Jordan: It’s a little like waiting for the other shoe to drop, if the precise timing of dropping that shoe was being debated in media and political circles, and was being strategized about in the back rooms, in the corridors of power in Ottawa.
News Clip: Stephanie Plant, a researcher of Canadian politics at the University of Ottawa, believes the Liberals want voters to settle down after heading back to work following summer vacations. So the prime minister might not drop the writ until closer to the September 15th deadline.
News Clip: The shorter the time frame, the shorter the sprint, the faster he could get his messaging and his team out there.
Jordan: It might seem as if the past few months have been an endless cycle of polls and predictions and reaction to them and people scratching their chins and wondering what it all means. But really, we haven’t even started yet. As soon as the writ drops, the polls will come daily and they will come from everywhere. Some of those polls will be bang on. Some of them absolutely will not be. That will apply equally to analysis of them. So rather than try to dig into which polls are good and which are bad and who’s got it right. We will take a more scientific approach. When you factor all of those numbers together. What stands out? What’s just noise? Leaving out these screaming weekly headlines, what has actually shifted in the Canadian political landscape this summer? And if you happen to be a normal person with a job and a life outside of the election bubble, what do you really need to pay attention to over the next couple months? I’m Jordan Heath Rawlings, and this is The Big Story. Philippe J. Fournier models Canada’s elections at 338canada dot com and at McClain’s dot ca. So, Philippe, we talked to you way back in April. What have you seen shift in your projection model since then?
Philippe: Well, actually plenty, since the last time we talked together. Jordan, the conservatives were way ahead because mostly because of the SNC Lavalin story and Jody Wilson-Raybould just announced that she would be an independent candidate along with Jane Philpott, and so back in April and May the conservatives held 4 to 6 point lead on average in the polls, and they were heading to a victory. And then, well, summer happened. Summer happened and it seems like the Conservatives wasted an opportunity, especially in Ontario and Quebec, to grow their base into, uh, and inflate their numbers. But we have seen several polls from professional firms, Angus Reid, IPSO, and all these guys, were on the field this summer and they showed, especially in Ontario, the Liberals gaining ground back and the conservatives slipping. So right now in the popular vote projection, we have a statistical tie. But a statistical tie means most likely a Liberal win because there is so many votes concentrated for the conservatives in Alberta, that it inflates, their popular vote number, but it does not increase their seat count. So right now, I would say that the most likely scenario would be a liberal plurality, probably a minority, but it could also be a majority. And so we’re on the starting line.
Jordan: How quickly can those things change? Because, you know, we’re all wary of polls a couple of months out of the election, and to your point, they just changed over a couple of months, even when you’re working with kind of the, ah, the compilation of all the polls that are taken, how fast to those things tend to move?
Philippe: Well again, we have to be careful, and I know there’s a lot of public mistrust about polling. I would attribute that to, uh, not really understanding what polls do. Polls do not predict the future. They’re a snapshot of the present and usually of the recent past. And so what the numbers that we have right now indicate that yet Justin Trudeau is the favorite to win the election by not that much. But that’s right now. And so you know, next week there’s going to be the first debate, most likely the first debate of the season hosted by McLean’s, and you know, it would only take one bad performance by Trudeau, or one really good performance by Andrew Scheer or Jagmeet Singh, or Elizabeth May to completely during this thing around.
Jordan: When you’re dealing with a compilation of poles like that and you know, to your point you mentioned one debate that’s coming next week, how do you go about attributing what’s driving that? For instance, I’m thinking, Ah, now of the recent SNC Lavalin ethics report, or maybe of the liberals highlighting Andrew Scheer’s past comments on gay marriage. Can you drill down into, Ah, your data and try to model what’s driving it?
