Thursday, April 12, 2012

Neck-and-neck federal race

Two federal polls were released last week showing a very tight race between the governing Conservatives and the opposition New Democrats. While the two surveys differ on what kind of momentum the Tories have, both show that the NDP is making significant gains, primarily at the expense of the Liberal Party.
The Léger poll is the most recent, and is also the most striking. It places the New Democrats ahead of the Conservatives with 33% to 32%, a gain of seven points for the NDP since Léger's last national survey of Feb. 28-Mar. 5. The Tories, meanwhile, are down two points.

The Liberals slip five points to 19%, back to where they were in May 2011. The Greens are up one to 8% while the Bloc Québécois is down one to 7%.

The New Democrats lead in Atlantic Canada with 49% (+18), Quebec with 47% (+20), and British Columbia with 34% (unchanged). They are running second in the Prairies and Ontario with 26% support in each (a loss of seven points in the Prairies and a gain of three in Ontario).

The Conservatives lead in Alberta with 61% (+2), the Prairies with 49% (+8), and Ontario with 39% (unchanged). They place second in British Columbia with 30%, a drop of six points since late February/early March.

The Liberals are second in Atlantic Canada with 20% (-17) and in Alberta with 17% (+5), while the Bloc Québécois is second in Quebec with 29% support, a loss of two points.

Atlantic Canada has huge variations in this survey, so the region's results can probably be discounted. The big NDP leap in Quebec is not so unusual, however, as other polls have indicated that the NDP has made a major rebound in Quebec with Thomas Mulcair as leader. That the NDP has supplanted the Liberals as runner-up in Ontario is good news for the party, as they need to make more gains there. Overall, however, the Conservatives are in a strong position thanks to their big leads in Alberta and Ontario.
Those big leads were echoed in Harris-Decima's older poll, taken between Mar. 22 and Apr. 2. Part of the survey was conducted before the Mar. 24 NDP leadership convention, but probably not enough to have skewed the results very much.

Harris-Decima keeps the Tories in the lead with 34%, a gain of three points since their previous poll of Mar. 8-19. The New Democrats are up four points to 32%, while the Liberals are down five points to 19% support.

The Greens are up one to 8% while the Bloc is down two to 6%.

There are fewer unusual results in this Harris-Decima poll, at least in terms of the variations since their last survey. Both Harris-Decima and Léger agree on the situation in Ontario as well as in Quebec, while giving the NDP the lead in B.C. and on the East Coast.

But the Conservatives lead in Alberta with 54% (-4), the Prairies with 45% (-2), and Ontario with 41% (+8). They are trailing the NDP in British Columbia (30%, -3) and Atlantic Canada (also 30%, +4).

The New Democrats lead in British Columbia with 44% (+9), Quebec with 39% (+13), and Atlantic Canada with 36% (+2). They place second to the Conservatives in the Prairies (34%, +3), Ontario (26%, -3), and Alberta (19%, +6).

The Liberals are only tied for second in Atlantic Canada with 30%, a loss of three points, while the Bloc is down 10 points in Quebec to 24%.

Both of these polls would result in a similar situation in the House of Commons: the Conservatives with a plurality of seats but the NDP and Liberals able to combine for a majority.

With Harris-Decima's narrow Conservative lead, the Tories take 147 seats to 119 for the NDP, 36 for the Liberals, five for the Bloc, and one for the Greens.

With Léger's narrow NDP lead, the Conservatives take 134 seats, the NDP 132, the Liberals 34, the Bloc seven, and the Greens one.

Regionally (Léger first, Harris-Decima second), the Conservatives win 14/12 seats in British Columbia, 27/27 in Alberta, 23/19 in the Prairies, 67/73 in Ontario, 1/5 in Quebec, 1/10 in Atlantic Canada, and one in the north.

The New Democrats win 15/21 seats in British Columbia, 1/1 in Alberta, 2/7 in the Prairies, 23/22 in Ontario, 66/58 in Quebec, 24/9 in Atlantic Canada, and one in the north.

The Liberals win 6/2 seats in British Columbia, 3/2 in the Praires, 16/11 in Ontario, 1/7 in Quebec, 7/13 in Atlantic Canada, and one in the north.

If we take the best and worst regional results for each party, we get a range of between 128 and 153 seats for the Conservatives on these numbers, with the New Democrats sitting between 108 and 143 seats. The Liberal range is between 24 and 46 seats - even a best-case-scenario results in only a minor gain of seats for the Liberals.

