Wednesday, July 27, 2011

NDP wins fourth mandate in first Manitoba projection

The Manitoba projection model is now completed, and projects a New Democratic majority government in the province, with the Progressive Conservatives forming the Official Opposition.
However, the NDP wins this fourth mandate with fewer votes than the PCs.

The Progressive Conservatives are projected to have the support of 44.3% of Manitobans, compared to 39.9% for the NDP.

The Liberals stand at a projected 11.5% support, while the Greens are at 4.0%.

But because of greater vote efficiency, the New Democrats manage to elect 33 MLAs to Winnipeg, compared to 22 for the Progressive Conservatives and two for the Liberals.

This represents a loss of three seats for the NDP, a gain of three seats for the PCs, and a gain of one seat for the Liberals compared to their current standing in the Legislative Assembly.

Outside of Winnipeg, the Progressive Conservatives take the majority of the seats with 16 to the NDP's 10.

The PCs take 10 seats in the southeast, six in the southwest, and none in the north, while the NDP takes four seats apiece in northern and southeastern Manitoba and two in the southwestern part of the province.

In Winnipeg itself, the New Democrats still dominate with 23 seats to the PCs' six and the Liberals' two.

Understandably, as the NDP is down by 4.4 points several of the seats they hold in the projection are marginal.

The NDP has a five-point or less lead in five seats, while the PCs are leading by such a margin in two seats and the Liberals one.

The PCs are trailing by five points or less in five seats, while the NDP is behind by five points or less in three.

This puts the NDP range at between 28 and 36 seats. At the lower end of the scale, the NDP is in danger of being defeated by the PCs while at the higher end it puts them on par with their current caucus. For the PCs, they could win as few as 20 seats or as many as 27, indicating that at this point they stand a very good chance of improving upon 2007's results.

The Liberals range at between one and two seats. This is the difference between the party holding on to the successor to the riding of Inkster after Kevin Lamoureux's departure or not.

Now, Manitoba's ridings have been re-drawn. Some of the ridings have changed completely while others have changed only a little bit. Many thanks go to Kirk Vilks, who calculated the transposition of votes using the poll-by-poll breakdowns. In the chart to the left, incumbents are determined either by the sitting MLA who is running for re-election in a given riding, or by the winning party in each riding after the transposition of votes.


  1. Goaltender Interference27 July, 2011 12:18

    I have nothing for or against any political party in Manitoba. I just want to point out that any system where a political party can get come second in votes and form a majority government is just simply flawed.

    Why do squiggly lines dividing riding boundaries have a greater impact on the outcome of an election than the actual number of votes that a political party receives?

    There are plenty of alternatives, some better than others. But our current electoral system is indefensible.

  2. What maths have you been using to get these numbers? I would like to see the transposition of the votes, as they're not available. (One guy I spoke to had them, but since he was paid to do it, he couldn't give them to me). I'm not sure if that's the case for you.

  3. I explained in the post. As they were kindly given to me, they are not mine to give out.

  4. "There are plenty of alternatives, some better than others. But our current electoral system is indefensible."

    I was wondering how long it would take for the FPTP bashing to begin !

    Our system is perfectly defensible.

    Its based on community representation. Tories dominate rural areas, NDP have a smaller edge in urban communities.

    Any crazy, complicated, rigged system that the FPTP bashers would come up with would disregard the wishes of urban ridings.

    Rural voters would be running everything.

    I prefer a balanced system where ALL communities are heard.

  5. A significant point not noted, is that in the lowest NDP range, after a Speaker is appointed the NDP would be a minority government and could be defeated. With a billion dollar deficit upcoming thanks to profligate spending promises made without Legislature debate or approval, the chances for Selinger to survive that scenerio are mighty slim.

  6. I would very surprised if the Liberals even saved their deposit in Tyndall Park. With Kevin Lamoureux out of the picture the Liberal vote there will almost certainly evaporate (see federal election results in Esquimalt-Juan de Fuca post Keith Martin)

  7. Erm, DL is right on that - because Tyndall Park is not really the successor to Inkster, The Maples is closer to its successor. If anything, The Maples would be the likely Liberal-held riding, not Tyndall Park. The latter contains more of the old riding of Wellington which was a NDP bastion, than it does Inkster. The Maples contains more of Inkster.

  8. Looks to me like the more highly populated part of Inkster is in Tyndall Park, not The Maples.

  9. Also, the more solidly Liberal part of Inskter (60%+) is in Tyndall Park.

  10. If that's true, then it still doesn't matter - Inkster is still dwarfed by the Wellington portion, is it not?

  11. Seems that 2,500 fewer people voted in Wellington than they did in Inkster. That explains why Tyndall Park appears to be more heavily weighted towards Inkster.

  12. Wellington also seems to have been incorporated into three other districts.

  13. To give that 2,500 number some context, about 4,300 people voted in Wellington while 6,800 voted in Inkster. There were 1,600 more Liberal voters in Inkster than there were NDP voters in Wellington.

    It looks right to me.

  14. The Tyndall Park riding contains about 5800 votes from the previous election. Of that about 2100 were from Wellington, and about 3700 were from Inkster. About 85% of Liberal vote in the new Tyndall Park riding were from Inkster, while only about 45% of the NDP and PC vote in the new Tyndall Park riding were from Inkster. If everyone voted the same as in the last election then the result would be a close win for the Liberals.

    I would agree with DL though, I think the loss of Kevin Lamoureux will probably remove enough Liberal support for the NDP to take the seat.


