Friday, October 23, 2009

Projection Update - 137 CPC, 98 LPC

Another week, another projection update. Only a small change this time, with the Conservatives gaining one seat from the Liberals in Ontario.This puts the Tories at 137 seats, the Liberals at 98 seats, the Bloc at 49, and the NDP at 24. The Conservatives have also gained 0.4 points nationally, while the Liberals have lost 0.3 and the Bloc and Greens have lost 0.1 points apiece.

In British Columbia, the Conservatives have gained 0.3 points and lead with 38.3% and 21 seats. The Liberals have lost 0.2 points and stand at 25.0% and 10 seats. The NDP has gained 0.1 points and is at 24.6% and five seats. The Greens have lost 0.1 points and are at 11.5%.

In Ontario, the Conservatives have made another big gain. They're up 0.5 points and lead the province with 38.8% and 51 seats, up from 50 seats last week. The Liberals are down 0.6 points and one seat, and are now at 35.8% and 44 seats. The NDP has gained 0.1 points and stands at 15.1% and 11 seats. The Greens remain steady at 9.8%.

In Quebec, The Bloc Quebecois has lost 0.1 points but maintains the lead with 37.0% and 49 seats. The Liberals have lost 0.4 points and are at 28.1% and 18 seats. The Conservatives have made a significant 0.5-point gain and are at 17.8% and seven seats. The NDP is down 0.2 points to 10.5% and one seat, and the Greens are up 0.1 points at 6.2%.

In terms of other large regional movements, the Conservatives are down 0.3 points in Alberta and up 0.5 points in the Prairies. The Liberals are up 0.3 in Alberta and down 0.4 points in the Prairies. The NDP is down 0.3 points in Alberta and 0.3 points in Atlantic Canada.

The Conservatives have had a good week, they've only dipped in Alberta. And their gains in Ontario and Quebec were big. The Liberals had a bad week, but not disastrous. They're up in Atlantic Canada and Alberta, but the losses in Ontario and Quebec are important. The NDP made some tiny gains in BC and Ontario, but their losses in Alberta, Quebec, and Atlantic Canada hurt them, especially in Alberta and Quebec where they have but a toe-hold. The Greens remained remarkably steady, while the Bloc took a tiny hit but grew their lead over the Liberals by 0.3 points.

I get the sense that things are going to return to normalcy, if we can call it that. The Liberals will claw their way back up to about 30% and the Conservatives will dip back to below 40% but above 35%. It is obvious now that Michael Ignatieff took a big hit by trying to force an election. I'm not convinced that the idea of an election itself was the problem, but rather that the Conservatives haven't done enough to deserve being booted out of office and, in any case, Ignatieff did not make a strong enough case as to why he should replace Stephen Harper.

Now that the electoral debate is over, I think a lot of angry swing voters who used to answer "Liberal" to polls will go back. But enough will stay with the Conservatives to keep them comfortably ahead.


  1. I don't think as many swing voters will swing back to the Liberals this time. When the Liberals did this before (back last June and last December) they backed down very fast. This time they didn't back down (initially) and that left people sitting on a Tory vote and not feeling too bad about it. I suspect that while there may be a few who drift back, the numbers won't move very much at all. It was lunacy to try and force an election to begin with.

  2. Eric are you trying to start a flame war and get called "conbot".

    Again great site on keeping the site update frequently, I may not agree with the algorithm (not because I am voting CPC) because other projection sites have the numbers differently.


  3. I still think that a lot of people answering polls in AB are losing support for Stelmach, not PMSH.
    We will wait for polls to come out, after the Nov leadership review for Stelmach to see if things change.
    I know when called I always say Green, just to improve their numbers. Wont vote for them but they need all the help they can get.
    I don't think I am the only one who doesn't tell the truth when asked.

  4. I know I have hammered this point on ad nauseum, but I'll say it again. Most of the seat projections make sense based on the projected popular vote numbers. But once again, BC makes little or no sense to me. In the last election the Tories took 22 seats in BC with 44% of the vote and the NDP took 9 seats with 26% and the Liberals took 5 seats with 19% (they were actually very lucky to get 5 seats with such a low popular vote but that's another story). So I still don't see how you can have Tory support down six points, NDP support down about one point and Liberal support up six points and have the NDP with only five seats in BC. When I take the results in BC in the last election and assume that the Tories drop about 6% in every riding and the Liberals gain about 6% and the NDP is very slightly lower, I get the Liberals winning back two seats on the north shore of Vancouver and being in a dead heat with the NDP in Vancouver Kingsway. I also get the Liberals winning Saanich-Gulf Islands - but there was a unique situation there last time with the NDP pulling out and the Liberals almost totally monopolizing the anti-Tory vote. Next time with the NDP running someone and with May running for the Greens - that seat is probably unwinnable for the Liberals. I also get the NDP winning Surrey North and being in a dead heat in North Vancouver Island.