The New Democrats are only slightly above the 31% the party achieved under Jack Layton in May 2011, but the extra point or two makes all the difference, particularly when the Tories are down six to nine points. The NDP is in a strong position on the two coasts and looks capable of keeping (or even increasing) their representation in Quebec. But the Conservatives still have the advantage thanks to their wide lead in Ontario and the clump of seats they hold claim to between B.C. and Manitoba. Until the NDP can start to whittle away the Tory holdings in the West and in Ontario, they will have no hope of toppling the Conservatives without the help of the Liberal Party.

90 comments:

  1. I'm not sure if this NDP uptick is such a surpirse. It'll be interesting to see if they can hold this level, or build, until the next election. As bad as it looks for the Liberals, there may be a ray of sunlight for them in this. First, despite a terrific NDP surge, they are at about the same level as the election , not worse. And they should also get a bit of a bounce after their leadership decision.

    ReplyDelete
    Replies
    1. A Rae of sunshine...

      Delete
    2. Yes, Rae has always been an enormously popular figure...

      Delete
    3. More popular than you.

      Delete
    4. but I'm not planning a run at the PMO... there are things that Rae and the Liberal party ought to take into account, one being his foul reputation in Ontario...

      Delete
  2. I was wondering how you calculate the seat projections without knowing the candidates. I read your post on how you calculate it during an election, but between elections?

    ReplyDelete
    Replies
    1. No changes, just that the incumbents are assumed to all be running again.

      Delete
  3. It is becoming increasingly clear that the NDP has replaced the Liberals as the potential alternative for the Conservatives. This is at least true for 3 of the 6 regions of Canada. A few more points in Ontario and the NDP will complete the circle...

    ReplyDelete
    Replies
    1. Yes I agree on that and suggest that in fact a de facto merger is going on??

      Without anything formal Liberals are slowly moving to the NDP. Now is that because of despair about their own party or because the NDP is deliberately moving towards the centre??

      Good questions.

      Delete
    2. There is no "de facto" merger going on, what are you talking about?

      The more likely case is that more Conservatives are supporting the NDP than before, and maybe there's some post-2011 Liberals in the mix, but polling has shown the LPC pretty steady between 18-25% in the last few months.

      I don't get people - the NDP must win Conservative votes, not Liberal votes, in order to win. Con-ser-va-tive votes. That's where pay dirt is.

      Delete
    3. The NDP will NEVER win Reform votes Volkov.

      They can win Liberal votes by moving towards the centre which is exactly the trip Layton started them on and Mulcair continues.

      Much as I'm a Liberal I'll vote for any party that is capable of taking the CPC out of Govt. Right now that's one choice !!

      Indeed the NDP may be picking up Progressive Conservative voters but much less than they are getting from the Liberals. Just look at the Quebec numbers to see that !

      Delete
    4. It is clear that when the NDP rises, so do the Conservatives. Harris-Decima only had the Conservatives at 31% in their last poll so it is clear that Liberal support has bled equally to the NDP as it has to the Conservatives. A merger of the Liberals and NDP will only ensure additional Conservative majorities.

      Delete
    5. The NDP will never win Reform votes? I dunno about that. A lot of Reform votes came from the NDP to begin with.

      Delete
    6. Peter I wouldn't be so quick to say the NDP can never win reform voters, especially on the prairies. Reform was created out of disgust for scandals, deficits and people feeling they are being taken advantage of. I'm hearing rumblings of this about the current path of the federal conservatives. The NDP could definitely make gains and I know lots of people who jump between NDP and Cons, skipping the Libs altogether.

      Delete
    7. Not all Conservative voters are past Reform voters...

      Besides, as it was pointed out, Reform took a lot of NDP voters in 1993. Easily go the other way, especially out west.

      As well, Anon 12:44 - correlation does not equal causation, yes?

      Delete
    8. "The NDP will NEVER win Reform votes Volkov."

      Why do you say that? Who do you think the Reform party voters used to vote for before the Reform party? Take the fromer NDP homeland, in Saskatchewan. In 1988, the NDP took 10 of the 14 ridings there (and 44% of the vote) compared to 4 seats for the PCs (with 36% of the vote). 20 years later, in 2008, the Tories took 13 seats with 54% of the vote, compared to no seats for the NDP with only 26% of the vote (the Liberal vote has stayed fairly constant in the mid-teens).

      So there's a province that 24 years ago voted solidly NDP and which now is almost as loyally conservative as Alberta. Those people who are voting for the Tories in Saskathchewan, 24 years ago they voted NDP.

