Friday, April 15, 2011

Liberals slip back two seats

It was a whirlwind 24 hours, but the Liberals are now back to 73 seats after this morning's Nanos poll brings them back down to earth. As that poll is just replacing yesterday's in the projection, and a new Mustel poll for British Columbia does not shift things much out west, there is little real change in the projection. But the two close ridings the Liberals took back yesterday have slipped just out of their fingers again.
The Conservatives have picked up 0.1 points in the projection are are now at 39%. They have been between 38.6% and 39% since March 29th, demonstrating just how little the numbers have moved in this campaign. They also pick up one seat and are back to 152.

The Liberals are unchanged at 28.2%, but small drops in Ontario and Quebec mean they are down to 73 seats.

The New Democrats are unchanged at 16.8% and 33 seats, while the Bloc Québécois is up 0.1 points to 8.9% and one seat to 50. The Greens are unchanged at 6% support.

We are supposed to hear from EKOS today, so look for tomorrow's projection in Le Devoir to see what effect it has on the numbers.

Since there wasn't much new to add to the projection, it has changed very little.

Atlantic Canada has had the biggest shift, with the Tories dropping 0.4 points to 37.3% and the NDP gaining 0.4 points. They are now at 18.9%. The Liberals are steady at 38% in the region.

In Quebec, the Conservatives and Liberals have dropped a little to the benefit of the Bloc Québécois, up 0.2 points to 36.9%. This is due to the 37.3% result from Nanos today.

And in Ontario, the Conservatives have widened the gap again by 0.3 points.

The seats lost by the Liberals are the ones they gained yesterday: Brampton - Springdale and Brossard - La Prairie. Parm Gill (Conservative) and Marcel Lussier (Bloc Québécois) are now favoured in those ridings.

This week I've been focusing on the Ontario ridings which could go from the Conservatives to the Liberals. But now that the trend has been reversed, at least for today, let's look at the ridings liable to swap in this province and others, based on which parties are gaining and losing.

In British Columbia, the Conservatives have dropped a little while the Greens have gained a little. The only riding where the two parties are going head-to-head is Saanich - Gulf Islands. However, I still have the Tories 10.2 points ahead in the riding.

In Ontario, the Conservatives are within three points of the Liberals in Brampton West (0.5 point gap), Kingston and the Islands (1.2), and Eglinton - Lawrence (2.7), and Mississauga South (2.7). There is, actually, little room for Tory growth at the expense of the Liberals in the province, though a gain of four seats would put the Tories at 156.

In Quebec, where the Bloc has now gained while the Liberals and Tories have dropped, Gilles Duceppe's party is within three points of their opponents in Beauport - Limoilou (Conservative, 1.7 point gap), Papineau (Liberal, 2.2), and Montmagny - L'Islet - Kamouraska - Rivière-du-Loup (Conservative, 2.3).

And in Atlantic Canada, where the NDP is making headway at the expense of the Conservatives (at least, numbers-wise. The Tory vote could be going to the Liberals and some Liberal vote could be going to the NDP, or some other combination of these shifts from one party to another), there is very little prospect of seat gains for Jack Layton. The NDP is within 13.4 points of the Tories in South Shore - St. Margaret's (Nova Scotia) in the projection, while St. John's South - Mount Pearl in Newfoundland & Labrador is a close three-way race.


  1. Hi Éric, nice work. I was wondering if you could do an overlay of the trendlines from 2008 to compare the party progress and outcomes?

  2. I'm hearing that your three way race in SJSMP is correct but that Fabian Manning is in an up hill battle in Avalon, all according to party polling.

  3. Interesting. If you can get your hands on it I'll keep it under my hat!

  4. Yep, Quebec looks part of the set now. :)

  5. Hi, does your model have any way for compensating for third-place candidates who seem bent on throwing the game, possibly to avoid vote-splitting?

    I am thinking about places like Ajax-Pickering where the NDP candidate hasn't yet started his campaign (according to media reports he has been on vacation out of the country for the past two weeks) and the NDP-held riding in Edmonton where (again according to the media) the Liberal candidate is deliberately running a minimalist campaign.

    Realise this is highly speculative and probably hard to model, but ignoring such phenomena also seems likely to skew the results.

  6. Because it isn't verifiable and there is no way to measure it, there is no way to model it.

  7. Is it just me, or is Giles' numbers in Laurier-Sainte Marie slowly dropping? He's down under 47% in your projection.

    It's not likely to change much, but if the voters think about strategically voting there, it might make him sweat a bit. I live in the riding and I am hearing a lot more english on the street than 3 years ago.


  8. Very true about some parties doing little in certain ridings. In my riding the NDP candidate is unreachable by email or phone so far, has no website, has yet to put up a single sign, and skipped the first debate.

    However, often a large part of that is due to ridings where there is little support thus candidates start becoming 'paper candidates'. Where I live the Greens passed the NDP a few elections ago and the gap is growing thus I suspect the local NDP has given up (their candidate wasn't picked until days before the deadline).

