Thursday, August 20, 2009

Conservative Majority Possible?

If an election is called in the fall, and probably if it is called in the spring, the Conservatives will be asking for a majority government. The party senses that Canadians are fed up with minorities.

I agree that they probably are, but that does not translate into wanting any majority. Canadians want a majority - with their own party in power. It could attract a few undecideds, if the Conservatives are in the lead, but it won't attract more committed Liberal or NDP voters. They are unlikely to be okay with a Conservative majority. And Quebecers voted Bloc in 2008, in part, to prevent a Conservative majority. I'm not so sure that this is a winning strategy for the Conservatives, as Canadians have been "scared off" from the Tories in the past when talk turned to a majority government.

Whether Canadians want a Conservative majority is one question. Whether the Tories can realistically get a majority is another.

The 2008 election was a perfect storm for Stephen Harper. The Liberals had a weak leader and had the least amount of support in their history. The NDP was strong enough to split the vote in some areas but not so strong that they ate into Tory chances, and the Bloc Quebecois was unable to wipe out the Conservatives in Quebec. It is hard to imagine a better scenario.

Their 143-seat win was the best they've managed. And aside from the NDP victory in Alberta that brought the party down to 27 seats and the trouble in Newfoundland & Labrador that limited the Conservatives to nine seats in Atlantic Canada, Harper's troops did the best they could do in every region. If you take the top Tory performances in their history since 2004 in each of the regions, you still only get to 144 seats - 11 seats short of a majority.

So what would it take to push the Conservatives to 155 seats?

Looking at the closest ridings in the 2008 election, there are 11 seats in the country where the Tories finished second with less than 5% between them and the winning party. In order to give the seats to the Conservatives, as well as give them their best historical performances, their popular vote in each of the regions needs to be nudged upwards. So what kind of gains do the Tories need to make between now and election day?

In short, huge gains.

First, let's look at their most likely areas of improvement. In the Prairies, they need to maintain the 22 seats they currently hold (and gain one more than they are currently projected to win). Alberta is a no-brainer, they merely need to win back the one NDP seat in Edmonton. In terms of the projection, they need to win back two seats and improve their vote by 4.7 points. That shouldn't be a problem in this province. In the North, they need to win two more seats than they are currently projected to win, and improve their vote to 48.5% from the currently projected 29.9%. Their best result, in 2008, had them at 36.2%. With only three ridings, it should not be too difficult to push their support up by 18.6 points, especially considering the Prime Minister's recent trip. Lastly, the party needs to gain three seats over the current projection in Atlantic Canada and improve their vote by 12.4 points. They currently stand at 29.1% but need to be at around 42% to be in the neighbourhood of 11 seats.

Quebec, Ontario, and British Columbia pose some greater challenges. In Quebec, the Tories need to move into a position where they can keep their 10 seats. This means improving their vote by about 8-points, which will not be easy. They've been at 24.6% before (in 2006), however, so it is not impossible. In British Columbia, the party needs to take 26 seats (six more than they are currently projected to win) and about 50% of the vote (about 13 points more than they are currently projected to have). That will be a difficult task, as their 44.4% in 2008 was a high watermark. Lastly, in Ontario, the party needs 56 seats. This means 15 more than they are currently projected to win. From their current 35.3% projected support level and their 39.2% support level in the 2008 election, the Tories need to end up at over 41% in the province. That would seem to be the most difficult task of all.

So, if the Tories can manage to make all of these gains, they will end up at 155 seats. Over their 2008 election total, they need to win four more seats in British Columbia, one more in Alberta, five more in Ontario, one more in Atlantic Canada, and one more in the North.

In the face of a more capable Liberal leader who also benefits from an NDP leader whose shine is starting to wear off, and a steady Bloc in their province, it would seem to be impossible to envision such a change of fortune for the Conservatives. Can they really do better than their best-case-scenario win in 2008?


  1. When are those new seats coming to Ontario and Alberta?

    If they can hold off an election till then,they may have a better shot.

    I agree that it seems unlikely for them, but in an election many things can change, and Iggy is far from proven in the heat of battle.

    Perhaps they can play on fears of a coalition, or trip up the opposition by forcing them to answer under what conditions a coalition would happen.

    Again I don't think it's likely, but I dont see them losing to many votes from announcing this. Did antone think they did NOT want a majority last time?

  2. You are ignoring voter backlash for the coalition "IF" they are framed as triggering another election for a power grab.

    You also ignore the POP in Ontario for the Liberals Premier Ehealth scandal and Toronto Mayor damage.

    Those will have an impact in their strongholds for bringing out the Liberal vote.

    Note the demographic and the residence in Sept/Nov for the Liberals. That demographic will not show up to vote for a Right Wing Liberal who supported George Bush (Bloc/NDP progressive vote).

