Thursday, March 31, 2011

Liberals and NDP trade support in new Nanos poll

Polling for the campaign is still a little slow off the mark, but Nanos Research has been wasting no time. Today we've been treated to the second report from their daily tracking, and it shows a shift in support that is verging on the statistically significant.

This tracking poll was in the field as recently as yesterday, a day that had the Liberals talking about giving Canadians the opportunity to top up their CPP contributions, the NDP about raising corporate tax rates, and the Conservatives... Why can't I ever remember what the Conservatives proposed?

You might think I'm trying to be funny, but I'm not. This is the second day in the row I've wracked my brain to come up with the previous day's Conservative campaign pledge, only to come up with nothing. And I watch CPAC, Newsnet, CBCNN, RDI, and read the news online throughout the day. Why isn't it sticking with me?

The Liberals have started the campaign strongly, coming out with big proposals that capture attention. The NDP, too, are very specific about what they want to do and have managed to hit a populist chord. But the Tories haven't been as clear. Often they are repeating the relatively uninspiring things that were in the budget, and until more recently Stephen Harper's campaign speeches were focusing more on the "coalition" than on what his party is proposing to Canadians.

Yesterday, the debate over the debate, an NDP candidate dropping out, and problems with Tory volunteers grabbed more headlines than anything Harper had to say. This is a problem.

A quick search tells me that the Conservatives promised to conclude free trade deals with the European Union by 2012 and India by 2013.

Could this be one of the reasons the Liberals are improving in the polls? Yesterday's numbers for the Liberals must have been very high to see such a huge change in support, or their numbers on Monday were very low.
But it isn't the Conservatives who are suffering. In fact, in this poll of 1,200 Canadians (21.7% of whom were undecided) conducted for CTV and The Globe and Mail, the Tories have gained 0.7 points nationally.

The NDP is the party on the decline, at least in this poll. They've dropped 3.7 points to 15.9%, a significant number as with their levels of support the margin of error for the change in their results in these two polls is about 3.4%.

The Liberals have gained four points and now stand at 32.7% (their highest result in my model). However, the margin of error for the Liberal results in these two polls is about +/- 4.4%, so this gain is just inside the MOE.

Generally speaking, Conservative numbers are solid and consistent. But what is going on below them is far more noteworthy.

Regionally, the Liberals are doing very well in the West. They are up 4.1 points in BC and now trail the Conservatives by 32.3% to 42%. In Alberta, Saskatchewan, and Manitoba, the Liberals are at 33.7%, a gain of 9.6 points since yesterday, only a few points behind the Conservatives (47.5%). Is this really happening? We need some more polling data to figure that out.

What is startling in this poll is the Conservative performance in Ontario: 47.5%. That is a gigantic number for the Tories. Here, the Liberals remained steady but the NDP dropped, while in Quebec the Liberals are up three points to 25.8%, a very high result for them. The NDP dropped 4.3 points to 13.5%.

And in Atlantic Canada, the Liberals are soaring at 47.9%, up 8.5 points. These are big swings that could, admittedly, be due to the sampling margins of error, but we are also in a campaign and things can change quickly.

I did not do a full seat projection for this one poll, as I can't use the Prairie numbers. But the Conservatives would win as many as 61 seats in Ontario with this poll, compared to 30 for the Liberals and 15 for the NDP. In Quebec and Atlantic Canada, however, the Liberals would make up for their Ontario weakness: 17 seats in Quebec and 22 in Atlantic Canada. The end result would probably find the parties generally where they were when the government fell, with fewer seats for the Bloc Québécois.

In addition to this national poll, I included the riding level polls from Segma Recherche into the model. The poll was conducted just before the campaign began. It had the Bloc ahead in Beauport - Limoilou, Louis-Hébert, Charlesbourg - Haute-Saint-Charles, and Québec. In all but Louis-Hébert the lead was 13 points over the Tories. The Conservatives lead in Louis-Saint-Laurent by nine points. Check out the PDF for more details.

Note: Nanos Research asked respondents "For those parties you would consider voting federally, could you please rank your top two current local preferences?"


