Friday, September 18, 2015

2015 Federal Election Link Round-up, Week 7

So last week wasn't the tipping point. It seems we're back to a three-way race, and maybe the closest point of it so far in this campaign.

Week 7: NDP 31.4% LPC 30.0% CPC 29.6% GPC 5.5% BQ 3.1%
Week 6: NDP 31.7% LPC 30.3% CPC 29.1% GPC 4.9% BQ 3.5%
Week 5: NDP 32.4% LPC 29.9% CPC 27.2% GPC 5.1% BQ 4.1%
Week 4: NDP 33.9% CPC 28.4% LPC 27.9% GPC 5.0% BQ 3.8%
Week 3: NDP 32.7% CPC 29.4% LPC 28.0% GPC 5.1% BQ 3.9%
Week 2: NDP 31.8% CPC 29.8% LPC 27.5% GPC 5.4% BQ 4.6%
Week 1: NDP 34.7% CPC 29.6% LPC 26.8% BQ 4.6% GPC 4.0%

Friday, September 18, 2015

- I wrote about that internal NDP poll showing Justin Trudeau trailing in his own riding. Then the Montreal Gazette and Mainstreet Research went and scooped me with their own poll for the riding.

- The Poll Tracker and the riding projections have been updated.

- Your daily Nanos. Compared to their last independent three day sample, a drop of 1.9 for the Tories, a gain of 1.4 for the Liberals, and an uptick of 0.9 for the NDP. So, par for the course.

- Riding polls from Forum for University–Rosedale, Ajax, and Toronto Centre. Note that the Toronto Centre and University–Rosedale polls were conducted in August.

Thursday, September 17, 2015

- My analysis of today's numbers.

- The Poll Tracker and the riding projections have been updated..

- It is polling day, apparently. Polls from Mainstreet (BC only), EKOS, and Environics. Something for everyone.

- On the latest episode of the Pollcast, I talk with Shachi Kurl of the Angus Reid Institute. We go over her new poll comparing platform promises.

- I broke down the race in Atlantic Canada on Power and Politics last night. With maps!

- Today's Nanos shows virtually no change at all (we're talking well under a point per party) from their previous independent three-day sample. Forum's poll is looking more recognizable as the three-way race everyone is seeing, but it also means they have recorded a serious drop for the NDP and a modest increase for the Conservatives. And some B.C. riding polls from the Dogwood Initiative and Insights West.

Wednesday, September 16, 2015

- The Poll Tracker and the riding projections have now been updated.

- In the first of a series of articles in which I will be breaking down the country region by region, I took a look at Atlantic Canada.

- Hier, sur Les voies du retour (1:18:30).

- Today's Nanos has another back-and-forth. But compared to its previous independent three-day poll, it is showing a 2.7-point drop for the Conservatives. Also, some interesting riding polls from Mainstreet Research for Ajax, Spadina–Fort York, and Calgary Confederation.

Tuesday, September 15, 2015

- Yesterday's ridings to watch: Oakville and Oakville–North Burlington.

- Your daily Nanos and a new Ipsos. Nanos showing little change from its previous independent three-day poll, apart from a modest Liberal drop. Ipsos is showing the NDP down a little.

Monday, September 14, 2015

- A double-header for today's ridings to watch: Oakville and Oakville–North Burlington.

- On the latest episode of the Pollcast, guest Greg Lyle from Innovative Research Group breaks down his latest poll, going beyond the three-way race.

- The Poll Tracker and the riding projections have now been updated.

- I spoke about the three-way race and the differing polls on Power and Politics on Friday and on The House on Saturday.

- A few polls out this morning: from Nanos, Innovative, and Abacus. Three-way race across the board.

187 comments:

  1. What happened with the Greens' numbers? They dropped quite a lot, even though, from what I see, they have a score at least as high as the Bloc in every poll.

    ReplyDelete
    Replies
    1. They're polling about where they were at the start of the campaign. They're poised to win the one seat they ever had a realistic shot at.

      What drop do you mean?

      Delete
    2. The Greens numbers spiked a bit right after the Macleans debate when Canada actually got to hear what Elizabeth May had to say. Since then, she hasn't been included in debates and they media isn't talking about her much. Lack of publicity mostly.

      Delete
    3. When I posted my comment, the average for the Greens was at 2.1%, which I found suspect since it gave them a lower average than the Bloc. But it has been corrected since then.

      Delete
    4. A poll (I'm sorry I don't have more details) had the Greens at 32% on Vancouver Island, NDP at 39%, Liberals and Tories 15% each on the Island. If these numbers are correct Murray Rankin (NDP M.P. Vioctoria) is in trouble!

      Delete
    5. I find the national numbers for Greens odd. Here in Thunder Bay the support is sky high for Bruce Hyer and I expect him to be re-elected, the NDP is fighting hard but have a candidate who pretty much admitted he will do whatever Mulcair says thus people here won't vote for him (very, very strong desire for local reps who actually represent us here). The Vancouver Island poll and I'm hearing of lots of support in Guelph. Yes, this is pockets only but strong ones which if polling firms are hitting those areas should be pushing up Green numbers a bit.

      Delete
    6. John_Northey,

      I am also told the Greens have some high hopes perhaps even momentum in the Yukon.

      Many people assume the NDP has the Island locked up but, a quick back-of-the-envelope calculation of the NDP returns on the Island for the 2011 election demonstrated only 35,5% of the vote went NDP and they only won 3/6 seats.

      It is certainly possible (though not likely) the Greens could win 3 additional seats on the Islands; Victoria, Esquimalt-Saanich-Sooke, and Cowichan-Malahat-Langford. I know Denise Savoie's resignation rubbed some the wrong way as she resigned almost immediately once she qualified for a parliamentary pension. Murray Rankin's people were desperate for volunteers during the early days of the campaign, in what is thought as a solid NDP riding. I think we will see at least two Greens elected come October 19th.

      Delete
  2. Just to let you know: the hyperlinks for the House and P&P are reversed. Aside from this minor error, great job all around.

    ReplyDelete
  3. Before people start getting too worked up about the possibility of a 3-way tie where the LPC and NDP need to talk to the GG to form government, I'd like to remind everyone that there's no more reason to expect these polls to hold until election day than there was to expect last week's polls to hold until election day.

    All the movement recently tells us that the electorate is fairly fluid, so there's opportunities for everyone, but that's about all we know.

    ReplyDelete
    Replies
    1. I agree they will not hold. Most voters are going to go to the opposition leader who is leading in the last week and wipe out Harper.

      Delete
    2. That's conjecture, Jimmy.

      Delete
    3. Empirical evidence at the door conjecture.

      Delete
    4. Ira,

      The Liberals and NDP do not "talk" with the GG. Only the Government is allowed to initiate conversation with the Crown. The Crown may call upon the Liberals or NDP at some point but, first they will have to defeat the Government on a confidence vote. The convention: a sitting premier has the right to meet the House: if they win a confidence vote they continue; if they lose a confidence vote, they will be given the courtesy of resigning, failure to do so will likely result in the termination of his commission and dismissal.

      Delete
    5. This is incorrect. Anyone may provide unsolicited advise to the GG.

      Delete
    6. Mapleson wrote: "This is incorrect. Anyone may provide unsolicited advise to the GG".

      That is absolutely untrue-Anybody has the right to petition the Crown-only the Government has the right to advise the Crown! If your theory was correct then why pray tell did Dion not become PM in 2008? Why did the Crown listen to the advice of the sitting PM instead of the leader of H.M.'s Loyal Opposition?

      More importantly, it is the Crown not the politicians who initiates the conversation: One is "called" then "asked" to form a Government. A documentary on the Queen for the Golden Jubilee shows the Queen asking Tony Blair to form a Government (I believe the Documentary is called "Queen and Country" from the BBC. However, I may be incorrect on the title).

      How could responsible government work any other way, Mapleson?

      Delete
    7. You seem to conflate the terms advise and direct to act. Dion and Harper both spoke to the GG in 2008, but the advise of Harper was followed.

      So responsible government only works when the Head of State only hears from one side of the scales and must weigh them according to only that one set of advise and biased facts?

      Delete
    8. Dion did not speak with the GG in 2008! That is why he was forced to write a letter! Dion was never "called" to give advice!

      Yes, responsible government only works when the Head of State follows the advice of the elected Government! Basic stuff really basic stuff. After an election during the interregnum period the Crown may call upon all party leaders or none and may or may not seek advice. The Crown by convention gives the sitting PM a chance to meet the House. Once a Government has secured the confidence of the House the Crown must follow the advice of his elected ministers only! Failure to do so would deprive the House of its right to judge confidence.

      Delete
    9. I used "speak" in the liberal interpretation as any communication, including a written letter. Just like I was use it for a Press Release from the PMO.