    IF (a big IF), the popular vote in BC was actually CPC 38% and Libs and NDP about 25% each as our projection suggests, I get NDP 11, Liberals 8 and Tories 17.

  5. DL,

    Seat projection is very funny. In my opinion other projection site give the Liberal less seats (10-15)

    I think this algorithm is too "efficient" for a specific party at the expense of the NDP/CPC.

    Wasted votes in AB mean nothing for CPC as they do in Montreal proper or Toronto proper.

    If the Liberals gain 2% in AB it wont help either. The gap is too big for seat losses.

    Looking at "close" riding less than 1000 votes vs the perfect storm scenario.

    Pundits guide has a great site with close ridings.

  6. I think your confusing the seat projection with the popular vote projection that Eric has done. His popular vote projection puts a lot more weight on older polls as opposed to being based purely on the most recent polling. If his popular vote projection was right and the Liberals got 30% of the vote, they probably would get about 98 or so seats. That's about what 30% got them in 2006.

    My issue is with what his popular vote produces in the way of seats in BC and I just can't figure out ANY formula whereby the Tory vote drops 6%, the Liberals vote rises 6% and the NDP stays about where it is - and somehow you get the NDP losing half its BC seats. That would only make sense if there were a whole bunch of seats in BC that the NDP had won very narrowly over the Liberals with the Tories a distant third - but there aren't ANY seats in BC that fit that description.

  7. DL,

    I'll take another look at British Columbia some time in the future.

  8. DL,

    You just repeated my exact point. The Liberals are strategically benefiting from seats gains at the expense of other parties.

    We don't have riding by riding polls. We have 45 swing ridings etc.

    I don't get it either.

    If you lost by 5,000-10,000 votes you are safe.

    If you barely won the move may help but without a local poll it is pure speculation.

    905 Liberals leads have shrunk from 10,000 to less than 500 votes in several cases.

    I will try to find Oct 2008 vs Oct 2009 polling show the numbers have gone up for for the CPC in 905 and Toronto. I think the CPC have moved up by over 5%. That would wipe out most of 905 ie Brampton and a 1-2 416 I think. Some in Missauga would be safe if they had leads of 5k+ but Paul Szabo might be vulnerable.

  9. --- "without a local poll it is pure speculation."

    What? You say that as if we projectors don't know what the exact situation is in all 308 ridings!

  10. Eric,

    I prefer to use the computer models developed by the UN that predict AGW.

    or at least dust off my Vic 20 to end the debate of Mac vs PC.

  11. CanadianSense,

    lol, global warming science is such junk isn't it ?

    No warming since the 90's and yet every year the warnings are getting more and more dire.

  12. Jesse I never argue about this new religion.

    I just want to know how much is it going to cost, who is paying for it, who/how are they going to enforce these new rules to save us from ourselves.

    Sadly no one will respond regarding those minor details.

  13. Eric,

    "Don't start."

    Yikes, I take it someone is a true believer?

    The science is far from settled.

    And then the economics, as CanadianSense refered to, is even more debatable.

    A lot of very talented economists suggest we make no effort to cut back on C02 production because things like child poverty and disease kill more people in the developing world every day then climate change ever will.

    Carbon heavy industrial development, as we all know, raises the standard of living exponentially. A lot of starving African, Brazilian, Chinese, and Indian children are willing to make the trade off. Environmentalism IS borderline racist in its refusal to support human development.

  14. Eric here is my Vic-20 moment.

    Last year Ekos Polled Metropolitan Cities including Suburbun area

    The current EKOS Poll a HUGE difference in the Metro Ridings
    Page 5
    Oct 2009 vs Oct 2008

    Libs had a 20% lead in 2008 and Now CPC are 10% higher can you look at those numbers again.

  15. Re: Jessie

    "Environmentalism IS borderline racist in its refusal to support human development."

    Soooooo the Conservative government's position that less developed countries should cut back on their emissions along with us is racist, and the old Pro-Kyoto crowd's position that we should cut back first is less so?

    That seems to be the logical outgrowth of that statement.

    Personally, I'd rather leave the label of racist to actions based on race; not economics or pollution.

  16. Kevin,

    I guess you've never heard of disparate impact? Britian and the commonwealth, as well as Europe, have already gone through industrialization and reaped its benefits. And now we're telling people from China, India, Brazil, Africa that they're not allowed to do the same thing.

    You can see how they would think it racist. Or the strong keeping the weak down.

