Wednesday, March 18, 2015

A contest of regions

In my column for the CBC today, I look at the latest numbers from the projection, which was updated late yesterday afternoon. Please check it out.

Those are the macro numbers. What about the micro numbers?

I thought it would be interesting to break the projection down to the sub-regional level to try to get an understanding of how things are playing out. Seeing the numbers in Ontario, Quebec, and British Columbia are fine - but these provinces are far from monoliths. In fact, every province is.

What I've done below is use the riding-by-riding projections to get an estimate of regional support. But pay attention, as there are some caveats.

This is not the same as the polling averages for these regions, but rather how the national polls translate into regional support levels. So, it is not necessarily reflective of how things are really looking in every part of the country. What it tells us instead is that, based on wider trends, these are how these regions are likely playing out. It is meant to give us a general idea, rather than a specific measure of public opinion.

In effect, it is using the projection model to project regionally, rather than by riding. But it isn't as simple as swinging the national polls and their provincial breakdowns to the regional level, as the riding projections take into account incumbency and the like. If a region has seen a lot of incumbents retire, that has an effect that a simple swing would not capture. By-elections, too, are also part of the equation.

As to the definitions of the regions, I used the ones presented here on Wikipedia. It is perhaps not a perfect division of the country (I would have done some of these differently), but it is good enough. Let's get to it.

Starting in British Columbia, the Conservatives are strongest in the B.C. Interior, where the New Democrats are running a decent second. The Tories are also ahead in the Fraser Valley and the Southern Lower Mainland, but here it is the Liberals who are running a close second.

The Liberals have the edge in Vancouver and the Northern Lower Mainland. The Conservatives and New Democrats are in a close battle for second, though even here we're looking at a sub-regional race: the NDP in Vancouver itself, the Conservatives further out.

The only regional lead the New Democrats hold in British Columbia is Vancouver Island, where they are narrowly edging out the Conservatives. But look at the Greens here - they are certainly in the race in this part of the province.

Alberta is far more complicated than the wide lead the Conservatives hold would suggest. Certainly the Tories are ahead through the province. But that leads runs from a very comfortable 46-point edge in rural Alberta to a slimmer margin of 13 points in Calgary.

The Liberals are clearly running second in Calgary by this measure, but are in a closer race with the New Democrats in and around Edmonton.

The Conservatives are well in front in rural Manitoba, where the Liberals trail in second at some distance. In Winnipeg, however, the tables are turned. There, the Liberals are comfortably ahead, with the Conservatives in second.

The definitions that Wikipedia had for the regions of Saskatchewan were not very satisfying, and I suspect they are a relic of the old urban-rural seats the province used to have. With their definitions, the Conservatives lead in both the south and the north, with the Liberals and NDP in a close race for second in the south (which includes Regina) and the NDP ensconced alone in second place in the north (which includes Saskatoon).

A more reasonable regional breakdown would be between rural and urban Saskatchewan. What I have done here is define Regina and Saskatoon as the six ridings with those cities in their names. When we do that, we see the race is much closer, with the Conservatives narrowly ahead of the NDP and the Liberals not too far behind. In the rest of Saskatchewan, the Conservative lead is not challenged.

The Ontario numbers are interesting in that they show the wide regional variations that exist in the province, with 30 points separating the best and worst performances for the Conservatives, 24 for the Liberals, and 19 for the NDP.

The Liberals have the edge in Ottawa, suburban Toronto, central Toronto, and Brampton - Mississauga - Oakville. Only in central Toronto do the New Democrats supplant the Conservatives for second place.

The Tories, meanwhile, are in front in eastern Ontario, central Ontario, Hamilton - Burlington - Niagara, midwestern Ontario, and southwestern Ontario. In these regions their main opponent is less consistent. The Liberals are second in eastern, central, and midwestern Ontario, while the NDP is vying for that position with the Liberals in the southwest and the Niagara Escarpment.

The New Democrats are only in first place, and in this case they are tied with the Tories, in northern Ontario. And even there the Liberals are only two points behind.

Southern Durham - York is a close race between the Liberals and Conservatives.

