Mainstreet: At What Point Do We Have To Take This Seriously?

On Mainstreet’s Federal Ontario Sample

It is probably sensible to start this column with things that are indisputably true, before we get into conjecture and pseudo-intellectualism posturing as opinion. Mainstreet is undeniably a good and important part of the Canadian polling fabric. They are a firm that has filled an important role in the public discourse, enabling better coverage of the state of play of cities and smaller provinces where reliable, public polling is either nonexistent otherwise or extremely sporadic. Were Mainstreet to cease to exist, or to become more like Ipsos and only poll Federally and the occasion big province, we'd be much worse off for it.

What is also impossible to deny is that what Mainstreet and Frank Graves are seeing - huge swings back to a genuinely competitive election - are almost too good to be true, as a Liberal. Mainstreet’s latest Ontario-only Federal vote intention question has the Liberals winning Ontario by 3%, which is essentially 2021 redux. Frank Graves is making the (entirely reasonable) point that a lot of people who said that his polling could not possibly be true should probably be at least open to the idea of it, especially given Mainstreet’s Ontario sample was 1200 people (done during their in-writ daily Ontario polling) and fairly robust.

It’s also worth noting that Mainstreet’s Ontario polling has seen the PCPO tick up since the last time they released Federal Ontario vote intention data and the Liberal lead is up a point. That’s decent evidence that it’s not Mainstreet’s federal data being in some way contaminated or influenced by the provincial question, because the margins went the other way. That's by no means a guarantee of anything but it’s worth keeping in mind.

But I think the most important thing about this Mainstreet data, and the EKOS data that came before it, is that I desperately want it to be true. I want more than anything on earth for Poilievre to be botching this. I think Poilievre would make a very bad Prime Minister if this country was foolish and ill-headed enough to make him PM. I can’t then suddenly deny that belief when it comes to figuring out what the fuck to do with these contradictory polls. I want to believe the end of Poilievre is nigh. The case for these polls being reasonably close to accurate is good. And yet I can’t shake the feeling that the only reason I’m even indulging this madness is because it’s the outcome I want.

The case for this being real movement and at something resembling this scale is actually remarkably easy to make. In 2021, when Mainstreet and EKOS showed Conservative strength, it didn’t last this long. The great polling weirdness wasn’t a nearly month-long march in one direction, and it wasn’t a situation that had this easy of a hook to the news.

When I had David Coletto on the podcast, we talked about whether Trudeau resigning is a plausible explanation both for the direction of the movement and for the level of movement seen, and we basically came to the same place - that some movement was plausible, but not 20%+ of it. But whatever the “true” amount of movement, it seems likely that there has been clear movement. And it’s been likely that there was movement because the fundamentals were pointing to there being a singular factor exacerbating the crisis.

I praised Justin Trudeau last night for the tariff response and think he genuinely did a very good job in tough circumstances, but his leadership was an anchor for the LPC. I don't want to make this a victory lap but let’s be frank - everything I wrote about the PM needing to resign has been validated by these polls, and every time Mainstreet or EKOS or even the panel pollsters showing meaningful swings show these kind of results I feel an entirely vain surge in my chest because I was right. That so many of the people now breathlessly sharing Franky and Quito’s work despite calling me an Indian or Russian backed hack for saying that getting rid of Trudeau could produce a bounce back hasn’t been ignored.

The actual answer is that many are setting themselves up for disappointment, insofar as Poilievre will likely win a comfortable majority government. I think those pretending that things are quite rosy enough to see a Hung Parliament or some form of Liberal Government are engaged in a level of cherry-picking as to qualify for an Olympic level event. That so many people are willing to ignore the good pollsters who see a serious but by no means this big swing because they like the Liberals is concerning. But I get it.

The other part of this that I cannot ignore is the chance that my biases mean that I will always be late to believing this is real. The honest truth is that it is easier for me, as a Liberal and a progressive, to predict good things for the right, because any time I say anything nice about the chances for a left wing government or party I will inevitably be yelled at for biased hackery along the lines of my preferred party. That criticism used to be accurate about me, and it’s something I zealously try to guard against.

It would be substantially easier if everybody agreed the national lead was down to roughly 12-15 points, which is roughly what you get if you do a straight average and extrapolate a national number from Mainstreet being L+3 in Ontario. (Fournier has the lead at 19% nationally, but he’s not including the Angus Reid Carney As Leader numbers and not extrapolating from the Mainstreet Ontario-only numbers. If I were him I’d probably make the same choices but both drag down the average to some extent.) But since when do we care about what’s easier, eh?

If this is just rambling incoherent nonsense I’m sorry, but it’s also a rare case where the honest answer is I have no idea. I want desperately for Quito and Frank to be right, but I also worry that any emotional capacity invested in those outcomes is just hope masquerading as analysis. I have no idea. But what is clear is two pollsters are seeing the Liberal Party not merely in recovery but in legitimate contention for another term. Whatever my doubts - and they absolutely exist - that isn’t anything to be scoffed at. And pretending these polls are meaningless garbage is as much nonsense as pretending they’re gospel truth.