Monday, March 18, 2013

B.C. NDP hits majority support

The B.C. New Democrats have moved above 50% support in British Columbia for the first time in the projection, thanks to a new poll from Ipsos-Reid that gave the NDP 51% support province-wide. The party is now forecast to have a 97.7% chance of winning the popular vote on election night and, if the election were held today, a 97.6% chance of winning the most seats.

Since the last projection of Feb. 25, which incorporated all polls in the field up to Feb. 22, the New Democrats have picked up 2.8 points and now lead with 50.4%. The B.C. Liberals increased by 0.8 points to 32%, but dropped one seat to 21. The NDP picked up that seat and are projected to win 63, though the ranges for both parties have widened to 48-75 for the NDP and 8-37 for the Liberals. That puts the New Democrats very safely in majority territory.

The B.C. Conservatives slipped 1.2 points to 9.6%, while the Greens fell 2.3 points to 6.3% support.

The New Democrats improved their position in Metro Vancouver, increasing by four points to 51.4%, while the Liberals were down 1.1 points to 32.5%. The two parties swapped three seats, with the NDP now up to 28 in the region and the Liberals down to 11. Both parties gained on Vancouver Island, however, with the NDP up 2.7 points to 55.4% and the Liberals up 4.3 points to 29.5%. The Greens, who did poorly in Ipsos-Reid's poll, slipped 6.2 points to only 8% on the island (though their range puts them as high as 11.5%).

In the Interior and North, the New Democrats dropped by 0.4 points to 43.8% while the Liberals increased by 2.2 points to 33.6%. The Liberals took two seats from the NDP in the projection in the process. This remains the one region of British Columbia while the Liberals have the greatest chances of salvaging things - that was made easier by a 1.2-point drop by the Conservatives to 13.7%.

Ipsos-Reid has not been in the field since Nov. 26-30, and much has happened in the intervening four months. Nevertheless, the changes in support were within the margin of error (or would be with a random sample): the NDP gained three points, while the Liberals dropped three points to 32%.

The Conservatives and Greens were unchanged at 9% and 7%, respectively. Support for independents and other parties was only 1%, also unchanged from November.

The New Democrats held an eight-point edge over the Liberals among men - and 29 points among women. The NDP was also ahead, by a margin of 51% to 33%, among British Columbians over the age of 55.

Regionally, the NDP increased to 56% on Vancouver Island and 53% in Vancouver, but slipped to 43% in the Interior and North. The Liberals were down in Vancouver to 31% and on Vancouver Island to 31%, but held steady in the Interior and North with 34%.

The personal ratings of Christy Clark remain troublesome, as only 30% of British Columbians approve of her performance. Her disapproval increased by six points to 65%, while Adrian Dix had a 51% to 40% split. Potentially problematic, however, is that his disapproval rating increased by six points.

John Cummins had an approval rating of 16%, while his disapproval rating fell by seven points to 44%. Jane Sterk of the Greens was steady at 23% approval to 20% disapproval.

If we remove the "don't knows", Cummins has the worst approval rating at 27%, compared to 32% for Clark, 53% for Sterk, and 56% for Dix.

But the poll did have some rare good news for Clark. While Dix beat Clark by wide margins on who was best able to protect the environment and deliver programs and services to British Columbians, Clark was narrowly ahead on the questions of economic management (40% to 32%), creating jobs (38% to 33%), and managing the province's finances (38% to 34%). The silver lining for Dix is that among older British Columbians (i.e., those who vote in the largest numbers) the margin was far narrower on managing the economy and creating jobs, and he scored better than Clark on financial management.

So, the theme of the economy remains a potentially strong one for Clark. But her advantage is still quite small. Dix did much better on being able to deliver programs and services (54% to 24%), which is significant considering that 24% of British Columbians said that health care was the top issue (more than any other issue), and that 10% identified ethics and accountability as most important. Nevertheless, the economy (21%) and the deficit/budget (10%) scored highly as well.

But if the economy was the only vote-deciding issue, and with the upcoming election now front and centre in provincial news, Clark's Liberals would be doing a lot better. Their time might simply be up.

39 comments:

  1. This poll was conducted Mar. 8 to 12, too soon, I think, for the plethora of anti-Dix radio ads I've been hearing recently to have much affect. It'll be interesting to see if that helps or hurts the Libs. I was polled by Angus Reid late last week, so that may give us a slightly better idea if the attack ads are having any influence.
    Oh yes, almost forgot, the ads I heard were on an oldies rock station, so targetting, I suppose, us old farts in an attempt at influencing those of us who vote most.