The problem for the Liberals in Quebec is demonstrated by this breakdown. Though they lead province wide, at the regional level that translates only to a lead in western Montreal and northern Montreal - Laval.

They are also in a close race with the New Democrats in eastern Montreal and Montérégie, south of the metropolis.

The NDP's support is relatively uniform, and in a divided landscape they are ahead in central Quebec and Laurentides - Outaouais - Northern Quebec. In the Eastern Townships they are neck-and-neck with the Conservatives.

The Tories have the edge in eastern Quebec, Côte-Nord - Saguenay, and Quebec City, generally where they are expected to pick up seats. Their lead in eastern Quebec is a bit distorted by the region stretching to the south shore to Lévis.

Finally, the Bloc Québécois is ahead nowhere in the province, though is not far behind in central Quebec, eastern Montreal, and the Montérégie.

The Wikipedia regional definitions for Atlantic Canada don't go beyond the provincial level, so these numbers aren't as useful. We sometimes get provincial breakdowns in the polls anyway. But as a proof of concept, it works as the numbers generally line-up with what the polls are showing in each of these provinces.

The Liberals lead in all four, but have a closer race in New Brunswick, where the Conservatives are strongest.

Let's try to break it down a little more. In Newfoundland and Labrador, for instance, the Liberals would be ahead in Avalon and St. John's with 43% to 34% for the NDP and 17% for the Conservatives. In the rest of the province, that would be 67% for the Liberals, 19% for the Conservatives, and 9% for the NDP.

In Nova Scotia, the Liberals would be ahead of the NDP in Halifax with 47% to 28%, with the Conservatives behind at 16%. On Cape Breton, it would be 64% for the Liberals to 19% for the Tories and 10% for the NDP, while in the rest of Nova Scotia the Liberals would be ahead of the Conservatives with 46% to 30%, with the NDP at 15%.

And in New Brunswick, the Liberals would be ahead in the urban ridings (Fredericton, Moncton, Saint John) with 43% to 31% for the Tories and 17% for the NDP, with the margin narrower in the rest of the province, with 40% for the Liberals and 32% for the Conservatives to 19% for the NDP.

Local battlegrounds

Looking at the map this way, we get a different picture of the electoral race. Simplified, there are three battleground provinces: British Columbia, Ontario, and Quebec. But breaking it down by region, we see that there are many more battlegrounds.

The Liberals and Conservatives are fighting over the Fraser Valley and Southern Lower Mainland in British Columbia, Calgary in Alberta, Winnipeg in the Prairies, the outlying areas of the GTA in Ontario, and New Brunswick in Atlantic Canada.

The Liberals and NDP are contesting central Toronto and northern Ontario, as well as the francophone areas around Montreal and St. John's in Newfoundland and Labrador.

The NDP and Conservatives are at odds too, over Vancouver Island (where the Greens are also present), urban Saskatchewan, southwestern and northern Ontario, central Quebec and Quebec City.

And some parts of the country, particularly in Quebec but also parts of Ontario and British Columbia, have areas where three (or four!) parties can pose as legitimate winners.

It makes for a complicated electoral map - one that we usually only get to see from 30,000 feet. Zoom in and the picture gets fuzzier, but an impression of how things are lining up can be seen.


  1. Three quick questions:

    1) In the aggregate polls you show the weighting by polls in a bar rather than a number. A number would make it easier to establish how much your aggregate is EKOS driven. Right now it would look to be over 50% with Forum accounting for say 20%. Would you add the weight number please.

    2) You have Nanos listed as having done polls. I have not found these polls that you have included in the aggregate. You are not using the preferred PM that is the only pick one of Nanos polls, are you? Not a question of who would you vote for.

    and one more: You have the Liberals taking over 50% of the vote in Avalon, NL despite kicking Scott Andrews out and likely facing a law suit from him for lose of job on hearsay non-legal, non-scrutinized accusation.. AND the CPC actually getting significantly more votes in 2011 in the new boundaries of Avalon.

    You have this with 95% certainty that the Liberals retain this poll. I can't imagine that the Liberals polling in NL has increase since Danny Williams ran the ABC campaign. Has it?