      If the Head of State only follows the advice of the sitting Government, then there is no point in the position. The GG followed Harper's advise in 2008, but didn't have to do so. What happened in 2008 was that the GG following the advise of the PMO deprived the House of its right to judge confidence.

      Delete
    10. Of course the Crown only follows the advice (noun) of the sitting Government-that is how responsible government works! The Government wins a confidence measure and is entitled to solely advise (verb) the Crown. The Government is then responsible to the House for those actions and continues in office with the right to advise the Crown until defeated on a confidence motion and or resignation.

      There is a large difference between speak, write, hear, listen and talk. It is telling M. Dion was not invited or called to Rideau Hall for a conversation. It demonstrates the Crown was not prepared to follow his advice and was not prepared to ask M. Dion to form a Government. Since, the Crown can only follow the advice of its ministers, it would have been inappropriate for Madame Jean to read the letter from M. Dion. Now, if Mr. Layton and Topp had understood how government formation works in Commonwealth Realms and quietly defeated the Government the Crown would not have been obligated to follow the advice of the outgoing prime minister but, would by convention had the ability to speak and seek advice from other party leaders although until the Coalition succeeded in winning a confidence vote in the Commons the Crown would not be obligated by convention to follow advice. In 2008 the Crown acted legally on the advice of the prime minister who had the confidence of the House of Commons-that's how responsible government works!

      The Crown by convention was bound to follow Mr. Harper's advice to prorogue the House in December 2008, this is because Mr. Harper controlled the confidence of the House. The opposition parties had not defeated the Government on a confidence motion and Harper had recently won a confidence vote on the Throne Speech.

      Delete
    11. That is how responsible government works!?! Repetition doesn't make your points any more salient or true. If you read the Letters Patent of King George VI (1947), the Governor General can act "with the advice of Our Privy Council for Canada or of any members thereof or individually, as die case requires, to exercise all powers and authorities lawfully belonging to Us in respect of Canada..." The Privy Council is an appointment for life, so in 2008 the GG could hear/receive/read the advice of Stephane Dion as he was appointed in 1996.

      The Head of State is a separate role from the Legislative branch in order to provide a balance to possible misuses of procedural powers. From GG.ca: "One of the governor general’s most important responsibilities is to ensure that Canada always has a prime minister and a government in place that has the confidence of Parliament. In addition, the governor general holds certain reserve powers, which are exercised at his or her own discretion. The governor general also presides over the swearing-in of the prime minister, the chief justice of Canada and cabinet ministers. It is the governor general who summons, prorogues and dissolves Parliament, who delivers the Speech from the Throne, and who gives Royal Assent to acts of Parliament."

      The GG is not bound to only follow the advice of the sitting Government. If you reach into the depths of Canadian history there are two large examples of using that reserve power contrary to the advice of the sitting Government: the resignation of Sir Tupper and the King-Byng Affair. The difference in 2008 and 1926 is that King did not command a plurality of seats and wished to dissolve parliament rather than prorogue it.

      Delete
    12. Mapleson,

      I am sorry you are incorrect. The Crown is honour bound to follow the advice of the sitting Government until said Government is defeated on a confidence motion in the House. How could responsible government work any other way?! It can't! You need to read some Nova Scotian history since, that is where responsible government began! Or pick up an excellent book by Dr. Andrew Heard from Simon Fraser University called: "Canadian constitutional conventions: The marriage of law and politics". I am sure you will find it a good read and it shouldn't cost that much, maybe $20.

      Secondly, The executive is a separate branch of government in America! In Canada the Crown is part of the legislature-The Queen in Parliament and part of the judicial system; pardons and clemency are meted out by the Crown. Prosecutions are entered in the Queen's name! We have a very different system of government from the States, you seem to have confused the two.

      The Crown could have accepted M. Dion's advice but, to do so would have contradicted the most fundamental principle of responsible Government-The Crown is bound to act on the will of the elected House, and violated the long held constitutional convention that the Crown only acts on the advice of Her ministers. Since, Mr. Layton, Topp, Dion and Duceppe failed to understand the Government first needed to be defeated in the House before the process of government formation could begin, Mme Jean had no other choice but, to follow constitutional convention and the advice of the leader who enjoyed the confidence of the House-Mr. Harper.

      Both the KIng-Byng affair and Sir Charles Tupper's dismissal prove how responsible government works. King was defeated in the House on censure motions and refused to resign. Tupper by contrast refused to recall Parliament but, had lost the election and was using his position to fill patronage appointments. Since, both Sir Charles Tupper and WLMK had lost the confidence of the House (Tupper refused to recall Parliament but, had lost the election to Laurier and the Liberals who commanded a small majority in the Commons) the Crown acted to implement the will of the House and dismissed the sitting prime ministers.

      Why you may ask was Tupper dismissed and a similar fate did not befall Harper in 2008? Harper had won a confidence motion in 2008, the Speech from the Throne. Tupper, on the other hand, refused to recall Parliament and submit his Government to a vote. Since, it was unclear in the Tupper example if he held confidence the Crown was well within its right to dismiss Sir Charles. In 2008 by contrast it was clear Mr. Harper enjoyed the confidence of the House and the suggestion provided by the opposition leaders that he no longer held confidence was merely speculation since, the opposition was unable to defeat the Government.

      Delete
    13. Pete,

      You are well versed in the details, but seem to have skewed the macro-focus. The Crown does have multiple roles within our system of government, but the roles are separated. This is like being the King of Scotland and the King of England by personal union prior to the Acts of Union. They are the same person, but they have different roles and powers.

      The Crown as embodied in the Governor General has exactly the powers that I have said. While I will read your suggestion, conventions are not legally binding, whereas my quotations come directly from the source.

      The point is that the GG could receive advice as given, but chose not to follow it. Until Harper there was no convention of successfully using procedural maneuvers to remain in power. Now there is, because it was legal and the GG saw it as in the best interests of the nation.

      Following the best interests of the nation is responsible government and the reason that we have a Senate and GG. To blindly follow the directions of one party leader and his sycophants is the path to dictatorship, not responsible democracy.

      Delete
    14. Mapleson,

      I'm trying to be nice but, really..."The roles are separate"? I have no idea what you are trying to suggest. No. They are not separate, the actions may be distinct but, the role of Monarch is held by a singular individual and one who has a singular occupation-Queen. The Queen-in-Parliament or the Queen as Head of the Church of England has the same duty as she has every day: to uphold and carry out the vows made in her coronation oath. Including governing the peoples of Canada "according to their laws and customs"! In other words Canada is governed by law (statutes) and customs (conventions and traditions).

      No. The point is it would be inappropriate for opposition leaders to submit untendered advice to the Crown! If the Crown needs advice they will ask for advice, the Crown usually does not wish to be bombarded by the equivalent of political junk mail from opposition leaders who have failed at their supposedly primary task-defeating the Government!

      If the Crown could receive advice from any privy councillor think of the chaos: no one would be able to discern the Government's policies from policies implemented by the Crown through the advice of privy councillors without (presumably) the approval of Parliament. The Crown would no longer be subject to the will of Parliament-instead the Crown would be the executive and Parliament little more than a consultative body!

      The Government always sets the legislative agenda and so there is a long history of governments using parliamentary procedures to advance or hold up legislation, you may have heard of closure; where the Government closes debate on a particular bill. In addition you forget that only the Crown has the right to invoke the particular procedure in question-prorogation. Once again there is a long history of prime ministers using prorogation to suit their political as well as legislative needs.

      No. No. NO. NO. Responsible government is a particular system of government whereby: The Crown follows the advice of the political party leader who commands the will of the elected lower House unless and until the Ministry is defeated on a confidence motion or resigns! The Government is responsible to the will of the House through this system!

      Delete
    15. I will leave you to your delusions. What I've said is explicit in the Letters Patent.

      Delete
    16. Mapleson,

      You cite the 1947 Letters Patent but, fail to understand the meaning or context. When "privy council" is used in the Letters Patent, the meaning is not the full council but, the executive committee thereof: The cabinet. The Letters patent invoke "privy council" as it is a formal document and formal language is used.

      Why you think it is constitutionally valid to unilaterally change the rules and traditions of parliamentary government in Canada is beyond me? Your ideas of how responsible government is exercised in Canada is bizarre and undemocratic.

      The Ministry, who represent the Crown, must be responsible to the elected House for responsible government to work, therefore, the Crown must follow the will of the House as expressed through the wishes of the Ministry that holds the confidence of the House for responsible government to work and for Canada to be democratic. it really is common sense but, I suppose not everyone is gifted with it.