      I'll give you another, local, example. For 20+ years Ed Broadbent cleaned up in Oshawa. That riding was as safe an NDP riding as you could find in the country. However, since the mid-1990's provincially, and the merger of the right federally, it has gone solidly PC/Conservative, with NDP finishing a distant 2nd (at best). Now, that still is a solid blue collar union riding, and yet it's one of the safest Tory ridings in Ontario. And I'll tell you why, it's because the blue collar men who used to vote for Ed Broadbent back in the day, now vote for the Tories. There's a host of reasons for that, provincially, the NDP killed themselves (and probably fatally wounded their federal cousins) with affirmative action in the 1990's (no surprise that the Mike Harris Tories first took Oshawa in 1995). Federally, the Tories have done well their in part due to their opposition to the gun registry (apart from being a strong union town, Oshawa also has one of the higest levels of (legal) gun ownership in southern Ontario - drive through it during deer season, you'll see why).

      The point is that you have to take off your ideological blinders. Both the NDP and the Reform party have common roots in grassroots opposition to what were seen as elite drive policies (typically imposed by the Central Canada "business" party, namely the Liberals). They both appeal, or in the case of the NDP, appealed, to an "outsider" mentality. Part of the success of the Conservatives (and their Ontario cousins) is that they've been responsive to the grassroots issues that the NDP used to feed off of, to attract voters who, 25 years ago, would have been sure-fire NDP voters, but who now are rock solid Tory voters.

      If the NDP wants to form a government, they'll have to figure out a way to get those voters back.

      Delete
  4. I wouldn't be surprised if this NDP rebound was mainly due to the fact that now they are the only opposition party with a permanent leader - the obvious attraction point for non-partisan anti-Harper voters. Things may change a lot once Liberals choose their new leader, especially if it will be competitive (not simply crowning Bob Rae).

    ReplyDelete
    Replies
    1. I agree with your last point, the Liberals need a new face. While I recognise that Rae is considered a good politician, I don't think he's the key to a Liberal come back.

      Delete
    2. Both you guys keep telling yourselves that. Sorry son, you're done.

      Arthur Cramer

      Delete
  5. +1000 on ignoring the Atlantic Canadian results in Léger. The NDP hitting 49% would require wiping out the Green vote, not quintupling it.

    ReplyDelete
    Replies
    1. I'm not sure about this. At least in Nova Scotia, the NDP government has recently introduced a good-news budget showing that the province is on track to return to a surplus next year, combined with (extremely modest) tax cuts. Demonstrating what is widely regarded as prudent fiscal management may be drawing in supporters of the other main parties, remembering that the right wing in Nova Scotia is much more Red Tory than Reform. That phenomenon is possibly playing out federally as well as provincially.

      Delete
    2. Undoubtedly the NDP is doing very well in the region - every poll shows that. It's just the huge NDP and Green numbers that make the results more like an impressionist painting. The sample size, after all, was 73 people. The MOE is almost 12 points on a sample of that size.

      Delete
    3. I'm an NDP member from NS, and am well aware of the climate here. If the numbers were NS-only, I'd be more forgiving. However, NFLD, NB, and *especially* PEI have completely different climates.

      Delete
    4. Perhaps, but pretty dramatic things are happening outside of NS as well:

      http://www.nbndp.ca/node/636

      Delete
    5. Saint John was the NDP's second strongest federal riding in NB (Behind Acadie-Bathurst, where Yvon Godin has won since 1997) in 2011. I WOULD like it if Elsie Wayne's old stomping ground turned orange, mind.

      Delete
  6. Thank you, Eric for posting about the Leger Marketing results as well as the Harris-Decima results for the federal level. I know I requested it a few days back, I was just anxious to hear your analysis on it. Greatly appreciated!

    ReplyDelete
  7. The only thread that keeps the Liberals alive at all is that there are still a few "zombies staggering around" and supporting the Liberals in Ontario. The NDP has three years to go for the kill and lure most of those people into the fold. There will be intense pressure in 2015 on what's left of the Liberal vote in Ontario to go NDP to stop Harper

    ReplyDelete
    Replies
    1. And that three years is crucial to getting rid of this appalling CPC Govt.

      I'd suggest everybody pitch in and work for this !

      Delete
  8. The conservative party scored 35% on april 29th, 3 days before the May 2 election.

    Today they scored 34%. Up 3% from the last poll.

    In the intervening 5 polls... they scored 34% once. And between 30 and 34 in the others.