    Of course, deep environmentalism isn't intentionally racist in theory, just in practice.

    This line of thinking says that all human life is not an absolute value and must be balanced against the wellbeing of the eco-system. If humans starve so that a rare species of frogs live so be it, frogs are important too!

  17. Kevin,

    the Conservative critique of Kyoto is that if we ACTUALLY want to stop emmisions then all countries need to be on board. Its logical and makes sense. Kyoto would have done nothing but move pollution offshore.

    Of course, we all know the developing world won't ever really agree to this.

    Which is kinda sorta the point because the Conservatives aren't really planning to do anything serious about CO2, thank goodness for that.

  18. Drop the "racist" line. Otherwise, in practice, the war in Afghanistan is racist.

  19. Eric,

    Did you have an opportunity to review the sesmic shift in the metro suburbun areas according to EKOS polling I linked?

    Are those captured and reflected in your analysis?

    Last year it was a 20% lead in Toronto for Liberals and now the CPC at 10% ahead.

  20. --- "Did you have an opportunity to review the sesmic shift in the metro suburbun areas according to EKOS polling I linked?"

    Yes, but it is just one poll.

    --- "Are those captured and reflected in your analysis?"

    Indirectly, yes.

    --- "Last year it was a 20% lead in Toronto for Liberals and now the CPC at 10% ahead."

    No, it was a 20% lead in the city of Toronto, the lead was smaller when you include the suburban areas of Toronto, which EKOS does in its recent polling.

    Obviously, with the Tories polling over 40% in Ontario, their performance is being captured in the projection.

  21. Eric the dissappearance of the Liberal lead has NOT been a single poll just released.

    The gap or lead has been diminishing for several months.

    I can link those EKOS Polls.

  22. It's all represented in the projection.

  23. Eric,

    "Drop the "racist" line. Otherwise, in practice, the war in Afghanistan is racist."

    That makes no sense. The victims of the fighting are all middle eastern sure but that's because they're the only people fighting. So you have a representative sample.

    The scope of the fight against global warming is world wide and yet western society is to reap the rewards of modern life and the developing world is being told not to.

    Its beyond obvious that the haves are white and the have nots aren't.

    Sigh, I guess when celebrities stopped singing "save the children" and switched to "save the planet" we should have known the children were being thrown under the bus.

  24. Never knew the Japanese were white!

    Learn something new everyday.

  25. Eric,

    Sure and parts of Africa and other places in the world with a history of colonialism or trade were also industrialized, benefiting from their interactions with Europeans.

    But the major population centres of the world today, China, India, Brazil, Indonesia are not white and are being told not to develop. Its very troubling.

  26. I don't think the NDP is likely to do so badly as you are projecting them to, or the Liberals as well as you think.

    If I understand your models correctly, you put significant weight on whether a party has historically done well in a riding. However, the NDP has been increasing its strength rapidly - from 19 in 2004 to 37 in 2008 - and in general those gains would not have been predicted by that kind of a model.

    I've been using a model that adjust the vote in each riding by the amount that the polls for that province/region have shifted from the results of the 2008 election.

    I've gotten much less drastic changes than your own: 135 Conservative seats, 89 Liberal, 35 NDP, 44 Bloc, and 2 Independents.

    Your results would have the NDP losing multiple BC seats that they won with over 40% in 2008. It seems like too big a change to predict based on a loss of 1.5% in BC poll and 2.7% in Canada overall - even with a big Liberal recovery. Most of the Liberal gains in BC seem to be coming away from the Conservatives.

  27. Ipsos Reid out:

  28. Despite what Bricker says a 15 point lead would IMO translate into a 170 - 180 seat, CPC win. A nine point lead in ON would certainly allow the CPC to break into Toronto and Ottawa. Likely Vancouver and Winnipeg would fall victim too. Although regional breakdowns are not given for the Maritimes and BC or Saskatoba I feel fairly confident that the CPC would gain seats in MB, the Maritimes and BC as well as ON. As to if this kind of spread could be sustained through an election campaign I don't know. Perhaps Eric can tell us what happened the last time there was a gap between the first and second place party in the 15% range, say 13%- 17% in term of a seat counts.

  29. In the election last year the Conservatives were almost 13 points ahead of the Liberals and that still wasn't enough to get them a majority and if the Ipsos poll were the results on election day and Tory support in Quebec went down to 18% and the BQ went up to 42% - it would probably mean the Tories losing about half of their Quebec seats.

