04.26.2011 01:27 PM

Democratic Space: Cons 161 seats; Libs 57; NDP 53

Hoo boy, this is unpleasant. Who says they’re right? Who says they’re wrong?

41 Comments

  1. JenS says:

    Who the hell knows? I really think this election is defying all traditional wisdom in terms of polls. And frankly, the frequency of the slightly breathless poll and seat projection updates makes me think we should move to much shorter campaigns. Like, say, four days. It’s as much as I can handle.

  2. MontrealElite says:

    If true, there’s going to be a fortune in “Don’t blame me, I didn’t vote for him” merchandise down the road.

    • nic coivert says:

      True.

      A result like this will cast all current leaders into ignominy. Jack for doing it, Ignatieff for letting it be done, and Harper for what he will do (and has done). Harper, of course, will spend millions of taxpayers money to convince the public otherwise but he’ll never be loved, only loathed. I expect our democratic system to be systematically altered in an attempt to align it with the American system. That will naturally aid the continual and gradual unification of Canada and the U.S., decreased Federalism and increased power to the provinces will abet the dissolution of Canada as well.

      Good morning America.

      Last night a friend of mine who does Biochemical research said he suspects they’ll end funding of anything that isn’t product based. In other words you’ll only get funding if you’re doing product research for a corporation. No more curiosity research, and this is what gets the results, creates innovation. He said the current model he’s working under was created by the Liberals under Chretien. He also said if Harper gets in he’s seriously considering moving to Sweden to do research.

      Canada, Harper’s brain drain.

      • que sera sera says:

        Interesting comment about Sweden. If Harper gets in I am investigating returning there (ancestral homeland) until Canada returns to her senses or until I die, whichever comes first. Harper is everything I loathe about the USA and rightwing fundamentalist yobbos.

      • myntje says:

        I don’t need to love the Prime Minister. I’m a Conservative and unlike most of you I didn’t Iove Jean Chretien. I already have a husband and a best friend and children to love. I just want a Prime Minister who’s competent. Jean Chretien was competent. Would Jack Layton be? I don’t think so. Ignatieff? He’d have to prove himself. I know Stephen Harper and contrary to all the caracitures, he’s taken us through the present economic woes rather well. Our son is a research scientist and prof at U of Miami and they’re jealous. So it’s Harper for me.

        Good luck with the merger but I’m afraid Jack will be heady with power and hard to deal with. You know he smiles while stabbing you in the back. Witness the English debate and his attacks on Ignatieff about a petty attendance record. He was like a peacock. Smarmy.

  3. Nastyboy says:

    I only believe polls that show my team winning. Anything else is obviously in error or the company is in the bag of one of the other parties I disagree with. And then the only poll that counts is on election day. And if my party isn’t voted in, then Canadians are idiots.

  4. AndrewOpala says:

    @Nastyboy is correct
    ’nuff said

  5. Is this what you mean by pulling together?

    • MM says:

      Warren can smell a sinking ship. But by all means, keep rowing!

      • George says:

        refresh your memory Jason – Warren predicted this outcome well before this charade was forced on Canadians. First mistake was ignoring your base, second mistake, continuing to ignore your base, third mistake, continuing to ignore your base and pretend all is well.

  6. Dan says:

    Who the hell knows.

  7. artwilliams says:

    Beaches-East York LEANING Liberal, Etobicoke-Lakeshore LEANING Liberal. On my!

    • Mike_02 says:

      Democratic Space leaning, leaning… oops, face plant.

    • MM says:

      Ignatieff’s own riding as “too close to call”? It has been Liberal for 32 of the past 43 years. Won’t look good if Ignatieff has to spend part of the last week of the campaign saving his own seat.

  8. ed_finnerty says:

    election prediction project has a much lower projection for the cons especially in ontario

  9. Supernaut says:

    92% overall prediction (with an even larger margin in the TCTC ridings) leaves a lot of room on either side over 308 seats. Seeing as they don’t publish a methodology, I’d consider it sketchy. I’m more inclined to trust 308’s numbers.

  10. Dr.J says:

    My conclusion unless a bomb goes off within the CPC was around 165 – 175….the Liberals in second but down no more than 10 seats and the NDP gaining no more than 15. Afterwards, Iggy and Gilles will say good-bye to federal politics as well as alot of other opposition older MP’s. (AKA Liberals)…However, I agree with JenS for the most part..”Who the Hell Knows”

  11. Steve B says:

    I have been running the poll data through my own program and have received results similar to what others are saying.