    ReplyDelete
  2. Hi Eric,

    I don't put much credence in this poll (it seems to be a bit of an outlier), but I wondered if you had reviewed it and thought it should be included in your forecast:

    http://bc2013.com/wp-content/uploads/2013/03/Press-release-BC-Poll-March-14.pdf

    ReplyDelete
    Replies
    1. I've decided that because of MRIA's censure of that firm, I will be excluding them from inclusion until they prove themselves with a good campaign result.

      Delete
  3. The anti-Dix radio and TV negative ads have been running ad neuseum since late January...any poll conducted March 8-12 would have totally any effect they might have had.

    ReplyDelete
  4. It seems to me, the folks don't really like any of the leaders so they have given up basing their decision on leadership. They tell me CC has a very bad gender split against her even though she is a woman premier, women don't like her. What are the origins of this?

    Women hate cutbacks to health or education but this seems a little more viseral and personal.

    ReplyDelete
  5. In BC we don't vote people in....we vote people out.

    ReplyDelete
    Replies
    1. No different from anywhere else. There's a reason it's a Canadian politics maxim that oppositions don't defeat governments, governments defeat themselves.

      Delete
    2. BC went twenty years without re-electing a Premier. I think that's pretty unusual TS.

      Delete
    3. Ryan,

      I don't think it was quite 20 years. from 1991-2001 we didn't re-elect a premier: Harcourt resigned, Clark won the 1996 election then resigned, Ujjal was turfed from office in 2001.

      S.F. Tolmie

      Delete
    4. Don't forget Bill Vander Zalm who was chosen as leader and then won in the general election in 1986, then resigned and was replaced by Johnson, who then lost in 1991. So that's a span from October 1986 to May 2001 without a Premier winning re-election. We had a revolving door of 6 premiers in those (almost) 25 years.

      Gotta love BC politics.

      Delete
    5. I think you mean 15 years...

      Delete
    6. Ahah! Math, my old nemesis.

      I guess when people throw around the 20 years thing they mean Bill Bennet in 83 to Gordon Campbell in 05.

      Delete
    7. I'll certainly grant that BC seems to have an unusually scandal-ridden political culture that contributes to something of a revolving door in the premier's office.

      Delete
    8. Ryan,

      Bill Bennett was re-elected for a third term in 1983.

      Delete
    9. I'd pin it partially on our parties being particularly prone to turn on their own leaders at the drop of a hat too. There have certainly been some legitimate scandals, but there have been a few cases like Mike Harcourt's too IMHO, where a non-story led to a party turfing it's own leader.

      Delete
    10. Harcourt resigned not so much due to internal party pressure as public (media) perception. IMHO Harcourt made the right decision. The party was tarnished and someone within the NDP had to accept blame. It was honourable of Harcourt to be the fall guy and I have no doubt his resignation was instrumental toward the NDP re-election in 1996.

      Delete
  6. It appears that the only long term salvation for the liberal party is to have the Liberal Party reduced to non-part status in the next leg. Anything else leaves control in hands of those that let us all down even top leadership changes in interim.

    ReplyDelete
  7. "John Cummins had an approval rating of 16%, while his disapproval rating fell by seven points to 44%."

    As someone who was once represented by John Cummins federally, I can tell you that that 16% is very generous.

    ReplyDelete
  8. Charles Harrison18 March, 2013 21:03

    What happened to the Greens on the Island?

    ReplyDelete
    Replies
    1. The same thing that always happens this time in the election cycle unfortunately. Disaffected NDPers go from identifying as Greens to returning to the NDP fold as the election date draws closer. The BC Conservatives have the same trend happening too.

      Delete
  9. I can't recall when a political party polled over 50% (except maybe in Alberta). What about federally? Can't think of anyone.

    ReplyDelete
    Replies
    1. BC Liberals took more than 50% in 2001. Now, that's not an example I want repeated by the NDP, but yah.

      Delete
    2. Mulroney took more than 50% in 1984 too, as did John Diefenbaker in 1958.

      There have been other polls with the NDP over 50% in the last couple years in BC though.

      Delete
    3. Correct, although barely -- Mulroney PCs got 50.03%. Diefenbaker had a somewhat larger majority. Before that, my guess is you'd have to go back to 1910s before the rise of 3rd parties.

