Sunday, December 12, 2010

Conservative jump in new Ipsos-Reid poll

Ipsos-Reid released a new poll on Friday, their first since the beginning of November. It shows a big gain for the Conservatives, with the New Democrats taking an equally big step backwards.Compared to Ipsos-Reid's last poll, the Conservatives have gained four points and lead with 39%. That is the highest score for the Tories in my model. Meanwhile, the Liberals are steady at 29%.

It is the New Democrats who've lost ground, down four points to a bare 12%, the lowest in my model.

The Bloc Québécois is up two to 10%, while the Greens are down two to 9%.

About 11% of respondents in this telephone survey were undecided, virtually unchanged from the last poll.

The breakdown by gender is notable. Ipsos-Reid has Conservative support among males at 47%, a huge number. The Liberals are only at 24%. But among women, the Liberals lead with 33% to the Conservatives' 31%.

There's been little change in Ontario, where the Tories have gained only one point and lead with 39%. The Liberals are unchanged at 34%, while the NDP is down three to 14%. The Greens, meanwhile, are up one to 11%.

In Quebec, the Bloc has gained six points and leads with 41%. The Liberals are down two to 22%, while the Conservatives are up the same amount to 21%. That is a very good result for the Tories in Quebec. The NDP is down four points to 7%, a very bad result.

The Conservatives have gained 10 points in British Columbia and lead with 44%. We've been seeing good Tory numbers in BC lately, so despite the big jump this isn't all that unusual. The Liberals are up three to 26%, while the NDP is down six to 18% and the Greens five to 12%.

The Liberals lead in Atlantic Canada with 44%. The Conservatives are up 11 points here, to 39%.

In Alberta, the Conservatives lead with 65%, followed by the Liberals at 16%, down six.

The Conservatives also lead in the Prairies with 49%, up 10 from early November. The Liberals follow with 28%, while the NDP is down seven to 19%.

With this poll, the Conservatives would win 25 seats in British Columbia, 28 in Alberta, 20 in the Prairies, 54 in Ontario, nine in Quebec, 10 in Atlantic Canada for a total of 147. That is 13 more than the projection from Ipsos-Reid's last poll.

The Liberals would win nine seats in British Columbia, none in Alberta, six in the Prairies, 41 in Ontario, 14 in Quebec, and 22 in Atlantic Canada for a total of 94. That's three fewer than last time.

The Bloc Québécois would win 52 seats in Quebec, three more than in early November.

The big loser of this poll, the New Democrats, would win only two seats in British Columbia, none in Alberta, two in the Prairies, 11 in Ontario, none in Quebec, and none in Atlantic Canada for a total of only 15 seats, 13 fewer than in Ipsos-Reid's last poll.

It's only one poll, of course, but these Ipsos-Reid numbers give us a look at what a one-on-one contest between the Conservatives and Liberals could be like.

This poll adds further evidence that the recent change we've seen in the numbers, that the Conservatives have moved out of their habitual five-point lead to something more substantial, is real. Growth in Conservatives support in British Columbia and Atlantic Canada in this poll, echoing some of the other polling results we've seen over the last few days, seem to be the cause. Why that is the case, however, is not clear.


  1. it's only been about a week since your last projection update, but this slew of polls showing big movement in the Atlantic and BC possibly warrant an early update. Things sure have changed in the past 1-2 weeks.

  2. Layton needs a wedge fast because progressives are moving towards the Liberals.

    He should focus in on Ignatieff's support of the Iraq war, his writing on torture, and the current war in Afghanistan.

    Put out a commercial saying Ignatieff = Bush.

  3. The reason for the Tory lead is not all that difficult to figure out. They are providing Canada with the best government we have enjoyed in almost 50 years.

  4. What's interesting is the Conservative support in BC and Alberta is the same as in 2008 and in Saskatoba their support is probably 1 or 2 points lowers. The Liberals however have seen large gains in all the west replacing the NDP as the alternative. In Ontario and Quebec the Conservative and Liberal numbers are the same as their 08 results. Atlantic Canada is the only place with different results and both the Conservatives and Liberals are up.