    I don’t think there is so much an issue of pollsters being in cahoots with a political party – usually those results are kept confidential – however it is very strange that one polling firm – Ekos – is publishing results that are very different than that of the other agencies.

    Ekos was the first to run with the NDP surge in Quebec – which was confirmed within hours by Ipsos Reid (I think); however they also continuously publish support for the Conservatives 3 to 10% lower than the other firms; and to me, that is a red flag indicating a flaw in their methodology. I don’t say this because I’m partisan – and I am – but because in statistics, when you look at a data set, items outside of the cluster of data points are indicators of an error in sampling or something – and can be excluded.

    Ekos’ is, as far as I know, the only polling firm to use their pRticular methodology of data collection, and the fact that they’re the only firm which publishes numbers outside the norm tells me that there is a problem.

  12. h holmes says:

    The problem for the NDP will be wasted votes.

    Same problem they have everywhere.

    They poll stronger than they deliver.

    Won’t be surprised if their actual voter turnout percentage is more like 19-21%.

    They just don’t have the people in Quebec to bring out their votes.

    Elsewhere , identified NDP votes vote less frequently then CPC or LIB votes.

    • The Doctor says:

      On the other hand, if the NDP surge in Quebec is real and significant, then they will benefit from the opposite phenomenon, i.e., the fact that our first-past-the-post system favours geographic concentration (up to a point where it’s then wasted, Exhibit A being CPC support in Alberta). I think that’s why Ekos had such a significant number of seats for the NDP, because they projected a big # of seats in Quebec for them (53, I think I saw).

    • Namesake says:

      We’ll know in a week, of course, but there’s several reasons why EKOS’ results are probably more accurate than those they’re disagreeing with:

      – their sample size is generally 2 to 3 times that of their rivals (and thus have smaller MOE’s); and,

      – and very close when compared to those with similarly large samples (e.g., with the Leger’s, at:
      http://www.cbc.ca/news/interactives/cv11-poll-tracker/ or for the Libs & CPC at least on the large (6,000) internal poll that Pollara did for the Libs in January http://www.bloggingtories.ca/forums/post93142.html

      – in addition to using a random sample (unlike, say, INTERNET POLLS), their methodology DOES differ in three key respects which can all enhance their reliability: they include cellphones; they’re short & sweet, just on the political stuff (not part of an omnibus survey, e.g.), which reduces respondent burden and encourages more participation from, um, normal people; and since they use a computer script, not human interviewers, people may be more likely to answer honestly rather than worry about what’s more socially acceptable.

      – in the proof is in the pudding dep’t, they were the only ones to call the Toronto mayoral election fairly accurately (which some were saying was “too close to call). True, they underestimated Ford by a few points which they gave to Joey Pants, but they did have him way ahead of Smitherman, whom they were bang-on on:

      here’s the last EKOS poll released on Oct. 22, three days before the election (w. field dates of Oct. 13-21), vs. the actual results:

      Rob Ford 43.9% (actual: 47.1)
      George Smitherman 35.6% (actual: 35.6)
      Joe Pantalone 15.0% (actual: 11.7)
      All others 5.5% (actual: 5.6)

  13. jack says:

    threehundredeight.com shows different numbers but who knows. But I will say that with libs happy to do post mortems before the final tally, the vote for them will naturallly go down. Many projections have the ndp only picking up 5 or 6 seats. This so called “surge” of a few percentage points (mostly in Quebec) will result in a seat or two in Quebec but more second place finishes to the Bloc. The sad part is the vote splitting will be greater and give harper the majority unless the libs get more coordinated and stop the analyzing and show asses and elbows instead of how smart they think they are. What is clear is that people are looking for an alternative to Harper.

  14. Robert K. says:

    Miles to go before we sleep……

  15. AndrewOpala says:

    I went through some of their (http://www.democraticspace.com) old election results and they look pretty bad … the had a lot of error in every respect (percentage votes cast, seats won). Their website doesn’t say much about their methodology but says a 92% accuracy in their model. The real question is “What does the Oracle of Pollara – Michael Marzolini say?”

  16. Dan F says:

    A blind monkey could get 80% right. Keeping in mind that most seats are very safe for one party or another, it isn’t really going out on a limb to say that the Conservatives will win 27 of 28 in Alberta. The real test is how they do with the 20% of seats that are close. If they get 90% right, it means only about a 50/50 chance on close seats; nothing to write home about.