Philippe: Well, that’s a very good question, Jordan and I have to tell you that in science, usually, when do you want to see the effect of variable? You change only one variable, and you keep everything constant. In social science and in political polling, that’s not always possible. And so I’m always very, very cautious to attribute cause and effect, to events and polling because we really do not know for sure. You know, the exception was maybe the SNC Lavalin story because it just occupied so many media cycle of last winter. And we saw the Liberals dropping support for eight consecutive weeks after the Globe and mail story was out, and so that was pretty straightforward cause and effect. But, uh, the the ethics commissioner report came out on August 14th I believe, and we’ve had six or seven national polls that was on the field, that were on the field after that date, and we have not seen a dent in liberal support, so it doesn’t seem to have any effect. And the, uh, you said about the well, the gay marriage and the social issues and the abortion story. Abortion story was a big story in Quebec. I’m not sure how big it was in Canada because we’ve had polls on the subject. Leger last year pulled the Canadians on their views on abortion. And, for instance, Quebec, 85% of Quebecers feel that the abortion issue is closed, that issue is done, and so there is no ground to be made. There is no votes to be gained for the conservatives, you know, touring the country, talking about abortion. But there’s plenty of ground to be made by the liberals talking about the conservatives talking about the abortion.
Philippe: Um, and so straightforward cause and effect, I’m very careful with that, but for sure, the past few weeks have not been good for Andrew Scheer. Thankfully, for him, he has a lot of times to get, get back up.
Jordan: One of the reasons we like talking to you is because you’re you’re not focused just on the national polls. Although those play you know those play a role in voting intentions. What have you seen recently in the regional polls? That really stands out to you when you’re putting it into your model?
Philippe: There’s been some very interesting local numbers .I would start my home problem, if you don’t mind. In Quebec it was really interesting is that when the SNC Lavalin story occurred in February, voting intentions in Quebec– it was the only place in Canada where voting intentions pretty much remained stable. The Liberals did not gain, but they didn’t really not really lose, either. On we start a conservative playing up a few points, but it was, you know, at the expense of, you know, the fact that there’s a lot of undecided, and it was always within the margins of error, but in Quebec the voting intentions really didn’t move, except for the Bloc Quebecois who, uh, you know, they had a new leader, and so they gained a few points. But outside of that, we saw changes in Atlantic Canada. Atlantic Canada was all liberal was all red four years ago. But when we look at the numbers now, well, the liberals are still in front, but it’s not dominating lead like it was in 2015. and so they should lose a few seats in the region, especially in New Brunswick. But again, you know, and I know it’s an old cliche, but Ontario, well, Ontario has the most seats and has the highest population, and so it’s gonna be a battleground, especially in the suburbs of Toronto. The 905 area has about 30 district’s, and I checked my numbers just this Sunday. Out of those 30 District’s 19 are either toss ups, or leaning, and so so that’s the election right there. You know, it’s it’s just as good a one a 14 seat majority in 2014 and their 19 toss up seats in the in the 905. So if let’s say, the conservatives do a little better than we think and they win, most of those seats, you could see well, very diminished, Ah, liberal in government. Or you could you know, the liberals outright, losing power. So every every scenario is on the table. Albert and Saskatchewan one did not move much. We know that the conservatives are very popular here. I had a few ah readers from Alberta saying, Well, We want to know will the Conservatives win 33 seats or 34 seats.
Philippe: Right now, that’s a question right now in Alberta. Finally, if we go out west to British Columbia, while their rural parts of the BC should be mostly conservative, Vancouver will be a battle between the Liberals and the NDP, With Green maybe playing spoilers, and Vancouver Island should be very interesting. Not necessarily for who gets to win power, but the Green Party, well, their, Their numbers are really good on Vancouver Island, and there are seven districts over there, so they should be in good position to make few gains there. What surprised me most was the NDP numbers. Thea NDP in Ontario, is down, in Quebec it’s getting wiped out. It’s not doing well in Atlantic Canada, and so when we look at the seat count for the NDP, they won 44 seats four years ago, and now they would be lucky to get 1/3 of that right now.
Philippe: And the numbers could be wrong by a few points. And of course I will adjust when new numbers are published, but right now, it’s really not looking good. In fact, I was I was told by a reader that really noticed that right now the Bloc Quebecois is expected to win more seats than the NDP.
