Thursday, October 8, 2009

Projection Update - 135 CPC, 100 LPC

A big jump this week, with the Conservatives getting into a "stable minority".The Conservatives have picked up six seats and are now at 135. The Liberals have lost five and are at 100, while the NDP has lost one and is at 24. The Bloc Quebecois is steady at 49 seats.

A few bad weeks of polling has finally begun to take a toll on the Liberals in the projection. What was a trend is now a political reality.

The Tories gain 0.8 points nationally. The Bloc has gained 0.1 points, while the Liberals lose 0.7 points and the Greens 0.1. The Conservatives now have a 4.1-point lead over the Liberals with 34.8% to 30.7%. The NDP is at 15.7%, the Bloc Quebecois at 9.3%, and the Greens are at 9.1%.

In British Columbia, the Conservatives have gained 0.7 points, the Liberals have lost 0.6, and the NDP and Greens have lost 0.3 points each. The Conservatives now lead with 37.8%, followed by the Liberals at 25.5% and the NDP at 24.7%. The Greens are at 11.4%. The Conservatives are projected to take 21 seats, the Liberals to take 10, and the NDP to take 5.

In Ontario, the Conservatives gain a whole point as well as three seats. The Liberals lose those seats to the Conservatives as well as 0.8 points. The NDP has gained 0.1 points and the Greens have lost 0.3 points. The projection is now 38.1% and 49 seats for the Tories, 36.6% and 46 seats for the Liberals, and 15.1% and 11 seats for the NDP. The Greens are at 9.7%.

In Quebec, the Bloc Quebecois has gained 0.4 points while the Conservatives have gained 0.5 and one seat. The Liberals have lost 0.7 points and one seat, the NDP is down 0.2, and the Greens are down 0.1. The Bloc still leads with 37.1% and 49 seats. The Liberals are next with 28.8% and 18 seats, then the Conservatives with 17% and 7 seats, and finally the NDP at 10.7% and 1 seat. The Greens are at 6%.

There was a seat change in Atlantic Canada, where the Conservatives gained one and the NDP lost one. Other big vote changes include a 0.4-point Liberal loss in Alberta, a 0.4-point NDP loss and a 0.5-point Liberal loss in the Prairies, a 0.6-point Conservative gain and 0.4-point Liberal loss in Atlantic Canada, and a 0.3-point Liberal loss in the North.

The Conservatives are safely on their way to a repeat of the 2008 performance, while the Liberals still look to increase their caucus, mostly at the expense of the NDP.


  1. I still find your projection for BC doesn't make sense. In the 2008 election the Tories had 44% of the vote and the NDP 26% and the NDP got 9 seats. Now our projection has Tories at 38% and the NDP at 25% and you have the NDP at 5 seats. Now granted you have the Liberals making a bit of a comeback compared to last year, but the only NDP held seat that is at all vulnerable to a big Liberal BC comeback (and I'll believe it when I see it) is Vancouver-Kingsway - so I just don't get it.

    I won't quibble about the rest of the country since i think that the seat projections numbers make sense if you input the popular vote numbers you project. But BC just doesn't make sense.

  2. Are you saying the NDP should have more seats?

  3. Has anyone seen the city breakdown for the new EKOS poll ? This is unheard of:

    Vancouver (MoE 10.22)
    Conservatives: 44.9 (+2.7)
    Liberals: 25.8 (-0.7)
    NDP: 18.2 (-2.5)
    Green: 11.1 (-1.4)

    Toronto (MoE 5.59)
    Conservatives: 44.8 (+5.7)
    Liberals: 35.6 (-7.7)
    NDP: 10.3 (-0.9)
    Green: 9.4 (+2.0)

    Montreal (MoE 5.35)
    Bloc Quebecois: 37.2 (+2.7)
    Conservatives: 19.8 (+5.6)
    Liberals: 24.4 (-7.3)
    NDP: 8.5 (+1.2)
    Green: 10.2 (-2.1)

    I think this answers the question about whether Stephen Harper is getting super majorities in old ridings or finding new found support in opposition ridings pretty clearly.