    On top of that the vote split would give the tories 147 seats. Short of a majority, but definitely close enough that a few votes could flip the last 7-8 seats.

    So HD would seem to think that the CPC is on an upswing, near the highest levels that HD has polled them at in the last 10-11 months.


    What would these 2 polls look like under the new electoral map with 30 more seats??

    ReplyDelete
    Replies
    1. What is with all these people popping up lately spouting a variation on this same talking point (i.e. that you have to basically add a couple percentage points onto every pollster's CPC result)?

      Delete
    2. Con spin ... say it often enough, it becomes the truth.

      Delete
    3. Actually, adding two points would not be enough for the Tory spinmeisters showing up here lately. No poll can refute the irrefutable fact that the Conservatives must be in the lead!

      But wait, BREAKING NEWS, a Postmedia exclusive: the Right's favourite pollster, Ipsos Reid, which polled the CPC at 38 and rising before the last election, has just released a new poll confirming the polling trends analyzed above (albeit with laughable spin on Harper's personal popularity -- down 6 points but "holding strong"):

      http://news.nationalpost.com/2012/04/12/conservative-popularity-sinks-but-stephen-harper-approval-holds-steady-poll/

      Too bad there are no regional numbers...

      Delete
    4. I don't remember adding anything to anything...

      I simply pointed out that contrary to the Leger (and now Ipsos this morning) the conservative numbers in the HD polling are rarely higher.

      The anti_tory propaganda chip in your head is malfunctioning if you look at that post and see adding. Or you are displaying the same math skills that slow you to add the 4 opposition parties votes together to get a majority.

      I don't deny that the tories are down in Leger and Ipsos, I just think it is interesting that given they are down in those 2 polls, why the up in HD polling?

      Delete
    5. I think you raise a good question! As for the comments above, they were not really targeted at your post as much as at "1 in 20" and a few others posting here lately who expressly DO NOT recognize any drop in Tory support. You were collateral damage. :(

      Delete
    6. It happens :)

      Delete
  9. Ipsos Reid has NDP and CPC tied also. There s a definite trending down for the CPC and up for the NDP. Its not an outlier. "Ipsos Reid senior vice-president John Wright attributed the Conservatives’ declining fortunes to weeks of enduring controversy, including the robocall scandal, an uninspiring budget and last week’s auditor general’s report on the troubled F-35 stealth-fighter program."

    ReplyDelete
  10. Now Ipsos comes out with Fed poll confirming the trend: Cons 34, NDP 33, Libs 21

    ReplyDelete
  11. And now with the Slash and Burn killing of jobs and services the Tories can expect to slip further down the numbers.

    ReplyDelete
  12. 1 in 20 outlier13 April, 2012 10:54

    Right before the election versus the latest polls:

    NDP support

    Leger: current 33 - April 28 2011 - 31
    Harris Decima: current 32 May 1, 2011 - 30
    IpsoReid : Current 33 april 28,2011 -33

    These polls are saying that the NDP are almost exactly where these pollsters had there support going into the last election.

    That is a spectacular concept for the NDP as the 2011 is by far the highest water mark (at least 50% betetr than previous). No one would expect that they could hang on to this level of support seeing the weak MPs that they elected and the loss of their once in a lifetime leader Mr. Layton.

    Right now with the same support as they had going into the last election they would get roughly the same number of MPs elected (102) as they did last election.

    The F-35 and auditor's report and leadership bump effectively cancelled out the loss of Layton and the weak NDP caucus.



    These polls won't result in them getting 119-132 seats andy more than they did when Eric predicted the same going into the last election.

    PS....... just way too much use of anonymous ... how hard is to provide a name--- poster 1 for instance.

    ReplyDelete
    Replies
    1. I never projected the NDP to win between 119 and 132 seats during the 2011 campaign.

      Delete
    2. When the NDP wins a majority, are you still going to be blabbing on here about how it's impossible for it to happen?

      You do realize people like you said the same thing about the Labour Party in the UK in the early 1900s, right?

      You need to wake up to the new reality that is Canadian politics.

      Delete
    3. Right on Adam!

      Art Cramer

      Delete
    4. Nice irony in the name 1 in 20... this must be the fourth poll in a row now that you're trying to argue has some problem.

      If the NDP couldn't hold their support because of the loss of Jack Layton, they wouldn't have it now. If their caucus was so weak that people would lose faith, it would have happened by now.

      The reality is that a full year after the election, the numbers are holding firm - not really evidence of NDP weakness, no matter how many times you try to spin it that way.

      - Ed T.