    Adscam Trial in Quebec in September.

  3. Those don't create a better situation than last year.

    Harper launched an election in 2008 despite his own legislation stopping that sort of behaviour, and people didn't care (after week one, people don't care about who started the election).

    Dion's reputation was destroyed before the election while Ignatieff is still relatively respected.

    Dalton McGuinty isn't equated with the federal Liberals (who are high in the polls anyway.

    That young demographic is just as likely to vote now as they were in October 2008.

    Adscam won't help the Tories, as those votes will go to the Bloc and in any case the Liberals and Conservatives aren't competing in the same regions in Quebec.

  4. Ad Trials details in September MSM will help all parties except the Liberals. More Liberals will stay home and less progressives will drift to Iggy.

    You ignored the framing vs the coalition question. The Coalition did not exist until after the election. The CPC numbers went to 46% nationally. (Ontario/West 50-60%)

    Did Harper get every single party leader exiting his residence to declare no support for his agenda in the Fall in front of the National Media? Did you miss those statements from each leader?
    Dion's reputation vs Iggy's and their polling? Did his reputation take a hit during the campaign? How did the performance/platform of the party leader affect the turnout of their base?
    Is it possible in your OWN bias you blame Harper for calling the unwanted election for political advantage and the ROC who are not as partisan witnessed the no-confidence from the leaders of the opposition?

  5. Simmer down. I just disagree that the situation will be as good for the Conservatives as you think it will be. I really don't see a situation where the Liberals can do worse than 2008 - that was their worst performance in their history! For all his faults, Ignatieff is still a much better party leader than Dion, and I think he will be much better as a campaigner.

    Just look at the numbers. Harper needs to make his majority in BC and Ontario all the while keeping his seats in Quebec and Atlantic Canada. Nothing in the polling data indicates all of these things are possible.

    As for the coalition, Ignatieff would have an easier time distancing himself from that since it was Dion's idea, not his. And at the end of the coalition talk people were starting to warm up to the idea, so I don't think it is as much as a vote-winner as you think it is.

  6. Certainly everyone is entitled to their own spin Ad Rem. Hard #s are fine but spin is the spice of political junkies such as us.

    I disagree that Ad-Scam, E-Health, or Mr.Millers buckleing to CUPE will have any major effects on an upcoming election.

    I do think however that an election where the ballot question is "Do the Conservatives deserve a majority?" can only benefit the CPC.

    To fully disclose my political leanings, I am about as Red a Tory as you will find, and feel the choice of Iggy as leader was an offering by the Liberals to voters like me.

    He has been underwhelming so far, and has not shown me that he has the strength of leadership to keep the more leftist elements of his party in line.

    In short, I don't think that Iggy will pull very many votes from the CPC, and will probably be attacked hard from the left as "to far right".

    It could very well result in a worse preformance in the next election then the privious, depending on the questions and policies raised.

    If Iggy puts forward high-speed rail, national child-care, and EI reform, he will be in trouble.

    I am also awaiting with intrest any mention of a coalition govt., and am skeptical that Canadians have "warmed up to the idea" as Eric suggests.

    One last note (sorry for the length) Does anyone know of the status of the new seats for Ont, Alta, and B.C.? I am having trouble finding it.

  7. From Wikipedia:

    On November 14, 2007, the government of Canada announced a bill that would see the creation of 22 new ridings - ten in Ontario, seven in British Columbia and five in Alberta.[1] After protests from Ontario that it would still not receive a fair number of seats under the new system, this was revised in December 2008 to give Ontario 21 new ridings.[2] The new districts would take effect after the results have been gatehered from the 2011 census. They will thus most likely appear in 2014, and apply to subsequent elections.[3]

  8. hmmm, I looked thru wikipedia and somehow missed that.

    2014 is a long time to wait for a majority

    Thanks for doing my leg work on my lazy research.

    On a side note Bruce Anderson has a decent article in todays Globe and Mail which encapsulates my views pretty well on why the majority question is a good one for the CPC.

    Check it out if you're interested, I'm not sure of his political leanings, but it seemed resonable to me.

  9. I am looking at your prediction and you have the Liberals winning in AB and NDP losing?

    You also have the Liberals winning all 3 seats in the North.

    Can you explain how the NDP and PC lose their seats to the Liberals.

  10. They lose their seats by more people voting for the Liberals than the NDP or the Conservatives.

    The polling results are at least as good as 2006, so it should be no surprise that the Liberals will be picking up 30 seats or more.

  11. For a longer answer, in Alberta the Liberals have increased their support by 8 points, and are close to their 2004 result where they won two seats. The NDP have lost a point and the Edmonton win was close to begin with. The Tories are also down five points in the province.

