting for their second choice because he has better chances than their first choice.

But in practice, there is scheme, "the absolute majority" rule, which works pretty well. It is this:
Each voter is asked to rank all candidates, say you might give 0 to Barr, 1 to McCain, 2 to Hillary, 3 to Obama, 4 to Nader or whatever. Now, for each pair of candidates (say Barr vs Hilary), the winner is computed on the basis of a hypothetical vote with only those two candidates. If someone wins all of his hypothetical head-on matches he will become president.

Theoretically this might not work because it is possible that e.g.
Hilary>McCain
McCain>Obama
Obama>Hilary

but in practice it will almost always work.

[DrTodd13]

What Helene is describing is called Condorcet voting. The theorem that all systems of voting (except dictatorship) violate some basic desirable principle of elections is called Arrow's Theorem. In practice, while it is theoretically possible to not vote your true preference and have that influence the election (under Condorcet), the information necessary to determine how to manipulate is extremely difficult to get and even if most people had it they couldn't figure out how to use it. The stupid approach of putting Obama first and McCain last with super-whackos in the middle is not guaranteed to increase Obama's chance of winning. In this sense, it is good that the obvious tactic will often fail. I don't think it makes a lot of sense to combine Condorcet with an electoral college. For it to be really useful I think you'd want to have it a unified national vote rather than spread out via states. Plurality voting (you get one vote among tens or hundreds of candidates) is close to the dumbest possible voting system. A revolution is a necessary but insufficient condition for Condorcet to be instituted because as long as they are in power the two party duopoly have no interest in allowing people to express their true preferences.

There are several good and justifiable mechanisms to use should there not be a clear Condorcet winner.

[luke warm]

at the time hillary had her interview on o'reilly, the dem nomination was still very much in doubt... i watched because i wanted to see how she handled that arrogant egomaniac, and i came away convinced that i'd vote for her.. i was very impressed and still think she would have made the best president of the bunch... i actually think she's politically as close (if not closer) to mccain than she is to obamba, and i believe she'd be a more formidable commander in chief

[jdonn]

I still find the way she does politics extremely dispicible, but you may be glad to know it's looking very likely she will be secretary of state.

[akhare]

DrTodd13, on Nov 14 2008, 06:33 PM, said:

What Helene is describing is called Condorcet voting.

I smell a conspiracy here .

The latest Microsoft MSDN magazine has a very good feature on a voting methods and one that satisfies the Condorcet technique as well:

http://msdn.microsof...e/dd148646.aspx

Condorcet Technique and Schulze Method
The Condorcet technique for collaboratively determining the best option from a set of choices was developed primarily as a reaction to the head-to-head paring problem of the Borda count method. The Condorcet method is very simple. It requires evaluators to rank all alternatives, and then a comparison of the results between each possible pair of alternatives is performed.
...
One of the most interesting systems that satisfies the Condorcet principle has a variety of names, including the Schulze method, the clone-proof Schwartz sequential dropping technique, and the beat-path method. I will call this technique the Schulze method.

[pclayton]

Vote for who you want to. There's no agenda here, except your own personal choice. If you like Nader, Perot, or Bob Barr better than the top 2, then its your prerogative.

What I can't stand is those that say you have to vote for 'x', because the 'x' party needs you. You aren't throwing away anything, since your vote is yours to begin with.