      Delete
    5. Yes Ed T. with Environics, Robins, Forum, Harris-Decima, Forum again, Leger, and now Ipsos-Reid having reported the same essential tie in support, then I guess we are up to "8 in 20"... and counting.

      Delete
    6. Robbins... not a real pollster. (atleast here in Canada, they release almost no polls)

      Leger - Tory 32(-2), NDP 33(+7)

      IpsosReid Tory 34(-3), NDP 33 (+4)

      Forum - Tory 35(-2), NDP 35(+7)

      Environics - Tory 30, NDP 30. Difficult to trend since they haven't released a poll since 2 weeks before election.

      HarrisDecima - Tory 34(+2) NDP 32(+3) (and the highest the tories have polled in HD since election)


      I don't know Jeff. I see the NDP strength. But most of it isn't at the expense of the Conservatives (who are showing a very slight weakness to be fair).

      A bump for the Muclair NDP to be sure... In what has been a very bad media month for the tories.

      It will be interesting to see if the NDP can hold their lead over the next 2 opposition leadership elections, and the 3 years until the next election.


      As far as the tories go, I am not going to be concerened about tory movement in the margin of error (down in 3 polls, up in 1) until a year out when the other leadership campaigns happen.

      I think its the liberals that should be cringing at the NDP up where they are. They are still pulling as much or more liberal support than tory. And the liberals are still/again languishing at the 20% mark or so in most polls. They need to make some noise before the NDP replace them on a permanent basis.

      Delete
    7. LPC no doubt has more to worry about in these polling developments, but the CPC is down from even its election level support which barely squeaked out a majority, and Harper's leadership numbers are dropping like a rock (leading indicator?).

      The regional shifts are even more important. NDP rise in BC is not at all coming from the CPC? How about Ontario? And what do these Tory declines, especially in Ontario, suggest for the stability of the newfound Conservative majority voting coalition? Harper has written extensively about his aspirations for to unify prairie reformers, Ontario progressive conservatives and the Quebec right. That coalition, held together by discount duct tape in May 2011, has not cemented after the election but rather is coming apart fast. And Harper knows it. Thus the anti-Rae ads (one reason anyway) and the 'moderate' budget *relative* to the deep, long-term transformation of government he would like to achieve with that elusive stable right-wing majority voting coalition.

      Delete
    8. All of which has been repeated ad nauseum since Harper won a minority in 2006.

      And yet, all that falling apart has resulted in a majority government.

      As another commenter around here is fond of saying. "The Harper slide is on!!!"

      Delete
  13. 1 in 20 outlier13 April, 2012 11:57

    TS.Apr 12, 2012 08:33 PM

    What is with all these people popping up lately spouting a variation on this same talking point (i.e. that you have to basically add a couple percentage points onto every pollster's CPC result)?


    No TS it is basically analysis of polls 101.

    If you don't want to have a critical analysis of polls beyond the MSM then why do you come to a site devoted to poll analysis?

    For whatever reasons it seems that the bulk of the polls have been underestimating the CPC results at the election.... by about 4% points... well outside the standard deviations that statistical science allows for in a properly constructed sampling.



    Here is my straw man argument.

    A group of people are polled and vote that water freezes at 4 degrees Celsius.

    Prediction are made as to when water freezes .

    It gets colder and colder and it actually gets cold enough for water to freeze and low and behold it freezes at 0.

    The polls were wrong.

    It warms up the water melts and the same polling methodoloy is used and once again it comes up with the results showing that water will freeze at 4 degrees.


    Do we look at the polls and say they must be right this time. Or do we say that last time you said water would freeze at 4 degrees it actually froze at 0. I think it will once again freeze at 0.

    ReplyDelete
    Replies
    1. One case of the polls under-estimating the support of Conservatives by about two or three points is not the same as "water freezes at 0 degrees celsius".

      Delete
    2. 1 in 20 - I agree that Canadian pollsters need to do a better job at using likely voter models, but just because that skewed things against the Conservatives in 2008 and 2011 doesn't meant it will skew against the Conservatives in 2015. After all, in 2006 the polls underestimate Liberal support, and were bang on for the Conservatives.

      I do think it's fair to eye these polls with a bit of suspicion, and it's give a bit more weight to pollsters with better track records (Nanos and Angus Reid for example). I don't think it's wise to assume 2015 will be the same as 2008 or 2011.

      What you're arguing for is like looking out the window and seeing storm clouds and then saying "I don't need an umbrella! It was sunny yesterday and the day before, so it can't possibly rain today." A wiser man would look at the storm clouds and bring an umbrella.

