Friday, September 25, 2015

2015 Federal Election Link Round-up: Week 8

Less than four weeks to go! Here are the unweighted weekly averages:

Week 8: CPC 30.8% LPC 30.7% NDP 28.7% GPC 4.9% BQ 4.0%
Week 7: NDP 30.7% CPC 30.0% LPC 29.8% GPC 5.6% BQ 3.3%
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 25, 2015

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

- Riding polls in Manitoba and Saskatchewan from Mainstreet.

- My analysis of the latest polls from yesterday. Is the NDP slipping? Nanos today argues against that, but Forum argues for it. The debate last night may make the question moot.

- I was on Power and Politics last night talking about the state of the race and the odd discrepancies in the polls yesterday.

- Nanos and Forum today show the same three-way race that they've been showing for awhile now. And an Alberta poll from Insights West with some interesting regional breakdowns.

Thursday, September 24, 2015

- It is pollmageddon today. Here's Nanos and Léger, showing the same old race. Here's EKOS, showing something unrecognizable. Here's Mainstreet in Ontario, and a few places in Quebec. And here's Forum in four ridings in Toronto.

Wednesday, September 23, 2015

- The Poll Tracker has been updated, and the riding projections will be updated shortly. In the meantime, check out the new feature Stephen McMurtry has added to the interactive projection map. It allows you to toggle between each party's high projected range, giving you an idea of what the map would look like if each party maxes out its potential wins. These numbers represent the likely ranges, and not the max ranges.

- My next regional spotlight is on southwestern Ontario.

- Your daily Nanos. Compared to the previous independent three-day sample, the Liberals are up a bit, the Conservatives are up a little, and the NDP is steady. It's the dance of the thousandth percentage points.

Tuesday, September 22, 2015

- On the latest episode of the Pollcast, Phil Authier from the Montreal Gazette joins me to talk about that whole NDP Papineau poll.

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

- Yesterday's polling panel with Dimitri Pantazapolous on Power and Politics.

- Polls from Forum and Ipsos Reid yesterday, and Nanos this morning.

Monday, September 21, 2015

- I took a stab at gauging the race in terms of whether opinion was firming up or not. It's not.

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

- Your daily Nanos. Compared to the previous independent three day sample, the NDP is down about two points, the Liberals are down a little less than that, and the Conservatives are up about two points. Taking the entire campaign in three day blocks, the Tories have registered, from start to today 26%, 31%, 31%, 29%, and 31%. For the NDP, it has been 31%, 30%, 30%, 31%, 29%, and for the Liberals it has been 33%, 32%, 30%, 31%, 29%. Not a whole lot of movement, except maybe a swing from the Liberals to the Tories in early September.

- A few Atlantic Canada riding polls from Mainstreet Research.

- My second regional spotlight went up Saturday. This time the focus was on Saskatchewan and Manitoba.

- I was on The House Saturday, talking about debates and the economy.

217 comments:

  1. As usual Nanos has it wrong . CPC trails not leads

    ReplyDelete
    Replies
    1. And you know this how...?

      Delete
    2. How exactly do you know this? Or is it just a gut feeling?

      Delete
    3. Just because you want something to be so, doesn't make it so. Also, for one pollster to have one party ahead by one point, and another pollster to have a different party ahead by one point, means the two polls are in agreement. It's essentially a three-way tie, and has been for most of the campaign.

      Delete
    4. This comment has been removed by the author.

      Delete
    5. People post on here all the time about how x pollster is wrong...and their preferred (assume they like the numbers better) is more accurate. This is nothing new

      Delete
    6. There is a difference between declaring a pollster wrong and recognizing the historical bias that comes with their numbers, which is generally a by-product of methodology.

      Delete
    7. Sigh, wanting the Tories to tank does not make it so. I'm voting Liberal this time but I don't have partisan blinders on. Any of the three major parties is poised to eke out a minority at this point.

      Delete
    8. Valid point, Ryan Brooks. While we can discern some general trendlines that have changed since 2011 (Conservatives down, Liberals up, etc), recent election results should serve as a lesson that none of us really knows what will happen. The picture won't be clear until at least early October, or maybe mid-October.

      Delete
  2. Just as a matter of being careful to couch language in accurate terms, you write: "Not a whole lot of movement, except maybe a swing from the Liberals to the Tories in early September." The data accompanying this shows an NDP drop of 1.8%, a Liberal drop of 0.7% and a Conservative gain of 1.2%. Given the Conservatives gained more than the Liberals dropped, either your statement is somewhat inaccurate, or there is other data that you could helpfully point us to.

    ReplyDelete
  3. Peter, Nanos has the best record for polling accuracy. The amount that the CPC is up is not really significant unless it is the start of a trend.

    ReplyDelete
    Replies
    1. Actually, Angus Reid had the best results in 2011.

      Delete
    2. Earl's probably remembering 2008, when Nanos was best.

      Delete
    3. Actually, while Nanos was the most accurate in the 2004-2008 period, it was NOT the most accurate in 2011. So we don't know which pollster will be the best this year until the night of October 19.

      Delete
  4. The first numbers after the economy debate give a small edge and quite a few seats to the CPC.

    ReplyDelete
  5. Nanos poll weighting does not appear to be consistent, at least in terms of how Poll Tracker is presenting the respective poll weightings. Prior to the latest update, the NDP and Cons had the same seat projection - and now a 16 seat shift in favour of the Cons. Either get rid of the "one of a kind Nanos polling methodology" or show some consistency in respective weightings.

    ReplyDelete
    Replies
    1. The weight of the aggregate to recent polls and the fact that Nanos polls daily now...means the aggregate is dominated most influenced by Nanos daily polling more than anything else.

      Delete
  6. So basically a three way tie. It will either stay this way and Harper will be out with a loose coalition or he will be out because one of the others break out. The Conservatives have no growth so Harper's only chance is to hope for record low voter turn out. I think the Conservatives are in big trouble or they'd be in the lead by now.

    ReplyDelete
    Replies
    1. The vote is still 4 weeks away. In a campaign of normal length, we'd have just started.

      Do you usually call the election 4 weeks out?

      Delete
    2. 5 weeks was the "standard" time for the last few elections, IIRC. (Or 36 days, given that the PM traditionally visited the GG on a Sunday.)

      Delete
    3. Liberals pull ahead of NDP, as support in Ontario grows, CPC third: Ipsos poll

      Harper has no growth and is in big trouble.

      Delete
    4. "Do you usually call the election 4 weeks out?"

      Depends on the situation, but I certainly called Prentice in trouble before he called his election in Alberta and was extremely surprised when he went ahead and did it anyway. Who was going to defeat him and by how much wasn't clear. That he was grossly out of touch was clear by his absolute refusal to do anything on the taxation rate for corporations even as he upped taxes on everything else.