      Delete
    17. Mapleson,

      How could responsible government work as you suggest; How could the Ministry be responsible to the House if the Crown need only follow the advice of any privy councillor most of whom do not sit in the House of Commons? You have not explained how your theory is democratic or responsible? Please do.

      Delete
  4. why do you give no weight to the Nanos polls from 9/9 and 9/10? it seems to penalize the conservatives in your projection as in those two polls they were in 1st and 2nd respectively and fell back to 3rd in the poll from Nanos that you include.

    ReplyDelete
    Replies
    1. Because they are not independent samples. Nanos is rolling three days of polling into each poll, so every three days a Nanos poll is given 100% weight because it is separate from the previous three-day sample. In between, weights are reduced to 33%.

      Delete
  5. you also continue to give weight to polls from 9/4 that have the CPC way down in the mid 20s... why?

    ReplyDelete
    Replies
    1. They aren't polls. They are online panels. Gee the same panel participants don't like the Conservatives over and over again, and they keep being counted as separate polls by Eric's model.

      Delete
    2. The polls that have had the CPC in the mid-20s were by Forum (IVR) and Nanos (telephone), and were not done online.

      Delete
    3. Hey shoshana, did you pick up the baton from bede as resident CPC whip?

      Delete
    4. I mostly criticize methodology and point out that voters and people who answer polls have been significantly different enough times (all over the world) to not be reporting on them like they are gospel.

      We don't have a Conservative party to speak of in BC and I commented several times during the last BC election that there was no way the NDP was that high.

      Again, I am hearing stories of university students being secret Conservatives, and that's in Vancouver. I worked the last federal election. Lots of voters who could only say 2 words to me, "Stephen Harper." I had to very diplomatically and carefully take them through the choices (none of which of course are Stephen Harper) to let them know the leaders of all the parties of the candidates on the ballot.

      There are too many people out there who won't talk about voting Conservative and/or who don't speak English but vote. At least in BC that's true.

      Delete
    5. Poor little ol' Stephen Harper, eh? He's just misunderstood. Doesn't get enough press. If only the young people could hear directly what he has to say...

      Delete
    6. Poor little Stevie is running away with chirumenga! The NDP is headed for third, your ilk should be afarid very afraid! the Liberals are unlikely to help defeat the Tories should the either the NDP or Tories win a plurality.

      Delete
  6. *This* is the Westminster parliamentary system, where the Prime Minister is the "first amongst equals" and MPs control the Commons and their party: http://www.abc.net.au/news/2015-09-14/malcolm-turnbull-wins-liberal-leadership-ballot-over-tony-abbott/6775464.

    What would it take for Canada to return to our roots?

    ReplyDelete
    Replies
    1. Australians seem pretty upset by that, it would appear. While Abbott wasn't well-liked by the international community (since he was perfectly willing to let ships full of refugees sink), Liberal-voting Aussies liked that about him.

      Delete
    2. Abbott made a lot of enemies. His loyalty to the coal barons meant that he was deliberately attacking the solar power industry in Australia, which is *booming*. Didn't help the coal industry, which is collapsing due to market forces anyway! He made a whole lot of other similar decisions where he made enemies and gained nothing; his sexist antics are infamous at this point.

      Frankly Turnbull gives the Liberals a decent chance of winning the next election in Australia. Abbott was sure-to-lose by now due to the number of enemies he'd made. A bunch of "never vote for Abbott ever" voters are willing to vote for Turnbull.

      Delete
    3. The evidence I've seen is that most Australians are very happy to see Abbot gone.

      Delete
    4. I think I saw early reaction, which understandably would have been the people most upset by the events.

      Turnbull's initial speech would have made me pretty happy were I Australian.

      Delete
    5. I believe the majority of Aussies were happy with Abbott leaving.

      Abbott was always a polarizing figure in his home nation. His disapproval rating is hovering around 60%

      Delete
    6. Wish the CPC would take a page out of their Australian friends book. They'd be better off if they did.

      Delete
    7. The idea that the PM is primus inter pares is an 18th century concept. Whether or not such a concept was put into practice in the 18th century is debatable. "Cock Robin" a.k.a. The Rt. Hon. Robert Walpole, a.k.a. the First Earl of Orford certainly had little problem unilaterally dispensing patronage during the 18th century while exercising the prerogatives of the Crown.

      Above writers make reference to Australia. I generally agree that if leaders were chosen by the parliamentary party or caucus, it would improve a PM's responsiveness to the House. Whether reducing the franchise from the tens or hundreds of thousands down to a few dozen is fair or democratic? is another question. However, lets be clear: Parliament does not choose the Ministry. The Crown appoints the Prime Minister and cabinet unilaterally! Parliament may approve or reject this choice through confidence measures but, they never pick, appoint, choose or elect the Ministry! Technically, the House doesn't even dismiss the Ministry; After losing a vote the Crown will always give the PM the option to resign, failure to do so by convention will lead to dismissal through the termination of a PM's commission.

      Delete
    8. The Crown has unilateral power, but rarely utilizes it. Generally speaking, we have new PMs within a seating Parliament based upon the resignation of the sitting PM as the party leader. However, in more democratic systems, party elections are possible at the drop of a hat, if and when the party leader loses the confidence of their party members.

      The Australian Liberal Party elects all Cabinet positions and this is the list given to the Crown for appointment. It's just makes the process more transparent.

      Delete
    9. No we have new a new PM during a sitting Parliament due to the resignation of the prime minister, and the election of a new party leader who can command the confidence of the House which is secondary.

      Delete
  7. Online Panels of the likes of Innovative are the most problematic of all so called polls. They are sampling the same people over and over again.

    Its not surprising that the online panels have the Conservatives in mid 20's in Ontario, which no one else is.

    It's extremely problematic to be counting these in aggregates.

    ReplyDelete
    Replies
    1. I'm registered in two firms for online panels and, since the beginning of the campaign, only one of those sent me an invitation to a political poll, and only once. And it has done more than one of those since the beginning of the campaign. So no, it's not the same people answering over and over again. While the list may not be probabilistic, they still do a random sampling of it and, therefore, overall, it should compensate and give something close to probabilistic. Actually, those polls are the exact reason why we should do an aggregate.

      Delete
    2. Pollsters may have tens of thousands of people registered. They have demographic data about them. They then sample from the population (the entire panel) to get a representative sample based on these demographic data. They are not based on the responses of the same people over and over.

      Delete
    3. Given the right "random" sample (to the extent that's ever possible), there's nothing intrinsically wrong with going back to the same people repeatedly. Arguably, it gives better information on trends, even if some unknown but constant bias may be wired in.

      Based on Thierry's comment, though, the point is moot.

      Delete
    4. No John, it's completely lacking in reliability to poll the same people more than once (aka you would not get the same result from different people) and it's not valid (aka its an unrepresentative sample) because they are self selected to begin with, aka they volunteer to answer online questions and be in a panel.

      It's heaping bad on top of bad methodology.

      Delete
  8. The major problem with the polls right now is that Ontario is all over the map with the Conservatives anywhere from third to first with a swing of about 10%.

    All the polls agree BC is a three way race, Atlantic Canada is owned by the Liberals, Quebec by the NDP, the Prairies by the Conservatives.

    Ontario is a hot mess in the polls and is the most seat rich province. The swings between the polls could give a majority to any of the major parties.

    There is essentially nothing to go by right now.

    ReplyDelete
    Replies
    1. Ontario is very regionally messed up right now. Cons very popular in the 519 area, NDP up north, Liberals in Toronto, 905 area a Lib/Con mix, plus pockets of Green support in Guelph & Thunder Bay. What does it all mean? That if a pollster doesn't balance those areas properly they can get wild swings just due to which zone they polled the most.

      Delete
  9. Seriously, the polling companies could just poll Ontario and BC if its close.

    ReplyDelete
  10. Interesting campaign so far. There has not been significant movement in the polls outside the margin of error and I doubt there will be in the weeks ahead.

    The only movement to note is that the NDP is trending downwards at the expense of the Liberals.

    Justin Trudeau has done well to claw himself back into the race. Thomas Mulcair has been on cruise control so far, but I am sure the NDP campaign will begin to ramp up to address the slow but steady increase in Liberal support.

    Like them or hate them, we have three strong, savvy leaders. I doubt any of them will concede much ground with five weeks left in the campaign.

    ReplyDelete
    Replies
    1. "we have three strong, savvy leaders"

      I know it's delivered in a spirit of fairness, but I can't agree. Harper is strong only because he's narrow in focus and because he's had the backing of the mainstream media since about 2002 (they conveniently minimised the significance of his presidency of the far right National Citizens Coalition, for instance). Trudeau is only as savvy as his class background makes him - and even that he manages to undermine. He's an incoherent thinker, but he's been well managed by his handlers. Of the three, Mulcair is the only one who actually fits the description, and no, I don't particularly like him. I've met him a couple of times and he operates in a crowded room like a shark - but that does make him a capable leader.