  1. The Tory campaign reminds me of Martin's rope-a-dope strategy in 2006 (how well did that work...). They are sitting back and hoping the Liberals will shoot themselves in the foot, but it isn't working (and unlike Martin, this is a 1-month campaign). Harper needs to take more than 5 questions from reporters, and needs to offer some substance instead of repeating the same lines we already know from his attack ads.

  2. "The NDP ... the margin of error for the change in their results in these two polls is about 3.4%. ... However, the margin of error for the Liberal results in these two polls is about +/- 4.4%, so this gain is just inside the MOE..."

    Hold on now just a minute.

    The MOE reported for the poll as a whole is 3.2%.

    It is true that the MOE goes down as the 'true' level of support for a party moves away from 50% ...

    ... which would translate to an MOE of about 2.9% for the Liberals and about 2.5% for the NDP.

    However, this MOE is meant to encapsulate the possible sampling error from the 'true' result.

    In other words, it would be consistent with both polls being correct (within the MOE) if the NDP's true level of support were unchanged at 18% -- both 15.9% and 19.6% would be within the MOE of 18%.

    It seems worth cautioning how easily we should declare results to be "significant".

  3. Henry,

    There is an MOE for variations from one poll to the next. This is what my statement was referring to.

  4. I am drawing on the conclusions from this paper:

  5. Since the Nanos poll is a rolling poll it could be dropping the results from Monday night that might explain this surge in Liberal support we see today. It may not represent an improvement in the Thursday results by themselves.

  6. That is very true. I did make a mention of that. We will have to wait and see.

  7. Hosertohoosier stop being a nervous nelly.

    Didn't you hear that the CPC is going UP in the polls still ?

    What we're seeing is a three party system (in the ROC) moving towards a two party system.

    Since the NDP is on the left its going to be the Liberals eating most of that support.

    But the CPC is still picking up some crumbs.

    And the NDP support going to the Liberals isn't helping them take CPC seats in Ontario.

    Ignatieff is eating Layton's lunch. He's taken all of his positions. He'll get most of his seats.

    But in doing so he's abandoned the centre of Canadian politics.

    If Ignatieff gains +15 seats in this election and Harper gains +8 will Ignatieff have really "won" ??


    Harper will be that much closer to a majority.

  8. Ok.

    I only had a chance to skim it but it looks like it is properly taking things into account. So, if you're using his formulae, you're probably on somewhat reasonable footing.

    MOEs are widely misunderstood, so I thought some clarification was in order. Apologies for misunderstanding.

    You might want to use comparable wording for the Liberal change as you did with the NDP "margin of error for the change".

    Just a thought.

  9. It's complicated to express correctly, which that paper mentions as well. Will continue to try to do better!

  10. Day 6 of the campaign and people write off Harper. The Tories are in this race for the longhaul and even at 32% the Liberals still have less than half the seat count of the Tories. This election will be won in Ontario and BC.

  11. As is often the case, it is all in the presentation and as the only numbers that really matter from the poll are the following as it is generally recognized that the election will be won or lost in these two regions ONT and BC.

    ONT 03/29 CPC 43.0%, 03/30 CPC 47.5% up 4.5% with fewer undecided ONT 22.3% down to 21.9%;
    ONT 03/29 LPC 32.9%, 03/30 LPC 32.2% down 0.7% with fewer undecided ONT 22.3% down to 21.9%;
    ONT 03/29 NDP 20.6%, 03/30 NDP 16.3% down 4.3% with fewer undecided ONT 22.3% down to 21.9%;

    With the Conservatives increasing their vote in Ontario by 4.5% and a declining Liberal support aligned with fewer undecided, it is a big up tick for the CPC and shows further decline in the Liberal brand in vote rich Ontario.

