So, the byelection was held on Monday night and in Brandon-Souris Larry Maguire of the Conservative Party of Canada won by 391 votes over Rolf Dinsdale of the Liberal Party.
Congratulations Mr. Maguire.
First things first, Forum Research’s polling in the riding was horribly inaccurate. The closest poll in fact was the straw poll done on Friday on 880 CKLQ radio’s Feedback call-in show. Had the polls been more accurate, there is a possibility that people who vote strategically may have changed their votes from what they actually voted. That is the strength of polls, they let people form a strategy in a first past the post system. It is also the downfall of polls, they let people form a strategy in a first past the post system. The question is, does a horribly inaccurate poll actually affect the outcome? I don’t know, but it seems a good question.
There have been various articles in the media about the results nationwide and how they show a Liberal increase and a Conservative decrease in support, however, one article in the Winnipeg Free Press by Deveryn Ross shows how the Liberal Party may have defeated itself in Brandon-Souris. Interesting read.
So, now that we have actual numbers, I’m going to go back to see how a preferential ballot may have affected the outcome of the race in Brandon-Souris. It has been suggested to me that a preferential ballot would unfairly benefit the Liberals because they are the second choice of many people. In Brandon-Souris that is probably true. Firstly, is that necessarily unfair? Secondly, is that necessarily the case? There are some ridings where the Liberals are a distant fourth; like Brandon-Souris in 2011.
Now to be honest, a preferential ballot in 2013 would have probably favoured the Liberals in Brandon-Souris. As a current Liberal supporter it could be argued that my advocating of a preferential ballot would show personal bias. However, I have been an advocate of the preferential ballot since before 2003 when I was a member of the Progressive Conservative Party, one of the two precursors to the current Conservatives. My desire to see a preferential ballot has been steady for over a decade.
So, I’m going to approach this with various thresholds in mind. The standard way that a preferential ballot would work is the first to get 50% + 1 votes in a particular round would be considered the winner. Now, 50% +1 at first glance seems reasonable, because it is looked at as the most democratic, a majority. It could be argued however that a majority of voters isn’t necessarily what is needed, just something that would be considered a “good showing”. It could be argued that Mr. Maguire’s 44.1% of the vote should be enough to win, and I actually may agree with that. So, we’ll look at three levels 40%, 45%, and 50%.
So, round 1 with a preferential ballot. At 40% threshold, since both Dinsdale and Maguire are over 40%, Maguire wins. At 45% no one has reached the threshold, so you go to round 2, as at 50% threshold.
In round 2 we are going to assume that it is reasonable that all of Godon’s supporters would pick Maguire as their second choice. In round 2 that means that Maguire would be at 45.1% of the vote. At 45% Maguire wins, at 50% we go to round three.
In round three we assume that 25% of the Green vote goes to Dinsdale, and that 75% goes to the NDP. We would now have Liberals at 12152, Conservative at 12476, NDP at 3053. That is 43.9%, 45.1%, and 11% respectively. At a 50% threshold we still have no winner. So we now go to round 4.
Round 4. We’ll throw the Conservatives a bone here. For some reason, there are 150 Greens that select Conservative as their third choice, and 200 NDP pick Conservative as their second choice. Assuming the rest of the votes have the Liberals as their next choice, we end up with Conservative at 12826 votes, and the Liberals at 14855 votes.
Dinsdale wins with 53.66% in round four.
Now there are a few assumptions made in these scenarios. Turnout at the polls is exactly the same, and that everyone numbers the candidates in order of preference. An “x” beside one candidate would still count until that candidate failed to make a round. It is an interesting exercise though.
I personally do not know what the threshold should be with a preferential ballot, but I do know that the possibility, however small, of a candidate winning with no more than 21% of the vote in a 5 person race is unacceptable. We need to change something.
To me, based on percentage of the vote, Mr. Maguire’s win in Brandon-Souris is entirely reasonable; so I would be comfortable with a 40% or 45% threshold in a 5 person race.
However, first past the post in it’s present form is still broken.