      Delete
    3. 1 in 20 outlier13 April, 2012 13:01

      I totally agree .. one case of the polls can wrong .... this should happen 1 times out of 20..... Not 19 times out of 20 and not all on the same side of what turns out to be the correct answer.

      Delete
    4. 1 in 20 outlier13 April, 2012 13:13

      I know that his will not be accepted as valid for a lot of people here since it was put out by fox news and shows Romney ahead of Obama.

      The thing that I find inteseting is the established bias of the polling firms


      "The FOX News poll is based on land-line and cell phone interviews with 910 randomly chosen registered voters nationwide and is conducted under the joint direction of Anderson Robbins Research (D) and Shaw & Company Research (R) from April 9 to April 11. For the total sample, it has a margin of sampling error of plus or minus three percentage points."

      link:

      http://www.myfoxphoenix.com/dpps/news/fox-poll-romney-edges-obamas-approval-rating-dpgonc-km-20120412_19079415

      ---

      In an attempt counteract unavoidable bias the poll and methodology is constructed by a Republican and a Democratic polling firms. These polling firms do not analysis the results..... that is done by an obviously bias media.... It seems much more open and a better way to do things....

      In Canada I would like Ekos, Nanos, HD to construct a poll in conjunction with COMPAS ....and let CBC or CTV or the G&M or the Sun or the Star do the analysis.

      Delete
    5. Those US firms are, as you point out, directly affiliated with political parties. This is not the case in Canada, those firms you mentioned are not affiliated with any political party.

      You write as if media in Canada reports the opinion of pollsters and that's it. I don't do that, and a lot of columnists don't do that. Pollsters usually include a preamble to their numbers as analysis (and why not?), and reporters report that analysis along with the numbers. Reporters report, they do not provide their own opinion.

      Columnists, pundits, and analysts do, however. All the time. This is not unusual.

      Delete
    6. And I know COMPAS is your go-to firm for balancing the results for some reason, but they are not an active polling firm that publicly releases VI results to the media.

      Delete
    7. 1 in 20 Outlier13 April, 2012 13:53

      Just using COMPAS as the balancing firm because in their limited sampling as they were the only firm to over estimate the CPC support.... and they did it by a lot.

      It caused you concern as to whether or not to include them in your estimation model.... if you would have weighted them at 50% and the rest of the polsters at the other 50% then you would have had the seat totals really close.

      If COMPAS came up with a poll that said the CPC had 46% I would assume that reality would be close to 40.

      Delete
    8. While that might be a good guess, it's likely you'd be wrong. I'll say it again (and again and again) - just because COMPAS was +6 on the results DOES NOT MEAN they are always going to be +6. They could have done the same people three times and gotten three different results.

      And in any case, any discrepancies from 2011 are very likely exceptional to the 2011 campaign. Any dynamics in play that skewed the polls will be different in 2015.

      Please, please, please, no more about this.

      Delete
    9. That should have been: "They could have done the same POLL three times and gotten three different results."

      Delete
    10. I'll ask you again 1 in 20 - What about 2006? When polls underestimated the Liberals and were bang on for the Conservatives?

      How do you know that next time it won't be the NDP or Greens or Bloc or Christian Heritage or the Pirate Party that polls underestimate?

      Delete
  14. Brian Nicholson13 April, 2012 14:00

    I have never seen so much smoke about so little fire. The pollsters give a snapshot at the time of the poll contacts. On Election Day, many factors can impact results by a few percentage points. I am not surprised that the Conservatives received a higher percentage of votes than pollsters predicted as in the last election, Conservative voters had a higher enthusiasm for voting than the opposition, led by the Liberals. Thus CON vote goes up marginally at the ballot box and the Lib vote down marginally, creating a four point spread. In Quebec this became evident by the bump to the NDP and the reduction of the BQ. Expecting identical bumps next election is foolhardy. IMHO the push to change government may actually bump up the anti-Con vote so these current polls may be overestimating CON vote. Only time and campaigns will give us the real answer.

    ReplyDelete
  15. Peter this "awful" CPC government is IMO doing a pretty good job of running the country. We need to balance the budget and there are only two ways to do that- higher taxes or spending cuts. Since they took power in 2006 the Harper Conservatives have, until this last budget, always increased spending at rate far exceeding the rate of inflation. They have added over 30,000 civil servants.