      In like manner, I have called for a long time that Harper will be out since (a) he's been tone deaf since his 2011 majority (b) there is no BQ factor any more and so one way or another the Liberals and NDP will come to some tacit agreement to kick the Cons out. To do otherwise is to invite political retribution in the next election which is likely in 2017 at the latest.

      Delete
    5. I might add that a lot of the noise is coming from one polling firm that keeps asking the same people the same question. If it's not a new sample it shouldn't count.

      Delete
    6. So polls that show the CPC sliding are wrong or bogus. Harper can't get above his ceiling of 30% because he has no growth. He is done.

      Delete
    7. I think you should be cautious about ruling Harper out. He wouldn't be running again if he thought he didn't have a chance. He still looks confident & energized and his own numbers have stabilized. While the Conservatives are down from 2011 and the Liberals are up, we don't know for sure what the next Parliament will look like.

      Delete
    8. The people of this thread will be very surprised at the results come election day. The idea the Tories have a 30% ceiling is so arcane it is bizarre perhaps even delusional. Here's a News Flash: The NDP is polling at 25% in English Canada!

      Delete
    9. "The people of this thread will be very surprised at the results come election day. "

      Or not.

      Delete
  7. A good question is why did the Conservatives shift from 3rd to 1st. Right now most of the shifting is just statistical noise but if one is looking for reasons, the past bit has seen a lot of talking about stuff that many 'on the left' see as 'racist' but 'on the right' would see as common sense ... 'old stock', fewer refugees, stronger security, etc. I wonder if more soft liberals are shifting Conservative now over that stuff. Trick for Liberals is not to lose too many that way and gain more from the NDP.

    ReplyDelete
    Replies
    1. Forum just released firms up the trend, which is saying something, considering it's Forum, and how far off the rest they were in early days.

      Delete
    2. With 7 weeks behind us, but still 4 more to go, it's hard to say.

      Is it fatigue? Are people losing interest in the election?

      Or are they just becoming interested now?

      Are those two answers even the same region to region?

      If this campaign ever makes sense, it will likely only be in hindsight.

      Delete
    3. Its not that big a shift. It's still just noise at this point. It has to be maintained or increase to not be noise.

      Delete
    4. Harper has no growth and is slipping. The Libs are beginning to pull away. No growth for Harper spells defeat.

      Delete
    5. "A good question is why did the Conservatives shift from 3rd to 1st."

      No one shifted from anything to anything. Almost all movement has been well within the margin of error, and that's only at a 95% confidence level; given the number of polls, 99% would be warranted, and then all are equal and all movement is within the MoE.

      Delete
    6. JImmy - the most recent Forum poll notwithstanding?

      Delete
    7. Forum poll is already stale. New numbers show Harper slipping below his ceiling of 30%

      Delete
    8. The ceiling he routinely exceeds in the Nanos numbers?

      Why are you so eager to mischaracterize the poll results?

      Delete
    9. Ira, he's irrationally ABC. Just take the liberal blather and move on, like we do with shoshana for the other side.

      Delete
  8. With the average published today (for the seventh week), my model gives:

    133 CPC (+9)
    114 NDP (-6)
    90 LPC (-3)
    1 GPC (even)

    By region, it gives:

    Atlantic (even)
    20 LPC
    7 NDP
    5 CPC

    Québec (even)
    60 NDP
    12 LPC
    6 CPC

    Ontario
    55 CPC (even)
    43 LPC (+1)
    23 NDP (-1)

    Prairies (even)
    16 CPC
    6 NDP
    6 LPC

    Alberta
    31 CPC (+4)
    2 NDP (-2)
    1 LPC (-2)

    British Columbia
    19 CPC (+4)
    15 NDP (-4)
    7 LPC (even)
    1 GPC (even)

    Territories
    1 CPC (+1)
    1 NDP (even)
    1 LPC (-1)

    ReplyDelete
  9. I would advise you update your numbers to incorporate the Ipsos poll that came out after 4 pm. That provide a better reflection of current voter intention.

    ReplyDelete
    Replies
    1. Huh? Ipsos is showing the same statistical three-way tie that Nanos is. Only minor change is NDP down within margin of error and Liberals up within margin of error. How is it a better reflection?

      Delete
    2. Why would you believe that an online panel that asks the same people each week the same question would not be skewed? This is not a random sampling of the Canadian population and the same error is repeated each week instead of asking new people erasing a possible non representative sample from the previous, they keep asking the same people who are obviously biased. They are coming in 4-6 points off other polling firms and its not surprising. Online panels are not polls.

      Delete
    3. I wouldn't say that the Ipsos poll is garbage, but I would point out that the numbers for Alberta (CPC at only 39%) just don't make sense.

      Delete
    4. Which just washes out the Forum poll. Methodology is important in that discussion however, and Eric is more qualified than you and I to comment on that.

      Delete
    5. I find it difficult to take Ipsos seriously when the statement on GlobalNews was the demonstrably false "It’s the first time since the campaign began that the NDP hasn’t been the most popular party. "

      Delete
    6. shoshana, you seem to be willing ignorant of how actually online panels work. It's not the same people each week. The panel has tens of thousands of members, and a small percentage are asked each time.

      It is equal to saying that telephone polls are biased because the same people are willing to answer their telephones each week.

      Delete
    7. Yeah their numbers don't make sense. I don't know if they will actually conduct a real poll right before the election, (probably) or they will increase how many people are interviewed each week and lessen the panel numbers, but to count these numbers is absurd.

      They will change their methodology and then claim their polls are fine. Or they will start changing their sample. They have to.

      Delete
    8. Ipsos is only referring to their polls when they say that. And in their dataset, they were right.

      Delete
    9. Mapleson, around 20 million Canadians have phones, very few are a part of Ipsos' online panel. And yes it is the same people every week. That's why its called a panel. Some people may choose not to answer in a given week and/or may be factored out to meet census data, but it is the same people. There may be a couple of thousand people on the online panel and much less per province like Alberta or BC. So its a sample of a sample that remains the same that was not randomly generated but self selected. You have to be willing to answer a poll a day for a long long time before you are "selected" to be in the online panel. it is not your average Candian.

      Delete
    10. There are 150,000 members of Ipsos's poll (so you have a 1/150 chance of taking part), and Ipsos also does 'river-sampling', which comes from outside the panel.

      Delete
    11. Ipsos is called garbage by those who are biased and don't like the results. It is just as legit as any other.

      Delete
    12. Telephone polling response rates are around 3% to 5%, which gives a total pool of respondents around 600,000 to 1,000,000. Compared to the Ipsos panel (over 200,000 members) it's not too different of a candidate pool.

      The chances of being picked for any given day are under 8.6% and the chances of being picked twice is 0.7%.

      Delete
  10. Is there a graph anywhere like the Poll Tracker, but showing semi-transparent colours to represent the margins of error? Rather than everyone looking at the thin lines squiggle about, maybe we should see how much the broad bands have been overlapping each other throughout the campaign.