    In the North, the Liberals have increased their vote share by 20% nationally, while the Conservatives and NDP have dropped by 10%. That puts the Liberals in a good spot to win the three seats in the North.

  12. Can you list where you are getting those Polling numbers for the North?

    Looking at Punditsguide the growth in the NDP/CPC vote in the North do not support your numbers.

    This looks like more fantasy than proper math. (Just my opinion, based on actual elections growth since 2004)

  13. How does this new survey help explain those 30 seat gains for the Liberals?

    October 14,2008 62.4% (did not vote for Harper)

    Aug 2, 2008 Nik Nanos found 58.9% won't vote for Harper.

    A 2.6% improvement for the CPC?

    The Left are STILL not united. Strategic voting?

    Is your overall impression of a minority government situation positive, somewhat positive, somewhat negative or negative?

    53.9% Positive, somewhat positive
    37.3% Negative, somewhat negative

  14. For the North, since there are no polling numbers, I average, with different weights, the last three elections and the proportional national change since the last election. That is where the Liberals have increased their support by 20% and the NDP and Tories have decreased their support by 10%. Because of this vote change, I calculate that the Liberals will be in a position to win all three seats.

    I'm just going by the numbers I have in the model, I don't make any judgements of my own in terms of who wins or loses. This site is non-partisan, the polling numbers are the only input I use, and those are non-partisan as well (taken as a whole).

    --- "How does this new survey help explain those 30 seat gains for the Liberals?"

    The findings of one poll proves nothing. Taken as a whole, the Liberals are up in every region. Nationally, they are projected to win 32.7% of the vote, quite a bit more than last year's 26.2% of the vote. It is impossible that they wouldn't win a significantly higher number of seats with an increase of support of 6.2 points.

    Canadians do seem to want a majority government, as you show, but they want a majority of their own party - not just any party. Liberal voters don't want a majority if it is a Conservative majority, just like Conservatives don't want a majority if it is a Liberal majority.

  15. We have "real data" from general elections for the last 3 election in the North. The data show the Liberals are in a serious decline in every election up North. There is no polling data to suggest otherwise regarding the North. Using the other regions to project a liberal growth in the North is not sound math.

    Your model can not ignore the 3 General Elections results/trends without polling data to back up your growth for Liberals up North.

    It looks like "fuzzy math".

  16. Think what you want, I'm just trying to take the present polling numbers and use them to make a projection in the North.

    We also don't see "serious decline", we see one bad election in 2008. I'm afraid that isn't enough.

  17. Your reasoning in suggesting the growth in the Liberal POP is supportable in Regional Polls by Ekos, Angus etc.

    Your desire to equate that data to the North is just faulty.

    Voters in Ontario may have increased vote for Party A, polls prove it.

    That may or may not be reflected in another province or territory.
    Without actual Polling for the North the most logical method is to use the existing data.

    NWT riding 2008 47.7% turnout.Liberal 13.6% NDP 41.1% Grn 5.5% CPC 37.6%

    Your model gives the Liberal a victory? Your kidding right?

  18. You have chosen to ignore the negative trend for the Liberals up north via REAL Data from 3 General Elecions to push your "model" as being a better indicator of the voter intention?

    The same model that had Liberals win 5 seats in AB last week?

    I am suggesting your model is deeply flawed at the benefit of the Liberal Party.

    My bias is I like math and numbers to have a pattern.

    You model has glaring flaws that are not open to review or debate.

    I admire your effort and work on this site. I just can't understand how you arrive at those conclusions.

  19. Obviously, I know that changing support levels in one part of the country doesn't mean that they will necessarily improve in another part of the country.

    But if, taken as a whole, the country has shown more support for one party and less support for two other parties, the conclusion which can be drawn is that it is more likely this will be the case than otherwise. It makes much more sense to assume a particular riding will go along with the trend than it will buck the trend. Otherwise, it is not a trend - more ridings than not have to show an improvement in Liberal support because the Liberals nationally have increased their support.

    The three ridings in the north are highly volatile in that individual candidates can play a huge role. So the fact the Liberals did so poorly in the Western Arctic in 2008 doesn't mean they will do poorly again next time around. With a better national campaign and a good local candidate, they have every chance of taking the riding, which they have held in the past.

  20. Ad Rem, please take your hostility elsewhere. The model is not meant to favour one party over the other. That you see a bias is likely the fault of your own bias, as demonstrated by your ridiculous Ignatieff cartoon.

    My model is open to review and debate, as is very easily shown in the change I made to the Alberta projection. That was the result of a simple error on my part, not because of any bias. I have made changes in the past as well, as you would know if you took some time to read through the site.

    You're asking me to change my projection because you don't like the result. If I gave all three ridings in the North to the Conservatives you wouldn't have cared a bit.

