Wednesday, January 7, 2015

December 2014 federal polling averages

Last month was the busiest December in federal polling since 2010, with six national and one Quebec-only poll surveying some 17,000 Canadians being conducted. The results show the Liberals holding on to the lead for the 21st consecutive month, the Conservatives closing the gap, and the NDP suffering one of its worst months since before the 2011 federal election.

The Liberals averaged 34.9% support in December, a gain of 0.2 points over their performance in November.

The Conservatives picked up 0.4 points and averaged 32.4%, maintaining their highest levels of support since February 2013.

The New Democrats, however, dropped 1.4 points to 20.3%, their lowest level since March 2011.

The Greens were up 0.2 points to 6.2% (the party has been at 6% for three consecutive months now), while the Bloc Québécois was at 4.3%. Another 2% of Canadians said they would vote for another party.

Considering that the biggest shift was worth only 1.4 points, December appears to have been a month of stability. But the aggregate masks the underlying numbers. In November, the Liberals were polled at between 34% and 36%, the Conservatives between 30% and 33%, and the NDP between 18% and 24%. Apart from the wider range for the New Democrats, this is a group of polls all saying the same thing.

December, however, has been much more volatile, with the Liberals between 32% and 41%, the Conservatives between 31% and 34%, and the NDP between 17% and 24%. The Conservative range was still tight, but the NDP's range grew a little and the Liberal range grew a lot, pointing towards anything from a big lead to a second-place finish. We should keep this in mind: while the aggregate suggests a stable trend line, the individual polls published in December are far less clear.

Click for full-sized version
The Liberals led in British Columbia with 34.6%, a gain of 2.5 points since November. The Conservatives were down 3.4 points to 28.7%, while the NDP was unchanged at 23.6% support. The Greens were up 0.9 points to 11.1%.

In Alberta, the Conservatives dropped 3.2 points but still led with 54.3%. The Liberals halted four months of decline with a gain of 2.3 points to hit 23.2%. The NDP was up 0.3 points to 14.4%, while the Greens were down 0.6 points to 4.8%.

The Conservatives were also in front in the Prairies, picking up 3.4 points to lead with 42.7%. That is their best showing since June. The Liberals had their best result since November 2013, picking up 1.6 points to reach 33.8%. The NDP, however, was down 3.1 points to just 16.4%, their lowest result in Saskatchewan and Manitoba since February 2009. The Greens were down 3.1 points to 4.8%.

Ontario is the closest race in the country, with the Liberals narrowly ahead with 37.9% support in December against 37.4% for the Conservatives. That represented a drop of 0.2 points for the Liberals and a gain of 1.7 points for the Conservatives, their best since January 2013 and the third consecutive month of increase for the party. The NDP was down 2.9 points to 16%, their worst since February 2011, while the Greens were up 1.2 points to 6.5%.

In Quebec, the Liberals were up 0.1 point to 32.3%, while the NDP was down 0.7 points to 29%. The Bloc Québécois picked up one point and averaged 17.9%, its best result since July. The Conservatives were down 0.7 points to 15.6%, while the Greens were down 0.1 point to 4%.

In Atlantic Canada, the Liberals were up 1.6 points to 54%, their best result since May, while the Conservatives were down 1.9 points to 22.4%. The NDP fell 2.3 points to 16.8% in the region, their lowest level of support since before 2009. The Greens were up 1.1 points to 5%.

With these levels of support, the Conservatives would likely win 139 seats, with 130 going to the Liberals and 65 to the New Democrats. The Greens and Bloc would each win two seats.

Since November, this represents a gain of six seats for the Conservatives, two for the Liberals, and one for the Bloc, with the NDP dropping nine.

The Conservatives picked up nine seats in Ontario and one in the Prairies, but dropped one in both Alberta and Atlantic Canada and two in British Columbia.

The Liberals gained two seats in Quebec and one each in the north, Alberta, and Atlantic Canada, but dropped three in Ontario.

The NDP dropped six seats in Ontario, three in Quebec, and one in the Prairies, but gained two in British Columbia.