    ReplyDelete
    Replies
    1. There's actually one of those (sort of) on the Wikipedia page tracking the polls for this campaign.

      Delete
    2. https://en.wikipedia.org/wiki/Opinion_polling_in_the_Canadian_federal_election,_2015

      Dom

      Delete
    3. Sweet. Dom/undermedia and Ira, thank you very much! (I tried to make something like this myself in Excel, but failed miserably. :)

      Delete
  11. Approximately 33 per cent of Canadians prefer the Liberal Party, according to the poll – three percentage points ahead of the NDP, and six points ahead of the Conservatives.

    The Liberals enjoy 41 per cent support in Ontario, a sizeable lead over the Conservatives

    A crushing blow for Harper
    The Liberals’ bump in the polls follows Trudeau’s strong showing at the Globe and Mail debate on the economy last week. Most respondents to the Ipsos survey say Trudeau, despite lower expectations of him than the other two, won the debate.

    No growth for the Conservatives is now killing them.

    ReplyDelete
    Replies
    1. I like your enthusiasm Jimmy, but the LPC needs to continue the numbers for at least a week for it to a legitimate shift.

      Delete
    2. But how many prefer Trudeau? And therein lies the Liberal problem. All the polls that ask have him 4-5% below his party in support. That's not good.

      Delete
    3. Whats really not good is the CPC have no growth. They are stuck at 30%. No one prefers Harper beyond the base.

      Delete
    4. The problem for Harper is that he has no growth. He can't get a majority. Without a majority he is done. The opposition is going to push him out.

      Delete
    5. shoshana, it's because of Trudeau that the Liberal Party is competitive again. If it wasn't for him, they wouldn't be in contention at all. You're misreading the poll numbers.

      Delete
    6. Which makes no sense to me. I dislike nearly everything Trudeau says. From his muddled stance of deficits, to his care bear economics, so his short-sightedness on defence procurement, to his incoherence on C-51.

      I intensely dislike Justin Trudeau. I would choose Gilles Duceppe to be Prime Minister before I would choose Justin.

      Delete
    7. But Ira that's been so obvious from Day One. !! Let's face it you are severely blinkered.

      And Justin was right on the defence issue as a den Tandt piece in today's National Post shows!!

      Heart of conservatism saying Justin is right !! WOW !!!

      Delete
    8. There's no benefit at all to declaring, right now, that we'll abandon the F-35.

      We're committed to buying zero planes.

      Trudeau's stance on the F-35 smacks of Chrétien on the EH-101. Chrétien won that election, but cancelling the EH-101 was a terrible decision.

      But at least Chrétien actually saw a real thing he was trying to avoid. Mulroney had committed to buying the EH-101, so Chrétien had to pay a penalty to get out of it.

      Not so with the F-35. Harper has committed to buying ZERO planes. So why the rush to cancel the program?

      Mulcair's position on this makes vastly more sense (as it does on most issues when compared to Trudeau).

      Delete
    9. Politics, unfortunately, is rarely about what makes the most sense, but rather what makes the largest emotional appeal. It's the fundamental flaw in representative democracy and the human condition. That's why smoking is still a thing.

      Delete
    10. Jimmythedeke, says "Harper has no room for growth" yet, the Conservatives are up 4.6% points in Ontario since July according to the polltracker, the Tories are up 3.3% points in BC since July according to the polltracker and the Conservatives are up 5.4% point in Alberta since, July.

      Jimmy, your hypothesis simply is not born out by the facts (data) which show Conservative support rising. Harper is well on his way to re-election.

      Delete
  12. Looking at he Poll Tracker update, this is where moving support to the Liberals helps reelect the Conservatives. The Liberals are running a very effective Conservative reelection campaign.

    ReplyDelete
    Replies
    1. Harper has no growth. He will get minority at best and both opposition leaders have now vowed to bring him down immediately. Harper is done.

      Delete
    2. To claim the Liberal surge will help the Conservatives is blatantly false.

      For instance, the Liberals are surging in Ontario. This support is concentrated in suburban GTA and the Ottawa area where the NDP is traditionally weak. The NDP is a non-factor in ridings such as Vaughan Woodbridge, Ajax or Nepean.

      On the other hand, the NDP lost some ground in BC to the Conservatives.

      Delete
    3. Geenrally speaking, if he Liberals ride high, hey take from the NDP, but no enough to defeat the Conservatives, who strengthen relatively to the Liberals and NDP in such circumstances. The NDP, with a broader base, if riding high, are stronger against the Conservatives.

      Delete
    4. Unless the NDP shows it can beat the Conservatives in Ontario, I don't see how Liberal growth in Ontario helps the Cons. Most experts say the Cons fear Liberal growth in Ontario.

      Delete
    5. No one is likely to win a majority, so : Liberal strength in Ontario - and elsewhere - largely comes at the expense of the NDP, but because the NDP has the overwhelming lead in Quebec, the Liberals can't quite eke out a win, preventing an NDP win, but ensuring one for the Conservatives

      Delete
  13. As of Thursday, September-17 ...
    ********************************

    46% chance NDP will win the most seats.

    36% chance Conservatives will win the most seats.

    18% chance Liberals will win the most seats.

    ..... (please don't shoot the messenger :-)

    ReplyDelete
    Replies
    1. ** correction: previous post should state -
      -------------
      ' 39% chance Conservatives will win the most seats. '

      Cheers all.

      Delete
  14. The real trend that is not being reported on is that since the election was called, the NDP has gone from 35% to 29% and the LIBS have gone from 27% to 31%. The CONS have not budged. The bottom line is that voters are linking up behind Trudeau as the anti Harper candidate.

    ReplyDelete
    Replies
    1. Depends on the part of the country.

      In Ontario, the Liberals are generally seen as the clear anti-Conservative choice. The exception is Southwestern Ontario, which is currently a wasteland for "big-L" Liberals.

      In Quebec, BC or Saskatchewan, the NDP is seen as the anti-Harper choice. Of course, there are pockets of Liberal support in those provinces too.

      Delete
    2. Big J, Interesting point. So what happens somewhere like Guelph where the Conservative vote share is down, but so is the Liberals (they won this riding in 2011). Does the split of the Liberal vote off to the NDP (who stil may not hav enough to win) mean a locally popular Conservative Candidate, Gloria Kovacs, who also ran in 2011, mean she wins?

      Delete
    3. But if the NDP contains to level off Nationally, and remains behind in Ontario, Quebec journalist Chantal Hebert says its possible Liberal (or BQ) support could rise in Quebec at NDP expense.

      Delete
    4. shoshana, Guelph has been Liberal since 1993. There is a strong Green base, so the Liberals can win so long as they stay proportionally above the CPC. It's about equal to May losing her seat in BC.