      Delete
    2. Harper's supporters would not agree that he has had the backing of the mainstream media since 2002.

      Why do you think SunNews was a thing?

      Delete
    3. "He's had the backing of the mainstream media"?

      When was the last time you read the Globe and Mail? Margaret Wente or Jeffrey Simpson are always slagging the PM. I wish they were as critical of all the leaders as they should be; “A government should never pretend it can replace the private market. It does not work, It didn’t work in England. Up until Thatcher’s time, that’s what they tried".-Thomas Mulcair. A more critical Fifth estate would opine that Mr. Mulcair is in the wrong party, that his 15 dollar a day child care plan conflicts with his above stated values, that he is all smoke and mirrors. Alas, instead they fawn at Mully and call what is at best a technical recession or more likely a rounding error-somehow threatens to turn our economy into a psuedo-dust-bowl economic malaise-it is ridiculous! Perhaps it explains why you have taken the sensible choice to discontinue reading the Globe?

      Delete
    4. Mulcair can be a chameleon & a shape-shifter and untrustworthy. Not sure those are good qualities.

      Delete
    5. Ira, I know Harper's backers wouldn't agree, but they'd be wrong. You can do an objective analysis, and on balance, the mainstream media have supported the Conservatives more than the Liberals and clearly more than the NDP. One concrete example, in the 2011 election, the Globe and Mail editorial board decided to endorse the Liberal Party. The newspaper's owners, however, stepped in and changed the endorsement to support the Conservatives. The vast majority of media outlets openly backed the Conservatives, few backed the Liberals, and none backed the NDP. The uniformity isn't too surprising since Canada has among the highest concentration of media ownership in the First World (exceeded only by Australia, where Rubert Murdoch owns virtually all of it).

      Delete
    6. Pete, I'm impressed you found as many as two people to back up your claim. Next to the scores of those (reporters, editors and publishers) who back the Conservatives, that's pretty remarkable. Find me two mainstream columnists who openly back the NDP.

      Delete
    7. Gerry Caplan supports the NDP He writes for the Globe on an infrequent if regular basis. Bill Tielman writes fawningly of the NDP for the Tyee and previously wrote for the Georgia Straight supporting the Socialsit party.

      Delete
  11. So it appears that the NDP support has leveled off during the election. It started at 35% and has dropped to 31%.

    ReplyDelete
  12. Thierry Soucie miss your seat projections.

    ReplyDelete
    Replies
    1. I'm really busy and since there are a lot of polls, I just didn't know which were worth doing and which weren't. And my August projections were not published (don't know why). I'll post the weekly average projections if you want, and try to be more active on that front.

      Delete
    2. Okay, I forgot that Eric's weekly tally at the top doesn't include regional breakdowns, so I'll actually use the poll tracker dates the most closely associated with the end of the campaign weeks.

      First week (August 10):

      126 CPC
      121 NDP
      86 LPC
      2 BQ
      1 GPC

      By region, it gives:

      Atlantic
      20 LPC
      7 NDP
      5 CPC

      Québec
      57 NDP
      13 LPC
      6 CPC
      2 BQ

      Ontario
      56 CPC
      38 LPC
      27 NDP

      Prairies
      16 CPC
      6 NDP
      6 LPC

      Alberta
      27 CPC
      4 NDP
      3 LPC

      British Columbia
      19 NDP
      15 CPC
      7 LPC
      1 GPC

      Territories
      1 CPC
      1 NDP
      1 LPC

      Delete
    3. Second week (August 19):

      128 CPC
      128 NDP
      81 LPC
      1 GPC

      By region, it gives:

      Atlantic
      14 LPC
      10 NDP
      8 CPC

      Québec
      60 NDP
      12 LPC
      6 CPC

      Ontario
      55 CPC
      38 LPC
      28 NDP

      Prairies
      16 CPC
      6 NDP
      6 LPC

      Alberta
      27 CPC
      4 NDP
      3 LPC

      British Columbia
      19 NDP
      15 CPC
      7 LPC
      1 GPC

      Territories
      1 CPC
      1 NDP
      1 LPC

      Delete
    4. Third week (August 26):

      132 NDP
      120 CPC
      84 LPC
      1 BQ
      1 GPC

      By region, it gives:

      Atlantic
      20 LPC
      7 NDP
      5 CPC

      Québec
      63 NDP
      8 LPC
      6 CPC
      1 BQ

      Ontario
      52 CPC
      39 LPC
      30 NDP

      Prairies
      14 CPC
      8 NDP
      6 LPC

      Alberta
      27 CPC
      4 NDP
      3 LPC

      British Columbia
      19 NDP
      15 CPC
      7 LPC
      1 GPC

      Territories
      1 CPC
      1 NDP
      1 LPC

      Delete
    5. Made a mistake for the third week, should have been August 24, don't know why I took the 26. Anyhow, it still gives an idea.

      Fourth week (September 1):

      130 CPC
      119 NDP
      87 LPC
      1 BQ
      1 GPC

      By region, it gives:

      Atlantic
      19 LPC
      7 CPC
      6 NDP

      Québec
      61 NDP
      10 LPC
      6 CPC
      1 BQ

      Ontario
      54 CPC
      41 LPC
      26 NDP

      Prairies
      16 CPC
      6 NDP
      6 LPC

      Alberta
      27 CPC
      4 NDP
      3 LPC

      British Columbia
      19 CPC
      15 NDP
      7 LPC
      1 GPC

      Territories
      1 CPC
      1 NDP
      1 LPC

      Delete
    6. Fifth week (September 8):

      123 NDP
      113 LPC
      101 CPC
      1 GPC

      By region, it gives:

      Atlantic
      23 LPC
      6 NDP
      3 CPC

      Québec
      65 NDP
      9 LPC
      4 CPC

      Ontario
      55 CLC
      47 CPC
      19 NDP

      Prairies
      14 CPC
      8 NDP
      6 LPC

      Alberta
      27 CPC
      4 NDP
      3 LPC

      British Columbia
      20 NDP
      15 LPC
      6 CPC
      1 GPC

      Territories
      2 LPC
      1 NDP

      Delete
    7. Sixth week (September 13):

      124 CPC
      120 NDP
      93 LPC
      1 GPC

      By region, it gives:

      Atlantic
      20 LPC
      7 NDP
      5 CPC

      Québec
      60 NDP
      12 LPC
      6 CPC

      Ontario
      55 CPC
      42 LPC
      24 NDP

      Prairies
      16 CPC
      6 NDP
      6 LPC

      Alberta
      27 CPC
      4 LPC
      3 NDP

      British Columbia
      19 NDP
      15 CPC
      7 LPC
      1 GPC

      Territories
      2 LPC
      1 NDP

      Delete
  13. Hey Eric I was wondering what your thoughts are now how polls are treating "undecided voters". Looking at a few polls it seems that pollsters are removing the true undecided from the sample and recalculating the percentages for the numbers reported as top line horse race.

    This seems weird to me. I'll take the latest Abacus poll as an example http://abacusdata.ca/wp-content/uploads/2015/09/Abacus-Release-Election-Wave-3-Release-1-FINAL.pdf According to the poll 20% are "Undecided", the general rule of thumb is that more often than not the undecided break against the incumbent, lets even ignore that hypothesis for now and just look at the numbers in Abacus crosstabs. Abacus says that off these 20% undecideds, in the 2011 election 3% voted for the Tories, 19% voted Liberal, 17% voted NDP and 56% seemingly did not vote. If the undecides were to break by that same proportion it should add 3 points each to NDP and Liberals. It leads me to wonder if the polls might be overestimating Conservative support at this point in the election

    ReplyDelete
    Replies
    1. I think portioning undecideds out proportionately is the best way to go, rather than to make guesses at how undecideds will behave. Better not to add another potential source of error if you don't have to.

      Delete
    2. Undecideds, more often than not, tend not to vote. Simply pretending they're not there is the best way to go.

      Delete
    3. How about just leaving them as undecideds and reporting it. That would be the best way not to introduce error.

      Delete
    4. The number of undecideds is sensitive to changes in methodology. It would make it much harder to interpret the data without offering much in the way of new information.

      Excluding them is a simplifying assumption.

      Delete
    5. If there were an industry standard for the question formation and reporting, it would be a good way of dealing with it, but the difference in methodology means you don't report the same figures across polling firms.

      Delete
  14. It seems portioning out undecideds proportionately doesn't eliminate the potential source of error. The potential source of error is inherent. Portioning them out proportionately is just one way to deal with the potential source of error, but we don't have any good reason to imagine it's the best way of dealing with it. Portioning them out based on their past voting record has to be just as good a way to deal with it, if not better.