    Where they are going to lose support, perhaps permanently is in issues on the fringe. Robo calls is one. The F 35's is another but more importantly is one I read about today. Apparently the esteemed Rob Anders is either medically ill or doesn't get enough sleep. There is a photo out today of him apparently asleep in HOC. More worrisome is that Anders has been appointed to HOC committee that deals with gun controls. Repealing the Long Gun Registry was something we knew was going to happen. Anders though is talking about reviewing other gun control legislation. That we didn't know about.

    The increase in the age at which one can collect OAS and GIS was something we didn't know about, as well. It is the kind of thing that I as a CPC supporter don't like. People will as it currently stands receive CPP at 65 but will have to wait until 67 to collect OAS. So there's a very attractive two pronged issue for the opposition in 2015. On one hand they can speculate that if re-elected the Conservatives will do the logical thing and raise the age at which one can collect CPP, while promising to restore both OAS and GIS to 65.

    This government may lack the finesse to be re-elected with a majority. Three years will give people a chance to truly evaluate Harper and his government. Their saving grace may be the reluctance of those outside of Quebec and especially those in Ontario to trust the NDP with the levers of power.

    Earl (I'm not signed in)

    ReplyDelete
  16. Earl

    "Peter this "awful" CPC government is IMO doing a pretty good job of running the country. "

    Quite bluntly that's major BS. They are rapidly and efficiently destroying this country. And the recent budget has us well on the way to real disaster.

    So I reject your entire hypothesis !!!

    ReplyDelete
    Replies
    1. Trudeau already destroyed the country by preparing the bedrock for on-going separatism in Quebec. HIs Constitution is an abject failure since, it did not accomplish its goal: making French Canadians feel at home in Canada and ending the Quebec independence movement.

      Delete
  17. Abacus/Sun poll for Alberta just before the debate. 46-29-12-10. Gap widens another 5 %

    http://abacusdata.ca/2012/04/12/alberta-politics-wildrose-leads-pcs-by-17/

    ReplyDelete
  18. the real question is, "running the country" for whom? If you're talking about big corporations and the wealthy, then the CPC is doing a "pretty good job", but for everyone else, it's a different matter entirely...

    ReplyDelete
  19. "a pretty good job of running the country" - should read "a pretty good job of ruining the country."

    ReplyDelete
  20. Peter it is my opinion the the Harper government is doing a good job. It is yours that they are not. That doesn't mean because I disagree with you on one thing I can't agree on others unless I have a closed mind. I would expect better of you.

    ReplyDelete
    Replies
    1. And since I have absolutely NO idea who you are do you think I really care ???

      Anonymity is great but don't expect the same treatment a real person gets.

      Delete
  21. Guys, can we not get into partisan debate? Part of the reason I love this site so much is that we can debate the polls without getting into bitter partisan bickering.

    ReplyDelete
    Replies
    1. Seconded - there are plenty of places on the Internet more suited for that kind of discussion.

      Delete
    2. Thirded, If there is another angle to interpreting the polls, i would love to hear it. Otherwise, I don't want any partisan bickering either.

      Delete
    3. Agreed

      Delete
  22. Interesting piece by Chantal Hebert in today's Toronto Star.

    Basically her thesis is if Smith wins it takes a lot of Harper's base away from him.

    http://www.thestar.com/news/canada/politics/article/1161314--hebert-why-stephen-harper-s-biggest-battles-will-be-with-conservatives

    ReplyDelete
  23. Éric, do you know where I can find your individual riding projections from the 2011 federal election?

    ReplyDelete
    Replies
    1. 1 in 20 outlier15 April, 2012 11:01

      http://www.threehundredeight.blogspot.ca/2011/05/final-projection-conservative-minority.html

      The riding by riding break down is an imbedded image.

      CPC 142 (36.4) actual 167 (39.6)
      NDP 68 (27.3) actual 102 (30.7)
      Liberal 60 (22.8) actual 34 (18.9)
      Bloc 27 (6.7) actual 4 (6.1)
      Green 0 (5.6) actual 1 (3.9)

      Here is the link to Eric's post election analysis

      http://threehundredeight.blogspot.ca/2011/05/projection-vs-results.html

      "Clearly, the final projection was wrong. It under-estimated both the Conservatives and the New Democrats and over-estimated the Liberals and the Bloc Québécois. While I was not alone in making this error, I humbly recognize that of all the projections mine was among the worst.

      Of the 308 ridings in the country, ThreeHundredEight.com correctly called 234 of them. That's an accuracy rate of 76.0%, which is absolutely unacceptable.