So it remains a close race between the Liberals and Conservatives, all the more so in the context of who would come out ahead in the seat count. But it was a bad month for the New Democrats, who were below 17% in every region except Quebec and British Columbia.

Normally, one would expect lower NDP numbers to benefit the Liberals, but it hasn't worked out that way (the Liberals' widest lead in 2014 came when the NDP was at 22%, in 2013 it was when the party was at 23%). Perhaps the Liberals are picking up a lot of support from the NDP that is making up for the voters they have lost to the Conservatives, or perhaps some New Democrats are moving over to the Tories. No matter what is going on, it appears that low NDP numbers are not necessarily a boon to the Liberals.


  1. Just came from your article at CBC.... from the comments there it appears I have been wrong about your leftist political leanings....

    In the CBC comments you are a right wing stooge directly appointed to the State run Broadcaster by Harper's minister of propaganda.

    1. I am only a partisan for numbers and charts and pretty colours.

    2. I wouldn't expect a response, but Éric, I haven't noted any real tendancies on your part that could be described as either "leftist political leanings" or some equivalent on the right. If I had to guess, I'd say straight-up Liberal, but you're obviously scrupulous about letting your personal predelictions show...

    3. I know people won't believe me, but I honestly don't care very much. I find the academic exercise of following the trends far more interesting than the actual politics of it all. I'll vote like everyone else, but I certainly don't live or die by a party's fortunes.

      You know what I want the results to be in the 2015 election? Exactly what I forecast them to be. I care about that infinitely more than who actually wins.

    4. I don't think your writing biased in any fashion Eric but, I have to say that for whatever reason I feel your projection favours the Liberals. 11 seats on the Prairies and 15 in BC? I don't think it reasonable to expect such Liberal success out West without a divided right.

    5. The numbers are the numbers - maybe the Liberals won't get as much support as the polls suggest in BC and PR, but that isn't what the seat projection takes into account.

    6. That's very interesting, Éric. Now that you've said it, it makes perfect sense and doesn't surprise me at all. But it's odd for me, since, as you could guess, I'm of a different bent - I'm interested in the party trends only in as much as they help me get an idea of where the politics themselves are going... and that being said, I'm not interested in the political parties per se, but much more so in how they (individually and together) affect society. It's politics and society that are of vital interest to me.

    7. "Exactly what I forecast them to be. I care about that infinitely more than who actually wins."

      That is a quixotic goal.

      You currently are functioning as an interpretation tool to translate the poll results into seat counts.

      If the polls are wrong then you are going to be wrong.

      This attitude can explain how you could look to be a Liberal. Most of the polls in the Harper era have consistently over stated the Liberal support.

      For your seat counts to be correct you have to hope like heck the Liberals get the kind of support that the polls show them to have.

      This is further exacerbated EKOS whose numbers of polls and size of their polls dominate the polling landscape.

      EKOS consistently seems to under poll CPC and over poll Liberals and Green.

      For your seat projections to be correct the big Green break through that EKOS has been reporting for the last 10 years has to actually be reflected in how people vote.

    8. I'm glad you understand that if the polls are wrong I am not at fault, but most people do not. I hope to avoid that trouble, but also to be able to tell the story of a campaign accurately. When the polls are wrong, it invalidates a lot of the work I do.

      The thing is, and I have looked at this, there is inconsistency in how the polls over or under estimate various parties. It is impossible to guess which way they will go in 2015.

      This is why I have been relying more on ranges in recent elections, and I will be doing the same this year.

      (As for EKOS, lately they've shown some of the weakest numbers for the Liberals, though I agree they are usually too high for the Greens).

    9. Huh - that only goes so far. I mean, your model does attempt to give greater credence to polls that have been more reliable in the past. So it's not *just* about being a "translation" of the polls to seat counts.

      I understand that you weight polls based on past performance, but do you ever discount (or magnify) specific aspects of polls? e.g.:

      a) Historically, Pollster X is reliable in Atlantic Canada, but appears to be just throwing darts when it comes to the rest of the country - I'm going to use those results for AC, but nowhere else.
      b) In the last *four* elections, Pollster Y has overestimated Party Z's numbers, both relative to other pollsters and relative to the actual results. I'm going to tweak those number, in keeping with their historical performance.