    1. Nanos released some national voting intentions numbers a few weeks ago in their weekly index report.

      Yes, the Liberals polling in NL has increased substantially. Liberals took 47% in 2008 in NL, latest CRA poll has them at 64%. Other polls since end of November have them between 58% and 64%.

      And I strongly doubt locals will be taking Andrews' side on this, but we'll see.

    2. Eric

      Are you implying that the folks in Avalon don't believe in due process and the rule of law?

      Mr. Andrews was fired from his job with anonymous, hearsay accusations that did not stand up to scrutiny in any court.

      Senator Colin Kenny is still political active as a senator who used to be Liberal despite far worse allegations that were not strong enough to convict him in a court of law although it did not make it to a court of law. He was cleared by a 3rd party investigation an a senate committee.

    3. Rest assured, I'm quite confident that the good people of Avalon believe in due process and the rule of law.

    4. Seriously? You are coming with anonymous hearsay talking point on this site? An Independent inquiry is not anonymous hearsay.

      I expect the Lib vote will go up after Trudeau's excellent handling of the situation.

    5. MP Andrews did say today that he was satisfied with the independent inquiry and its findings. Including that he needs to be more aware of this Newfoundland friendliness and that he has learned a great deal about personal space and the feelings of others.

      Unfortunately, it seems from media reports about what had triggered this inquiry is that this friendliness includes a bunch of unwanted behavior including following a woman home, forcing his way through her door, groping her, and grinding his pelvis against her, then later accusing her of being a c**k tease. I'd say that is most unfriendly. No doubt this has not been tested in court but he seems to see this behavior is a problem is accepting the inquiry and not contesting its main summary points.

      I seriously doubt that this issue is going to change how people vote in any way. But really parties dropping MP's from their party is an internal matter, it does not require a court of law to make findings. There are a number of conservatives who have been booted out as well without any criminal convictions, and of course one even made justice minister actually having actually broken laws

    6. Eric it's really odd how Ottawa is this Liberal bastion in Eastern Ontario and once you are outside the city the CPC completely dominate even with some really lousy candidates.

    7. Your opinion of the candidates aside, that isn't very odd. The Liberals generally do better in cities and the Conservatives generally do better outside of them.

  2. Eric, Thank you again for an excellent discussion.

    I agree the numbers have pretty much flattened out in the first few months of 2015.

    The flattening may indicate that the CPC has mined out what could be gained with C-51 and tough on crime. It seems they need something more to get to 39.6%

    We are in interesting times.

    1. George

      Ekos that provides over 50% of the aggregate has the CPC at the same place they were when they won the majority.

      last election ekos had the Cons a 33.9 to 34.5 in the last 5 polls before the election.

      In the last 6 polls over the last month n=18,000 EKos has the Cons at 33,32,31,32,33,35 average 32.7.

      In answer to the Statistics hypothesis:

      The level support for the Cons is at the same level for the last month as it was at the last election.

      This hypothesis has been proven true by EKOS.

    2. BCVoR, you use a lot of numbers and mathmatical words, but you fail to show equalivence. First for all, the last 6 EKOS polls were not in the last month, but released since February, and those numbers were 32.9, 32.2, 31.1, 32.3, 32.0, 35.0, average 32.58.

      Next, only one (the oldest) of those numbers is equal or higher to the lowest of your 5 polls prior to the last elections. The average is a 1.3% less than the lowest.

      Equally, your final 5 EKOS polls for 2011 are wrong, being 34.6, 34.5, 34.8, 34.0, 33.9 average 34.36.

      Finally, your n count is per item not culmative. Statistically speaking, the CPC has significantly less support than in 2011.

    3. BCVofR it's not the first time you are trying to compare the results of the Cons. VS 2011. However, they is nothing similar. In 2011, unless exception, Conservative were always way ahead of the Liberals. No one had any doubt they would form the government. Yes they did have bad polls but many of them between 36 and 38 and even a pool at 43 a few months before elections. This time, with all the bad press with Trudeau since a few months… they didn't had one pool over 35. If Conservative remain in power, so far, a Majority government is fairly unlikely. IN 2011, 39% for Cons. was a surprise, however they did had pool around that from time to time. Since 2012, Cons. had no pool over 35…. Basically, different story! They actually need something more if they want to re-live 2011 election and get a majority. But it will be hard since many peoples lost faith in them so they need a way to get that trust back.