    In about two weeks time we will see some real live votes in ballot boxes when we have the four federal byelections. If the Tories win CCMV it will be the expected result, but we should look at the margin compared to 2006 (the last time there was areal election in that riding). If the Tories lose by a lot in Kamouraska, it will be evidence that they will be playing defence in Quebec. IF, the Tories were actually going to start gaining seats out west, on paper New Westminster-Coquitlam should be low hanging fruit. But the consensus of the pundits etc... is that it will be an NDP hold - so if the Tories can't win a seat they lost so narrowly last time - where will they make gains?

    This poll seems to be at the high end of the scale for the Tories with a 15% gap given that Nanos and Ekos have more of a 10 point gap - so we shall see. Keep in mind that three days before the 2006 election there were polls showing the Tories 15% ahead - and they ended up only 6% ahead.

  30. DL the CPC were not high in the Metro centres Van, Toronto, MTL.

    Compare Ekos 2008 vs 2009.

  31. YOu can't be 15% ahead in a poll for that poll not to have you having significant support in Vancouver, Montreal and Toronto. and the Tories already hold over half the seats in the Vancouver area - when Ekos reports numbers for "Vancouver" that is the entire metropolitan area including tons of heavily Tory seats in the suburbs. It will be a cold day in hell before we see the Tories win any seats within about 200 miles of Montreal. AS for the the GTA, I don't know what the popular vote was in the GTA in the last election, the Tories did win a number of suburban Toronto seats last time. I suppose that if the overall Tory lead in Ontario went from 6% lead they had in 2008 to the 10% lead some polls are showing, they would probably gain a handful of marginal seats. The thing is once you get past 3 or 4 Liberals seats in the GTA that were close calls, the Tories come up against this wall of seats that went Liberal by really big margins.

  32. DL,

    No matter how the four by-elections go, I wouldn't put much weight on them as predicting national trends.

    By-elections often produce results that are not reflective of what might have happened in a general election.

    This can happen for several reasons:
    - low turnout
    - voters know the by-election will not affect the overall balance of power in Parliament, and so may feel more free to vote in some unexpected way
    - without an active national campaign, by-elections can be more prone to swing on riding-specific issues and the candidates

  33. DL,

    nice try but Martin is exactly right.

    In fact I saw the pundits predict the NDP hold the BC seat and take the Nova seat because of the popularity of the NDP provincial government.

    The BQ are expected to hold the two Quebec ridings.

    So one can understand why a partisan like yourself would have an interest in exagerating the importance of by-elections.

    I can just hear the narrative now - NDP wave sweeping the nation, Layton to form official opposition against Tory minority.

  34. Jesse the CPC will win 2 of the four by elections.

  35. Byelections do tell us something that polls don't. They tell us about how people vote when an active campaign takes place and people are forced to actually think about what they are going to do.

    Its true that they have idiosyncrasies - but the same could be said about the race in all 308 individual ridings.

    If we look back at recent federal byelections, they each tended to tell us something about the political situation at the time. For example, the Liberal loss in northern Saskatchewan and near loss in vancouver Quadra was a sign of the impending liberal collapse in the west. The Outremon byelection was a sign that things were not good for Dion. The very existence of a united Conservative party in Canada is largely attributed to a byelection in southwestern ontario where the CA came in third and convinced harper that there was no hope for his party on its own.

  36. What we can be sure of with the by-elections is that the party that wins will say it is a good national sign for their party and the party that loses will say they are just by-elections which don't mean much nationally.

    You can already see it in the way you commenters are positioning the by-elections!

  37. Eric,

    don't worry, not matter what happens i'm sure we will all agree that its a bad sign for the Liberal party of Canada.

  38. I wouldn't go so far as to say that by-elections never tell us anything, but they are not a great signal of national trends in the next general election.

    Let's look at another 2007 by-election: Roberval-Lac-Saint-Jean:

    In 2006 the BQ won this riding by a margin of 45.2% to 37.2%. In 2007, Conservative Denis Lebel won this riding in the by-election by a whopping 59.7% to 26.8% margin. And yet no serious analysts were predicting that the Tories would start sweeping scores of Quebec francophone ridings with outright majorities.

    "Its true that they have idiosyncrasies - but the same could be said about the race in all 308 individual ridings."

    But the point is that you are trying to interpret what might happen in a set of 308 based on what happens in 4 (based on dramatically different circumstances).

    It's perhaps an overly simplistic parallel but imagine trying to interpret the political opinion of a village of 300 people by only interviewing 4 of them.

  39. DL wrote:

    "It will be a cold day in hell before we see the Tories win any seats within about 200 miles of Montreal."

    That's a bit extreme, isn't it?

    200 miles is a big enough radius that it would include the Tories' Quebec City-area ridings as well as Lawrence Cannon's Pontiac riding.

    I would agree, however, that Montreal and its suburbs are not very fruitful ground for the Conservatives.


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.