      I would suggest that there would be statistical justification for doing either of those things. On the other hand, choosing to do that could come off as a political choice.

    10. No, I don't make adjustments like that. I always try to do everything uniformly according to the 'rules' of the model. I do everything I can to keep my own subjective judgement out of the calculations.

      Past elections have shown that pollsters don't always over- or under-estimate a party in the same way. And in many cases, we're talking about completely different methodologies. Does it matter, say, how Ipsos Reid did things in 2011 now that they have gone from telephone to internet polling?

  2. I think the polls will be proven wrong at election time.

    Your modeling the polls to the election results gives the Liberals 26 seats in the West.

    That would set a record for the most that the Liberals have ever gotten.

    That does not make sense.... Trudeau should be filling stadiums at rallies and raising bucket of cash.

    Meanwhile the Alberta right comes together. The BC liberal party maintains its right wing financial responsibility, Wall is the most popular Premier in Sask and the Manitoba NDP is nearing break up status.

    Especially in BC I think that the pollsters need to clarify that the Liberals being polled about are the Trudeau Liberals and not the keep the NDP at Bay right wing coalition Provincial Liberals.

    I am a staunch Provincial Liberal and understand the difference.

    1. Note that the West in general will set a record in 2015 for the number of seats at stake.

    2. Your logic isn't very sound. The Conservatives and New Democrats both won record numbers of seats in Ontario and Quebec respectively. Why can't the Liberals do the same in the West?

    3. When the NDP won their record seats in Quebec Layton was like a rock star. He dominated the news not just politically.

      The Cons winning record seats in Ontario was a gradual momentum thing where they were increasing their totals over a series of 4 elections.

      Right now I do not see that for the Liberals in the West.

      Trudeau is not a rock star in the West... even by political standards... and there has not be a gradual increase in support for the Liberals in the West over the last few elections.

      The most significant area of Liberal support in the West is the Provincial Liberals. This is not a federal Liberal proxy as are the Ontario Provincial Liberals.

      Stockwell Day and many other prominent federal Cons actively campaigned for the provincial Liberals.

      The closest green shoot would have been the infiltration and take over of the Alberta Cons by the liberal Redford. That did not turn out well.

    4. The PCs won seats in Quebec in the past, so that's hardly new.

      The NDP performance is Quebec was certainly new, but everyone agrees that was an unforeseeable surge. 26 Liberal seats in the west would be a staggeringly high number.

      That said, BCVoR is wrong about the polls being "proven wrong". The polls today aren't preducting the election outcome. That's not even what they're trying to do. There is no possible election outcome that would prove these polls wrong.

    5. BCVOR, "Wall is the most popular Premier in Sask"…? Well, he's actually the only premier in Saskatchewan, unless they've changed to a multi-premier system of government since I moved away.

    6. Matt Astell,

      The Tories won a record number of seats but not a record proportion of seats in 2011 in Ontario. Mulroney won a higher proportion of Ontario seats in 1984 even though the Tories won 6 less seats

    7. Everyone needs to pay more attention to Eric's very first response - there are *way* more seats in play in the west than ever before. 25 seats would be a great Liberal performance, but not heroic. It would pro-rate to 22 seats last election, or 16 in 1968 (actual: 24). Failing to account for the increase in seats is making a mountain out of a foothill.

  3. Interesting observation that NDP vote erosion is no longer going to the Liberals. I still suspect that Trudeau attracted a substantial number of NDP voters after he won the leadership, but that may be all he's going to get from the NDP (which might well be enough).

    The Ontario chart shows the Greens benefiting from recent NDP erosion.

    There could be two classes of voters departing the NDP:

    1: Those who want their vote to change government.

    2: Those who want to make an environmental protest vote.

    1. Then both classes of voter are delusional.

      1. Voting for the Liberals just changes the names on the doors. history shows policies between the Libs and Cons are much the same.

