Angus Reid’s tweet this weekend has kicked off a firestorm, both because of its incendiary, anti-Trudeau nature and the fact that both in specific and in general his claims about over a hundred churches being burned down as some retaliation for residential schooling are untrue. It was at best a wildly ill-informed, at worst intentionally misleading tweet. It’s abhorrent, as is his crap half-apology.
But this isn’t the point of this column, because it’s a point that shouldn’t require an entire paragraph, let alone a column. What is interesting is the notion that some are clinging to that this is a stain on the Canadian polling industry writ large, or that his polling is somehow biased against the Liberals accordingly. I tweeted this week that Angus Reid’s polling is friendlier to the ONDP than the Ontario Liberals, which is true. They had the NDP winning with 36% of the vote in January 2022 when most others had them in a dogfight for second in the mid 20s. I don’t regard their work as top tier.
But the idea that pollsters’ personal beliefs meaningfully affect their results, or are somehow why the Liberals are in trouble, is nonsensical. It’s worth being very specific here - there is no meaningful difference in opinion polling based on pollster methods, partisan/ideological belief of the public-facing head of the company, or anything else. The Liberals are fucked, and it’s not because the pollsters are lying to us.
..
The easier way to make clear that pollsters aren’t lying to us is to point out that plenty of the same pollsters having the Liberals federally in the shits are the same pollsters giving the provincial left good results. Mainstreet, who just had the Liberals in third nationally, had Bonnie Crombie at 30% last time they were in the field, a recent high, and they have Susan Holt on track for government in their latest New Brunswick poll. Nanos’ early August New Brunswick offering was essentially in line with the Narrative Research polls from May.
Abacus’ 2023 Alberta polling was basically bang on the money in the end, but if anything his post-scandal poll was too good to be true for the NDP. Not exactly what you’d release if you were conspiring to elect Conservatives. (If you come at me with “#Actually yes you would because that scared Conservatives”, you need to keep that energy for Mainstreet’s Calgary polls in 2017, which is endlessly and falsely used as evidence of Mainstreet’s biases.)
Even getting into an ideological fight is nonsensical - Angus Reid’s biggest, most obvious house effect is pro-Ontario NDP, and their polls were consistently worse for the BC Conservatives as they rose from also-ran to plausible government. If he was putting his thumb on the scale I doubt he’d be doing so for Marit Stiles, Andrea Horwath, and Kevin Falcon. Quito Maggi is a member of the Ontario Liberal Party. Frank Graves, for all his bravado and his tweets, has the CPC lead in his last release at 15.6%, and that’s because EKOS has a slightly lower NDP reading than some others. There’s no conspiracy here.
If you want to split the polls by polling method, the average lead from the online pollsters (most recent poll, September field dates) is 20.6%. Of the three telephone or IVR pollsters, it’s 20%. If you slice it by pollsters who follow me on Twitter it’s 20.5%, versus 19.5% for firms whose proprietors have me blocked on Twitter. (No, I don’t think this is a meaningful distinction, but I’m making a point.) There is nothing here.
Mainstreet, for all the shit it gets, stuck its neck out with two polls showing the Bloc winning LaSalle when everyone, myself included, said it was Craig Sauve and the NDP’s to lose. Their poll in Elmwood was essentially bang on the money - if anything, they didn’t have the Liberal vote collapsing far enough. There’s just nothing here to all the conspiracies and all the just asking questions that are thrown at the Canadian polling industry.
What we have is a government that is very unpopular due to a combination of two things - bad material circumstances, and the direct link between those bad material circumstances and their actions. The effects of the government’s decision to increase immigration levels so dramatically has deflated wage gains compared to what American counterparts have seen in the bottom quartile of the income distribution, contributed to youth unemployment above 10%, and raised housing costs. These are material circumstances that outweigh whatever gains the government has given non-parents of young children on the carbon tax or by removing interest on student loans or other benefits.
If you are in this government’s zone of interest, yes, this government has been worth it. But it’s absolutely the case that if you are childless and don’t own, this government has been a disaster for you. It’s not a shock that the young are the group swinging against the government the most. This is why the government is losing badly.
No amount of bitching about Doug Ford’s rent control will change the truth that if you are 22 and want to move out of your parents’ house you will pay significantly more for your first place (or first post-Uni space) than you would have in 2015. Should we vote out every NIMBY councilor in the country too, and defeat Doug Ford? Obviously. But denying that federal decisions have raised the cost of housing is as much disinformation as what Angus Reid posted.
It is comforting to blame what the Liberals are going through to the fault of algorithms and misinformation and lies, but it’s not. Does Poilievre lie? Yes, but so does every politician. It’s not “both-sides” to point out that Justin Trudeau has routinely played fast and loose with the truth, going back to his years long claim that Canadian workers hadn’t gotten a raise in 30 years that was A) false at literally any other timescale when he started saying it and B) not even true by the 2015 election, when he kept saying it. It’s not bad faith hackery to point out that the people who defend Mr. “The story in the Globe is false” and “2015 will be the last election under First Past The Post” are not exactly committed allies to the cause of truth and accuracy. It’s fucking reality.
You know when else the Conservatives lied about the carbon tax? In 2019. You know who owned the National Post and the Citizen and the Gazette and the Province then? Postmedia. And yet, then the country was on board with carbon pricing and the broad Liberal agenda. Now it’s not. It’s not all propaganda and disinformation. It’s economic realities.
The polls aren’t wrong. There isn’t a bias. The reason to condemn Angus Reid’s tweet isn’t that it is proof of some grand conspiracy or because it shows a biased industry, the reason to condemn it is it’s contemptible. There’s no conspiracy, there’s no tangible pattern, there’s just a government about to destroy its legacy and kill all its successes by refusing to change course. That's the story we should be paying attention to, and trying to pressure the government to fix. If you find it more satisfying to dunk on the contemptible than to try and fix the structural problems the Liberals face, then you’re more committed to a good time than to fighting for progressive values. And that’s fine! Just don’t pretend you’re doing something you’re not.
I disagree about the bias in polling but not towards or away from one political party or another. I think the polls are more of a measure of where respondents get their information than how they would vote if there was an election today. People like to consider themselves part of a community and when their media of choice (mm, sm) keep telling them that someone one is hated, that's how they respond when asked. Unless and until the results are analysed statistically for media bias polling will continue to "get it wrong". Polls are a measure of media effectiveness and spread rather than of voter intention or voter like/dislike.
I was wondering if you would write about this one. On Twitter you made the comparison with Frank Graves. I think this comparison is quite wrong. Frank Graves has been clear about his personal preferences, but I have never seen him spread lies or disinformation to further his personal beliefs.
Angus Reid is less clear about his personal preferences, but as it turns out he quite willing to knowingly spread lies to damage Trudeau. And when he is called out, he offers a non-apology and basically doubles down on his lies. I think there is a significant difference and the remaining credibilities of the two pollsters are not even close.
Now, does that mean that there is a country wide conspiracy to skew the polls? No, but at the same time there is no reason to take anything that comes from Angus Reid seriously as he demonstrated that he is willing to lie knowingly to achieve his personal preferences.