  5. PT, in fairness, some of the gains for the Liberals out west involves a return to their "normal" level of support (i.e., roughly what they had during the Chretien years). If they really want to challenge the Tories, they have to move above that level.

  6. According the the latest Leger poll out today - the NDP has caught up to the Liberals in Quebec with 21% (Tories are fourth) and the NDP is firmly in second place among francophones. Just when you think the federal Liberal meltdown in Quebec can't get any worse - it gets worse. The Liberals are now at 14% among francophone Quebecers.

    How long before Martin Cauchon drops out of running again and goes back to his law practice?

  7. @Dio

    It is possible to argue that Harper's leadership is more consistent than Martin. But 50 years??? Hard not to take your message as random propaganda.

  8. Carl,
    The NDP has pretty much been the alternative in Western Canada for several years now. All I'm saying is they're now being replaced by the Liberals, even if it were the Liberals who were succeeding here previously.

    DL said: "How long before Martin Cauchon drops out of running again and goes back to his law practice?"

    How long before the Liberals smarten up and relaize they need someone like Martin Cauchon to become leader?

  9. "because progressives are moving towards the Liberals"


    There are few more pompous and annoying terms in politics today than referring to the left as "progressive".

  10. The highest CPC value in your model?

    How far back do you keep data? Nanos put CPC support at 39.5 on Dec 13, 2009 (that's the most recept CPC total above 39).

  11. Ira, I certainly don't use year-old polls in my model.

    The oldest polls I have go back to June.

  12. So Ipsos has the NDP at 7% in Québec the same day that Léger has them at 21%? MOE is large at the provincial level, sure, but that's a remarkably different number. One polling house has three times the volume of support in Québec for the NDP that another pollster has. Strange.

  13. "Just when you think the federal Liberal meltdown in Quebec can't get any worse - it gets worse."

    Let's call a spade a spade: the Bloc Quebecois is dominating ALL the other parties. They have won a majority of seats in Quebec in five consecutive elections and would win a near-record number of seats if a sixth election were held. Which party comes in a distant second is of interest to spin doctors only.

  14. @goaltender interference

    a small point, but who comes second is also of interest to anyone who has a stake in whether or not the Conservatives have the ability to win a majority government. If they could push towards 25%, it would almost guarantee them a majority given stable support in other regions.

  15. Darrell Bricker here from Ipsos Reid. Just a point of caution for your readers. This seat projection model, while entertaining, is not endorsed by us. We stand behind what our results say about the political environment, but we DO NOT endorse what the model implies about seat counts.

    Projecting seat outcomes based on data like these, even when it's done with multiple polls, is a bit of a fool's errand. After the 2006election we absolutely swore them off, or at least publishing them. We do run a model, but it's really just for our own internal analysis.

    Also, on election day we run a live seat model based on the on-line exit poll interviews we do. But, we've developed a way to associate the actual interviews (in 2008 there were 36K) to the ridings that respondents voted in. If you're going to do a seat model, you almost have to do that. And, we don't even publish this model.

    This doesn't mean that I won't be watching 308. I like political entertainment as much as the next geek. And, who knows, you may have built a better mousetrap. But, beware a failure. Your peers won't be kind.

  16. Bryan wrote,

    "If they could push towards 25%, it would almost guarantee them a majority given stable support in other regions."

    Fat chance.

    Even if the Tories got 25% in Quebec they might be able to win maybe two or three seats in addition to the ones they hold now.

    There are simply too many ridings where their candidates finished well behind the winner in 2006 and 2008.

    That doesn't make the Quebec seats irrelevant -- indeed if they could gain a couple in that province rather than losing a couple, that might at least give them some hope of cobbling together enough extra seats nationwide to get a majority.

  17. Éric - I didn't think you did use year-old polls. But your remark about the highest CPC value placed a firm upper limit on poll age in your model, so I thought I ask for a more precise answer.

    And you gave me one. Thank you.


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.