Friday, October 16, 2009

New AR Poll: 14-point Conservative Lead

Angus-Reid has released a new poll taken between October 13 and October 14 and involving 1,003 Canadians. The national result:

Conservatives - 41%
Liberals - 27%
New Democrats - 16%
Bloc Quebecois - 8%
Greens - 6%

Strong result for the Conservatives, who have gained from the last AR poll. The Liberals are actually steady at 27%.

I have to take off for the rest of the day and so don't have time to expand on this poll right now. I'll post more about it tonight or tomorrow. Hopefully we'll also have the Harris-Decima details at that time.

33 comments:

  1. Here are the details. Looks to me like another 170-180 seat CPC majority. AR confirms growing CPC support in Quebec. Also CPC leads by 25 points in BC and an amazing 16 points in ON. CPC lead is again larger in SK/MB than in AB.

    Details:

    http://www.angusreidstrategies.com/polls-analysis/opinion-polls/conservatives-flirt-majority-ignatieff-momentum-plummets

    ReplyDelete
  2. EKOS, Strategic Counsel, and now Angus too. When's Nanos going to do one?

    ReplyDelete
  3. Earl,

    That's the second poll showing stronger SK/MB numbers than AB numbers. I think its time we should stop dismissing this as impossible.

    With these sort of numbers it might be possible to attract a top notch candidate to take on Ralph Goodale and make a real race out of it.

    In Manitoba Anita Neville and Jim Maloway are vulnerable but the other NDP ridings look safe.

    ReplyDelete
  4. I'll leave it to Eric to do a projection from these numbers, but I don't think this would get the CPC anywhere near 170-180 seats. If every region voted EXACTLY as this poll suggests - the Tories would gain nothing in Atlantic, maybe gain 1 seat in Quebec, LOSE a seat or two in Alberta, possibly gain a seat in Man/Sask and gain a couple in BC - so in other words up about three seats outside Ontario, then the question becomes how many seats could the Tories realistically gain in Ontario if they actually carried the province by that margin. By my reckoning if there was a net 4.5 point Liberal to Conservative swing in Ontario (ie: Tories go from 40% to 45% and the Liberals drop from 33 to 29%) it would shift about 15 seats - so we would be looking at about 160 seats. That's IF these numbers held on election day - which is a big "if".

    ReplyDelete
  5. There actually is a reasonable explanation for why the Tories could be getting less support in Alberta than in Man/Sask. Let's keep in mind that right now the provincial Tories under Stelmach are very unpopular and the recession seems to hitting Alberta particularly hard. So there could be an anti-Tory mini-wave developing uniquely in Alberta.

    ReplyDelete
  6. DL,

    there is clearly confusion between the provincial Progressive Conservatives and the federal Conservatives.

    However from the provincial polling I saw last night on the news it is NOT helping the NDP/Liberals.

    9% NDP, 18% Liberal and I can't recall exactly but the Tory's were in the low 30's and the Wildrose was in the high 20's I believe.

    So we're not seeing a shift to the left. We're seeing a split on the right and a movement towards an even more Conservative option.

    My expectation then is that these voters will come home when they realize its a federal election and they're voting for Harper not Ed Stelmach.

    ReplyDelete
  7. DL,

    the Cons losing more seats in Alberta? C'mon.

    I'm gonna go out on a limb here and predict they sweep Alberta AND Sask.

    ReplyDelete
  8. I'm not saying what I think will actually happen whenever we have an election (in 2010? 2011?), I'm saying what would happen if the popular vote in Alberta was EXACTLY what this poll says. I acknowledge that is a very hypothetical situation. IF Tory support actually fell to 51% and Liberal support rose to 21% and the NDP stayed about where it is in Alberta - the Tories would lose Edmonton Centre to the Liberals and maybe Edmonton East to the NDP.. I'm not saying I expect that to happen on e-day, but I also don't expect the Tories to get 68% of the vote in ManSask despite what that Ipsos poll says.

    ReplyDelete
  9. No way AB will elect any liberal federally. The problems with eddie will not translate to PMSH. We remember the NEP, the finger, why should I sell your wheat, and I hate Albertans.

    ReplyDelete
  10. "We remember the NEP, the finger, why should I sell your wheat, and I hate Albertans."

    Those were all much more recent memories in Alberta in 1993 and FOUR Liberals were elected in Alberta that year, plus there were a couple of other near misses.

    ReplyDelete
  11. DL,

    93 was a wave election that featured a very fractured right with the PC/Reform split.

    If you combine PC + Reform votes all four of those Edmonton ridings that went to the Liberals would have been Conservative.

    ReplyDelete
  12. But the Liberals did actually win something like 27% of the popular vote in Alberta in 1993 and they also won seats in Alberta in 1997, 2000 and 2004 and as we all know, the NDP took a seat in 2008 and could be competitive in another one next time too. There's always been a nucleus of three or four ridings in Edmonton that have political attitudes that are more like Winnipeg than Calgary.

    ReplyDelete
  13. Sorry guys, DL is right. With the Conservatives at 52% and the Liberals at 21%, the Conservatives would lose seats in the province.