    4. Using your numbers Ekos latest polls has a mean of 32.58 with a std deviation of 1.31972

      the 95% range for this data would be 32.58 +/- 2 std deviations or 29.94 to 35.21.

      Hey the mean of 34.36 plus all the EKOS polls taken just pre election 2011 fall into this range...19 time out of 20.

      Fairly certain that the Conservative support has not statistically changed from last election until now according to EKOS.

    5. Arguing numbers with BCVoR is a questionable use of time. BCVoR's conclusions come before the numbers, so the only salient point is whether the rationalizations are logically coherent. Sometimes they are.

    6. BCVoR,

      Unfortunately, that is not the right test, but for future reference it's 1.96, not 2 Neither the margins of error nor the of the 2011 number were taken into consideration.

      To go beyond your misuse of statistics, just to put a final nail in this mordum argument, if you look at the 4 current polls in the last month (your words, but not your actions) and the 4 last polls before the 2011 election (for equal sample sizes), you get 32.125 support now with 0.75 and 34.475 with a 0.34 Using your personal choice, you get 33.625, which is remains less than two below the 2011 polling average.

    7. @2015 Federal

      You don't seem to understand what I am saying.

      1) EKOS polls dominate Eric's analysis. It accounts for well over 50% of the input to the Eric's seat generator and analysis.

      Can you agree on this fact?

      2) EKOS did polling right before the 2011 election.
      These polls showed the Cons with about 34% in the polls closest to the actual election.

      Can you agree on this fact?

      3) Right now in EKOS latest polls EKOS has the Cons with an average of 32.7 and a range of 31-35.

      Can you agree on this fact?

      4) Here is the big one. The difference between 34% and 32.7% is well within the margin of error.

      You have seen Eric (and Frank Graves and any pollster whose analysis you read) who you respect call this difference a statistical tie.

      Do you agree with this fact?

      5) Finally, and here is the big one, is that according to the EKOS polling the level of Cons support is at the same level that EKOS found it right before the last election.


      I did not say the polls were the same... The Liberals are obviously polling much better and the NDP much worse.....

      but according to EKOS the Cons have the same support as they did the last week before the may 2 , 2011 election.

    8. +1 to Thingamabob...this is about the only thing I do dislike about reading comments on the site, partisans positions staked out, and then searching for rationalizations to support predetermined conclusions...In which no evidence counter to that predetermined conclusion will actually do any good...

    9. I'm not sure why so much effort/convincing is being put into using past results to pre-determine future events in the same of supporting one cause over another. If you want to look to past event, when Kim Campbell took over the PCs they were polling pretty close to the then-Chretien opposition Libs until just before the election campaign got underway. I think we all know how that turned out. Even if the Reform/PC split of the day didn't exist, the right-of-centre parties still would have been trounced either way as in very few ridings that the Liberals won, was the combined Reform + PC tally > the Libs' victory. So based on all this historical data, I'm going to go out on a limb (while rolling my eyes) and say one of the NDP/Liberals/Conservatives will form the next heard it hear first!

    10. @BCVoR,

      I would dispute points #2 through #4, which leads to point #5.

      #2 - EKOS polling before 2011 was 34.6, 34.5, 34.8, 34.0, and 33.9. The average of the latest 3 was 34.6%, the latest 4 was 34.5%, and the latest 5 was 34.4%. Take your pick, but use a real number.

      #3 - EKOS polling since February is 32.9, 32.2, 31.1, 32.3, 32.0, and 35.0. The average of the latest 3, 4 or 5 is 32.1% and of all 6 is 32.6%. Again, you can selectively pick the highest number, but don't inflate it beyond that.

      #4 - The difference between 32.1% (now) and 34.5% (2011) is 2.4%, rather than the 1.3% you claim. And beyond current stated MOE of 1.8%.