      Delete
    5. There is a small chance for the Conservatives to take Guelph with less than 30% of the vote if Liberal/NDP/Green vote was split perfectly amongst each other.

      But Liberals are polling better in Ontario in recent weeks - so its that scenario is unlikely.

      Delete
  15. I want to see what effect the Sep 18 Forum poll has on the tracker once Éric adds it.

    ReplyDelete
    Replies
    1. 30.3% Libs
      29.7% CPC
      30.5% NDP

      This stalemate ensures no majority for the CPC and the opposition has vowed to push Harper out. The two together would have nearly double the seats of the CPC.

      Delete
    2. Jimmy you know that all pushing him out means is calling another election, and the GG may not agree to that if there is no meaningful shift in public sentiment aka another three way election. Whoever has the most seats will govern for at least 18 months.

      Besides the government can only fall on the budget and the Conservatives will wait until one of the parties is lower in the polls to bring one in and they will have to support it.

      If the Conservatives win the most seats they will govern just like any of the other parties.

      Delete
    3. No, shoshana, there's no need for another election.

      The problem in 2008 was that the opposition voted for the throne speech, and then they said they wanted to defeat the government. The proper procedure would have been for them to defeat the throne speech, which would have led to Prime Minister Dion and Deputy Prime Minister Layton.

      If Mulcair and Trudeau vote down Harper's throne speech, they'll be offered a chance to govern. That's how the system works.

      Delete
    4. Shoshana, there is no basis in law for your scenario. The GG does not have the authority to allow a party to govern without the confidence of the House. He/She can ask another party to try and earn the confidence of the House or it's time for another election, regardless of public appetite and likely outcome (which is a silly notion, since the outcome is only predetermined if it's rigged).

      Delete
    5. That's not factually accurate. They could fall on the Speech from the Throne (or any number of confidence motions). And if the GG does not agree to another election the only alternative would be to ask the leader of another party to form government.

      Albeit I think a King-Byng II is unlikely and no party will be in eager/able to conduct a 2nd election so swiftly so I imagine Harper would be able to muddle though for at least a little while but he'd have to put through some manner of supply bill if not a full budget so it's not like he'd have the ability to just dictate when.

      Delete
    6. shoshana, that is incorrect. Both Martin in 2005 and Harper in 2011 were defeated by non-confidence motions separate from budgetary bills.

      Delete
    7. JF,

      Quite right. The Crown needs supply. At some point Harper or whoever is PM will need to present a budget. However, one could govern without a full budget using last years number through the use of Royal Warrants for six or twelve months but any longer than that and the numbers will start to go wonky and a new budget required.

      Delete
    8. Half the reason for this long campaign is to exhaust the financial resources of the other parties. A Harper minority would want a new election as soon as palpable to the public.

      Delete
    9. Parliament is required, per the Constitution, to meet for at least one day per year. Therefore, any attempt to evade Parliament is constrained by that limit.

      Delete
    10. Yes Paul, but remember, Parliament has already sat this year.

      Delete
    11. Let me be more specific: there can be no more than 12 months between meetings, which means Parliament must next sit before June 19th, 2016.

      Delete
  16. Power And Politics Ballot Box Question

    September 21, 2015 we asked: Have you decided who to vote for?

    Here are the results:

    Yes: 694 votes (83%)
    No: 128 votes (15%)
    Not sure: 13 votes (2%)

    So that gives the lie in a way to all the polls !!

    ReplyDelete
    Replies
    1. The Ballot Box is not a poll, it is a survey of viewers.

      Delete
    2. In that case it was a poll Eric, sorry.

      Delete
    3. It did not have a random sample, and so is not meaningful.

      Delete
  17. Its painfully clear one of two things are going to happen. Either the CPC win a minority from a three way split (they have no growth, so a majority is completely out of the question), or, one of the opposition breaks out and wins a minority.

    I think the CPC will win a razor small minority and Harper will be swept out the door by a cooperative opposition who have pledged to do just that.

    ReplyDelete
    Replies
    1. It is not at all clear. The data support no conclusions as of yet.

      Delete
    2. Data shows clearly that the CPC has no growth and is stagnant.

      Delete
    3. That's a very different claim. And arguably true, depending what you mean by "has no growth". They've shown no growth.

      The data cannot show that it isn't there or that it won't happen, but the data can show that it hasn't happened, and it can fail to show that it is there, both of which are true.

      Delete
  18. I fail to understand the difference between No I haven't decided yet and Not Sure?

    ReplyDelete
    Replies
    1. Both are the same. Both are waiting to see who to vote for to topple Harper.

      Delete
    2. It's similar to the difference between atheists and agnostics. One is sure there is no God; the other is uncommitted on the existence of God.

      Some people are sure that haven't decided while others may have a preference, but not yet sure if that's what they will decide.

      Delete
  19. Update:
    --------
    41% chance NDP will win the most seats

    39% chance Conservatives will win the most seats

    23% chance Liberals will win the most seats

    18% chance the three main parties will win 100 seats each

    Cheers all.

    ReplyDelete
  20. I don't like the way Nanos, because of its rolling poll format, crowds the apparent results, shifting perception of the voter trends. Thoughts?

    ReplyDelete
    Replies
    1. I agree. I don't understand why they don't just release a new poll every third day. Most likely it's to get the page hits and views. I know it's the first thing I do when I wake up in the morning is check Nanos.

      Delete
    2. Same. It's mostly statistical noise that just gets presented as trends because two days in a row show growth for a party... but that is the same sample! *Groan* Even if we take only the independant 3 days sample, it's so close to each other that it doesn't represent much of anything. A poll a week would make more sense in my mind...

      Delete
    3. I looked at the list of all polls with Nanos removed entirely, and the impression one has of the polls is quite different, with only the last Ispsos, the last two Forum polls (conducted within three days of each other), an Ekos poll from (8 Sep) and the Mainstreet poll of 11 August showing anyone but the NDP leading. That's 5 polls out of 28. To be fair, I'll have to go back and put in every third Nanos poll to see how that looks... Impressions are a big part of party politics...

      Delete
    4. That's what I do. I've been maintaining a chart of all of the publicly released polls since 2005, and with overlapping rolling polls like this I always use only discrete samples.

      Delete
  21. "Less than four weeks to go!"

    Thank goodness! ;)

    Isn't this the longest federal campaign in 100 years? Normally it would already have been over by now.

    I'm wondering what toll it is taking on the candidates out there? They don't normally have to run for this sustained length of time.

    ReplyDelete
    Replies
    1. Actually, this is the longest federal campaign in Canadian history (78 days) barring the second, where the voting period was spread over three months. The next longest between dissolution and election was after the King-Byng Affair in 1926 (74 days) or 1980 with 66 days between writ and election.

      Delete
  22. My apologies - a slip of the mouse means I deleted the comments that were posted in the last hour and awaiting moderation.