    Also, none of the projections take into account the strategic voting that may or may not have an impact on the final results. There are half a dozen websites (or more) set up to encourage people to vote for the specific candidate that is most likely to beat the Conservative in each riding, and if that has an impact of even 5% in close races it will be enough to flip seats away from Harper’s majority, putting him firmly in minority territory.

  17. Craig Chamberlain says:

    Election-time polling manipulates voters and should be banned altogether. Let the media actually dig for a story! (They actually wouldn’t have to dig that far to find something to report.)

    • jack says:

      Hear Hear!! The wide variance in polling shows how bad the science of polling really is. They claim certain statistical reliability but its hooey. For a general population’s opinion, todays technology is way out front of the science of polling. Of course pollsters deny that but the variation in results proves it. If they were reliable, they would all be fairly close together. The media uses them to defend or put forward the own editorial bias or to even assist a party depending on their bias. I have asked some of the people calling (when it was not a robodialer) and they advised me that they get very very few people willing to participate in most surveys. Their results and accuracy depends on the randomness of the survey and they assume they have randomness, but fact is they are far from random. At best their sample set is limited to people with a land line that are home that answer 800 number calls and are willing to take part in surveys. Of those, 38% prefer Harper. Dan F is bang on when he says a monkey could get 80%. IN fact, the monkey could guess at the rest and still be as accurate as the poll itself. Get out the swing ridings, take the two parties in contention and let an elephant pick between them. It would be at least as good as the pollsters.

      • Dan F says:

        Appreciate your vote of confidence Jack. It also just occurred to me that these new NDP supporters (and in fact all NDP supporters this week) will be EAGER to talk to a pollster in the same way a recent convert wants to spread the news as far and as wide as possible. If dippers are spilling their guts to everyone they can talk to (including pollsters) and crotchety old bloquists are hanging up on them, I suspect that would also have an impact on the results. So many factors… so many undecided voters – perhaps the undecided voters are only waiting to decide which party to vote for to beat Harper? I could go on forever. Bottom line, roll the dice. Then do like Warren says and GET YOUR BUM TO THE POLLING PLACE & VOTE.

    • reformatory says:

      I’ve been saying that over and over again. BAN ALL POLLING during the Writ period. It skews everything and is used to manipulate. A true democracy would not tolerate it.

      • The Doctor says:

        That’s an interesting definition of “a true democracy” that you have there. A “true democracy” is one in which we don’t trust people to be able to digest poll results? Seems awfully nannyish to me.

  18. John Larocque says:

    Their predictions seem closer to 2008 than some of the numbers coming out of the other pollsters. We don’t have the NDP scooping up three dozen seats in Quebec, or the latest Fantasy Arithmetic from Frank Graves/Ekos, or a 45% CPC score in Atlantic Canada. Democratic Space has the Liberals ahead in AC and most Quebec Liberals re-elected, which “seems” about right, even if it goes against the grain. Election night things probably won’t start to get interesting until they start counting Ontario and Quebec.

    The narrative is almost conventional – a Conservative increase in Ontario (a minor bloodbath of 905/416), and a few other seats elsewhere, just enough to tilt Harper to Majority-land. I admire their caution, even if they may wind up underestimating some of the current polling trends (Jack-O-Mania).

  19. Raymond says:

    Nah.
    Too good to be true.

  20. George says:

    http://news.nationalpost.com/2011/04/24/analysis-%E2%80%98trust%E2%80%99-issue-backfires-on-liberals/

    Nothing new in this article that Warren hasn’t touched on here or in his columns.

  21. scott d says:

    The NDP look like they could take Davenport as these two stories say. The democratic space poll seems off with what is going on the ground which is why I hate individual riding polls, as they are often out of touch.

    โ€œIf signs could vote, NDP challenger Andrew Cash, a writer and former recording artist, would be a shoo-inโ€

    Davenport could flip orange after 49 years
    http://www.thestar.com/news/article/980060–davenport-could-flip-orange-after-49-years

    Road map to a potential NDP breakthrough
    http://news.nationalpost.com/2011/04/26/road-map-to-a-potential-ndp-breakthrough/

  22. Randy says:

    Good analysis on seat projectors http://www.punditsguide.ca/

    I wouldn’t trust any of them.

  23. Fluther Good says:

    And on May 3 read how every pollster will claim to be spot on!

  24. Adam says:

    Has anyone started a list of Liberal MPs likely to cross the floor to the Tories in the event the NDP forms the opposition? Sorry Warren I’m not ready to give up, but when you see Grits in BC trying to bail out the dinghy with a few beer cups…..

Leave a Reply to myntje Cancel reply

Your email address will not be published.