    The point infinitelightning is making is a big part of my theory as to why the polls overestimated the PCs and underestimated the Liberals in the ON election last fall.

    My hunch is that the same dynamic will play out in this election. Or maybe that's just my hope...

    ReplyDelete
    Replies
    1. I think Tim Hudak is why the polls overestimated PC support last fall.

      Delete
    2. Governments in power all over have been underestimated in polls regularly, Alberta notwithstanding.

      Delete
    3. I agree with that Ira.

      I think generally the trend throughout multiple jurisdictions throughout the Worls is that polls tend to underestimate Conservative/ consrevative party support because conservative voters have a highyer turnout than the general population and Liberals/liberals. That was certainly the case in the UK last May and in the BC General election of 2013. However, even the polls in Alberta this year underestimated PC support by approximately 4%. So with all due respect, I think the data can easily be maneuvered to support either point of view.

      Delete
    4. And I think Stephen Harper is why the poll will overestimate the PC support this fall.

      Delete
    5. It's not so much that conservative voters are more likely to vote, but that older voters are more likely to vote. In Alberta, older voters voted NDP in Edmonton and PC in Calgary and WR in rural areas.

      Delete
    6. At first glance people think Conservatives are underestimated in polling, and yes they tend to have better voter turnout, but in polling especially more than one week out its actually the party in power at is underestimated the most, regardless of affiliation. The exceptions to this are when you actually have a realignment of parties, aka the Conservatives splitting into 3 in 1993 and the Wild Rose essentially folding into the Conservatives in the last Alberta election.

      Delete
    7. "Wild Rose essentially folding into the Conservatives in the last Alberta election."

      What?

      Delete
  15. I speculated that the photo of the drowned boy on the beach would move the numbers, but I don't see any big change.

    Ontario still anybody's ball game.

    Press speculation that Trudeau would prop up Harper minority.

    ReplyDelete
    Replies
    1. Not to take anything away from the photo or trivialise the family's grief but, voters usually do not allow foreign policy determine their vote, domestic policies usually have far more influence on voters' preferences.

      Ontario still up for grabs but, the NDP seems to be in decline. The election is turning into a two party race with the Dippers as also rans.

      Delete
    2. The Liberals propped up Harper during his two minority terms, and regularly voted with the Conservatives during their majority (C-51 anyone?). So, for once, pretty solid press speculation.

      Delete
    3. According to the Innovative Research poll, only 8% of Canadians believe the story about the Kurdi's family. I think that goes a long way toward explaining the lack of any sort of shift in the polls.

      Delete
    4. "Press speculation that Trudeau would prop up Harper minority."

      Why would Trudeau do that? What press speculation are you referring to?

      Delete
    5. Trudeau will go through the motions of letting the CPC, if they get the most votes try to set up a government and even introduce a budget, which of course will be defeated.

      Only then will he approach the GG to try to form a new government. Waiting patiently a few weeks/months for your turn is not the same as "propping up a Harper minority".

      Delete
    6. Globe and mail oped 2? days ago

      Delete
    7. Craig Buntin,

      '"Press speculation that Trudeau would prop up Harper minority."

      Why would Trudeau do that? What press speculation are you referring to?'

      Because the Liberals and the Conservatives are largely in agreement on policy. Contrary to current expedient fantasies of the mainstream press and Liberal functionaries, the Liberals have far more in common with the Conservatives in policy terms than they do with the NDP. They would prop up the Conservatives - as they have done before - because they agree, in essence, with the Conservatives.

      Delete
    8. The Liberals likely don't want to be the lesser partner in an agreement with the NDP. Letting Mulcair be Prime Minister grants the NDP legitimacy they've never before had, and that hurts the Liberal brand.

      The Liberals might prefer to hold influence over a CPC government while marginalizing the NDP.

      That doesn't speak well of the Liberals from a governance standpoint, but it might make strategic sense if their objective is to preserve the brand (rather than provide a decent government).

      Delete
    9. Brian,

      Gordon Gibson speculated in the Globe that the Liberals are desperate to keep the NDP from Government; thereby attaining the right and perception of a "governing party".

      The theory being: Canada and all parliamentary systems can only support one dominant leftwing and rightwing party. A crowded left wing or field will result in the eventual disintegration and re-absorbtion of the smaller party into the larger, as was generally supposed to be the case in the United Kingdom with the Liberal Party. Or with the Social Credit Party in B.C. and the disappearance of both the provincial Liberals and Progressive Conservatives (I've obviously data-mined my examples, but, hopefully you get the idea even if you disagree).

      The best strategy in politics as it is in War is: If you can't win then, retreat and be able to fight another day-self-preservation. Giving the NDP the keys to Government whereby, they may not be defeated for sometime if the Tories are looking for a new leader would put the Liberals in a position of very little influence in the new Parliament and as others have said crown the Dippers with a legitimacy hitherto unheard of-Government!

      Delete
  16. To continue with the speculations of Jimmy and Ira, I claim that IF a few days before the election the top two parties are NDP and LPC, then LPC will win the most seats of the two.

    Why? because I'm pretty sure a good 5% of otherwise-CPC voters are so scared of an NDP government that they would switch and vote LPC.

    ReplyDelete
    Replies
    1. I agree with that speculation. I'm hearing it at the door all over.
      A lot of people are fed up with the present government and are very angry.

      Delete
    2. As a Laurier Liberal) (AKA, fiscally conservative/libertarian, socially liberal), I wouldn't touch Trudeau; Wynne have driven Ontario far to the left and Trudeau would do likewise for Canada. It would be fiscally irresponsible to vote Liberal instead of NDP in this election.

      Delete
    3. I think three small deficits are far from "fiscally irresponsible". If on the other hand we were to agree that three small deficts are bad, then Harper would have to be disastrous, since he's had 5 enormous deficits (six if we do the proper thing and do not count one time gains against the deficit this year).

      Reality is that one may or may not agree with the concept of three small deficits as proposed by Trudeau, but they are far from disastrous. As far as the Ontario "disaster", it's the province leading the country out of this Harper recession, while the conservative paradise out West is mired in deficit and recession.

      Delete
    4. @PolStats I don't think you are understanding how the accounting is working: infrastructure spending doesn't get counted towards deficit numbers; just look at Ontario where the debt is increasing by far larger amounts than the reported deficits, which are already enormous. All the infrastructure spending Trudeau wants doesn't get counted in his deficit numbers.

      Delete
    5. I don't know from where you get this idea that they don't count towards deficits. Trudeau has explicitly said that the small deficits will be put to use on infrastructure building.

      Perhaps you are referring to the fact that in accountanting infrastructure is amortized differently than regular debt. This is no different than the fact that is quite ok for a person to carry a 150K mortgage but not a 150K credit card bill spent on fancy toys such as F35s.

      How's that program for disastrous and wasteful spending?

      Delete
    6. @PolStats (1) What is normally called infrastructure (e.g., a house) is, per your comment, accounted for differently than regular spending. This is why the Ontario debt rose by $36 billion in 2013/2014 over 2012/2013 while the government claimed a deficit of "only" $12.5 billion.
      (2) Trudeau is explicitly saying he wants to invest in "social infrastructure" as well as building infrastructure.
      (3) Social spending normally count toward deficits but it is extremely unclear if Trudeau will, in fact, count it that way, since he is deliberately using the word "infrastructure."
      (4) Building infrastructure will not add to deficits but is financed through borrowing and as such will add to debt. Thus, when Trudeau is claiming a few small deficits, he is being deeply disingenuous. He is planning in excess of $100 billion in infrastructure which will not be recorded as deficit spending in his accounting. As such, debt will increase massively under a Trudeau-led government, even as he will claim to have only small deficits.
      (5) Even building infrastructure is not necessarily of value, and thus borrowing to build is frequently unwise. Just look at the building binge in the US from 1998 to 2008 for an example of this. Claiming you are building something of value, and recording it as such on the books, doesn't make it of value.

      Summary: Trudeau is making claims of small deficits but the experience from Ontario is that this is grossly false; debt in Ontario is spiralling out of control and our credit rating is dropping commensurately. Given the close relationship between Wynne and Trudeau, there is every reason to believe a vote for Trudeau is fiscally irresponsible.

      Delete
    7. "He is planning in excess of $100 billion in infrastructure".

      Let's clarify that figure. We are currently spending $65 billion under Harper. Trudeau has proposed we add another $60 billion to that. What is the cost of the F-35 program? $46 billion (which for some reason you do not consider fiscally irresponsible).

      So the actual cost of Trudeau's program, if we cancel the wasteful F-35 is $19 billion over ten years.

      You might or might not agree with the Liberal plan but it really doesn't register in the grand scheme of things. More money was wasted by Harper hosting the G-8 in Toronto with no infrastructure to show for it.