      2. If you don't think the NDP is the environmental party then I think you need to define environmental, because the Greens are no more better or worse then the NDP in that area.

    2. 3. Those New Democrats who might wish to tip a few seats towards the Tories in order to keep the Liberal/Conservative seat count close enough that they can maintain the balance of power in a minority Parliament, because that's the only way to pass progressive legislation while a Trudeau is in office.

    3. George Not Bush,

      Obviously Trudeau has not gained enough (former) NDP voters since, he trails in the seat count.


      I don't think the NDP an environmental party, the BCNDP opposed the carbon tax!

      Mulcair wanted to reduce Canadian oil production not for environmental reasons much less environmental concern but, in order to promote environmentally damaging industries in Ontario.

      I would define an "environmental party" as an entity that promotes environmental policies that reduce known carbon emissions, pollutants and invasive species. And promotes recycling and industries that have limited impacts on the environment.

  4. It's interesting to note that the Liberals lead in public opinion in the big 3 provinces, yet trail in the seat count in all 3 places. Talk about an inefficient electorate (like in 1979)...and an extremely efficient Conservative vote! The Bloc still out-polls the Cons in Quebec, yet the projected seat count is 9-2. If roughly the same numbers hold, it'll be interesting to see how that truly translates into seats come election night.

  5. And Nanos out today says Liberals lead again.

    1. I can't find a Nanos poll only Nanos' "party power index". Do you have the poll link?

    2. Cap it's on CTV national website.

      Nanos doesn't conventionally poll for this stuff. Talk to Eric as I don't think it's worth the effort to discuss.

    3. Peter is referring to the Nanos index, which isn't a conventional poll with voting intentions info. The number I pay attention to in that is the 'Best PM' tally, which most recently was 33% Harper, 30% Trudeau, and 18% Mulcair.

    4. This Nanos power index should be soundly ridiculed and denounced.

      To be able to generate a headline based upon a made up question were everyone in Canada can have 2 votes if they want is wrong.

      Nanos skips right over the question who would be your 2nd choice.... which would provide some value in the possibilities of a coalition.

      When you think about it the Liberals should have close to 100 % of Canadians willing to vote for them.

      A CPC loyalist would definitely not want to vote for the NDP ( see the BC example). So In this poll it would show me supporting the Liberal and CPC equally.

      There is no way that an NDP believer should be able to vote CPC.

      So even before the poll was taken the basic assumption would be that the middle party... the party with no policies of their own, the party that stands for everything and nothing, the Liberals should do extremely well.

      the Nanos power poll actually shows leakage from the hypothesis going in as it does not have the Liberals miles ahead.

    5. Eric went and looked at the preferred for PM file. That was quite a surprise re Trudeau-Harper nationally ??

    6. The PPP index is a strange and purposefully confusing metric-why don't they list the component criteria and results?

      Re: preferred PM the only surprise is that young Trudeau is doing so well. I expect Harper's lead to increase as we move closer to the election. Liberals can't expect people to vote for them when their only policy is to legalize weed.

      What is interesting about the PPP is that the Liberals are at their lowest level in a year which would seem to confirm the polls.

    7. BCVoR - I'm planning to vote CPC, but I would easily vote NDP if that's what I needed to do to keep a Liberal out of office.

      I understand the NDP. I disagree with them, but at least their positions seem to have some ideological basis. I get what they're saying. They're wrong, but I get it.

      The Liberals scare the hell out of me.

    8. @IRA

      so in the Nanos poll you equally support the NDP and the CPC.

      But chances are that you have never voted NDP in your life.

    9. @Ira

      I think you might not be so adamantly voting against Liberals if there was a real chance that the ideological pure NDP were close to getting into power and openly going to shut down private enterprise with huge tax increase and give all Public service unions 50% raises over the next 4 years.

    10. Actually, I voted NDP in the last BC provincial election. I wanted the Liberals to win the election, but I wanted Christy Clark to lose her seat, so I voted NDP. And it worked out just as I hoped.

      But I've never voted NDP federally, no. I've only lived in ridings where they either didn't have a chance or were guaranteed victory.

