The latest figures released by EKOS and fed into our projection model suggest that the PCs would end up with a solid majority government and bring the Ontario Liberals below 19 seats.The Ontario NDP, meanwhile, would gain a handful of seats and Andrea Horwath would become the leader of the official opposition.
This is something to keep in mind as the Liberals and NDP negotiate over the upcoming budget.
Perhaps the NDP's list of micro-demands are a very deft move, as Horwath's reasonable gains for various constituencies in the middle of a jobs crisis wouldn't be seen as precipitating an election. If she obtains them, people may thank her. If she does not, and the Liberals take a my-way-or-the-highway approach to the budget, causing an election, the Liberals will look bad for having nothing to offer on youth unemployment.
The prize of earning Official Opposition status might be very valuable symbolically for the NDP, even though it would actually represent a net loss of power for them.
On the other hand, Hudak's preferred policies are extreme enough to likely prompt the NDP and Liberal bases to pressure their leaders to find a solution to avoid an election, which looks set to lead to an even more austerity-prone government.
2 comments:
Can you give any commentary how the model concluded that the Liberal vote distribution would be so inefficient and the NDP suddenly is so efficient? It seems a departure from past history where the NDP is distributed too widely to garner seats.
Also, how did your model project a PC majority with such a thin popular lead? Does the model favour or bias incumbent governments, new/existing leaders?
I think a good starting place is to note that there are 10 races with the Liberals at close second, 8 PC close seconds and 3 NDP close seconds.
I don't think that it's so much a matter of efficiency as much as it is "progressive vote splitting". All of the gains for the PCs represent ridings where the NDP's share of the vote is equal or greater to the margin between the Liberals and the PCs. Simply put, the NDP is doing well enough to sap enough support from the Liberals in ridings across Ontario that the PCs are able to get a majority.
That's the case observed in Ajax-Pickering, Ancaster, Brampton-Springdale, Brampton West, Brant, Glengarry, Kitchener Centre, (Not Kitchener-Waterloo, interestingly, which the PCs are projected to re-take), London North Centre, Mississauga-Brampton, Mississauga East-Cooksville, Mississauga-Erindale, Mississauga South, Niagara Falls, Oak Ridges-Markham, Oakville, Ottawa-Orléans, Ottawa South, Ottawa West-Nepean, Peterborough, Pickering-Scarborough, Richmond Hill, St. Catherine's, Scarborough-Agincourt, Willowdale, and York Centre.
With that said, there are a number of problems with polling in Canada. One is that most times a "likely voter" model isn't even proffered. The EKOS one attempts to do so, rather clumsily, as has been criticized elsewhere.
The model implements some of the Threehundredeight.com discoveries, such as a bonus for an incumbent when the party is behind its previous province-wide scores. That bonus is set at 1.05 of the result of the model. So the results here are actually with a Liberal bonus.
Post a Comment