      Delete
    8. @PolStat
      (1) At what point did I say that $46 billion for F-35s is fiscally responsible?
      (2) That said, you're comparing apples and oranges, since the spending for the F-35s would be over 40 years; the proposed infrastructure spending is over a much, much shorter timeframe.
      (3) Trudeau's program costs what it costs irrespective of what is done with the F-35 program and to claim otherwise is totally false. That's tantamount to claiming that some car you want to buy is only going to cost $20,000, not $35,000 because you're not going to buy a $15,000 Jacuzzi. You may or may not need the Jacuzzi, but it doesn't change the cost of the car.
      (4) Further, you raised the issue of F-35s while completely ignoring my comments. Sorry, but you come across as a Liberal supporter, not as someone who is simply looking in a non-partisan way at what is being proposed.

      Delete
    9. From a budgetary point of view, you are only increasing the size by $20K by not buying the Jacuzzi. At the end of the day, it's a question of what level of taxes we are paying, by shifting from military procurement to infrastructure spending, the total debt is adjusted down.

      While I support infrastructure spending, I don't support deficit spending, so this one is a toss up for me.

      Delete
  17. Eric, how come the CBC poll tracker has both the Conservatives and NDP with 31% for the latest Nanos poll (Sep. 12-14). The detailed report has the Conservatives with 31.0% and the NDP with 30.4%.

    http://www.nanosresearch.com/library/polls/20150914%20Ballot%20TrackingE.pdf

    ReplyDelete
    Replies
    1. Today, Nanos has the CPC in third again with 29%.

      Delete
    2. Twas a typo, I fixed it yesterday.

      Delete
  18. Again, I believe there's simply no way the Conservatives can get anywhere near the seats projected in Ontario as it stands right now. I understand the methodology, but things have changed greatly in Ontario since 2011.

    As disliked as Tim Hudak was, there may be more disdain for Western based Harper in Ontario. And that's without the anger for the scandals of the Liberals and Wynne for the Provincial election factored in, so Harper may be far worse off than Hudak.

    The memories of Bob Rae's NDP that drove so many right of centre Liberals to vote for the PC's, is not there anymore, and has been over ridden by the wariness of the Conservatives that is not necessarily listening to Ontarians.


    The Conservatives have extremely strong support right now in rural area ridings, but polls don't take this into account when factoring in city votes. The same thing happened in the polls before the Ontario election and one of the main reasons why the seat total projections were so off. With such concentrative Conservative pockets in farmlands Ontario, the cities become very vulnerable for the Tories if percentages remain the same.

    As someone's who's lucky enough to travel across the province for work, the signage, conversations and information seems to suggest that Conservatives will do well in their tradition strongholds in the 613, it's up in the air in 705, but the Tories are in huge trouble in the 905 and 519.

    But you don't have to take my word for it. There are many other factors, and even Conservative operatives know this is happening as there's now an excellent article about it now on Reuters. http://ca.reuters.com/article/domesticNews/idCAKCN0RF0D120150915

    As a former federal PC campaign volunteer, I must say I'm not surprised at by this as the Conservatives went too far right for Ontarians. I wouldn't be shocked if Harper gets around the same number of seats that Hudak did.

    ReplyDelete
    Replies
    1. Nanos has CPC third again today. 5 points behind the Libs in Ontario. 29% isn't even close to what they need. They can't seem to get anything over 31%. No growth.

      Delete
    2. I wouldn't describe the Harper Conservatives are particularly right-wing, at all. The Chrétien-Martin Liberals were more right-wing. Several provincial governments have been way more right-wing.

      Delete
    3. Jimmy, they don't need growth. They just need to suppress voter turnout for the opposition.

      Delete
    4. As much as Nanos is reputable, there is quite a bit of volatility in their polling, and Ontario is all over the map in polls. Hard to gather much from that, other than its a 3 way race.

      Delete
    5. I wouldn't describe the Harper Conservatives are particularly right-wing.


      Laughable statement the Reform Alliance is not particularly right wing. They are Tea Party disguised as PC's.

      Delete
    6. Nanos has CPC third again today. It is starting to look like a trend.

      Delete
    7. Jimmy yes but I think actually to the Right of the Tea Party !! And so begotten to Big Business as to be a danger to the country !!

      Delete
    8. The Tea Party is largely an interest group for seniors and racists. They talk like they're right-wing, but then oppose any cuts to spending if that spending benefits them.

      A right-wing party should support spending cuts nearly everywhere. A right-wing party should oppose the sort of piecemeal tax code that benefits big business. A right-wing party should oppose subsidies and marketing boards. A right-wing party would have killed the dairy board to get us that free trade deal with Europe.

      The Harper Conservatives have dramatically increased spending. The "stimulus" of 2009 never really went away - that just became the baseline for spending going forward.

      Canada doesn't have a right-wing party right now, with the possible exception of the Libertarians.

      Delete
  19. So Eric, is it accurate to say that NDP support has leveled off during the election? It started at 35% and has dropped to 31%, yes?

    ReplyDelete
    Replies
    1. I would say that the NDP actually started at 31.8% since, in the first week, there were only two polls done, one with them at 39%, which definitely stands out and looks like a statistical anomaly. That would actually make the NDP very stable throughout the campaign for the moment.

      Delete
  20. What I'm seeing from the Nanos daily polls is "twitching" !! Each day a different leader and the margin is 1 !!

    So a House divided by three seems likely .

    ReplyDelete
    Replies
    1. If the vote were today, sure.

      The vote isn't today.

      What we're seeing from Nanos is an excellent demonstration of what "margin of error" means.

      Delete
    2. More interesting is that Harper is still on top for preferred prime minister 31% to Trudeau's 25% hmmm. And that for every party the consider voting for went down. Does that mean people are more decided or more confused????

      Delete
    3. shoshana, you conveniently left out Mulcair at 29% - the margins are pretty miniscule, nothing to draw asafe conclusion from...

      Delete
  21. The newly released CRA poll could definitely throw a wrench into the Atlantic Canada narrative: http://cra.ca/wp-content/uploads/2015/09/15-3-Federal-Press-Release-208572309834598230682.pdf

    Of course, this data is a month old, where the Liberals have trended upwards and the Dippers have dipped since then. But there are some interesting things here:

    First, Liberal dominance in Nova Scotia, a complete erosion of Conservative support, and a maintenance (if not slight increase) of NDP support. This likely means that the Conservatives lose ALL their seats here, and that the NDP may have an outside shot of maintaining Dartmouth Cole Harbor based on incumbency, while picking up South Shore-St. Margarets.

    According to the poll, Tory support has completely collapsed as well in Newfoundland and PEI. But NDP support has also risen, making St. Johns South a slightly more safe option, with a victory even possible in Avalon if the Independent MP Scott Andrews siphons off enough liberal votes for an NDP victory (Still a longshot, and would still require probably another NDP surge in support). But PEI is interesting as well, suggesting that the NDP has a chance of picking up Charlottetown (if you do a proportional swing instead of a flat swing), while the Liberals should still probably pick up Egmont (though this is in doubt).

    The biggest surprise of the poll is definitely in New Brunswick, where the NDP leads with 35, followed by the Conservatives at 31 and the Liberals at 29. This still probably means Fredericton and Madawasha go safely to the Liberals. It also means Saint John's-Rothesay goes more comfortably to the NDP. But Fundy-Royal and even Southwest New Brunswick now has an outside shot to go to the NDP (still rather unlikely, truth be told, with these numbers - NDP would need more like an 8 point lead over the Conservatives to make this happen). Moncton-Riverview-Dieppe would also put the NDP in contest to win this, although the Liberals would still be favored here based on a flat swing. Miramichi would likely stay with the Conservatives as would of course Tobique-Mactaquac.

    But still very interesting. Conservatives still have some strength left in NB to make it a non-complete blow out, but a small swing to the NDP could give them a really good night. Liberals are still extremely strong in Nova Scotia. While an upsurge in NDP strength in NB, PEI and NL could make for an interesting finish.

    ReplyDelete
  22. Can anyone comment as to the Conservative performance in the tracker? They appear to be pulling in the smallest third of the vote, and translating it into the greatest seats? Talk about an efficient spread!

    And they're doing it without taking the biggest share of Ontario, BC or Quebec?! That strikes me as... mathematically remarkable.

    ReplyDelete
    Replies
    1. The Conservatives don't waste votes losing. In areas where they don't win seats, they finish WAY behind.

      Their vote is actually more efficient now than it was last election because it seems they'll be winning Alberta's ridings by smaller margins. There's no real benefit to getting 75% of the vote, as the Tories did in SEVEN?! different Alberta ridings in 2011.