    BC 03/29 CPC 41.5%, 03/30 CPC 42.0% up 0.5% with fewer undecided BC 24.5% down to 21.8%;
    BC 03/29 LPC 28.2%, 03/30 LPC 32.3% Up 4.1% with fewer undecided BC 24.5% down to 21.8%;
    BC 03/29 NDP 21.5%, 03/30 NDP 19.4% down 2. % with fewer undecided BC 22.3% down to 21.9%;

  12. I suppose momentum can shift in a campaign for reasons that are not entirely clear, but its possible that concrete proposals and Harper's popularity could push NDP voters to the LPC. Why it is where it seems to be, I don't know and I suppose for just one poll I don't really need to put that much on its regional results.

    As for the CPC: Conditional income splitting and money for businesses? Like in the last election it seems they're just going with the same attack ads they've been using in the months previous while hammering one negative charge. Though a coalition doesn't seem to have the sting of a carbon tax and Ignatieff doesn't seem to be making unforced errors. We'll see.

  13. Nanos poll,

    It should be noted that the increase in Liberal support may be due in a large part with a support level of 25.0% in Quebec and that can be measured against the 20% to 24% average poll results from the seven companies polling the election for Liberal support in Quebec.

    There is a similar result for the Prairies provinces where the average level of support for the Liberals is at 19% to 22%

    Theses numbers may be the start of a trend, but given the history I do believe so.

  14. I can't see how it is fair to characterize the Conservative campaign as lacking policy substance. They have already announced several policy items, such as the income splitting policy (contrary to media reports, this would be implemented before the next election if current assumptions are correct), tax credits and temporary tax cuts for small business, advancing free trade negotiations, and if the speculation is true, there will be an announcement on the Labrador hydroelectric project today. Granted, some of these policies were already announced, but it's somewhat ridiculous to say that a policy is "not new" because it was announced a week ago in the budget. One can hardly say the Conservatives are not presenting policy or are lacking substance. This is not a personal criticism to the author, just thought the point needed to be made.

  15. Shadow, in a FPTP voting system the gap between the top two parties, and not their absolute levels of support is the best predictor of seats. The oft-cited notion of pundits that 40% is a magic number leading to majority-land is incorrect.

    The contrast between 2006 and 2008 illustrate the point well. Harper's level of support changed little, going from 36 to 37.8, however, the Liberal-Tory gap changed a lot (from 6 points to 13 points). The result was solid gains for Harper, and defeat for many Liberals. A return to 2006 levels of support, means a return to a parliament in which NDP+Liberal support is sufficient to form a government (whether via coalition or tactical agreement).

  16. Backing out of the one on debate with Iggy makes Harper look bad. He comes across as deceitful and scared.

    It's early in the campaign but Harper is slowly messing up as he has in the past. Restricting reporters to five questions a day makes him look like he's hiding.

    I've found nothing in what Iggy has said to date to make me think he's a better choice. Typical tax and spend guy. Harper though having a clear shot at a majority is wasting the opportunity.

  17. I can't wait for the Liberal platform to be released on monday !

    The economists have just been brutal. Especially on the almost universally ridiculed idea to raise corporate taxes.

    This Stephen Gordon tweet is priceless:

    Dear LPC: You cannot spend the same $1b six times. Signed, Arithmetic.

    I hope Harper announces spending CUTS at some point. LGR, Voter subsidy, etc.

    He'll be the only leader not going around promising billion dollar goodies at a time of record deficits.

    He needs to identify enough cuts to return to balance BEFORE the Liberal platform. (Assuming they even do a five year projection...)

  18. Hosertohoosier that's not the case.

    The best predictor for seats is provincial numbers in BC, Ontario, and Quebec.

    The vast majority of Liberal-CPC contests are in Ontario.

    Liberals stealing NDP support in western Canada and recovering in Quebec is just irrelevent, even if it does move the top line numbers.

    You and Earl need to calm down. Its early days.

    One little bump for the Liberals and everyone is freaking out.

    We don't even know if it was a statistical fluke or not for goodness sakes!


  19. "The best predictor for seats is provincial numbers in BC, Ontario, and Quebec. "

    Sure, provincial breakdowns provide a more accurate picture, but what matters there as well the gap in support, not absolute support levels. However if you look at the provincial level, the Liberal-Tory gap shrunk (or reversed in Atlantic Canada) everywhere but Ontario.

