Friday, June 12, 2009

Projection Update: Harris-Decima

Now that the details of the Harris-Decima poll are available, I've updated the projection.

Firstly, this poll would have resulted in a stable Liberal minority:

Liberals - 140 seats
Conservatives - 94 seats
Bloc Quebecois - 52 seats
New Democrats - 22 seats

The Liberals made significant seat gains from other individual polls thanks to their lead in British Columbia (33% to 30%), their big lead in Ontario (42% to 31%), and their even bigger lead in Atlantic Canada (45% to 30%). The 40% result for the Bloc increased their seat total to 52 seats.

In the short-term projection, the Liberals have gained enough seats to now be considered capable of forming a stable minority government. They are now up four seats to 133, while the Conservatives are done three to 107. The Greens have lost their seat in Atlantic Canada, the NDP is down one, and the Bloc is up one.

In the long-term projection, the New Democrats have lost a seat to the Liberals in Atlantic Canada. That puts the Liberals up to 121 seats, four up on the Tories. There were no big regional gains, but the most significant was the 0.3 point gain by the Liberals in Atlantic Canada.

Now that it seems both Layton and Duceppe have committed themselves against the government's economic update, Michael Ignatieff has to decide whether to send Canadians to an election. Based on the numbers this week, he would be Canada's next Prime Minister. But we've seen how quickly things change, and how individual polls can sometimes show leads within the statistical margin of error. The "Canadians don't want an election" theme is always overstated, after the first week the voters seem to forget about who forced the election. A summer election would be bad for voter turnout, and I'd predict an all-time low in that domain. Of course, Ignatieff can wait until the fall or next spring, but the old cliché is that a short period of time is an eternity in politics. The situation in the fall or next spring might not be as favourable as it is now.

Things for Ignatieff to ponder. If you're reading this Michael, I'd hold off. You have until 2012 to bring down the government, and I'm sure another opportunity will present itself. It's a bit risky right now.


  1. i disagree. if you're reading this michael, i very much hope you bring down this govt. sure it's "risky", but as far as we can tell the momentum has been behind the liberals lately. sure, we can wait longer until the numbers improve, but if the bloc and the ndp start to do worst in the polls as a result of the liberals doing better, one or both parties can, at least, abstain from voting against the govt and the harperites stand. waiting will also essentially hand over the air waves to harper and his attack ads to do more damage (there is an argument to be made that the liberals could be even higher than they are if not for the attack ads, as their support had been steadily growing up til that point). with the ndp and the bloc already lined up against harper, the stars have aligned, so to speak. now is the time.

    and finally, when it comes down to it, if you really think you can do a better job than the govt, and there's an opportunity to force an election, then do so and make your case to the voters. but the middle of the campaign, ppl will indeed soon forget who forced the election or even that there was an election just a year ago. i think the liberals have a decent shot at winning a solid minority if not a majority govt over the course of a campaign. i don't think canadians were really all that endeared by harper to begin with.

  2. There are other polls than the ones for voting intention and direct positive/negative impressions of the leaders.

    For example, there is a report of Nanos asking a very open-ended question about the perceived strengths and weaknesses of the parties, and reporting not on the strengths themselves, but rather on how the answers were given.

    Because this is such an indeirect approach, its intepretation becomes subjective. Nonetheless, it seems to be the sort of information which when factored into the regional subsets, might sharpen the seat projection somewhat if properly done.

    Eric, do you do anything with such polls? Or are your translations from "popular vote" to "seats" done purely on the electoral record? How do you do the seat projections, in any case?

  3. I saw that poll. I don't include things like that in the projection. I'm not sure how they can be incorporated. For example, May and Layton are relatively well-liked, and certainly better liked than either Harper or Ignatieff, but that doesn't translate to votes.

    The seat projections are based on popular vote projections, and measured against historical results.

  4. Just to let you know that I find this site very interesting and useful. Only stumbled on it recently but will let others know.
    Another issue for the Liberals will be whether they think they have got the party machinery in sufficient shape by this time. Things had deteriorated in the party for some time and Dion wasn't able to make much headway. I am impressed by how the party has built up its outreach to members and hence I expect some improvement in funding. However I am not sure that they are yet up to scratch in terms of competitive riding organizations.


COMMENT MODERATION POLICY - Please be respectful when commenting. If choosing to remain anonymous, please sign your comment with some sort of pseudonym to avoid confusion. Please do not use any derogatory terms for fellow commenters, parties, or politicians. Inflammatory and overly partisan comments will not be posted. PLEASE KEEP DISCUSSION ON TOPIC.