The Ontario election isn't until June 7. We often say that a couple of months is an eternity in politics. Kathleen Wynne, the current Premier of the province, better hopes this is true as she is currently significantly behind the PC of Patrick Brown. She'll need to rally greatly if she wants to keep her job. Or she can hope for a massive polling mistake.
Wynne suffers from being the least popular Premier in the country according to Angus-Reid. At 17% approval rate, this makes Donal Trump looks like a popular president (although it's an improvement over the low of 12% last year). It's therefore natural to see the first projections giving a huge advantage to the PC. Using the three most recent polls -all done this month- from Forum Research, Campaign Research and Mainstreet Research, the model is giving Patrick Brown over 87% chance of winning the most seats. Below you have the details:
 |
| In order: voting intentions; Seat projections with 95% confidence intervals; Chances of winning the most seats |
If you want the riding-by-riding projections, you can find them at the end of this post. Remember also that you can use the model yourself with the simulator.
The possible outcomes are given here:
Please keep in mind that this is the first version of the model. I'm still tweaking things out. In particular, it seems the model is very harsh on the PC in term of vote efficiency. This means the Liberals could win with significantly fewer votes. Or the PC could be denied a majority despite reaching the 40% mark. Why is the model so hard on the PC? Because the model doesn't use a uniform swing approach. Instead the model uses past elections' results to estimate regional coefficients of variation. Since 2003, the greater GTA has been more stable than the North or the Southwest, significantly and consistently so. If you use a uniform model, the PC is at 69 seats and therefore would get a majority. Please notice however that the swing in Ontario hasn't been very uniformed the last 3 elections.
Let's also remember that the OLP won a majority in 2014 with only 38.65%. And in 2011, Dalton McGuinty was only 1 seat away from a majority with only a 2-points lead over the PC. All that to say that nobody should be surprised at the electoral advantage the Liberals have. What I'm currently unsure of is the magnitude of this advantage. Between now and June, I'll keep monitoring the polls and projections and will likely make some adjustments. For now, if you think the model is too hard on the PC, then just boost the PC by 1-point in the simulator.
For Patrick Brown to get a majority, he'll need to break the GTA-wall the Liberals have been enjoying for many years. With the Liberals polling as low as 30%, this is well possible. Think of how Harper got many seats in the GTA in 2011. If he were to fail to get 63 seats or more, the NDP of Andrea Horwath would have a choice to make. Would she follow Andrew Weaver from the BC Green and support the party that finished second? Hard to say right now.
Quito Maggi from Mainstreet says that the numbers point to a PC majority. I agree with that based on his poll and his regional numbers alone. The PC is higher in Toronto and the GTA than what my projections suggest. But I'll wait and see before making adjustments. As usual, if I see a pattern in polls not matched by my projections, I'll make changes. With that said, the model does account for the fact that if the Liberals fall too far (like below 30%), then their resilience in the GTA will be gone.
Quito Maggi from Mainstreet says that the numbers point to a PC majority. I agree with that based on his poll and his regional numbers alone. The PC is higher in Toronto and the GTA than what my projections suggest. But I'll wait and see before making adjustments. As usual, if I see a pattern in polls not matched by my projections, I'll make changes. With that said, the model does account for the fact that if the Liberals fall too far (like below 30%), then their resilience in the GTA will be gone.
For now, my personal read of the situation is that the Conservatives are clearly ahead and favourites but a majority isn't guaranteed. It's absolutely possible, just not a sure thing. The Ontario Liberals aren't easy to beat (honestly, the Liberals everywhere in Canada are often very resilient, just look at Quebec or BC where Christy Clark, despite low approval numbers, was a couple of votes away from a majority last year). I wouldn't be surprised if the numbers were to become a lot closer as we approach June. With that said, the situation in Ontario is ripe for a change. You have a governing party that has been there for years and is very unpopular. You have a housing market that has gone crazy over the last few years, possibly upsetting the middle class in the suburbs like we saw in the Lower Mainland. Add to this big policies like the rental regulations last year and the recent $14 minimum wage and you have all the ingredients for a heated campaign.
If there is a desire for change, let's also remember that there is an established third party in Ontario. The BC NDP are stable compared to four years ago and will likely look to try to grab some of the discontent voters in the GTA. Voters that have showns at both the federal and provincial levels that voting Conservatives is possible for them but far from natural.
Anyway, this should be it for now. I don't want to make too many comments on the three polls themselves. They have different methodologies and Forum has the Liberals much lower than the other two. But 5 months from the election, I won't start to worry too much about that. Things can and will likely change between now and June.
Detailed projections:
Detailed projections:
Ontario Proj 23-01-2018 by bryanbreguet on Scribd