      I was represented federally by, in succession, Diane Ablonczy, Svend Robinson, and Hedy Fry. Those weren't competitive races.

      As you say, I wouldn't vote NDP in that case, but they don't advocate those policies. I'm talking about the NDP we have, not some fictional bogeyman.

    11. Ira,

      I don't understand your logic. You did not want Clark as your MLA so you voted for Eby but, you did want Clark as premier? What is your rationale? Surely, you must have realised that if Clark won the election but, lost her seat she'd still be premier and it would be unlikely she'd resign as leader?

      Both Fry and Robinson's ridings are/ were swing ridings and competitive races neither were typical safe seats. Robinson’s seat was rarely in danger of turning Liberal red,the4 race was always between Conservatives and NDP with 1993 being a possible exception, Svend never won by landslide numbers but, often comfortably by a few thousand votes. Vancouver-Centre is always competitive and the NDP usually finishes second.
      If you generally consider yourself a Conservative voting NDP in Vancouver-Centre or Burnaby was counter-productive as the Conservatives have a good chance of winning both seats.

    12. Ira

      "I understand the NDP. I disagree with them, but at least their positions seem to have some ideological basis. I get what they're saying. They're wrong, but I get it."

      followed by

      "I'm talking about the NDP we have, not some fictional bogeyman.

      Either the NDP stand for something which I understand or as in the 2nd quoted statement second statement you regard them as a variation of Liberal.

      You have the inconsistency of a well paid upper middle class government employee. :)

    13. @bede - Yes, I never have voted NDP federally, because I've never been in a position where it seemed necessary. At no point have I seen the Liberals win a seat in a close victory over the NDP.

      I didn't want Clark as my MLA because people who reside in the Premier (or Prime Minister's) riding are often poorly served by their representative. I also thought Clark should be punished for her idiotic statements. But I also vastly preferred her to Adrian Dix as a possible premier. So yes, I knew she'd run (and win) somewhere else, and that was my preferred outcome.

      @BCVoR - I don't think the NDP we have resemble the NDP you describe. They're not talking about massive nationalisation schemes or command economies or wage controls (well, the Saskatchewan NDP has done that fairly recently, but they're also in no danger of winning an election). The NDP of today appear to stand for a sustainable activist government, and that's far preferrable to the endless gladhanding I would expect from Liberals.

      But I still prefer the Conservatives, because I like lower taxes, free trade, and fewer restrictions on my life.

    14. Ira wrote: "I've never been in a position where it seemed necessary. At no point have I seen the Liberals win a seat in a close victory over the NDP".

      I guess it depends what one considers "close" but, Fry usually wins her seat by 10% or less.

    15. Well it's moot now, as I've moved to Edmonton-Leduc.

  6. Éric,

    Au Québec, le cœur a toujours ses raisons que la raison ne connaît point.

  7. B.C (no) voice of reason said:
    "Your modeling the polls to the election results gives the Liberals 26 seats in the West.
    That would set a record for the most that the Liberals have ever gotten."

    That is incorrect. In 1993 the Liberals won 6 seats in B.C, 4 in Alberta, 5 in Saskatchewan and 12 in Manitoba for a total of 27 seats.

  8. The Liberals also won 27 seats in the west in 1968. Per wiki (,_1968)
    16 seats in B.C
    4 in Alberta
    2 in Saskatchewan
    5 in Manitoba

    1. Yes, Trudeaumania and the historic split in the right produced those results.

      What is the cause this time?

    2. The Liberals also won a larger share of the popular vote than the polls indicate today, 42% in BC for example or 10% better than today.

  9. 26 seats in 2015 = 25% of Western Canadian seats
    27 seats in 1968 = 40% of Western Canadian seats
    27 seats in 1993 = 31% of Western Canadian seats

  10. "Yes, Trudeaumania and the historic split in the right produced those results.

    What is the cause this time?"

    I see P.C voters as those who swing back and forth between the Liberal and the Conservative Party.

    1. Is the Liberal party is so weak and bereft of ideas they need to claim "progressive conservativism" as part of their ideology? This is historical revisionism writ large-shameless!