      Delete
    4. Dief got 53.66%!

      Mackenzie King got 52% in 1940

      Sir Robert Borden took 57% in 1917

      Delete
    5. Borden's a bit iffy though isn't it? Since it was a coalition government of Conservatives and Liberal-Unionists.

      Delete
    6. The Saskatchewan Party in Saskatchewan won the last election with over 60%

      Delete
    7. Brad Wall's Sask Party is consistently up there. The last election they won with over 60% IIRC

      Delete
  10. Honnestly, I don,t get how, with a 20-points lead in the poll and being above 50% of voting intentions, why the NDP would only have 97.5% chances of winning. To me, this is really playing it too safe. it's like you want to be sure not to be wrong. But in this case, even if the polls were as wrong as they were in Alberta (which I think should provide us with a bound for how wrong polls can be), the NDP would still win...

    ReplyDelete
    Replies
    1. I'm afraid you're wrong, the difference between the results and a few of the final polls was a total swing of 19 points, more than the current 18.4-point margin I am projecting right now some two months before the election.

      And the model gives the NDP an "only" 97.7% chance of winning the popular vote because there have been cases of a swing of more than 18.4 points this far out before an election in the past.

      As to your other points, it would be foolish to assume 100% certainty of anything before the election campaign has even begun. There is no "playing it safe" here, the model spits out what it spits out. It is based on past cases, so if it has happened before it can certainly happen again.

      Delete
  11. Interesting to note that the UBC BC Election Stock Market has these results as of today:

    NDP - 42.1%
    Liberal - 32.2%
    CPBC - 14%
    Green - 9.2%
    Other - 2.4%

    http://predictionmarkets.ca/BC13.php

    "the markets deliver predictions of popular vote shares that are often more accurate than public opinion polls"

    ReplyDelete
    Replies
    1. You'll note, however, that these markets are about guessing the result of the election - not about gauging support right now. The current trading levels are all within the forecasted May 14 ranges on this site.

      Delete
    2. I believe (correct me if I'm wrong) that there's some evidence to suggest these betting markets perform as well as or better than polls.

      Delete
    3. The problem with prediction markets is that they are easily gamed. Nate Silver commented on this during the presidential election when it was obvious that Intrade was being gamed to make it look like Romney had a better chance than he did.

      I also challenge the assertion that the prediction markets are better at predicting the result than polling, especially in the pre-writ period. A prediction market is no more likely to be accurate this long before an election than polling is.

      Delete
    4. I disagree. Polling can at best give what Nate Silver calls the "now-cast", not a forecast. The typical poll question is "if an election were held tomorrow...", not "how do you think your opinion will change in the next few months?" The latter question would be absurd, because if I thought I would change my mind to support X, I would change my mind already.

      I think this distinction is especially important before the election campaign is in full swing, because the incumbent has higher media exposure than the challenger. (Admittedly, this is less of an issue in Canada than in the United States, because we have an institutionalized Official Opposition and shadow cabinet.) People who bet in the prediction markets may be able to consider leading indicators of public opinion, the signs that suggest how opinion is likely to change in the future.

      Delete
    5. That's a good point. I imagine the volume on this market is quite low, so probably quite volatile.

      Delete
  12. One BC govt leader Clark will be happy with Dix's win Glenn Clark that is!

    ReplyDelete
  13. Charles Harrison20 March, 2013 20:50

    I don't trust this poll in the Green numbers; my prediction using this poll and the Angus-Reid poll giving this poll more weight gives the Greens only 9.5% of the vote in West Vancouver-Sea to Sky!

    ReplyDelete
  14. Does anyone believe there will be a merger between the BC Liberals and the BC Conservatives to ensure a grander coalition of ideas occur or is the BC Conservative party here to stay?

    I don't think the election will be as much of a bloodbath than originally thought. The polls appear to be much closer now and by May I believe it will be within 10 points.

    If the NDP does win, it will most likely be a referendum on how well the BC NDP governs.

    ReplyDelete

COMMENT MODERATION POLICY - Please be respectful when commenting. If choosing to remain anonymous, please sign your comment with some sort of pseudonym to avoid confusion. Please do not use any derogatory terms for fellow commenters, parties, or politicians. Inflammatory and overly partisan comments will not be posted. Please keep discussion on topic.