Jordan: That was my next question, is you mentioned the Bloc a little bit and you mentioned the surprising performance. Ah, by the greens, especially out West. The last time we talked, uh, people were flocking to the greens, or at least more than they had been, because climate change was an issue that was clearly rising in the polls. Has that continued? Where are they right now and and how much of a spoiler could they play?
Philippe: Well, that’s a very good question. The Green Party. Well, the surge that they had in the spring was a modest surge, but it was a real surge. We saw them go from 5 to 6% all the way to ten to 12%. But it’s stalled there. Uh, they seem to have hit a ceiling in their numbers and, I can’t really tell you why other than the fact that right now since the vote projection is so close in Canada, at some point, those voters will maybe say Okay, so if we let if you vote green, maybe we’re helping electing a conservative government or a liberal government. And so they’ll have to take a decision because of our voting system. You know, the first past the post, you have to make a decision. If you vote for a third party that doesn’t have a chance to win your riding, you may help a worse option. Uh, in your mind, of course. I’m not saying which one is worse, which one is better. But if we look at the regionals for the Green Party of Canada, well we know that they’re doing well in BC as I mentioned there around the 16 17% on average in BC and elsewhere in the country well, they’re at the 10% threshold. Pretty much everywhere. Uh, 8% in Alberta, 7% in the praries, 13% in the Atlantic. Uh, so is that enough to win new seats outside of Vancouver Island? I am not too sure. The numbers don’t suggest that, but let’s say, for instance, we take Fredericton. There should be a very close race between the liberals and conservatives. Let’s say they split that in the middle with maybe 28 to 30% each. Yes, we could see a green party just going between and squeak a victory there. But as for the green surge, well, 10% is good for their tender, but it doesn’t win you any seats.
Jordan: What are the chances? And maybe I shouldn’t be asking a scientist for just speculate on the chances. But if the NDP collapse continues and the green stay where they are, what are we talking about In terms of balance of power, when you say one of the likely scenarios is a minority government?
Philippe: Well, allow me to take you put on my political analyst hat for a second. Uh, let’s say we have a minority liberal government and let’s say, as you said, that the NDP collapses to maybe maybe 10 or 12 seats. Well, is it possible for Jagmeet Singh to remain NDP leader? If you leave the party to the worst result in 25 years? Well, if you hold the balance of power, well, he could make a case that I did that he should stay, but it’s gonna be very hard if those numbers translate into ballot boxes on October 21st with the NDP. I do not see how Jagmeet Singh remains NDP leader. And so for the Green Party, Well, they’re expected to win five or six seats. Uh, so it would to hold a balance of power, The liberals would have to be very close to that majority threshold. So, you know, I we have not seen coalition in this country since Pierre Trudeau for did it in 1973. I was not alive, and I’m pretty darn sure that you weren’t in the live either. And so I’m not sure the culture of coalition is ingrained in the political in the political science in Canada. So we’ll have to wait and see. But you could, if, let’s say, for instance, with the Greens and the NDP say we will support your government. But we want electoral reform. Will the Liberals do that?
Philippe: That’s that’s a fair question. But if I were the liberal that we’ll just tell them well, no, not happening and who are you going to support the conservatives? They don’t want the electoral reform either, so they would have the greens and the NDP would have much, much more power, of course, if they could win maybe 40 or 50 seats. But if they’re just winning a dozen, then give them the left. You were a bargaining chip.
Jordan: The writ is expected to be dropped in a week or less or so. What changes when that happens in terms of your modeling and the work you’re doing?
Philippe: Oh, except that I will not sleep for the next six weeks. Uh,
Philippe: I told I told my wife well, it’s gonna be fun to see you sometime later this October. But I expect the polling companies to be on the field heavily right from the get go to see where you know where Canadians stand early in the campaign, we expect polling firms like Main Street to have a daily tracker that I will follow. So I’m gonna follow maybe 7 to 8, maybe nine polling firms. And so every day we will have new numbers. We will have to be careful to not confuse noise for signal because there will be a lot of noise for sure. And I’m gonna be very interested to see if the debates had an effect because again, four years ago, you may remember we had a triple tie on top of voting attentions with Tom Mulcair Stephen Harper and Justin Trudeau for the longest time. And when did the numbers really started to move? It was after the French debate on Radio Canada. It was September 24 so the campaign was already six or seven weeks long on. We had the first debate Justin Trudeau was really good, He had low expectations and it was really good. And from that point, the NDP slowly shedded support and the Liberals cruised to a majority. And so I’m I will be looking forward to those before and after numbers of the debates.