      P.C. voters only exist at the provincial level today speaking about them in a federal context is an anachronism. I understand Liberals live in the past, hence their leader and their platform that consists of 2 policies but, you can't lead if you're always looking behind for directions! Without ideas the Liberal party is as good as dead-that is why your prediction of 26 Liberal seats is likely to be incorrect come election time.

    2. If a liberal breakthrough in BC and the flat-lands happens (and that is a big if) historians will pin the cause on widespread loathing of Mr Harper. There is no question that a significant portion of Canadians dislike Harper. But it is a very open question whether that will translate into many western seats for Mr Trudeau.

    3. Liberals expect a breakthrough on the Prairies while they paint those provinces in pejorative insults "flatlands"?

      LIberals have no policy and no plan all they can do is insult Mr. Harper, yet, Harper is still Canadians preferred choice for PM.

      Canadians don't loathe Mr. Harper or dislike him-what Liberals fail to realise; Canadians are mature enough to differentiate between the person and his or her politics. The Liberals are in store for a very significant defeat if they don't start treating Canadians with respect and release a platform on how they will govern. Canadians will not vote for a party that is ill-prepared and so disrespectful they conceal their policies.

  11. Bede, I, like I'm sure most people here, have no idea what you are talking about.

    The best the Liberals could do in the west in the next election is around 20 seats.
    10 in British Columbia, all in Vancouver and the L.M
    5 or so in Alberta and Saskatchewan
    5 in Manitoba all in Winnipeg where the Liberals have mostly done well since 1988.

    1. Adam,

      Obviously news has not gotten around to the bloggers for hire in the Liberal party, The Lower Mainland doesn't exist-Hasn't existed for some years. Do you mean Metro Vancouver? The Liberals will not get 10 seats in Metro Vancouver because they are so out of touch and out of date they use anachronistic terminology they don't relate to Metro Vancouverites much less understand them. With a 30 point lead in the polls the Tories will sweep Alberta but the Liberals will retain Goodale's seat and win a handful in Winnipeg.

    2. I believe it was the GVRD that renaimed the Lower Mainland as Metro Vancouver. I prefer the term 'Lower Mainland' and I'm not going to stop calling it that because a bunch of politicians decided it should be called something else. Fight the Power!

  12. I'd really like to see that Calgary Mayor, Nasheed Nenshi, running for a Fed seat. Great guy.

    1. I find Nenshi's rhetoric disingenuous. When he became mayor, he talked all the time about how right-wing/left-wing divides didn't make any sense, and used snow removal as an example. "Is snow removal a right-wing or left-wing issue?" he'd say.

      And that misses the point. A left-wing approach to snow removal would be to use public workers to clear snow, and do so in an egalitarian way. A right-wing approach to snow removal would be to use private contractors (to avoid having to pay for crews when you don't need them, and to avoid necessarily using unionised workers) and to remove snow only insofar as doing so provides economic benefit. Compare Ottawa's snow removal program to that of Edmonton's. That's a right/left divide.

      Nenshi's rhetoric wins him votes, but it doesn't say anything meaningful about governance.

      Also, the last Calgary Mayor fared quite badly when he ran for federal office.

      Nenshi's recent statements on things like race and sexuality are excellent. I just wish he knew better how to manage money. When he was President of the Student's Union at the University of Calgary, he lobbied hard to prevent the publication of the SU's budget (which had been obtained by the Faculty of Engineering). Why weren't the students allowed to see how their money was being spent?

      Luckily, the Engineers has a satirical newspaper in which they were able to publish the budget. Nenshi was livid.


  13. Jan. 14: Conservatives slide for fourth week in Nanos Party Power Index

    CTV website

    1. Peter,

      Maybe you should go work for Nanos since, you seem to be promoting an obscure measure of voter preference.

    2. bede all I am doing is telling folk here that Nanos has pronounced again.

      And I don't believe his numbers really

    3. Peter,

      People have the internet-they can find this info out for themselves! What you are doping is advertising for Nanos! If you don't believe Nanos' numbers then what you are doing is false advertising by disseminating questionable information.


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.