Our current system of voting is flawed. Realistically, only the two strongest candidates have any chance of winning. A vote for third party candidates cannot affect the race between the two main candidates, and a strong third party contender can tip the election quite easily. When the vote is close, every third party candidate has a huge influence, as happened in Florida. As more and more people become dissatisfied with our current system, the popularity of third parties is likely to grow. Like Perot in 1992, and Nader in 2000, our elections are more and more likely to contain "spoiler" candidates. Voters who agree with the Green, Libertarian, Reform or other platforms, are torn. Should they vote for one of the two major candidates, the only ones who have a realistic chance? Or should they vote for the only candidate that actually represents their views at the expense of deciding who will win?

There is a solution. If we change our current antiquated voting method, voters will be able to support the parties they actually believe in and cast a meaningful vote at the same time.

IRV has received the most attention as a possible alternative to our current method. The way IRV works is that candidates are ranked in order of preference. When the votes are tallied, the candidate with the least total amount of votes dropped, and anyone ranking that candidate first has his or her vote transferred to their next listed candidate. If a voter only indicates one candidate, once their candidate is eliminated, their vote does not transfer. Unfortunately, IRV only displays true voter intent as long as third party candidates do not actually have a chance at winning. Consider this hypothetical example. There are three candidates, a Democrat, a Republican, and a Green. Each are polling about equally well. If in the first round, the Green has the least votes, then he will be knocked out of the race, and most of the votes will (probably) transfer to the Democrat. If, on the other hand, the Democrat is eliminated first, it is unlikely that most of their votes will transfer to the Green party, and instead the Republican candidate will probably win. In this way, voters who voted for the Green party and actually prefer the democrats over the Republicans will have helped the Republican candidate to win with their support of the Green candidate. Just as in the current election method. Furthermore, if some of the Republican supporters lied about their first choice, and rank the Green candidate first, and the Republican second, it could ensure a run-off between the Republican and the Green, of which the Republicans would be likely to win. The fatal flaw with this system is the same as with our current method, voters must lie about their true first choices in order to have a preferable outcome in the election.

Approval voting is the method in which voters are enabled to vote for as many candidates as they like, in effect allowing voters to both vote for their preferred third party candidate and the main candidate they prefer the most. (For example, someone could vote for the Reform party, the Libertarian party, and the Republican party.) This method is superior to IRV for two reasons. First, it could be used with our current voting system, by punching more than one hole in a ballot and counting both votes. Second, it does not suffer from the problem that IRV does, described above, in which it is beneficial to lie about a preferred candidate. The only drawback that approval voting suffers, is that it is difficult for a third party to actually elect a candidate, because it means that they would have to stop supporting the main candidate that comes closest to their beliefs. For example, assuming most Libertarians would vote for both the Libertarian and the Republican candidate, in an election between the two parties, the Republican will always win if the Libertarians vote for him as well as their favored candidate. In order for the Libertarians to win, they would have to vote only for the Libertarian, possibly costing the Republican the election. This method is better than our current method, and IRV because voters do not have to lie about their preferences to cast a meaningful vote. However, it does suffer from the aforementioned defect, while the Condorcet voting method does not.

The input for the Condorcet method works the same way as IRV; voters rank the candidates in an order from most preferred to least, with no ranking indicating no preference. However, Condorcet does not count the votes in the traditional manner. Instead, each candidate is compared in a matrix against each other candidate, and the candidate that has the most votes in individual comparisons with other candidates is the winner. It seems complicated at first, but this method is best illustrated and clarified with an example. There are 3 candidates, A, B, and C. Each voter ranks the candidates in order of preference. If a candidate is ranked first, it means he is preferred over all other candidates, if second, he is preferred over all other candidates except the candidate ranked above him, and so on. A three way race would generate 3 individual races. A against B, A against C and B against C. If a majority of voters preferred one candidate (say A) over both others, he would win flat out. A would win the race in A vs B and he would win in A vs C. B vs C is inconsequential because A beats both of them. Things become slightly more complicated if voters display a cyclic preference. That is, A is preferred over B, B is preferred over C and C is preferred over A. In this instance, a winner is determined by dropping the weakest defeat among the set. For instance, say A beats B by 20 votes, B beats C by 15 votes, and C beats A by 10 votes. In this instance, the weakest victory is C over A, so C is eliminated from the race. Now A wins.

Of the methods discussed, the Condorcet method is the only one that always displays true voter preference and does not ever require voters to lie about their true preference. Unlike the other methods, where it is sometimes advantageous to misrepresent your true beliefs, the Condorcet method will never reward dishonesty among the voters.

(For more information about the Condorcet voting method, see http://www.electionmethods.org)
a