    ReplyDelete
  23. Interestingly, as of today's adjustments, the NDP is down 5% from the start of the campaign, the Conservatives are exactly where they were, and the Liberals are up 6%, and have risen steadily through the campaign. And yet... in all the analysis that I have seen, this single salient fact seems to get no mention. In fact, the only mention this week Eric still has you speculating about a shift from Libs to Tories in early September.

    Personally, I think it's too early to draw any conclusions whatsoever. But many have been drawn, nonetheless, just not the most obvious one. I think this is where all of the observers have been dealing with unconscious "Just Not Ready" bias created by all those ads.

    ReplyDelete
    Replies
    1. The start of campaign polling numbers probably were a statistical exception and really is an arbitrary line in the sand.

      Delete
    2. This is exactly the point I made a few days ago. It stands out quite blatantly in the charts Eric shows and yet he has not commented on this clearly identifiable trend.

      Delete
    3. One problem with that narrative is that there have only been 6 polls that showed the NDP at 35 or better. 4 were from Forum and had them between 36 and 40%. Everyone else (with the exception of 2 polls) has had the NDP between 29 and 34%. Early in the campaign, there were fewer polls, so the Forum polls would be weighted more heavily and would have pushed NDP numbers up.

      Now the Nanos polls have more weight as they are both more frequent and use more recent data when they get added into the tracker. They have never showed the NDP anywhere near 35% (one poll had them at 32.7 but that was a high outlier for Nanos).

      From their standpoint, there has been a modest NDP decline (-1%) and a modest Liberal increase (+2%) since the beginning of the campaign. The Conservatives have been up and down but are more or less back to where they started.

      Nanos could be wrong and Forum may be right, but if so the story of the campaign then is a significant decline in the NDP vote (-10%) and a resurgence by the conservatives (+5% from the beginning and + 10% from their low point) and a steady Liberal party (+4 from the start but otherwise their current numbers are consistent with every other Forum poll since the election).

      When we speak to trends in the poll tracker we should be mindful of how different polls and the frequency of polls will shape them. Based on the numbers we are seeing, it is reasonable to think that there has been a modest NDP decline, a modest Liberal increase but that this is still a three way race.

      Delete
    4. Ryan, one interesting factor I'd add on to what you said is that even during the 2012 Mulcair NDP honeymoon period after winning the leadership, and during the Summer of 2015 after the Notley Alberta wave, the Mulcair NDP was never able to get leads as big as what the Trudeau Liberals had during their honeymoon period, or what the Harper Conservatives have had in the past.

      The NDP numbers, although respectable, have never been as high as what the Liberals & Conservatives have had at their peak periods, and perhaps that reflects a limited vote potential for the NDP afterall.

      Delete
  24. Eric, how do you split the number Nanos has for "Prairies" into AB and SK/MB for the regional averages in the Poll Tracker? Or do you not try at all? Thanks.

    ReplyDelete
    Replies
    1. I combine the projection into an AB+PR number, and then move the numbers in each region according to how the Nanos AB+PR numbers are different from that AB+PR projection.

      Delete
  25. So I'm going to make an election prediction today based on trends we can already see in the polls.

    The combined vote total of he Bloc and Green Party in the 2011 election was 10%.

    It is going to be significantly less than that in this election. No more than 5% possibly 4%, combined.

    This is because often these votes are the throw away votes of I don't want to vote for the Party that is going to win, so I will vote for one of the fringe Parties that have interesting things to say that I like but I know they won't win.

    This time the three way race makes the choice of top party much less certain and makes the votes more important in people's minds.

    The rise of the NDP and resurgence of the Liberals to tied status will decimate the Green and Bloc vote, even from the last election, and it was pretty low in the last election.

    So my prediction lets say 4.5% COMBINED.

    ReplyDelete
    Replies
    1. Shoshana, you're right about that. Although the BQ has gotten a minor bump lately and it's possible that Duceppe will be able to salvage a couple seats for the BQ, at the present time, the BQ is projected to win ZERO seats.

      Likewise, the Greens, who were polling in double digits in some polls earlier in 2015, particularly in B.C., have fallen in support and are currently projected to only win ONE seat -- the Elizabeth May one in Saanich-Gulf Islands. They have a shot in one or two others like Victoria, but they' re not as strong as they were earlier in the year, and it's not looking like converted Green MP Bruce Hyer is going to make it in Thunder Bay.

      So yes, bottom line is that the 4th & 5th parties are really struggling right now.

      Delete
  26. I'm also going to go out on a limb and say that at least the Nanos poll is going to show 33%, 32% 30%, for the parties a week before election day. I don't know which party is which but, that's what it will say and we may only see a slight shift indicating a trend that will continue to the voting polls 2 days before election day, again I'm not predicting for which party, just for a party.

    I also predict that advanced polling will be down because people will wait to make up their minds.

    ReplyDelete
  27. The MOE could easily explain the numbers, but if believed to the national decimal percentage point the entrail pickers can call a steady CPC with LPC making a faint gain at the NDP's expense.

    I'd agree that Trudeau in the debates is beating the low expectations set by the CPC ad campaign. Hoist by their own petard.

    Mulcair may have hurt himself with his "no deficit" promise. People have heard that so many times from all the major parties that it's like declaring "I'm a typical politician who will promise the moon if it gets me votes".

    It might possibly sell with young voters who have had less exposure to the "no deficit" mantra.

    There's something like 3200 votes for the weaker opposition party in 2011 that gave the CPC its majority. Vote Together now has 68,000 pledges to vote for the stronger opposition party in each riding. We will have to watch to see if VT sways swing ridings. How Eric builds that into his model will be a challenge.

    ReplyDelete
    Replies
    1. George, there's an interesting article from the Canadian Press today about those riding polls that says to get an accurate reading, its 10 grand a riding, and no one is putting up that kind of money. Accurate polls are expensive, especially micro-targeted. Although the Conservative Party thanks them for doing a lot of the work and shouldering the cost, they will be putting up the big bucks for accurate polls of which ridings to micro target.

      Delete
    2. I think the CPC hurt itself with its "not ready" campaign. Incumbent parties run on their record, like it or not. By launching such direct attack on JT there is a huge implicit admission that there is nothing on Harper's record worth voting for.

      "Vote for Harper because he's not JT" might work for an opposition party or for a young government but not for someone who has been prime minister for nine years.

      Delete
    3. Polstats not sure whether you can see the NDP "He should just leave" ad but of late they have added, right at the end the "Nice hair though" line, thus shafting the Tories "not ready" position !

      Delete
    4. Polstats,

      Focusing on the Liberal leader is a winning formula for the Tories; 2008 Dion the Puffin, 2011; Ignatiev Didn't come back for you. Judging by the polls it appears to have remarkable resilience, the Tories look set to win a fourth election in a row, the only question is whether it will be a majority or minority.

