Thursday, January 13, 2011

With bar set low, Greens have decent shot at victory

Whenever it takes place, the next election will be Elizabeth May’s second as Green Party Leader. After increasing her party’s support and getting herself into the nationally televised debates in 2008, her supporters will expect nothing less than the party’s first seat victory when Canadians are next asked to cast their ballots.

The rest of the article can be read on The Globe and Mail website.

This is the first of five pieces on each of the parties, which will determine their floor and ceiling in terms of seats. Ridings will be classified in five categories: fortress, secure, vulnerable, targeted, and potential.

Fortress ridings are those which are extremely unlikely to be lost by the party in the next election. Secure ridings are those which are unlikely to be lost, but which are within the greatest margin that was overcome in 2008. Vulnerable ridings are those which, according to current projections and special circumstances (i.e., Larry Smith in Lac-Saint-Louis), are at play. Targeted ridings are those that are held by other parties and are at play, while potential ridings are those that are within the margin that was overcome in Egmont in 2008.

Starting with the Greens means there are no fortress, secure, or vulnerable ridings. I've identified Saanich - Gulf Islands as the only riding fitting in the "targeted" category, while Bruce - Grey - Owen Sound, Guelph, and Vancouver Centre are in the "potential" category. This gives the Greens a ceiling of four seats.

As there won't be room to list all of the ridings in each category in my Globe pieces for the other parties, I will do so here.

I expect to post about yesterday's Angus-Reid poll later this afternoon.


  1. The next election will be May's second, not third election as leader.

  2. Picky point: Bob Bell withdrew as the Green Party Candidate in Guelph in September, 2010 so the Green Party does not have a candidate registered in that riding.

  3. Lou Arab,

    Damn, you're right. Argh! Will have it fixed.


    Saw that too, and will have it corrected.

  4. Vancouver Centre?

    In 2008 the Greens ran a star candidate there and finished fourth. A strong fourth, but fourth. I don't see them surpassing all three of the traditional federal parties all of the sudden when they couldn't pass any of them even with a star Candidate.

  5. It's just a ceiling. I don't think they can actually win in Vancouver Centre, but they are a party in the race.

  6. Ugh, I'm really upset at the mistake I made about it being May's second election. She became leader in the summer of 2006, after the election. I didn't check as I was certain she had been the leader in 2006, especially since it seems she's been around forever.

    Really sorry about that, folks.

  7. Ugh, I'm really upset at the mistake I made about it being May's second election.

    Don't get to bent out of shape. Relative to the amount of work you put into this site the error rate is extremely low.

    Thank You !!

  8. Éric.

    I really am waiting for your take on the seat analysis done in this fashion.

    Is this analysis tied in with your regular top down seat projection model based on the polls?

    I did the 1 hour version of this sort of analysis based on the 2008 results and came up with

    Party - absolute high - realistic high- realistic low - absolute low
    CPC - 190 – 167 – 124 – 103
    Lib - 121 – 99 – 55 – 33
    NDP – 51 – 44 – 26 – 16
    Bloc – 55 – 52- 40 - 33

    (details in the comments of the the December best worse case analysis post)

    I used the Vaughn Margin of victory of 14.9% to establish the in play ridings. I got beat up a bit on it a few weeks ago so I imagine your choice of Edgmont and the 4,200 vote shift or 23% will draw some fire.

    The 23% shift is very tenuous as it is such a small riding. Are you using the raw vote total or the % shift as your filter?

    Between the adjustment of the weighting given to EKOS in your overall prediction model and my pushing (nagging) this sort of analysis as a sanity check to your model I am getting a feeling of being listened to.

    I guess you have been able filter through my blatant partisanship to mine my nuggets of useful contributions. I hope I have not driven away too many people from your site.... but now you have a much larger audience at the G&M.

  9. While Egmont featured the biggest margin that was overcome, there were a surprisingly high number of ridings that went from one party to another with margins of over 15 points. Thus my choice of Egmont as my guiding result.

    But the "potential" seat wins are not what a party is likely to win, far from it. Once I get into the other parties, the focus will be on the ridings that are vulnerable and the ones that will be targeted. These will be based on 2008's results, my current projections, and the special circumstances of each riding.

  10. Hello from Saanich Gulf Islands --

    Your observations on this riding would be strengthened with reference to a very strong federal Liberal candidate -- Renee Hetherington. The previous Liberal candidate - Briony Penn - encouraged Hetherington to run, knowing her background in both environmental and economic fields would enable her to overtake incumbent Gary Lunn. (Penn came within an eyelash of winning in 2008.) The poll you cite in your posting deserves a second look -- the company that conducted the survey issued a correction shortly after the initial release. The clarification shows a high level of support for the federal Liberal party:

    Hope the link works!

  11. Alice over at Pundits Guide had an interesting comment when the Tory target list was released.

    "close margins the last time out only predict 1/2 the seats that change hands next time."

    So definetly pay attention to star candidates, even in races that weren't even close last go around.

    Ex. Tories are running really hard in the open seat Mississauga East – Cooksville.

    They're running what Paul Wells calls the "Polish Fantino". He's gone with the PM on a visit to eastern Europe so he's certainly a heavy hitter.

    Don't be fooled by the Liberals 17.6 margin of victory last time out.

    It should certainly make your list!

  12. Another poll to consider: paid newspaper readership.

    The Toronto Star is mandated to be a liberal newspaper in its founding concepts and charter.
    The QMI (SUN chain) is as far to the right as is the Star to the left.

    The Star has historically been Canada's highest circulation newspaper... sort of the media wing of the naturally governing party. Weekly circulation has dropped from just over 3.2 Million in 2007 to just under 2.2 million in 2009. The Liberals are short 1.1 Million pieces of propaganda in front of the voters each and every week.

    Meanwhile Sun media (Journal de Montreal, Toronto Sun, Journal de Montreal, Journal de Quebec, Edmonton Sun, Calgary Sun, Ottawa Sun and Winnipeg Sun) are bought by just under 4.2 Million people each week.

    Ezra Levant is read almost twice as much as Jim Travers!!

    The Liberals (and coalition) have to consider forcing an election prior to the March startup of SUN TV. SUN TV will wipe out the Liberal advantage the near daily appearances of Susan Delacourt and Jim Travers on CTV as unbiased commentators and opinion shapers.

  13. Éric, that's a good and fair analysis. Thanks!

    To break through in the next election, the Green Party execution will have to be a couple of notches better than historical efforts. I'm confident that we're going to see this happen.

    For the next year and a half, time is on the side of two parties: the Tories (because power begets power) and the Greens (because a shift in organization is under way). It's neutral for the Bloc, who are ready for an election whenever. It's against the NDP and especially the Liberals, who appear weaker and further from power as time goes on.

    That notwithstanding, there are currently no signs that Ignatieff will pull the plug before the mandatory election date.

  14. Eric, you need to check out the info at
    The data from the McAllister poll is seriously flawed, except for just two responses that were re-weighted to give a true picture of the riding. One corrected response shows May just behind Lunn, and Hetherington just behind May. The second corrected response shows the LPC with a big lead over the CPC, and the NDP and Greens trailing.
    None of the other published data from this poll is accurate, so you need to adjust your analysis in light of this information.

  15. Farwestie, you were confused and other polls and the the subsequent election bears this out. The seriously flawed pol was another Oracle poll, but it was simply weighted differently, and the question you cite didnt mention who was running; just parties.l If you look up the McAllister poll, you'll see that it was weighted correctly, using a refined approach which weighted age and gender within census divisions.

  16. For details see the comments on this blog.


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.