    We really need the system to change to encourage participation everywhere. Even if just to a 'preferred ballot' where you pick the candidates in a 'top choice' 'second choice' type method where the bottom ranked is removed, adjust votes accordingly, then the new bottom is removed and so on until you get 50% + 1 thus removing strategic voting which used to kill the PC/Reform parties and now hurts the Green/NDP/Liberals generally.

  9. One thing no one has picked up on is that that Tories dropped 3 points in Quebec in just one night. This poll would have only one night out of three that was post-French debate where Harper totally bombed. It will be interesting to see if a Tory meltdown ensues in Quebec after Harper (according to ALL polls) totally turned off about 90% of Quebecers.

  10. Wouldn't it be great for Elizabeth May to win a seat and hold the balance of power? Ah I'm wearing my rose coloured glasses to look at the crystal ball today.

  11. There would be no balance of power for Elizabeth May. She is a firm Liberal supporter. There was talk of her being in the Coalition cabinet under PM Dion.

  12. Last night on the CBC "At Issue" Panel the whole panel was talking about the NDP catching up to the Liberals in the low 20s.

    It was as though they (Chantal, Coyne and Mansbridge) totally believed the COMPAS poll. Alan Gregg did not provide a dissenting opinion. The only thing he could offer was that it wasn't over til its over.

    There are no recent polls other than COMPAS that has been published that would support the basis for their discussion.

  13. COMPAS had the NDP at 16%. I think they were using polls showing Layton had won the debate and the recent uptick in Nanos (and even EKOS) as the basis of their discussion.

  14. ...and now Forum has a brand new poll out post-debate:

    CPC - 36%
    Liberals - 25%
    NDP - 22% (!!)

  15. I do not have data on minimalistic campaign results but in Central Nova where the Liberals did not run against Ms. May, Mackay won by a bigger margin in 2008 than 2006.

    He had 1000 more votes for him.... Blue Dog Liberals?

    Ms May increased the combined Liberal-Green vote count from 11,000 in 2006 to 12,600 in 2008.

    The NDP dropped from 13,800 to 7,600.

    Using the accepted Liberal analysis of the 2008 election 6,000 NDP voters sat out the election.

    Had those 6,000 missing NDP voted for Ms. May she would have won a squeaker.

  16. COMPAS had the NDP a lot closer to the Liberals (8 Pts) than the Liberals were to the CPC. No other poll has shown that.

    They were talking like it was a race for second place with the NDP having momentum.


    Forum Research ( April 14 sample 2241)

    CPC 36 Liberal 25 NDP 22

  17. wow the forum poll was all taken last night and have the bloc at 27% in Quebec.... NDP at 23% in Quebec

    sketchy and selective on regional details in the Hill Times article.

    61% approved of the way the government is handling jobs and economy.

    and 50 per cent approved of the government’s ability to provide open, transparent and ethical government.

    Only 15% thought Ignatieff won the English debate.

  18. Elizabeth May worked for the Mulroney PMO. She'd be more than happy to have a cabinet seat if the CPC are at 154 seats and she has 1. Ooo la la, what a scenario. It isn't going to happen, but it's still fun to play.

  19. I have been reading this blog over the course of the campaign, and must say you are doing an impeccable job. However, there was one line made by Ignatieff during the English debate that should be addressed.

    Towards the end of the debate, Ignatieff became tongue-tied and criticized the Conservative Party for supporting "gifts to upper-middle class Canadians", and affirmed Liberal spending priorities would be diverted elsewhere.

    Although he may have misspoke, this may nevertheless warrant a change in your seat-by-seat projection. I presume this comment will cause support to shift from the Liberals to the Conservatives in upper-middle class ridings. If this is true, then post debate-polls indicating the Conservatives gaining at the expense of the Liberals may be limited to ridings with a significant upper-middle class population.

    What is your take on this?

  20. The NDP has been on a tear since the debate.

  21. Like BC voice, I to thought it a bit odd that they seemed to think the NDP was going to surge in momentum past the Liberals. Possible I guess though I think it somewhat unlikely. Is there anything in the polls to suggest this level of NDP surge?

  22. EKOS poll might need some amendments or explanation.

    The Other in Quebec dropped from 2% to 0. However not to worry Other passed the Green in Atlantic Canada hitting 10%. This was not noteworthy in examining trends :)

    My guess is that André Arthur separated his riding of Portneuf--Jacques-Cartier and moved to New Brunswick.... or Maybe NL

    Would have thought that would have been newsworthy.

  23. Before anyone gets too carried away, Forum had the NDP at 20% in their previous poll.

  24. There is something to be said for looking at the trend from each pollster. Forum has now had three polls since the campaign began. The Tory vote has gone 43 to 38 to 36 and the NDP vote has gone 19 to 20 to 22.

    There are still two weeks to go, but more and more looking at the polls as a whole - with the exception of that ridiculous Compass poll - none of the polls show the Tories in majority territory. In almost every poll they are losing ground in BC and Quebec and they are not opening up anywhere near the kind of lead they need in Ontario to suddenly start scooping up a lot of seats.

  25. Thank you for a really good site. The contributors, aside from some modest but transparent cheerleading and insight and perspective. Would love it if you could take the same data and run proportional representation results ( pick a model). As I said modest but transparent cheerleading


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.