    I have taken results from past elections and made a projection based on those past elections AND the current political situation in Canada. For you to claim otherwise is false. Yes, the Liberals demonstrated a significant drop in the region in 2008, but until we have another election to prove that this was, in fact, a real trend rather than the result of a bad election, I am going to continue making my North projection the way I have been doing. The Liberals did not sure any major changes in support in 2004 and 2006, leading me to believe that 2008 was an aberration, rather than a "trend".

    If you feel that my projection does not give the results you want, feel free to start your own projection site rather than to attack me as being partisan. I resent that and I am insulted by it. I have tried very hard to make this site non-partisan, and the projections are simply what the numbers give me. I don't choose what I want the projection to show.

  21. I have not attacked you. I question the algorithm you use.

    Your are wrong regarding the CPC winning all three. If your model stated that again without a Poll you discounted the General Elections.

    Bias is to decribe the model is faulty to favours one party in the North.

    My choice of cartoons is personal, if you had 3 Children an original Elmo that I could have sold for $300 you would be much more understanding.

    I have not once suggested the CPC will win all three seats. In fact if your model had CPC win those 3 seats I would still question your model.

    It is easy to suggest my choice of cartoons clouds my judgement.

    I am not making it personal and have not suggested your were a Liberal.

    I said the model has a bias toward the Liberals. (Thats all)

  22. I disagree that the model is biased towards the Liberals. I've used the last three elections to calibrate the model, and the Conservatives won two of them. And despite the fact that the Liberals and Tories are neck-and-neck in the national support level, the Conservatives still have the projected seat lead, and have held it for most of the year.

    In the North the Liberals may have the advantage in the projection because of the way it is setup. This was not intentional and in fact I would go so far as to argue that this is natural for the region as the Liberals have historically held the advantage here.

    The fact of the matter is that the last three elections in the North have actually been, on the whole, good for the Liberals. They won the three seats in 2004, two seats in 2006, and one seat in 2008. In the first two elections they had more than 40% of the vote. With such a significant jump in Liberal support nationally from the disastrous 2008 campaign, it is not unusual that the Liberals would have a good chance of taking back the three seats.

    Considering the amount of seats in the region and the fact that, if I'm wrong, it will only be by one or two seats anyway, I think you are attributing more importance to what you see as an error than you should.

  23. Eric did you look at the numbers for the NDP/CPC for those last 3 elections?

    I use Punditsguide as well and your cherry picking detail to validate your model.

    You also suggest the Liberals held the riding in the Past and you than deny the Trend and growth of the NDP/CPC/Green votes in those specific ridings for National numbers.

    You have ignored the trend and growth of the NDP/CPC vote in those 3 General elections.

    In 2000 Neck and Neck with NDP.

    The Liberals are FLAT none or little growth in those 3 elections.

    Yukon Lib CPC NDP GRN
    2008 45.8 32.7 8.7 12.8
    2006 48.5 23.7 23.9 4.0
    2004 45.7 20.9 25.7 4.6

    The Liberals vote has not collapsed because of 2008. The addition of the Green and growth in other parties make your case for the Liberals winning all 3 without any LOCAL Polls to counter the existing data (faulty).

    Thats it. The patterns your suggest are not there and you are using your (definition/model) of a "disaster" in 2008 to discount the growth of the other parties including the Greens.

    Are you introducing subjective data into your model or analysis regarding different weight?

  24. Yes, of course the model takes into account the results of the other parties and no, I am not "ignoring" their performances. I am well aware of the electoral history in the North.

    The NDP have been stagnant in the North while the Tories have made some significant gains - that is why they are very close to winning one of the three seats in the North. They just need to improve their national position a little bit.

    Ad Rem, I'm afraid you're really starting to annoy me, because you are attributing things to me that is not true.

    I am not ignoring trends, I am not ignoring the performance of other parties.

    I am averaging out the 2004 performance (less weight), the 2006 performance (a little more weight), and the 2008 performance (more weight), to get a result. Adding to that is the proportional change nationally to the 2008 performance (the most weight). That gives me a result. That result is compared to the necessary percentages in the North that history has deemed necessary to win three, two, one, or no seats in the North.

    That's it, that's all. The Conservatives are closer to the 0 seat level than they are the 1 or 2. The same for the NDP.

    That's it. Just numbers.

    I am not suggesting any patterns, I am not guessing at trends. I'm looking at the historical performance of the four parties since 2004 and then estimating what the current position of the parties would be by their proportional growth or losses nationally. You can disagree with using that method of estimating current support levels, but that is the method I've chosen because there is nothing else to do without using my own subjective analysis of the situation in the North.

    As this site is meant to be non-partisan and based on the numbers alone, I have not made any judgement call like that. I just punch the numbers into a uniform projection system.

    In the end, does it really matter? We're talking about three seats that are always anyone's guess.


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.