    You guys need to be more consistent.

    ReplyDelete
  14. Eric wrote:

    "With the Conservatives at 52% and the Liberals at 21%, the Conservatives would lose seats in the province [Alberta]."

    Surely "would" is overstating.

    Let's take Edmonton Centre as an example (Anne McLellan's old riding; likely the most winnable for the Liberals):

    In 2008, this "vulnerable" riding was won by the Conservative over the Liberal by a margin of 22634 to 12661 -- a comfortable margin by any definition.

    Now if we take the current Tory poll number for Alberta (52%) and compare with the 2008 election (65%) and apply that proportional discount to 2008 vote total, one could hypothesize the sitting Tory getting a mere 18107 votes.

    So even then, the Liberal in that riding would have to increase his/her vote tally by about 50% from 2008 to overtake the weakened Tory.

    Is that possible? How does one model, on a riding basis, a provincial Liberal increase from 11.4% to 21%? Surely it's unrealistic to assume the Liberal vote in Edmonton Centre would double.

    Perhaps one might instead assume that their total might increase from 27% (in that riding) to 37% -- which would still mean a loss to the Conservative candidate.

    Remember, Anne McLellan barely won this riding in 2004 when she 1) was personally well-respected 2) had the incumbency advantage and 3) had the profile of being Deputy Prime Minister.

    So, while I agree that under this scenario for Alberta (CPC: 52%, LPC: 21%), the Liberals could pick up one or two seats, I would argue that to say with confidence that they WOULD, is unwarranted.

    ReplyDelete
  15. Eric,

    I don't see where the riding is that the Liberals would gain their seats in.

    This poll, compared to the '08 results, shows the Conservative vote down 13%, Liberal vote up by 10%, NDP down by 2%, Green party up by 3%.

    Was any riding actually close enough that such changes would result in a Liberal victory or is this just an artifact of your model that's saying Liberals won seats before therefore they can do it again ??

    ReplyDelete
  16. Look, the Liberals double their vote and the Conservatives lose about 1/5th of theirs. And the Conservatives are still supposed to sweep? Come on.

    ReplyDelete
  17. Eric wrote:

    "Look, the Liberals double their vote and the Conservatives lose about 1/5th of theirs. And the Conservatives are still supposed to sweep? Come on."

    That doesn't seem persuasive.

    Doubling from extremely low levels does not guarantee anything. Even with a provincial result of 21%, there are really only a couple of ridings that would likely be in play for the Liberals. Much might then depend on how riding-specific factors played out.

    Keep in mind that in the 1987 New Brunswick provincial election the Tories won 29% of the popular vote and yet lost every single one of the 58 ridings.

    Again, I'm not saying that with this Alberta scenario, that the Conservatives would be guaranteed a sweep, I'm just saying that it is also far from certain that the Liberals would win any ridings.

    ReplyDelete
  18. Just for fun I put together a few examples of lack of success in seats (in a given province) in general elections.

    Various factors are different, of course, but I think that they are illustrative of how the threshold for winning any seats in a province is sometimes higher than one might assume.

    Comparing this poll for Alberta where CPC is 52%, LPC is 21% for a gap of 31% ...

    1972 Federal Election (Alberta results)

    Tories got 57.6%
    Liberals got 25%
    for a gap of 32.6%
    the Liberals won: zero seats

    1993 Federal Election (Ontario results)

    Liberals got 52.9%
    Reform got 22.4%
    for a gap of 30.5%
    Reform won: one seat (out of 99)

    1997 Federal Election (Ontario results)

    Liberals got 49.5%
    Reform got 19.1%
    for a gap of 30.4%
    Reform won: zero seats

    2008 Federal Election (Saskatchewan results)

    Tories got 53.7%
    NDP got 25.6%
    for a gap of 28.1%
    NDP won: zero seats

    So, with these kinds of historical examples, if an election yielded Alberta results of CPC: 52%, LPC: 21%, it does not seem at all absurd that the Liberals _might_ still come away with zero seats in Alberta.

    ReplyDelete
  19. BC Voice of Reason16 October, 2009 22:46

    Martin....

    You forgot about the strong strategic voting that Alberta does:

    1) Joe Clark in Calgary to prevent the Conservative party from being wiped out

    2 Annie McClellan as the Minister of Alberta pork

    I would venture to say that if there was an obvious strong Liberal minority with obvious strong candidates that would be able to deliver the big cheques, there would be a couple of Liberal seats in Alberta.

    As there is virtually no chance of a Liberal cabinet minister from Alberta.... without a coalition of at least the NDP/Liberal there will not be stronger enough candidates to run as Liberals to pull one out.

    No Liberal wins in Alberta this decade.

    ReplyDelete
  20. A couple of years ago if you had suggested that the NDP would win a seat in Alberta people would have said you were a lunatic. The NDP has a seat in Alberta despite only having 13% of the popular vote compared to the Tories 60%

    ReplyDelete
  21. Eric,

    the Conservatives could easily still sweep with these numbers depending on where in the province they are losing support. Assuming a uniform reduction the Liberals would win Edmonton-Centre.