      Delete
    5. I hadn't seen it, but I think it really disarms the attacks from Harper. It is always very interesting to read, by the actions they take, what the campaign personnel thinks about their own candidate. For example McCain launched the Palin hail mary, which indicated they knew they were about to lose.

      The Romney campaign kept on changing their attacks on Obama because according to their own polling nothing was sticking.

      Clinton out of desperation launched a bus tour of the south and went on TV to play the saxophone... and managed to turn the campaign around, but you could tell he knew he was loosing.

      Harper's campaign people have spoken loud and clear: there really aren't any good reasons to re-elect our own candidate, but look the other guy is young, so vote for me.

      Sad... really sad. Like the NDP ad says: if that's all you got leave, just leave.

      Delete
    6. @Formerly Carbonear Pete

      You write: "Focusing on the Liberal leader is a winning formula for the Tories;"

      This is exactly the type of flawed thinking that leads the present campaign to chase a formula that is not working. The CPC is looking presently at best at a minority, when they should be coasting to a majority.

      Lastly saying things such as "the only question is minority or majority " with the present polls is just a self-affirmation. The election is quite a bit away, so I'm not discounting a CPC majority just yet, but an NDP or Liberal led minority government are about as likely as a CPC minority, and orders of magnitude more likely than a CPC majority or a majority from any other party for that matter.

      Delete
    7. Formerly Carbonear Pete, Harper has the advantage of competing against 2 Opposition Leaders. He has never been able to win much popular support, unlike prior PC PM's of the past like Mulroney and others who were able to win large majorities with well over 40% of the vote.

      I think it's absolutely possible that Harper can win again. His opponents often underestimate him, in both the Liberals & NDP. But, keep in mind that the Liberals are doing very well under Justin Trudeau and that the NDP is doing pretty well under Tom Mulcair. Both leaders have put their parties in a position that neither party had several years ago.

      Delete
    8. shoshana, obviously we all need to be cautious when looking at riding polls. They have a larger margin of error and can be inaccurate. BUT, many of the riding polls coming out are actually fairly close to the regional trends & seat projections on sites like this.

      For example, the new Mount Royal poll showing the Liberals 15 points ahead very much matches up with what the Grenier model is showing, as well as the model by Bryan Breguet on his seat projection site. We don't know for sure whether these riding polls are accurate, but so far they are looking reasonable.

      Delete
    9. Mulroney and Dief are alone in the modern Era for achieving 50%, most majorities were attained between 37-43%

      Harper 2011 got just about 40%, and majority without QC. Removing QC from share of the vote and he topped 50% in the ROC

      Delete
  28. Éric, the Poll Tracker is now a daily addiction. I have two suggestions for the graph "Poll Averages":

    * Double the vertical axis. This comment has been made before. The current graph would be crushed even without the unprecedented asymptotic approach to three-party exact parity.

    * Fix the right side of the graph to October 19th, with the poll results marching closer to it each day. That would improve the visualization of change timing, assuming that any changes do in fact occur in the next four weeks.

    Together, those should be a couple of minutes for the web guru.

    Great stuff!

    ReplyDelete
    Replies
    1. I agree, and thanks again for all the great work!

      Delete
  29. Is this the poll by Leadnow (Vote Together) you are referring to?

    http://www.votetogether.ca/pages/localpolls/

    ReplyDelete
  30. I hope to hear about your analysis of the LeadNow riding-specific polling!

    ReplyDelete
    Replies
    1. Éric likely won't analyse those polls. He avoids polls commissioned by parties or interest groups.

      LeadNow has a clear electoral objective. They are not impartial.

      Delete
    2. LeadNow isn't impartial, true. It's a progressive group which is trying to figure out how to suggest to non-Conservative voters in each riding across the country how best to defeat the Cons.

      However, if they are commissioning respected mainstream pollsters (eg. Environics) to do their riding polling, it is still worth a glance to see if it matches up with other polls. And for the most part, most of the LeadNow polls are in ballpark of what other firms are showing.

      As long as someone isn't removing pages from a polling document and submitting a misleading poll to the press like the NDP did in Papineau, these polls can still provide us with some clues here & there.

      Delete
    3. Actually I looked through the LeadNow polls and wasn't impressed. From what I can tell they only asked which party people would vote for and didn't mention candidates names. A very big error for local polling as you often get local candidates who can shift the vote significantly. For example, in Guelph the Greens are running Gord Miller who has some 'star power'. Up here in Thunder Bay we have sitting MP Bruce Hyer and I can tell you during my provincial campaign I didn't have as many votes as he has signs. Clearly the name makes a big difference and should be used in local polling. Only someone who doesn't understand how to poll local ridings would make that type of mistake. LeadNow is probably more a Liberal or NDP front which will direct people to the party they support most while telling Greens they are wasting their vote.

      Delete
    4. I've seen the questionnaire, and candidate names were included.

      Delete
    5. Glad to hear that Eric. I emailed them but never heard back and the way they listed it on their PDF suggests parties only. I doubted anyone would do riding by riding polls without as like I said that would be a major problem.

      Delete
    6. The LeadNow reported numbers are fundamentally broken because they don't report Undecided. Without that, it's impossible to assess whether a race is too fuzzy to say anything meaningful about or is truly close.

      That said, the "Undecided" I talk to are wavering between any two of the Greens, NDP and Liberals, but never the Conservatives. If it's truly a coinflip between the Conservatives and another party before Undecided is factored in, that other party--assuming that the current numbers hold through to election day, a very big if--has it nailed.

      Delete
  31. Ding Ding Ding Folks! We have a winner. For the first time in this election cycle we have two polls in exact agreement!

    Leger's new poll had the Conservatives and Liberals at 31% each with the NDP at 29%, essentially exactly the same as Nanos. 31.4, 31.5, 29.1.

    It's about flippin time.

    ReplyDelete
  32. Replies
    1. I suppose one could say that unless other polls start showing agreement - and so far they're not - it can best be described as an outlier the same as that Forum poll that showed the NDP at 40%.

      Delete
  33. So Eric Curious as to why you would call Ekos unrecognizable, which s an actual random interview poll, and you are ok with Ipsos, an online panel which really shouldn't even include a margin of error and is not random.

    Conservatives who were "undecided" and parking their vote, have started answering the polls. it makes perfect sense. I don't know if it's true or accurate but it makes perfect sense and is completely "recognizable" not just in Canada but all over the world with incumbent governments.

    “It’s clear that there has been growth in some of the constituencies they won their majority with in 2011— people who had been sitting on the sidelines unhappy with them or whatever they were doing,” Frank Graves.

    ReplyDelete
    Replies
    1. Stop trying to see what you want to see and see what is there. 5 straight polls after Ekos say a tie or Liberal lead. Show three or four polls in a row with Ekos numbers and you have something. Until then it is an outlier.

      Delete
    2. Um... because it's radically out of step with what every other pollster is reporting about the national race?