      However, it was not the seat projection model that failed. The seat projection model actually performed very well - or would have had the popular vote projection model not missed the mark so completely"


      Sound a lot like Eric was saying the pollsters were out to lunch as a group AND/OR that his weighting and use of the data from the polls: his "popular vote projection model" for being very wrong

      Despite my partisan CPC slant my comments are meant to influence for the better Eric's seat projection model by using the polling data in a better way.


      Eric has said in the past that he has adjusted his model(weighting of polls to the vote projection model) based on my insights and suggestions.

      Delete
    2. No, I have not made any changes based on your comments.

      And your breakdown of my final projection is wrong.

      Delete
    3. Anonymous 12:21,

      The final projections are here:

      http://1.bp.blogspot.com/-MZAvcr87sDY/Tb4og5e6NAI/AAAAAAAAFDc/kfJ7oPKft4c/s1600/11-05-01+Ridings.PNG

      But it would be a good idea to read the post-mortem as well: http://threehundredeight.blogspot.ca/2011/05/projection-vs-results.html

      Delete
    4. Thank you very much!

      Delete
    5. Jeez, 1 in 20, you get more wrong every time you comment. Did you think we were too lazy to click on the link in the right-hand column to see the 2011 final projection and outcome vs. projection, or did you think we were too stupid to notice? Because it's got to be one of those things, and neither is very nice to think about your fellow commenters.

      Delete
    6. 1 in 20 outlier16 April, 2012 10:47

      TS

      I assume your insightful comment slops over to express the same disdain of Eric's post of Apr 15, 2012 08:48 AM

      ie He provided the same information.

      Have as good a day as you can with the hand that the good Lord dealt you

      Peace Out

      Delete
    7. 1 in 20, Eric did no such thing. In the post that you linked to, the correct numbers were there. Just own up and admit that either you thought you could pull one over on us, or you made a mistake.

      And please, keep your religion to yourself, and I'll keep my non-religion to myself. Thanks.

      Delete
  24. @Eric,

    As always, enjoying your site.

    Question re: Alberta Election:

    I've noticed that the Calgary Herald/Nat Post seem to be reporting a "dead-heat" from their polls, whereas the Sun Newspapers seem to be saying it will be a WR landslide.

    Are the polls actually that different, or is it just media spin?

    I also enjoy your analysis, so I look forward to hearing your thoughts on the most recent polls whenever you get a chance (I recognize that the Alberta readership is likely a smaller demographic that those following the national polls).

    Go Wildrose!
    40 Years is Long Enough!
    Kick the Bums Out!

    Wildrose Supporter (Former PC Supporter)

    ReplyDelete
    Replies
    1. It's because they are referring to THEIR polling. The Herald/Journal is relying on Leger, and their last poll showed a neck-and-neck race. The Sun is relying on Abacus, which shows a Wildrose lead. ThinkHQ (CTV), Campaign (independent), and Forum (freelance, last poll with G&M) also show Wildrose lead.

      So, it isn't media spin, just that each media outlet tends to relying primarily (or solely) on what the polls they paid for are showing.

      Delete
    2. Thank you for responding.

      So what accounts for the consistent "neck-and-neck" polls from Leger (I think this is the second such result)? Are they done in a different format from the others? Do you think that one kind of poll is more reliable than the others?

      In advance, I appreciate you taking the time to answer. I clearly don't have your expertise in this.

      Wildrose Supporter (Former PC Supporter)

      Delete
    3. Leger just had one that showed the tight race recently, and I think it had a lot to do with the fact it was taken over the Easter long weekend.

      Delete
  25. I was thinking of this link:

    http://www.calgaryherald.com/news/politics/Alberta+election+rival+leaders+Smith+Redford+still+neck+neck/6459314/story.html

    But you're right, on a closer read, it appears to be a bit of spin on the part of the Herald. The "leaders" are "neck-in-neck," but the WR still shows as in the lead.

    Thanks for your time.

    Wildrose Supporter (Former PC Supporter)

    ReplyDelete
  26. http://news.nationalpost.com/2012/04/15/allan-hunsperger-wildrose-blog/

    Has to hurt ??

    ReplyDelete
  27. Do we know anything about the regional numbers from the most recent Ipsos poll that shows a national dead heat?

    ReplyDelete
  28. I would like to be able to make my own predictions. Do you think you could offer a simplified (in other words no extras, just proportional swing. to the average person?

    ReplyDelete
    Replies
    1. You could try your hand with this forecaster:

      http://esm.ubc.ca/CA15/forecast.php

      Delete

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.