I do not believe IRV (Instant Runoff Voting) is even remotely a good election method and if you would like to discuss with me my reasons for believing this, you are welcome to write me. If you would like to learn about some of the problems with IRV, I would recommend visiting:

The Problem with Instant Runoff Voting
Instant Run-off Voting (IRV)
Computer Simulations Comparing MAM and Instant Runoff

I have only included the method on my site to more easily compare results with far better methods.

One of the key questions in an implementation of IRV, which allows candidates to be ranked equally, is how much equally ranked candidates contribute to their vote. Assuming that there is a ballot which looks like A=B>C. The 'Whole Votes' would grant 1 vote to both A & B. The 'Fractional Votes' method divides one vote equally among all the equally ranked candidates and would grant .5 votes to both A & B.

Another one of the key questions in an implementation of IRV is what to do in the case where there are two or more candidates tied for least votes. I have implemented several common methods. The 'Prior Round Totals' method looks at the vote totals from the prior rounds and if a decision still cannot be made, a random select is made (see demochoice.org). The 'Deterministic' method eliminates all tied candidates at the same time, unless they are the only remaining candidates. The 'Random' method simply selects a tied candidate at random and eliminates it.

Back to the voting calculator