      Delete
    3. The thing is, polls are only useful if they are actually able to capture something meaningful. So the numbers can't only say what you'd like them to say, but they also have to reflect reality more or less. The fact that you can create a narrative to justify what you want to believe is not proof of anything other than that humans are terrific at rationalizations. Because once you start rationalizing, you don't even need polls, accuracy or truth--just a nice story.

      Delete
    4. Ipsos is a random sample of the online panel. Technically, it doesn't include a MOE, but it is used for weighting purposes. Elsewhere on this site, Eric has looked at different polling techniques and the results are all comparable.

      Delete
    5. NO it wasn't. it was in steep with what several firms regional polls said. And I was questioning the term "unrecognizable." Those numbers are very recognizable, more so than any poll that has the Conservatives at 24%, when we know that Conservative support is rock bottom around 29%, across many polling firms and many years.

      Delete
  34. Btw Mainstreet with a sample of over 4000 in Ontario has the Conservatives at 38%, which makes EKOS' 35% nationally for the Conservatives "recognizable"

    NDP at 22%, Liberals at 34% in Ontario roughly translates to the EKOS poll as well.

    It's basically two polls in agreement.

    ReplyDelete
    Replies
    1. Mainstreet and Ekos are not in agreement. Well maybe they both have something that can help a dreamer think the CPC is on its way to majority.

      Remenber Mainstreet was a whopper outlier last time.

      Delete
    2. Mainstreet hasn't been doing a whole lot of national polling after that whopper, either. Sort of telling I guess.

      Delete
  35. Interesting that the poll weighting on Eric's tracker has the EKOS poll at <25% of the weight after having them at 70% for most of the year.

    Nanos poll is now well over 50%.. The aged out polls are not chronological but if the CPC > 31 they get aged out and older polls kept in their stead.

    If the unlikely event the EKOS poll is accurate would the seat projector have them at majority?

    CPC at 150 seats would have to be the point where it would be political suicide for either the Liberals or NDP to try for a PM having 100 seats.

    ReplyDelete
    Replies
    1. After Forum, EKOS is the pollster whose results I find most questionable and most all over the map.

      Both Forum & EKOS tend to show massive shifts towards the Liberals, NDP or Conservatives depending on the moment. Not long ago EKOS & Forum had the Liberals winning a Majority, and then the NDP winning a Majority. Now they seem to be saying something similar for the Conservatives.

      We have to leave open the possibility that EKOS & Forum may have picked up certain trends, but I'm a bit skeptical about these 2 pollsters at this point in time.

      Delete
    2. "CPC at 150 seats would have to be the point where it would be political suicide for either the Liberals or NDP to try for a PM having 100 seats. "

      I disagree. They would need a formal arrangement of some kind, but there is absolutely no reason why they could not or would not do this.

      Delete
    3. Eric is not in the business to make outliers push the stats in your desired direction.

      Delete
  36. Whatever else people might say about those EKOS numbers, at least they're not herding.

    ReplyDelete
  37. It's completely nuts !! Nanos today Friday

    Lib 32
    NDP 31
    CPC 29

    Check tomorrow and I'll bet they all change !

    ReplyDelete
    Replies
    1. It rather reminds me of the 2012 Republican primary: no one wanted Romney but whenever they looked for an alternative, it was worse.

      I think these polls reflect the fact that, outside of core support, Canadians don't like any of the three main parties. Regrettably, I think Canadians will not do what I expect to do, which is vote for someone other than the main three.

      Delete
    2. Lots of fog in the crystal ball today.

      Delete
    3. BTW Conservatives at 41% in the Prairies, not possible, so grain of salt there.

      Delete
    4. Like it or not, the clear trend is the Liberals are gradually inching upward.

      Delete
  38. Ok, ok, Ekos is now outdated and an obvious outlier. It was a nice dream for those who want to force the numbers to help Harper.

    Time to move on to reality and the 5 polls that have come out since then, either showing a virtual three way tie between the three main parties or Libs in the lead or tied with the CPC. Nanos has the Libs on the rise and CPC slipping, but still more of the same deadlock.

    Tomorrow is a new day.

    ReplyDelete
    Replies
    1. EKOS has a different methodology. Don't assume there's no bias there.

      EKOS may well have been acccurately measuring things as they were given that means of measurement.

      We won't know what to make of that result until we can make an apples to apples comparison to it, ideally with another EKOS poll.

      Delete
    2. EKOS = outlier. It is now clear after 5 more polls.

      Delete
    3. I would be cautious about assuming EKOS is an outlier. It may very well be given that EKOS (and Forum) have tended to be a bit over the map in recent months.

      But people often underestimate the support of the Harper Conservatives (eg. 2011 Election), so it would be risky to assume EKOS is totally wrong in showing Conservative growth.

      Delete
    4. 4 Nanos, a Forum and Leger polls have Libs in the lead.
      EKOS is a distant memory now.

      Delete
  39. You have the Libs marginally behind the CPC here, and marginally ahead on the CBC site. Not that it matters much given how small a difference and the margin of error, but could you get those two aligned, just for clarity?

    Thanks!

    ReplyDelete
    Replies
    1. The numbers here are weekly averages, with no adjustments being made. The numbers on the Poll Tracker are the 'official' projections.

      Is this confusing to people? Maybe I should take the weekly numbers down.

      Delete
    2. Actually, I had assumed that the numbers were the same, because you update the weekly numbers here every day as well, so it is news to me that they are not the same as the poll tracker. I'm not sure that it matters, but absent a clear explanation, I will always jump to my own conclusions!

      Delete
    3. It would only confuse people who don't bother to give it a bit of thought.

      We should want to confuse those people.

      Delete
    4. No, please keep them. As someone else has pointed out they show the only discernible trends in this sea of numbers and near tied parties:

      1) NDP has been bleeding support from the get go

      2) LPC has made small steady modest gains

      3) CPC is more or less steady not moving much in either direction

      Delete
  40. Eric, did the Poll Tracker change Nanos weighting methodology because both the 22nd and 23rd have weights (4%, 6%), but going back the others only every third is used as an independent sample.

    ReplyDelete
    Replies
    1. Those still have some weight because the new Nanos isn't a new three-day sample yet. So those two polls are representing one day of polling that is being dropped from the rolling poll.

      Delete
  41. For those who are open to additional, contrasting information to enhance future discussion, try this :

    Paul Fairie / Globe election forecast.

    Cheers all.

    ReplyDelete
    Replies
    1. I love competing quality opinions, but so far I'm completely unimpressed by the Globe and Mail effort. They first published their "Conservatives to win" forecast the day after the CPC had moved to second place in most seat projections.

      This was of course reflected on their next set of numbers. I mean, I'm ok with predictions such as "if the elections were held today" but those 4 week out forecasts are just hubris.