    But its also possible the Conservatives are losing more support in Calgary because of brand confusion with Ed Stelmach and his disasterous handling of oil royalties. This is where Stelmach lost seats in the provincial election, while gaining some in Edmonton.

    Without a regional breakdown of where the Conservatives are losing their supporters and where the Liberals are surging its just not possible to rule out a Conservative sweep.

    ReplyDelete
  22. Well, when you see my projection for this one poll tomorrow, you'll see that the Conservatives do not sweep the province.

    ReplyDelete
  23. Well Eric you might might be right about the Liberals and Alberta,but the preponderance of evidence shows that you're likely wrong. I'd say that there is a better chance of the CPC sweeping SK/MB than there is the Liberals electing members from AB given the polling numbers in AR and EKOS and IR. The NDP win in Edmonton in 2008 was an anomaly, not a ground swell. Sure strange things happen in every election but based on the numbers in the polls we got today it is highly unlikely that a Liberal would be elected in AB, IMO. You are the guy who told me that riding are not in play in your model if the winning margin is greater than 10%.

    ReplyDelete
  24. Eric,

    it'll be interesting to see what your model says. 1 Liberal, 1 NDP i'd expect.

    But remember:

    There are more things in heaven and earth, Eric, than are dreamt of in your model.

    ReplyDelete
  25. Earl,

    --- "I'd say that there is a better chance of the CPC sweeping SK/MB than there is the Liberals electing members from AB given the polling numbers in AR and EKOS and IR."

    No one is talking about other polls. We're talking about this one. Do I think the Liberals will win a seat in Alberta? No. Does this one single poll indicate that they would? Yes.

    Jesse,

    --- "There are more things in heaven and earth, Eric, than are dreamt of in your model."

    Stone the blasphemer!

    ReplyDelete
  26. I guess Eric I find your model to be different than my way of thinking. I think EKOS knows more than you frankly and they see CPC at 167 seats very close to my 170.

    What about the 10% rule Eric? Did it go out the window in AB?

    ReplyDelete
  27. shrug... tories go down 12 liberals go up by 10...

    in theroy anything that was within a 22% spread would be in play.

    By my quick count of the 2008 results that is...

    well none of them.

    except the NDP seat. And possible E-Sherwood park because of the independent who pull so many votes.


    That is if the 22% swing (if it actually is that much... Can't see it with Stelmach losing support to the right) is evenly distributed between seats. It almost definitely is not. Tho that would be an interesting statistical problem if it was :)

    I think there might be a seat in play for the liberals to get close... but even with those numbers I wouldn't expect them to win. The greens probably have as much chance as the liberals.

    And Linda Duncans seat I am not makin gany predictions.. it might go either way (tory/ndp).

    ReplyDelete
  28. Rahim Jaffer was m.i.a. and that is why Duncan won, plus Lib strategic voting.
    It won't happen again, I promise.

    Next election the West will tell Libs and Dippers what they think about the coalition.
    Because we can take our protest to the ballot box instead of the streets.

    ReplyDelete
  29. "I'd say that there is a better chance of the CPC sweeping SK/MB than there is the Liberals electing members from AB given the polling numbers in AR and EKOS and IR."

    If you look at the results last election in Winnipeg Centre and Winnipeg North you'd realize that the Tories would need to be at almost 80% and the NDP in single digits before either could go Conservative.

    ReplyDelete
  30. Earl,

    --- "I guess Eric I find your model to be different than my way of thinking. I think EKOS knows more than you frankly and they see CPC at 167 seats very close to my 170."

    EKOS is in the business of survey taking, not projections. I doubt their system is any more sophisticated than mine or others.

    And I put their poll at 160 seats, only seven away from what they said. So I don't think they know all that much more than me.

    ReplyDelete
  31. Earl,

    --- "What about the 10% rule Eric? Did it go out the window in AB?"

    No, the Liberals have a cap of 2 seats in Alberta. The NDP have a cap of 1.

    ReplyDelete
  32. One thing missing in talking about landslide Annie's win is the shennanigans that went on in that riding with voting lists ect.
    She had a full page ad in the paper, the wekend before the election, with pictures of so called supporters. My group was approached to sponsor said ad, for 275.00 and some nice words under a photo of our director. When she brought it up she said it would not cost us anything as the money would be returned as a donation after the election. I objected to this in and out scheme, as we had members from every riding in AB and most would object. We never went for it, but many did. She won by 6 votes.
    There was also a problem with voters registering at place of employment and also in their own polling district.
    Laurie worked for years to get this straightened out. And as I have said before, Kilgour could have won as a Rhino, he was a good MP and worked hard in his riding.
    Today's results in the WRA leadership will tell us how much trouble eddie is in. If rumors are true, we could be the opposition soon. That is a big IF.

    ReplyDelete
  33. Eric wrote:

    "EKOS is in the business of survey taking, not projections. I doubt their system is any more sophisticated than mine or others"


    For what it's worth, for the 2008 federal election, Ekos' seat projection was the closest of five major seat projecting organizations (the others were the UBC election stock market, the Laurier Institute (LISPOP), democraticspace and electionprediction.org).

    ReplyDelete

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.