HOW PREFERENTIAL VOTING SHOULD REALLY WORK
By Kevin Loughrey
The objective of preferential voting is to provide a measure of which candidate has the most support amongst the electors of an electorate.
The present system of preferential voting does not achieve this. The present system is, in a way, a variation of "first past the post" voting system. It is not a pure preferential system.
Let me explain:
At present the primary votes are counted and whoever received the least number of ballots with “1” next to their name is washed-out and their preferences, ie, the candidates on that ballot with a “2”, "3", etc, are distributed to the remaining candidates. The counting then moves onto the next candidate with the least number of ballots with a “1” in the ballots, and so on. The preferences for the washed-out candidates, that were expressed by voters on the ballots of the remaining candidates are lost, that is, these preferences do not, at some later stage in the counting, flow back to the candidates who were washed-out. This is where the system does not faithfully represent the preferences of the electors.
If you are confused at this point then you are not alone! Most people haven't a clue how the preferential system of voting works. I didn't until I ran for office.
Let me illustrate the point I am trying to make by way of an example. Here I show you a fictional (deliberately extreme) example in a table with 5 candidates running for a seat. In this example, the 5th candidate received fewer first preference votes than any of the other candidates but the ballots of the remaining candidates had, as their second preference, Candidate 5. In other words, electors were saying, if they couldn’t have their preferred candidate, they would happily accept Candidate 5. That’s true preferential voting. Instead, Candidate 5 is washed out and the preferences that electors gave to him/her do not flow back. They are lost.
Here is what should happen if we want a true indication of the preference of the voters. As shown in the table, there are 5 people running for an electoral division (called an "Electorate"). If a candidate receives a 1 on a ballot, they score 5 points. If they receive a “2” on a ballot then the score 4 points. And so it goes, a 3 = 3, a 4=2 and a 5=1 point and a blank box equals 0 points. All that is required is to then add up the number of points each candidate was awarded by the electors. Whoever received the most number of points wins the election. Where there is a tie, the person who received the most number of 1st preference votes, ie, a “1” in the box, wins. If this is still tied, it is the most number of 2’s and if this is still tied then on to the 3’s and 4’s until one of the candidates receives more of those preferences than the candidate they have tied with.
You can see in this example, the 5th candidate got the lowest number of primary votes but received the highest number of 2nd preferences of any of the candidates running. (Note that, if you add up 10+100+200+300 and 13,390 you come to 14,000 ballots, not the total of 15,000 of all ballots cast. That is because 1,000 ballots already had 1 in their box for the 5th candidate and so are excluded from the 2nd preference count for the 5th candidate.) This resulted in Candidate 5 having the highest number of points, 58,560, and therefore being the most preferred amongst the electors.
That is how true preferential voting should work.
I would like to obtain all the ballots for an election and see what the result would have been had this system been used. That would be a very interesting exercise.
What do you think?
Please let me know.