  4. Not really, because most of the movements are within the MOE, and those polling areas include the suburbs.

    Consider this: "Montreal" includes the island, Laval, Longueuil, and the surrounding areas. Even if the Liberals are at 24% in the city, they'll still win all of western Montreal and much of the central part of the island. If they're losing ground in Laval, the Laurentides, and Monteregie, it doesn't really matter.

  5. Eric,

    I am not saying what I think will actually happen at the end of an election campaign, but I'm saying that if I were to take your latest BC projection which would point to approximately Tories down 6% from last year, NDP down 1% and Liberals up 7%, then when I look at the 9 NDP held seats in BC, I get Vancouver-Kingsway becoming a Liberal-NDP tossup and the other 8 NDP seats stay NDP quite easily.

  6. Eric,

    Do you see anywhere the Conservatives could win a seat then ?

    They've been wanting to claim a "Montreal" seat for awhile now, for sheer symbolism.

  7. No, the Conservatives won't win anything near Montreal.

  8. Eric,

    that's too bad about Montreal. Speaking of Quebec I just saw Jean Lapierre on CTV discussing the upcoming by-elections in Quebec. He said Hochelaga was BQ no matter but that in Rivière-du-Loup the Conservatives have the best candidate and he's seeing a 7 point spread between the BQ and the Cons that they could close if people thought they might be electing a cabinet minister.

    Was there a local poll done or is he just polling numbers out of the air ??

  9. If Jean Lapierre says it then it must be crap.

  10. DL comments on the B.C. riding projection for the NDP. I think he makes a valid point.

    In most of the NDP-held ridings, the Liberals were a distant third (or worse) in the 2008 election and the Liberals have not been polling well enough to become a factor.

    So, that leaves:

    - Vancouver East which was won by the NDP candidate by a 54% to 17% margin (almost certainly a safe NDP seat)
    - Vancouver Kingsway (as DL mentions)

    Can the NDP lose some of the other ridings to the Tories? Maybe, but not by adjusting 2008 results by the current vote projection (where the Tories have dropped a bit more than the NDP, as DL observes).

    In other words, using the 2008 election as the baseline and using the projected B.C. vote percentages, the NDP would seem unlikely to lose more than one seat in that province.

    However, I believe Eric has explained (and he can correct me if I have misunderstood) that he uses all three elections (2004, 2006 and 2008) to establish his matrix of what percentage vote will translate into how many seats.

    If one were to compare the 2004 election with the current vote projection, one would find:

    CPC: 36.3% (2004); 37.8% (proj)
    LPC: 28.6% (2004); 25.5% (proj)
    NDP: 26.6% (2004); 24.7% (proj)

    In that election the NDP only won 5 seats.

    By that metric, projecting 5 seats for the NDP now does not seem so preposterous.

  11. Martin, you are right and i'm aware of that 2004 seats to votes ratio (though it should also be noted that I suspect that if Eric had used his matrix to project seats in 2008 there is no way that he would have had the NDP winning 9 seats in BC with 26% compared to the Tories 44%).

    The difference between 2004 and what is likely to happen in an election now is that in 2004 the NDP only had ONE MP running for re-election. Now they have 9 incumbents and most of these people have built up really big majorities in the last couple of elections and their seats can no longer be considered marginal. On paper, Burnaby-Douglas is marginal to the Tories, but if the Tories do well enough to win there, it would mean that the Liberals would be doing badly enough not to have much chance in Vancouver-Kingsway. Anyway you slice it, I think that 5 seats for the NDP in BC is way too low based on the current projected popular vote.

  12. I think BC is definetly a place where pre-writ and post-writ polls/election results strongly diverge.

    It seems like in between elections people in BC always say they support the Liberal party and the leader but then when an election rolls around most of the non-Vancouver ridings quickly become two way races between the NDP-Cons.

    In the past i've heard pollsters say the same thing but it would be interesting to see some historical data.


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.