      Delete
    2. What about all the seats the CPC won by a few hundred votes? There were plenty. They are all up for grabs now and likely not for the CPC. If the CPC can't get a 5 point lead with over 35% they are done.

      The opposition is going to break out one way or another right before the election or on elections day. Harper will need to combat that problem. That is going to sink him.

      Delete
    3. Close wins are efficient. Close losses aren't.

      The real wildcard is, I think, the Liberal support. It's so much higher than the last election that I'm not confident we can model it effectively. Look at a riding like Vancouver Quadra. Last time, the Liberals won it by a tiny margin. This time, Éric is projecting a total Liberal walkover.

      Is that much of a swing plausible? I don't know.

      Delete
    4. Eric has the LPC at 64,2% and the CPC at 18,3% while I have the LPC lower at 55,9%and the CPC higher at 24,7%. We may not exactly agree on the numbers, but in the end though, that remains a landslide.

      Delete
    5. I live in Quadra and have for many a year. I think the Tory numbers are low. Murray won by 4% in 2011. Redistribution has sliced off Shaughnessy (an upscale area) from the Riding but, demographics are changing and the large immigrant population are more Conservative voters than the previous generation. The NDP appear to be doing better than usual, judging by the sign count but, this is about the safest Liberal riding in the country and has been for the last 30 years-it's why Turner ran her in 1984, it was thought to be one of the few ridings the Liberals would win. I would think the Liberals will get close to 50% this time round with the Tory candidate in the 30% range.

      Delete
    6. I think my Liberal Quadra vote a bit optomistic. Joyce Murray will get 40%+ of the vote followed by the Tory who'll pick up close to 40%.

      Delete
  23. In the recent regional numbers, I'm seeing the NDP flat while the CPC and LPC show gains at the cost of the other. Interesting in that polls show very few CPC voters inclined to take a second choice while few LPC voters have the CPC as second choice.

    We are of course looking at very small shifts. In the bond market, these small percentages are worth serious money. With polls, I wouldn't bet more than a beer.

    ReplyDelete
  24. Um Eric, I don't know what Nanos poll you read this moring but the NDP are down 1.2 and the Conservatives are up 1.1.

    In party to consider the NDP is down 1.5

    And honestly these numbers mean that each night of polling is outside the margin or error swinging wildly back and forth for each party.

    A 1.5% percent change, which has commonly seen in Nanos polling in a single day is actually a jump of 4.% percent from three nights previous.

    A 4.5% difference will mean the difference between 3rd place and 1st place with possibly a majority (only for the Conservatives due to efficiency of their vote)

    So what can we take from these polls? (and oh yes Forum has the Conservatives well in front of the NDP this week)

    Either the NDP are in first place or third place. Either the Conservatives are in first place or in third place. The Liberals are probably in between in either of those scenarios or could come in second. Yep not much to go on.

    ReplyDelete
  25. Sorry Eric got it from their previous three day independent sample, but that includes up to 7 days ago. (6 nights) From 3 nights ago, night over night, Conservatives up 3.3 NDP down 3.6 to give the shift over a 3 night average seen today.

    ReplyDelete
  26. These rolling numbers from Nanos are fascinating. Are we seeing cyclical weekly fluctuations? Do people's voting preferences change depending how soon the weekend is?

    ReplyDelete
  27. Mainstreet released a BC poll as well today (Sept 17)
    http://www.mainstreetresearch.ca/battleground-british-columbia/

    ReplyDelete
    Replies
    1. Interesting poll, generally it shows what the poll averages show in B.C.: a tight NDP-Tory race. The Green numbers on the Island look a little low to me but, one poll doesn't tell us much.

      Delete
  28. And a new Environics poll proves the Conservatives are either in first or well back in third. Substitute other parties wherever you like.

    ReplyDelete
    Replies
    1. What stands out in that Environics data is the Green number.

      If it were an EKOS poll, that would make sense; EKOS has a long history of inflating he Green numbers.

      Delete
  29. Wish you'd just ban the usual suspects who transparently cheerlead for their own particular party in the comments. They're easy to identify, and it's very annoying.

    ReplyDelete
  30. Ekos hmm curiouser and curiouser. Well pollsters had better start applying likely voter models to their results and coming out with 2 numbers. It's all about GOTV and turn out at this point, and unless the Cons tick off their base before Eday that favours them.

    ReplyDelete
    Replies
    1. EKOS has long favoured the Greens, and historically they've underreported Liberal support to do it.

      It makes them a nice balance to Forum.

      Delete
    2. There isn't a go likely voter model for Canadian elections yet, so all you do is introduce a larger error. Eric has written about this in the past and it's why he ignores them.

      Delete
    3. I think it's fair to use past voting behaviour and questions about how to vote (aka where is your polling station, what ID do you need to vote), which is what is used in the US along with registration which is a non issue in Canada since you can register at the polls.

      There are likely voter screens used on polls that Eric is reporting, but the last one I read had a reported 85% absolutely certain to vote and we know that isn't true. It's more like 60%.

      We know absolutely without a doubt that younger people are not going to vote in the same proportion to their demographic and that seniors will outvote their demographic.

      Do the math on that in a 3 way tie. The Liberal vote may materialize at the polls but we already know the NDP vote will be lower and the Conservatives higher. It just will be.

      Delete
    4. As far as the 85% goes you have to remember the sample they get is composed entirely of people who are willing to answer a political poll and I don't think it's unfair to assume that the people who will spend time being polled about politics are more likely to vote then the people who just hang up the phone or say "No" when asked if they have time to participate.

      Delete
    5. shoshana, your basic falacy is using a US-style model for Canada. If we knew anything "without a doubt" there would be no need for polling or even elections. With time, we probably will evolve a decent likely voter model, but so far it's just increased the average error in the polling.

      Delete
  31. http://www.cjad.com/cjad-news/2015/09/17/justin-trudeau-in-trouble-in-his-own-riding-new-poll

    I know that polls commissioned by political parties aren't included but I was just wondering why Trudeau was so ahead in the riding projection despite him only winning by 8% (in the redistributed results) last election, is this mostly due to a star candidate modifier? Given that the Bloc have collapsed and Bloc votes tend to go more to the NDP than Liberals, is Trudeau losing his seat a realistic prospect with the current Quebec polling numbers?

    ReplyDelete
    Replies
    1. I have Trudeau way ahead in his riding too. The reason is, while the LPC dropped by 9,5% in Québec, he only dropped by 3%. Now, the LPC is bck up the 9,5% on last election, which means a lot of the gains should be made in his riding, especially since he's a party leader now. The Bloc is down 9,5%, which seems to have mostly gone to the LPC. Of course, there is a difference in regions as to where the Bloc's votes will go, but on the Montréal island, that would mostly be the LPC. And to top it up, the NDP is actually a bit lower than at the last election, by about 1%, so that makes them less competitive. Overall, this poll is... odd. While I could potentially conceive of a NDP victory in Papineau (by a very slim margin), an 11 point lead is, I believe, absurd.

      Delete
    2. No, it wasn't weighted properly. They oversampled 2011 NDP voters. If the poll was done properly, they would have sampled 28 per cent 2011 New Democrat voters rather then the 38 per cent they included in the sample.

      It's pretty clearly a propaganda poll, commisioned and released by the NDP. I think it's irresponable for major media outlets to report on party commisioned polls.

      Delete
    3. That isn't an indication that this wasn't a representative sample, people frequently misremember who they voted for. It happens on many polls. CROP is a reputable pollster.

      Delete
    4. Eric,

      Very good write up about this NDP poll! Well balanced reporting in my humble opinion. The poll and its serendipitous leak to the media certainly follow the NDP penchant for saying anything to get elected: fudge-it budgets, fast ferries, increasing spending without raising taxes, $15-a-day childcare (childcare being a provincial responsibility it is odd neither Mulcair nor his candidates talk about the need for provincial partnerships on this file). Us British Columbians have heard it all from the NDP-that is why we only elect NDP governments once every generation or less and it is why during the present campaign the Tories look like a safe bet for re-election albeit with a minority government.

      Delete
  32. IPSOS, Environics and latest Nanos have CPC in third. Harper was poor in the debates. Not great.

    ReplyDelete
  33. Eric, is there anywhere I can still find the regional breakdowns of the poll aggregation? I can't find it here anymore and it just seems to be the seat projections on the PollTracker.

    ReplyDelete
    Replies
    1. On the Poll Tracker, look at the first Poll Averages chart. There is a drop-down menu for you to choose one of the regions.

      Delete
    2. Thanks Eric. I actually found it a couple minutes ago, just before your reply. I was sure it was there, I don't know how I missed it. Gotta work on the old eyesight ... Thanks again for all this fabulous work!