    11. BCVR said "....but according to EKOS the Cons have the same support as they did the last week before the may 2 , 2011 election."

      Except the Cons were polling around 38% and the Libs were polling around 20% by Namos just before the 2011 election. Right now Namos has Libs 1 point in the lead. Nanos was very accurate in the 2011 election.

      Not looking so good for the Cons in 2015.

    12. +1 to Thingamabob

      The Cons are polling much worse than 2011 as far as Nanos is concerned. 7 points as a matter of fact. Nanos results were off by 0.9%. Also according gto Nanos,.Libs are polling 14% higher than just before the 2011 election.

      Not looking good for the CPC.

    13. When comparing two polls, you need to account for the MOE of both.

      If one poll says 40 plus or minus 3, and the other says 36 plus or minus 2, they overlap.

  3. Interesting addition to the CBC piece. Thanks and on to a Liberal victory !

  4. I am curious about why you have made specific divisions in Ontario.
    Southern Durham - York: What counts as Southern Durham, and why connect it to York region?
    Barmpton - Mississauga - Oakville vs. Burlington - Hamilton - Niagara. Where does Milton sit in this group? Why not Peel Region vs. Halton Region vs. Hamilton - Niagara. Does it make any difference?

    1. As mentioned in the article, I used the definitions from the Wikipedia article for the 2015 election.

  5. Eric An off topic question:

    Is there significant support for Coule pas chez nous?

    Student leader Gabriel Nadeau-Dubois donated $25K from his book on his last action.... not sales of his book but a literary prize to fund this action and then seems to have raised another $400K online.

    You personally have an idea of how hard it is to raise money online.

    This should be raising all sorts of alarms as the BQ got between 400k-900k in tax deducible contribution in a full year of fund raising when they were the largest federal political organization in Quebec.

    Any idea who is providing the money? Unions? Georges Soros? Professors? The Koch Brothers :) ?

    They are using the money to have a anti-pipeline protest in Quebec on April 11. They are trying to get Mulcair and Trudeau to stop the pipeline.

    This could be a major event.

    It is political suicide (ask former BC NDP leader and almost premier Adrian Dix) to come out fully against the pipeline.

    In today's climate of terrorist threats and much more open security concerns there will be more of a reaction and less public acceptance and tolerance to any kind of Black Bloc violence.

    Any thoughts as to the political impact and what to expect?

    The student won last time but in 2014 PQ who caved in to the students was booted solidly out of office with historic low support and an almost Conservative (by Quebec standards) Colliard and significant support to free enterprise CAQ were put in charge.

    1. I think anyone who wants to win seats in Alberta will need to be semi-pro pipeline thus the Liberals and NDP will waffle on it. I suspect many BC seats will demand local MP's to be against it - a big plus for the Green Party and a negative for the NDP/Liberals while the Conservative voters will probably not care - seems nothing will shake that 30% core support for them. In Quebec being anti-pipeline should be a big seller. I know up here in Northern Ontario it will be (lots of risk, little reward from that pipeline plus maybe higher natural gas prices as you remove a pipeline for that and change it to oil and that'll cut supply). Expect the pipeline issue to grow during the campaign. This election how each party reacts during the campaign to core issues could change everything.

    2. The recent epidemic of oil-hauling train derailments should increase pipeline support. Pipelines are a vastly safer way to transport oil per ton-mile.

      The oil is going to get produced, moved, sold, and burned. The question is, do you want it to blow up the occasional town along the way, or would you rather it not do that?

  6. I think you are out to lunch regarding the Liberal numbers in BC. Way out to lunch.

    1. Wow, geez. I think I have to re-evaluate everything. Thanks.

    2. They are a little low but not that far off.

    3. Yes clearly Eric you need to ignore the polling and just go with your gut. Because you know picking numbers out of going to work better. Seriously though polling numbers can be wrong the output is only as good as the input...but why would anyone blame Eric for that. Add random pulled out of the air correction factor based on gut feeling is not an analysis I would bother reading.