      Other suspect figures were a 16% chance that the BQ attains official party status (I put it right now at about 5% tops) and a 17% chance that all parties get 100 seats or more. I think this is about double that figure, i.e. around 30%. Though I expect that once it is clear who between the NDP and LPC is stronger people from the ABH movement will add a 2-3% vote bonus putting them into clear next government territory (though still likely not a majority one).

      Delete
    2. The Globe forecast model (and their headlines) are terrible.

      Their random simulation is ridiculous. I get the NDP winning Ottawa West Nepean while losing Niagara Centre. I got the Liberals winning Oxford in one scenario where they only garner 144 seats in total.

      Delete
  42. "Is the NDP slipping? Nanos today argues against that, but Forum argues for it. The debate last night may make the question moot."

    What does the debate may make the question moot mean?

    ReplyDelete
  43. Here's my take. Watching the various polls and the unending "noise" we seem to be seeing - very little actual movement in support - I detect a slight, subtle shift beginning, that, with four weeks left, will become more pronounced.

    It seems to me the primary reason behind these stalled numbers is the majority of voters who want a change in government (nearly 72%) have been waiting to see which of the two realistic alternatives - Liberal or NDP - had the best chance to defeat the Harper Conservatives. I believe that majority has decided on Trudeau's Liberals and we'll see a continued movement to their column in the remaining month.

    That said, unlike past elections, strategic voting in a number of very close ridings will be the determining factor in which party wins the most seats. As of today, my money is on a minority LIBERAL government.

    ReplyDelete
    Replies
    1. Sorry James, there is no basis for that in the distribution of the vote. The Liberal Vote is the least efficient and their GOTV machine is questionable. They would have to be quite far ahead of the other two parties to form a minority government.

      Delete
    2. Watching the numbers, hearing what I hear from the street (pushing for Greens) I think the NDP support is softer than most think. I've heard lifelong NDP'ers who are unhappy with Mulcair feeling the party is moving too far to the right, while non-NDP'ers are easily shifted to another party but are parking their vote hoping to knock out Harper. That suggests a lot of growth room for the Liberals if Trudeau doesn't blow it.

      If I was betting money though I'd bet on a minority Conservative that is knocked out quickly and Harper trying to force another election while the Liberals go nuts if it is called as they would've had house support given a shot. Harper outspends all other parties combined thanks to his super-long election this time and still can't get close to a majority as voters get angry and we see something break in an early 2016 election. Where it breaks is impossible to know right now but odds are we'd see a 1984/1993 type mess for the Cons (84 saw Liberals slaughtered, 93 the PC's).

      Delete
    3. There's not going to be another election if a Conservative minority is brought down quickly after this election, and I seem to recall Adrienne Clarkson herself said roughly one year would be the cutoff before she would even consider an election and there is enough convention to support this. Thus, the GG would be bound by that convention to offer the opposition leader a chance to form government is they were willing to do so. And as it stands now, both the NDP and the Liberals would probably have the backing of the other party to form a minority government similar to what happened in Ontario in 1985.

      Despite this, I have a feeling it won't stop Harper from kicking and screaming for an election up to and including using the same tactics against the GG as he did the Chief Justice of the Supreme Court when he didn't get his way, up to and including trying to fire the GG. The mess could potentially even be worse than King-Byng.

      Delete
    4. Agreed Nick. I fear that Harper will fight to force an immediate election should he not get confidence of the house when having the most seats, feeling that he could con voters into giving him the majority he wants/needs. This will be an election for the history books, more so than any since 1993 when the PC's were nearly wiped out and '84 when the Liberals nearly were. I see 2004-2011 elections as a big storyline dealing with corruption and how voters respond to it (which I'm sure will be fun for some professors to use). 2015 hopefully will be the end of it and we move into a new phase of 'who will run the country' as the Libs/NDP/Cons all go into a battle royal for a few elections to see if one can take over.

      What is interesting is I heard from a person who has always flipped Liberal/Conservative who is now voting Green and might push others who have gotten sick of the system to do so as well. I can dream of a groundswell of disgust with the big 3 to lead to something remarkable ala NDP 2011 in Quebec can't I?

      Delete
  44. Nanos Sept. 26

    Lib 31.8
    NDP 30.3
    CPC 29.6

    ReplyDelete
  45. has any one done analysis of first past post--using 2nd choice voter analysis.

    ReplyDelete
  46. Heard a couple of interesting things lately. First from a Tory organizer in Atlantic Canada who said they have 15 seats right now but he figures they will lose 10 at the vote. Most would go Liberal I think.

    Plus heard Mulcair on CBC's The House this morning and aside from nearly throwing up knew without doubt I can't vote NDP !!

    ReplyDelete
    Replies
    1. Peter...really?

      The need to share such political immaturity must be so profound. Others should know about your belly-ache and urge-to-purge, why ??

      Delete
    2. Atlantic Canada is a real sore point for the Conservatives.

      They may potentially be wiped out of Nova Scotia - with Central Nova being their best bet to hold a seat there.

      They will lose their sole PEI seat. It's laughable to suggest they will win a seat in NFLD (did Harper even visit that province during the campaign?).

      Best case for them is holding a few rural New Brunswick seats.

      Delete
  47. So how many angels can dance on a pinhead?

    From week 1 to 8:

    CPC +1.2
    LPC + 3.9
    NDP - 6.0
    G+Q + 0.3

    The Notley surge seems to have receded.

    Harper and Duceppe are blowing the niqab dog whistle.

    Mulcair accuses Harper of hiding the economy behind a niqab.

    CPC retains its loyal base, but 8 weeks later, remains 9% short of majority.

    Trudeau or Mulcair need to find an issue that will persuade voters wanting change that he's the one to do it.

    ReplyDelete
  48. Sunday Dec 27 Nanos has

    Lib 32.5
    CPC 31.5
    NDP 27.6

    Nanos poll: NDP slips to 3rd place nationally in wake of French debate

    ReplyDelete
    Replies
    1. A French-language debate that much of the French-language media believe Mulcair won.

      Delete
  49. While there has been a lot of shuffling of the pack between first, second & third place, the three main parties have been essentially neck and neck throughout, and IMHO there is not much likelihood this will change. What we should expect to see Oct 20th, then, is that there will not be a clear "winner" of is election, and it will be Parliament that decides who is Prime Minister. Does Mr Harper maintain the confidence of the House? If not, who does, and how is confidence demonstrated? Do we see unstable, ad hoc, issue-by-issue minority government (which has been the norm inCanadian federal politics to date), a formal, McGuinty-Rae model formal support agreement without a true coalition, or a true coalition government with multi-party Cabinet membership? Only time will tell, but I think Canadians will be pi$$ed off if we get an unstable minority and have to go to the polls again in a year or 18 months...

    ReplyDelete
    Replies
    1. If Harper cannot get a majority he is finished. The unstable minority/coalition boogeyman is not flying with Canadians anymore. Most just want Harper gone at any cost.

      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.