A couple of days ago, a poll by Forum got a lot of attention in the media essentially because it was showing a much closer race in Ontario after the recent budget from the Wynne Liberal government. Forum was also showing that many measures included in the budget were actually rather popular. The budget overall wasn't incredibly liked (a plurality still disapproved of it) but its effect on voting intentions was definitely in the OLP's favour.
This morning we have a new Mainstreet poll showing pretty much the exact opposite! How great. According to Mainstreet, as many as 53% of Ontarians are saying the budget is making it less likely they'll vote for the Liberals! (Note: the Forum poll actually had a similar result despite the fact the Liberals were up in this poll. Weird). More importantly, in terms of voting intentions, the Conservative of Doug Ford are now above 50% among decided and leaning voters! That's right, Ford and the PC could win a majority even if Ontario had a proportional system!
The table below shows the differences in variations before/after the budget for the two firms.
Hard to find a clear explanation as to why the two pollsters have such different variations. Their polls got conducted relatively during the same period, both before and after the budget. It's also surprising to see Forum showing such a decline for the PC as this party was usually higher in the polls of this firm than in others. It's possible we are simply reading too much into some numbers here and it's simply noise created by relatively small sample sizes. Still, Forum went from being the pollster with the lowest OLP to the highest (and almost the perfect opposite for the PC). As usual here, we will have to wait and see. Or go wild with your theories (alternatively, you can cherry pick the pollster you like more, either for valid reasons or simply because their numbers are showing what you want to see). Personally, I think Forum is experiencing one of their usual wild swings between their polls. Because the questions specific about the budget got relatively similar responses from both pollsters.
Using the Forum and Mainstreet polls (as well as others, slightly older and thus given a lower weight), we get the following projections. Remember that "chances to win" means winning the most seats. I'm not trying to predict coalitions or anything. It doesn't matter much for now since the odds of a majority are of 93.5%.
 |
| Voting intentions; Seat projections with confidence intervals; Chances of wining the most seats |
It's fairly stable with the previous projections mostly since the Forum and Mainstreet variations are cancelling out each other. If I was using the number of the Mainstreet only, the chances would obviously be at 100%.
I'm still finishing the model and will need to make some adjustments based on some unpublished riding polls I saw as well as adding the results of some by-elections, but details like these don't change the big picture: as it stands, two months before the election, the Conservatives of Ford are overwhelmingly favourites to win. And a majority is also very likely.
The Tories are dominating pretty much everywhere. They literally have a lead in every single region, including Toronto (the 416) where Ford appears to be quite popular. The North could go NDP but it won't matter much given the sweep of seats the PC could get in the 905 and elsewhere.
Let's put it another way: if the election was tomorrow, you wouldn't need a model to predict a Conservative majority.
It doesn't mean things can't and won't change. Mainstreet has some interesting numbers about the certainty of the vote.
24.8% of the Liberal voters are likely to change their mind and vote for their second choice (which is the NDP at 57%!). For the PC, it's 21.8% that are in this situation (although the main second choice is nobody - 53.1% - followed by the NDP at 29.8%. Yes, the NDP is the main second choice of the PC voters, not the Liberals). As for the NDP, its vote is, as expected, less certain with 30.6% who could switch their vote (the second choices for NDP voters are all over the place, with some who would go Liberals - 34.6% - some PC - 26.4% - or even to the Green - 17.6% -).
Second choices are useful but have caveats. They can change just as the first choices can. With that said, the current second choices tend to show that if the PC was to be stopped, the NDP might be more likely to do so. Imagine that the party of Andrea Horwath were to get some additional support from the Liberal voters (who have this party as clear second choice) as well as from the PC (maybe some centrist voters turned off by Ford). In this scenario, the NDP could climb to almost 30%. And that's only assuming that 57% of the 24.8% of unsure Liberals would switch. If the NDP was to get to 30% however, we could imagine that even more Liberals could be tempted to switch in order to block the PC.
The fact the NDP is the main second choice (excluding "none") of both the Liberals and PC means this party has a real potential for growth. But the NDP has had a high potential in Ontario for a long time without being successful in turning this potential into actual votes. Will this time be different? One thing going for them is the fact that Andrea Horwarth is the most popular of the three leaders. She's actually the only one with a net favourable rating.
I'm not saying the Liberals are out of this race here, I'm simply saying it is relatively easier to picture a path for the NDP to climb higher (and take votes from the PC) than for the incredibly unpopular Wynne's Liberals. The Liberal vote is also currently quite inefficient and things could get worse if Ford is indeed popular in the 416 and 905.
Alright, that's all for now. I'll try to make the last adjustments to the model by the end of next week. Just in time to start an actual tracking a couple of weeks before the election.
Ontario Proj Detailed April 6th 2018 by bryanbreguet on Scribd