    Is this a statistical blip? Possibly, but if you look at what is happening in the campaign, I think not. Michael Ignatieff is dominating the media coverage with daily policy announcements, and his personal levels are moving above Dion levels as the public gets to know him better (and it seems pretty plausible to me that Canadians would like Ignatieff more than Dion).

    In contrast, Harper's message has been met with solid rebukes - the Globe Editorial board criticized his coalition framing, and everybody jumped on the timeline for income-splitting. If Harper is already losing his lead despite very positive economic figures, and months of saturating the airwaves with attack ads (during the writ he is on equal footing with the opposition parties because of ceilings on advertising), I don't think he is likely to fare well.

  20. Here's the real problem for why the Liberals are gaining ground (though maybe not seats):

    Harper was walking into a majority for this election due to high poll numbers and there being no 'real' reason for the election in the minds of voters. 'Real' meaning - the economy was fine, so a non-issue; few people 'cared' enough about Harper & co's 'ethics' problems; no one outside of Ottawa thought there was a need for an election.

    So now:

    Harper pounds the drums about a 'coalition'. The only one being grilled about it is him! No one else. Sure it plays to his base, but they were voting for him anyways. He keeps bringing it up, over and over and over again. Every other party is putting something on the table. Ok, not everything is 'good' about that, but they're at least trying to campaign whereas he's just trying to say... nothing useful.

    Harper says 'ok, let's do a 1-1 debate with Iggy'. Iggy says "any time, any place". Harper backs out.

    Harper either has the wrong advisers around him, or secretly doesn't want the majority that was being handed to him on a silver platter.

    It's his election to lose, and it is already starting to look like he wants to lose it.

    The Liberals won't dig out of 2nd place, but they might get just enough of a loss to be able to drop Iggy. Hopefully they also drop Bob Rae and go find someone who could start rebuilding the party. It's a tough job, given that the Liberals win when they can get the right mix of left and right voters.


  21. Both Earl and Anon have referred to the Harper "backing out" of the proposed 1-1 debate.

    The problem is that claim doesn't match reality.

    Harper proposed several alternatives:

    “We can have a traditional debate of parliamentary leaders,” he said — a format that would exclude May because the Greens don’t have any sitting MPs. “We can have a debate that includes Ms. May in such a format. We could have a debate that includes every party that’s on the ballot.”

    So, we're talking about at least 4 options (8 if each option were carried out in both languages):
    - a debate among the parliamentary leaders AND
    - a debate among the parliamentary leaders plus May AND
    - a debate among all the party leaders (even fringe parties) AND
    - a debate between Harper and Ignatieff only

    Does anyone seriously believe that he was proposing ALL of those options to proceed?

    Mr. Harper was clearly proposing alternative formats for the two debates (one in English and one in French) that have already been scheduled.

    Mr. Harper suggested the fourth option. The opposition declined.

    Mr. Harper did not "back out" of anything.

  22. I heard that Harpers team is thinking about firing people because the campaign is going so poorly. The icing on the cake was when Harper backed out of the one on one and looked like a coward. The media is also rebelling because they do not want to regurgitate Harper's misleading and false statements anymore. That is why Harper has restricted questions.

    On the flip side, Ignatieff's campaign is going very well. A lot of nervous nellys in the Harper camp today.

  23. It looks like Iggy has to focus on Ontario, the provincial libs are dragging him down.

  24. Oops.

    My quotation was truncated, which may have made things less clear:

    “We can have a traditional debate of parliamentary leaders,” he said — a format that would exclude May because the Greens don’t have any sitting MPs. “We can have a debate that includes Ms. May in such a format. We could have a debate that includes every party that’s on the ballot.” But Mr. Harper then suggested a final option: “We could also have a debate between Mr. Ignatieff and myself, since, after all, the real choice in this election is a choice between a Conservative government or an Ignatieff-led government that all of these other parties will support.”

    The question still stands:

    Does anyone honestly believe that those four debate formats were contemplated as anything other than mutually exclusive?