  15. The "Liberal" vote in Inkster will prove to be about as real as "Liberal" votes in Esquimalt-Juan de Fuca. I think that unless Lamoureux has an identical twin with the same last name, you might as well deduct about 75% of the Liberal there right off the bat because he won't be on the ballot. The Manitoba Liberals are essentially a dead party and Lamoureux only ever won be being a quasi-independent.

  16. Eric said he doesn't use local intelligence.

    That's for elections predictions project.

    Even though its OBVIOUS that the loss of Kevin Lamoureux means that the Liberals can not hold that seat.

    Politics isn't math ! Eric you need to find a way to incorporate what people are telling you into your model.

  17. The Liberals not having an incumbent in Tyndall Park is taken into account, and their vote has been reduced accordingly.

  18. If there really is a 28-27-1 outcome in favor of the NDP, it would be hard to see any other outcome other than the Liberals (Assuming Gerrard is the only MLA) in some sort of coalition agreement giving the Liberals a few minor concessions.

    A couple other seats I would think should also be closer under the current projection: St Norbert has a strong Liberal candidate (who was a former PC MLA) that will likely split the votes with the PCs. Even if the NDP MLA is not re-offering, It would be highly unlikely that this seat will be that strongly PC. Another seat is Brandon West, due to the extremely strong candidate for the PC's (and former MP) retiring. This seat had also been won by the NDP strongly in 2003.

  19. Five 0 Six,

    The lower seat range would give two seats to the Liberals, so 28-27-2.

    The Tories' lack of an incumbent in Brandon West is taken into account, but you may be right that the vote shift could be more dramatic.

  20. Anonymous 15:04 shows that all he is interested in is Tory Govt.

    He won't accept any other version or thinking. Yet most agree FPTP really isn't Representative Govt !!

  21. Peter 12:25 should go back and realize that in this case FPTP helps an NDP GOVERNMENT IN MANITOBA.

    Sorry for the caps but c'mon ! Learn to read man !

    Plus where is the proof that "most agree" FPTP isn't represenative ! !

    Its the very defintion of representative. Its community representation !

    Geographic representation instead of just rep by pop.

  22. FPTP sn't representative of the population as a whole, but it is representative of 308 discrete pieces of population.

  23. "Geographic representation instead of just rep by pop. "

    And we have ?? Rep by pop. That's supposedly the current system.

    Not geographic but representing the POPULATION !!

  24. Goaltender Interference01 August, 2011 12:12

    "Geographic representation instead of just rep by pop."

    That's an argument in favour of FPTP? Legislators are supposed to represent people, not tracts of land. Highways don't vote, railway tracks don't vote, yet they determine who gets elected to the legislature? It makes no sense.

    "FPTP isn't representative of the population as a whole, but it is representative of 308 discrete pieces of population"

    Yes, 308 arbitrary pieces that have nothing to do with the way people live and therefore make their "representativeness" completely meaningless. When I lived in downtown Montreal, I would start my day in my apartment below Saint-Antoine Street (in St-Henri-Ste-Anne, a competitive riding), walk five blocks north to Rene-Levesque Blvd to work (in the Liberal stronghold of Westmount St-Louis), and usually have lunch a 10 minute walk away on St-Laurent Street (in the PQ stronghold of Saint-Marie-Saint-Jacques.) Those riding boundaries mean nothing at all to anyone who lives in or near them-- they don't represent neighbourhoods or communities or anything else. They are just arbitrary administrative conveniences to uphold the ancient voting system that we have. Their arbitrariness dilutes the representativeness of the legislature such that it can end up misrepresenting the whole province.

    I couldn't care less if FPTP benefits the NDP, the Liberals, the PQ or anyone else. The very fact that a voting system can be perceived as advantageous to one party rather than neutral between all parties, is perhaps the biggest argument against FPTP.

  25. Goaltender you are absolutely correct. Which was the thrust of my earlier comment.

    FPTP is simply NOT a fair election system !!

  26. Wrong Peter, just wrong.

    Its called communities of interest !

    Have more respect for the people who spend a lot of work on redistricting after every census.

  27. anonymous 04:24

    You just don't get it do you. People vote, not communities, not regions, not highways, PEOPLE !!

    The current FPTP system essentially disenfranchises great chunks of the country !!

    Just look at the last Fed election to see. Less than 40% voted for the CPC but they got a majority of seats because of this crappy system !!

    CPC should have got 123 seats for 40% of the vote, instead they got 165, You think that's either fair or balanced ???

  28. Peter what you are suggesting is an attack on Quebec !

    Its an attack on ethnic communities. Its an attack on Atlantic and rural Canadians.

    Communities DO matter. And Canada is a nation of many, many diverse communities.

    There's no such thing as a one size fits all, that's why its simply WRONG to say the CPC got 40% of the vote.

    They got various amounts of the vote in 308 ridings. THAT is our system.

    Its not a presidential system. Its a parliamentary system.

  29. Goaltender Interference03 August, 2011 19:59

    Your argument is spurious. The 308 ridings don't correspond to "communities of interest". They are just chunks of roughly 100,000 people divided up by Elections Canada. Unless you live in a very large, rural riding, then many of the people you interact with every day live in a different riding from you.

  30. GI

    Couldn't have said it better myself. Thanks

  31. Thank you GI though I think any discussion with Anonymous on this verges on farce as we have no idea who we are addressing.

    But FPTP has been proven faulty so lets address improving the system ??

  32. It's pretty rare to win ALL the close races, especially when your party is trailing in the overall popular vote by four points. The NDP would have to be extremely lucky to pull that off. I still think the signs point towards the PCs winning a slim majority on the basis of prying a few of those Winnipeg seats free.


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.

Note: Only a member of this blog may post a comment.