  7. With all this EKOS talk, I opened up their full report and interestingly, they break down support by current vote intention for leader approval/disapproval. It's quite interesting when you take out the own-party support. Trudeau is 35.2%, 51.8%, 12.9% for approval, disapproval, unknown (MOE 3.5%). Mulclair is 42.5%, 41.7%, 15.8% (MOE 3.4%) and Harper is 14.6%, 74.4%, 11.0% (MOE 3.7%).

    Assuming leadership approval correlates to undecides voting and/or changing voting preference, does this mean the CPC are much more limited in upswing potential? Statistically speaking Greens, Liberals, and Undecides like Trudeau and Conservatives dislike him. Greens, Liberals, and NDP like Mulclair and Conservatives dislike him. Conservatives like Harper and every other group dislikes him.

  8. Aside from the normal difficulties of small sample size at this level, I'm wondering how you feel about the reliability of polls overall given:

    1 - growth of mobile-only households that aren't contacted by polling firms (since this is a non-random characteristic, there is reason to believe that the voting intentions of cellphone-only households could be materially different from landline households.)
    2 - lack of updates to weighting information thanks to the elimination of the long-form census. (Harder to correct for non-random non-responses.)

    1. Right off the bat, these numbers above are based on the riding projections, so they aren't polls where sample size comes into play.

      Polling companies, for the most part, are calling cell phones so this shouldn't be an issue. There may be less of a response rate among cell phones, though, which could potentially be a problem.

      As for your second question, it is my understanding that the mandatory questions on the census would still provide enough information for pollsters to weight their responses.

  9. Hi Eric,

    With the strong liberal numbers in Nova Scotia is Peter Mackay at risk of losing his seat. Aside from 2011 results which were devastating for liberals across the country, Peter Mackay has only got victories between 40-46% the greens put up a good showing when the liberals worked out an agreement with them.
    Now with a strong liberal leader the province has warmed too, I am not here arguing Justin Trudeau's merits but it can not be argued he is very popular. Combined with this popularity, the strong liberal poll numbers province wide, the current defeat of the NDP government and a poor showing so far by Mr. Muclair, is this a seat the liberals could put into play and should Peter Mackay be nervous?


    1. For an answer to this and your Calgary question, you can check out the riding projections on this site. Click on the chart at the top of the page and scroll down.

  10. Are there seats in play for the liberals in Calgary?

    1. provides a riding-by-riding breakdown. There are five.

    2. Looking at these regional breakdowns I think it fairly unlikely 5 ridings are really "in play". With a 13% lead over the Grits the Tories are odds on favour to sweep the region yet again. The one exception may be Calgary-Skyview in the City's Northeast where a popular M.L.A. is running for the Liberals but, he is up against a little giant in Deepak Ohbrai who is well known and well liked among the South Asian community who make up a large cohort of the riding's population.

      Over all these "local" numbers are good news for the Tories and bad news for everyone else with the exception of the Greens on Vancouver Island and BC more generally. From my vantage point it sure looks like Harper is on pace to win a second majority government.