      Delete
  34. Nanos has CPC third again today Friday. 29% isn't even close to what they need. They can't seem to get anything over 31%. No growth. Poor showing at the debate. Harper is looking like he is not making any gains. He is not going to make it this time.

    ReplyDelete
    Replies
    1. Who are you trying to convince?

      Delete
    2. No one.
      Just pointing out the CPC has no growth and are going to be in trouble once one of the opposition breaks out.

      Pretty sure the "Old Stock" comment didn't help either.

      Delete
    3. What term would you use to describe Canadians who aren't new Canadians?

      It's as good a term as any.

      As I've explained before, the CPC doesn't really need growth in order to win. They just need the other parties' voter turnout to decline. Since that's the primary effect of CPC advertising, and we haven't seen the bulk of the CPC advertising yet (and they have no incentive at all not to spend the maximum allowed), I find your conclusion premature.

      We also don't know what effect on voter turnout the details of the Fair Elections Act will have. Low information voters may well disappear from the voting population as a result.

      We don't know. You're jumping to conclusions.

      Delete
    4. I sure would not call them Old Stock.

      Low information voters vote for code words like Old Stock, but Harper has now angered his Old Stock base by speeding up the refugee process. No win for Harper.

      Delete
  35. Well the economy debate is over and essentially nothing changed. Definitely no clear winner though I felt Trudeau came across best and Harper did nothing but spew the party line and where Mulcair was is a big question.

    ReplyDelete
    Replies
    1. The "Old Stock" comment was really bad too. Harper lost a big opportunity to stand out as fiscal superior and he did not do it.

      Delete
    2. Peter, shouting over others reflected poorly on Trudeau. That type of behaviour is more at home in a high school gymnasium setting.

      Glaring contradictions in his debate argument did not help Trudeau either. At which point he was encouraged to 'pick one' side of argument or the other. 'You can't do both'.

      Raising one's voice in a stacatto delivery, does not lend itself well to effective communication and winning a debate.

      Truly, mature leadership may come with more experience: another decade, istm. Until then Trudeau would be a national leader by brand name only: driven behind the scene by old guard liberals who would, effectively, be back in power, side-stepping due electoral process.

      Our 30+ member third party offers little more, istm,
      than an attempt to masqarade it's way into a national goverment power by way of an inexperienced junior member, powered by the "brand" cachet of the fathers' last name and the old guard hiding in the shadows. Canadians do not need this as much as we do need the return of Stephen Harper as PM.


      Delete
  36. We've seen nearly two months of polls where nearly all of the movement amounts to statistically insignificant noise.

    The electorate is acting like the campaign hasn't started yet.

    ReplyDelete
    Replies
    1. True. And when they do, Harper is done. He has no growth and one of the other parties will break out.

      Delete
    2. I must admit, I am also beginning to think all the polls across the board amount to is statistical noise. What is the electorate thinking? Confusion I imagine. What does that translate into? My take is that when people are confused they don't show up or go with the status quo, and both those scenarios benefit the Conservatives. But I could be wrong.

      Delete
    3. Its the CPC that will stay home. Harper has disappointed many conservatives. They won't vote. The voters are waiting to see which opposition party breaks out. Many are waiting for late in the campaign to decide who will beat Harper.

      Delete
    4. In some sense, the campaign hasn't started. The Conservative Machine is tried and tested and will definitely drive a larger proportion of its voters to the polls than the other parties. Conservative support amongst the elderly who actually get out and vote should not be discounted and they'll do slightly better than polls currently show. That said...

      Anecdotally, I'm hearing an absolutely unprecedented number of "anyone but..." voters. This mad-as-hell electorate is loosely parked with the Grits, Dippers or Greens for now. However, they are ready to answer the call of a light-from-above inspirational utterance or (more likely) desert in droves after the game-changing bozo eruption that will be identified as the turning point in years to come.

      That bozo eruption may positively affect Green outcomes, especially in BC if the loony moment belongs to the NDP. Its main effect, though, will be to kick off a self-feeding avalance of ABH voter to either red or orange. When it starts, we'll see a repeat of 2011 written large. That will swamp any minor efficiencies in the blue machine.

      Either that or there will be no unforced errors from here to election day and we'll remain balanced on the knife edge that leads to a minority parliament. If history is any guide, though, the safe bets are on the bozo. That's why Trudeau and Mulcair will play increasingly more cautious games with each passing week. They only wish that they could control their candidates equally well.

      Delete
    5. No Jimmy the polls all tell us that the Conservatives have the most committed voters. People who are confused or half hearted are much less likely to vote.

      Delete
    6. John I've been amazed by the number of people who I've talked to and who bluntly say, without being asked "Anybody but Harper" !!

      I don't think a lot on here realise the size of this movement !!

      Delete
    7. Most committed but no growth potential above the core. Recipe for failure. They can't win without growth potential and they don't have it.

      Delete
    8. Peter, I imagine you live in a city or work in a city larger than 100,000 people, that is in Ontario, or is Vancouver. It's anecdotal.

      Jimmy the growth potential for Conservatives is in their turnout and support for their leader.

      I direct you to the Nanos numbers on preferred Prime Minister. Stephen Harper is the only one whose number is close to the number for his Party.

      Trudeau's number is 5% off. He only gets 25% support and Mulcair's is 2.5% off. That means there's 5% Liberal vote that may not show up to vote off the bat, because they don't want Trudeau as PM. The weakest candidate for his party is not Harper. It is Trudeau followed by Mulcair.

      Delete
    9. shoshana try a small Ontario town of about 5000 in a solid Conservative riding. The winds of change are blowing !

      Delete
    10. They had committed voters in 2011. They now have no growth from the already committed and will lose all the soft vote that put them over the top in 2011. They will lose accordingly due to no growth.

      Delete
    11. John,

      I and many door to door canvassers are seeing a HUGE anti Harper anger across the country. Harper is in for a big surprise when the anger toward him is felt at the ballot box and all he has left is his committed base. That is why the polls show no growth. He is done. Change is coming on way or another.

      Delete
    12. Which town, Peter? How did they vote last election?

      We can find out. Elections Canada makes poll-by-poll voting results freely available. The riding might be Conservative, but that town may never have been.

      You're drawing conclusions based on unnecessarily incomplete information.

      Delete
    13. Wow Jimmy. Can you see the emotion clouding your judgement in your posts? One of the most vivid memories of the last BC election was the NDP commentator on the CBC who refused to believe even the voting polls as they were coming in and even as the election was being called as a clear majority for the Liberals was still saying it couldn't be possible.

      I'll give you another take on your anger at Harper at the door scenario. There are a whole bunch of quiet people out there who are could vote Tory (I don't know if they will) but I know they certainly wouldn't be able to voice that opinion up against the irrational and at times very disrespectful anger you cite. Yeah it's there. I'm not sure it helps your cause but I l know it keeps others with differing opinions quiet. I hear different stories from people, and I'm in Vancouver.

      Delete
    14. No Ira I am not making wrong conclusions. This town and this riding has been solidly Tory since about 2000. Get used to the coming change !

      Delete
    15. Wow shoshana. Can you see the emotion clouding your judgement in your posts? You think that magically voters will stop disliking Harper. He has angered most of the soft swing vote.

      I have no anger myself for Harper, its at the door and it ugly and prevalent. There are also a whole bunch of Con voters out there who could vote Tory but are staying home because Harper betrayed them and they won't vote for anyone else. Harper is in big trouble and he knows it. That is why his statements are getting more desperate. He has no growth and he will not PM in November.

      Delete
    16. Have you run the numbers, Peter, or is your evidence anecdotal?

      You're allowed to prefer Harper be defeated. I don't mind. He hasn't been anything like the Prime Minister I wanted him to be.

      But I want you to make data-driven evidence-based decisions. I want everyone to do that. I want whoever becomes Prime Minister to do that.

      Delete
  37. The Forum poll for Ajax mentioned on 19 Sep (http://poll.forumresearch.com/data/35b2a36c-15e6-4a28-a5a5-2cf75ffb6a90Ajax%20News%20Release%20(2015%2009%2011)%20Forum%20Research.pdf) has the wrong name for the Green candidate - very odd.

    ReplyDelete
  38. Hi Eric,

    I think you are doing great work. I know you are not a pundit, but I have noticed a trend on your riding projections with Thomas Muclair and his rating going down quite dramatically from where it was just a week or so ago. I am wondering with all the data and polls you are looking at, what is the odds that he could end up losing his seat?

    ReplyDelete
  39. Dear Eric,

    Here is a new Spadina-Fort York poll that may be factored into your riding projections:

    http://www.cp24.com/news/olivia-chow-s-lead-narrows-in-fort-york-spadina-new-poll-suggests-1.2592007

    ReplyDelete

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.