Jordan: For people who are going to be sort of casually following the election. At least it’s at least in the early weeks of the campaign, what polls would you suggest that they really keep an eye on because they’re gonna be inundated with them and it’s gonna be really hard to pull out. You know what’s important and what’s not, because I have a feeling there’s about to be a lot of noise.
Philippe: Oh, yeah, Oh, yeah, absolutely. And you may have noticed that how ah liberals have found old videos of Andrew Scheer saying, You know, weird stuff about education and gay marriage and and so they’re just dropping that into the public and it’s it’s fair game. I mean, you know, Andrew Scheer said those things and the liberals are waiting for the campaign to begin to just unleash on the Conservatives on and the Conservatives, I am sure will do the same for the liberals, but what polls should they look at. Well, I don’t want to preach for my own choir. I don’t want to toot my own horn. But you know, if you follow only one or two polling firms, you may not get the whole story. And so that’s why I built my website where I take all the polls and I I you know, I deconstruct them by regions and I rate them and explain to people what noise is and what uncertainty is and margin of error, because so many comments that I have online are Yeah, but Hillary was supposed to win and like well, yes, she what she did win. Polls measure the the popular vote. So we have to remember that. Making a steep projection is is really complicated and has high uncertainty, the totals that I will project on October 20th may be different from the total that are elected on the next day. But you have to look at the big picture. For instance, in Alberta in mid April, my model correctly had identified 82 out of 87 winners, which I think is great, and I was really happy with this result. I missed only five district. That’s what happened. People from those five districts says well, you don’t know what you’re talking about. Both schools are wrong.
Philippe: they were. They were 94% right, And so we have to manage expectations. There is uncertainty in those numbers. There’s noise that’s entirely natural, that the nature of polling, if you really want to have a poll, with no noise. Well, that’s calling an election, and we can’t have them every year, every week. That doesn’t make sense. So I know I tried to do the same way. I tell my students I teach my students about the scientific method I teach my students about what is uncertainty and what is the fluctuations in the data. You will see a lot of noise in the data, and that’s why my website is there.
Jordan: So how frequently I know you’re doing weekly ones right now at McClean’s and also at 338 canada dot com, Will you be continuing weekly through the election or upping the frequency?
Philippe: Well, starting, remember, first started daily.
Jordan: So you’re really not sleeping?
Philippe: I’m really not sleeping. I know I I think a few red bulls in a lot of caffeine in my brain. But no, seriously, I think there would be. They will be so much data available as soon as it comes out, I take a look at it, see if it’s serious. If it’s from, uh, from a renowned firm, if the data is reliable, and so I think they put in my model, and every evening I update the website, so I invite you our listeners to go to 3 38 Canada dot com. To find their districts. Every district has a very high uncertainty, but if you look at the great the greater picture, if you look at your province or even the whole country, well the uncertainty reduces its about maybe plus or minus 4% for the main party
Jordan: we will keep an eye on it. Thank you very much, Philippe.
Philippe: My pleasure Jordan talk to you soon.
Jordan: Philippe J. Fournier is an election modeler and a professor. He does his work at 3 38 canada dot com And at McClean’s dot ca. That was The Big Story. You want more? They’re at The Big Story podcast dot C A. You can also find our brother and sister shows on frequency Podcast network dot com. You may know that this weekend marks the opening of Tiff in Toronto, and if you are feeling movies this time of year, head on over and check out someone else’s movie hosted by Norm Wilner, I promise that it has your favorite films buried somewhere in its back catalogue. Claire Broussard is the lead producer of The Big Story, Ryan Clark and Stephanie Phillips are our associate producers. Annalise Nielsen is our digital editor, and I’m Jordan Heath Rawlings. Thanks for listening. Have a great weekend. We’ll talk Monday.
Back to top of page