  25. I think Harper wants a four-person debate because he wants to paint the debates as 'The Coalition' ganging up on him. That way, he can't really lose the debate - he can just paint it as illegitimate anyway as the other parties were all working together to assassinate his character.

    After all, it's not like Duceppe is going to do mucnh attacking of Layton, for example, right?

  26. Henry you are right technically. Perception is reality though in political campaigns. Harper was seen to be issuing a challenge and similarly seen to be "chickening out" when he backed away from the challenge. I happen to think Harper could handle Iggy. I think he is making a mistake in backing away from Iggy's challenge.

    Further I'm disappointed in the CPC campaign to date. Overly negative me thinks. I think some positive announcements would help as well as a more relaxed attitude towards the press would do wonders.

  27. In a 3-day rolling poll, if a party goes up 1%, it would mean that they were 3% better on the newest day than the day that was dropped ... the Liberals are up 28.8% on day 4 of Nanos over day 1 on the Prairies??

  28. Did the question at hand ask 'who would you vote for if there were an election tomorrow?'. Eric gave us the resukts of that question earlier on the day which gave a more accurate poll. Lets not get too excited at this 32%.

  29. HTH things aren't half as bad as you think they are.

    Here's a factoid from Pundits Guide. In 2008 44% of races were Con-LPC or LPC-Con. (Mostly in Ontario.)

    23% were Con-NDP or NDP-Con races. (Mostly in the West.)

    If the Liberals are pulling away NDP support in the West while not gaining anything in Ontario why is that bad news for Harper ??

    Its the opposite of strategic voting on the part of the coalition.

    It means Harper keeps Ontario and picks off NDP incumbents out west.

    Sure the Atlantic numbers aren't good but those races will hinge around local candidates and the CPC has some serious powerhouses running for them there.

  30. I hadn't really though about it before... but I'm not sure if its possible for Harper to create a better balance sheet even if he isn't promising expensive goodies. He's already committed to a few billion here and there himself, just on less politically popular items like jets and jail expansions. One can argue about the savings and efficiency of corporate taxes, but rolling them back provides fiscal room without crossing the public. There's not many places you can do that.

  31. I think the Conservative's strategy is to basically ignore policy and issues as much as possible and turn it into a mere popularity contest between Harper and Ignatieff. It's basically the same strategy that they used in 2008, in which they didn't even release a platform for the election and focused on blasting Dion's personality and his ideas. They make a few promises, but nothing substantive.

    They have certainly set the stage for such a strategy by their ads against Ignatieff in the run-up to the election.

    The reasons for such a strategy is simple: avoid frightening the moderates with a right-wing platform while not alienating the right-wing base that could leave the party (again) with a platform they think too moderate. The Conservatives couldn't win if they ran as a right-wing party, they need to run as a center-right party, so if they can just avoid talking about the issues and focus on their leader's personal advantage over his opponents, that is just what they're going to do. (As an aside, it makes one wonder how they expect to deal with a majority, their base, after agreeing to keep mostly silent for a long time, is going to ask for a lot of right-wing actions that might not be popular and if Harper doesn't cave in, then they might just reform the Reform or Alliance parties, split the vote and recreate the Liberal domination)

    Will it work? Well it did in 2008, but Dion had much less charisma than Ignatieff. It was easy to paint Ignatieff in a bad light before the campaign as he was mostly out of the spotlight and the CPC had the cash to put up ads to blast him. However, in the campaign, Ignatieff has the ear of the media, so Canadians can see and hear the real man.

  32. "Typical tax and spend guy."

    All governments tax and spend (although some run deficits and spend leaving tax for next government)

    "Mr. Harper suggested the fourth option. The opposition declined."

    That is so disingenuous.

    Harper's 'magic' has been to control message, and paint it as 'leadership'. This veneer is finally being peeled back a bit, partially due to increased media reporting of 'control' vs 'leadership'. This is his Achilles heel. Once 'leadership' is stripped away, what remains is 'control'.


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.

Note: Only a member of this blog may post a comment.