  11. I decided to try the same exercise to compare our results. So, after quite a bit of time, I have those results! Generally speaking, the differences are not major, but there are a few. Sorry for the formatting, it’s difficult to do in a comment.
    In the Atlantic, I have:
    PEI 63,6 26,3 3,6 5,1
    NFL&L 62,0 17,1 15,3 4,2
    NS 53,3 22,8 15,0 6,8
    NB 42,8 29,9 17,2 5,7
    In Québec, I have: BQ
    Eastern 26,9 31,5 14,2 5,3 22,4
    Côte-Nord – Sag. 19,4 32,0 22,6 5,2 20,8
    Québec City 22,9 28,8 27,0 5,0 16,0
    Central 23,8 19,5 29,0 5,9 22,0
    Eastern Townships 24,3 26,8 25,1 5,1 18,6
    Montérégie 27,4 13,4 30,2 5,2 22,3
    Eastern Montréal 29,7 10,2 29,9 4,7 24,7
    Western Montréal 45,1 24,4 15,1 5,9 9,4
    N. Montréal – Laval 43,3 13,2 20,6 5,0 17,3
    Laurent – Out – North 25,7 15,1 36,3 5,2 18,3
    In Ontario, I have:
    Ottawa 45,3 30,7 14,9 6,5
    Eastern 34,4 45,2 11,9 6,2
    Central 26,0 48,0 11,4 8,6
    S. Durham and York 40,7 41,3 9,9 5,3
    Suburban Toronto 46,4 26,8 19,6 5,2
    Central Toronto 45,5 12,8 32,9 5,8
    Bram – Mississ – Oak44,6 36,9 10,2 5,7
    Ham – Burling – Nia 31,0 38,0 21,6 6,2
    Midwestern 35,1 42,7 13,0 6,7
    Southwestern 30,5 35,8 25,5 5,6
    Northern 30,1 26,6 33,9 6,8
    In the Prairies, I have:
    Rural Manitoba 34,1 41,6 13,5 8,0
    Winnipeg 44,4 28,0 17,6 8,0
    Southern Sask. 28,7 41,9 20,0 7,1
    Northern Sask. 22,0 39,5 29,0 6,9
    Regina & Saskatoon 28,2 33,8 28,1 7,2
    Rest of Sask. 22,5 47,6 20,8 6,9
    In Alberta, I have:
    Rural Alberta 25,3 53,8 10,8 5,8
    Edmonton and env. 22,1 47,9 21,9 4,6
    Calgary 30,6 46,9 9,8 9,7
    In British Columbia, I have:
    B.C. Interior 24,2 36,9 25,9 9,9
    Fraser Val–S-Low Mainl 33,1 38,4 17,8 7,6
    Van. – N-Low Mainl 37,5 24,0 26,4 9,2
    Vancouver Island 22,5 22,1 31,0 21,6

  12. @Thierry, any chance you can do an Ottawa-Gatineau/Outaouais breakdown as a true reflection of the NCR? Les Laurenties and Nord du Québec have virtually nothing in common with the Outaouais.

    1. Ask, and you shall receive! No kidding, I, too, find that "region" a bit strange, as Laurentides would be better off with Lanaudières, Outaouais on its own (speaking of Québec, as a larger part of Canada, it does fit in nicely with Ottawa) and the North... with Laurentides and Lanaudières, I guess. Or on its own, but it has few seats. Anyway, here goes:

      Gatineau/Outaouais 28,6 17,2 38,1 4,4 11,6
      Ottawa/Gat/Out 37,9 18,6 30,6 5,7 5,8
      Ott+suburbs/Gat/Out 39,7 26,2 22,6 5,8 3,9

  13. I find the Newfoundland stat quite interesting. The only reason Newfoundland is doing well and is a have Province is due to Hibernia - which was put into place by the Conservative Mulroney. I guess its a matter of what have you done for me lately.

    1. So if Mulroney didn't buy the 8.5% stake in the development project three months before leaving office, then the other 91.5% would have just thrown in the towel and never developed it after oil took off this millennium? They only became a Have province in 2008, so it could have been either a Liberal or Harper claiming to be responsible if it hadn't been for Mulroney.

  14. I think the projection for Cumberland-Colchester will have to be updated from 2% Other vote, as Bill Casey has announced he is coming out of retirement to run as Independent.

    1. Casey is running for the Liberals.

  15. Right you are - the article I read on this was rather vague.

  16. On Atlantic Canada ridings:

    1. BCVoR is incoherent re Avalon. It requires no "due process" to throw a candidate nomination away, nor to ditch someone from caucus. Under the law it's entirely a party leader decision. It's bad law, in a bad electoral system, that has bad results. But there's lots of precedent for ditching candidates based on sniff tests, however unfair or vague, a leader of any Canadian federal party is a dictator of who runs.

    2. Re: Cumberland-Colchester this race will be decided on candidate factors, period. Bill Casey is running Liberal and this is a bellwether case for deciding if the future of NS progressive Conservatism is basically to merge with the Liberals as in BC and Quebec. If Casey wins, there will be more provincial PCs joining the Liberals, as the NS PCs will likely never hold office again. That because the federal Cons have made it near impossible to recruit good provincial candidates.

    3. Central Nova is also an interesting personality race with MacKay up against an anti-Harper veteran, in fact one of the ones Rob Anders gravely insulted.


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.