What is the significance of the latest polls?

Look at these numbers:

Nationally, however, the party appears to have stopped its week-long slide and come to rest with between 33 and 35 per cent support from decided voters - a shade below the 36 per cent of the vote it won in 2006, allowing it to form a minority government.

What does all this really mean? Very little, to tell the truth. First of all, polls are never accurate. To achieve any real representative accuracy or significance, polling companies would have to poll at least 3,500 people for each poll, rather than only 1,200, 1,000 or 800-900. Given the sample sizes in Canadian polls, all of them are for the birds.

Second, there is the issue of Canada's undemocratic and antiquated first-past-the-post system – i.e., the winner takes all, as they say. Even if the polling numbers are accurate, which they are not, and we assume that the Tories will get, say, 34%, they can still form a majority government. All they have to do is win by at least one vote in 155 ridings, and the majority is in place. Or to put it in more drastic terms, if they obtained as little as 10% of the votes in 155 ridings, and 10% happens to be the highest count for any of the candidates in the ridings, they would also win a majority.

Always keep in mind what happened in the provincial election in Alberta this March: with only 22% of the electorate supporting them, the Alberta Tories formed a majority government, holding 72 of 83 seats (!).

Finally, the above numbers may not be all that accurate, as I said:

University of B.C. political scientist Fred Cutler, a voting specialist, warned that polling numbers early in a campaign can be suspect and that the decline in support is probably exaggerated.

It is for this reason that it will be a long time before any of the mainstream parties will ever agree to change to a system of proportional representation. As they surely see it: Why fix it if it ain't broke?