Models, Polls, And The Liberals On Majority’s Edge

Pre-Budget Model Update!

Models And The Damage Done

Tomorrow is Budget Day, and since the state of the country’s politics is the story of the week and not the state of the country or its economy, we might as well be honest with what everybody is talking about. Because of it, I decided to update the model, and it’s a big one - the Liberals are on 172 seats, otherwise known as Majority Government.

Now, being at 172 in this projection is not actually important in a real way - it’s still obviously within any margin of error of minority government, in the same way previous forecasts just short of the majority line don’t rule out a majority. It’s incredibly close between majority and minority right now, and there’s nothing much between it. Just like April, still.

The situation in Quebec is unchanged - the House would be returned in the exact same numbers there, short of any departures for Bloc MPs who would rather run provincially next year. Ontario is essentially a wash - the model sees 3 Brampton seats flipping off that Liaison poll of the city last week, and Vince Gasparro is in real trouble in Eg-Law - but the Liberals get a couple seats back each in the Atlantic, Alberta, and BC. Nothing flashy, and absolutely nothing that I would take to the bank. The Liberals’ hold on majority government is by 0.2% in the tipping point seat (Edmonton-Riverbend, if anybody’s curious), which is why this is very blasé.

If you took out the EKOS poll which had the LPC up 16% and did nothing else, I’m sure the Liberals would fall back under the threshold. They wouldn’t come anywhere close to a CPC plurality, but yes, erasing the most Liberal friendly results from an average will move the average right. But if you take the most LPC friendly poll out and the most CPC friendly one out region by region, nothing changes. What I call the “Olympic Average” - how diving and figure skating is graded, with the best and worst scores taken out - doesn’t change shit here because Angus Reid was pretty far out there in a pro-Tory direction.

Right now, the fundamentals of the race are pretty simple - the Liberals lead in Ontario pretty much according to everyone. 9 in Leger and Liaison, 11 in Nick Kouvalis’ work, 7 in Pallas, 4 in a certain paywall pollster I don’t want to get on the wrong side of again, and 2 in Angus Reid. Abacus’ tie is the worst of the bunch, and that’s a sign that the Liberals won’t lose this election if the polls don’t move significantly.

The reason some other models are seeing a wildly different picture (though, interestingly, not 338, which is basically in line with me) is simple - there’s a clear risk that the NDP could be overstated, depending on the poll, by the nature of modelling. Because the NDP got 6% at the last election, a 3% improvement in their vote will turn into a 1.5x multiplier in proportional models, and artificially boost NDP candidates. Plainly, any model that has the NDP winning back seats they lost by 10-20% is suffering from an understandable but flawed statistical quirk, not the reality of the NDP’s position.

But if the NDP are taking seats - or even just splitting the vote - then yes, a Conservative government looks plausible. But being plausible if a specific set of circumstances that aren’t happening happen is the best I can say about the CPC, then they wouldn’t win an election held right now.

The other interesting factor is incumbency - the Liberals have new incumbents in many of their tightest races, which is going to be a boost in key seats especially in BC and Quebec. Part of the reason the Bloc never made that big of an inroad back is because the Liberals ousted incumbent Bloc MPs, and the loss of the Bloc personal vote and the gain of a Liberal one makes those seats harder for the BQ to gain back in the model. Similarly in seats like New Westminster-Burnaby, Burnaby Central, Victoria, Eg-Law, Carleton, and others, those new LPC MPs will benefit, which might save otherwise contactable seats. New incumbency will also help the Tories throughout Ontario and their (mostly NDP) facing gains in BC, which is another reason why the model isn’t moving much.

None of this is a guarantee of minimal movement going forward - if the polls radically change, so will the math. But right now the honest truth of the polling is that the Liberals came a hundred ish votes short of majority government in April and right now they’re on course for basically the same result. And given that, they’re going to be on course for the line between majority and minority government.

It’s also worth pointing out there’s still basically no chance the NDP are going to let this happen - they’ve already told the Globe that abstention is on the table, the start of a climb down and an easy way to square previous statements that it will hard to support an austerity budget while not causing an election. It’s been the obvious solution this entire time, which is why I said it’d be what happens two weeks ago.

The reason I was and am so adamant that there won’t be an election right now is because the only argument for one happening is either the NDP are idiots who will do it even though it’s a bad idea or almost comically false levels of cope to try and claim it’s a good idea for the NDP. It’s an interesting dynamic where everybody implicitly or explicitly thinks that the Liberals would benefit from an election, but yet the speculation is about how the opposition will find a way to do a thing that’s obviously not in their interests. I don’t think much of Don Davies, but it seems pretty obvious to me he’s not actually going to let the party do this damn silly thing.

The dynamic at the core of Canadian politics right now is we are a united country on what needs to get done - big majorities support national projects, there’s closer to a consensus on pipelines, the Ring of Fire, etc - and a big divide between who is better to achieve that. Carney is being given time to sink or swim, and people won’t judge him yet. If he isn’t given that time by the Opposition, he’ll likely win a majority to get the time to prove himself able to get shit done. If he doesn’t in a year, then we’ll likely be in a late Fall 2026/early Spring 2027 election all about how Carney’s all bark and no bite. But any election before Carney is given time to get shit done will see the Opposition go backwards, and they clearly know it since their reaction to Steve MacKinnon’s posturing is to call him a meanie and not to say “I look forward to an election at the earliest opportunity”. If Michelle Rempel thought an election was a good thing for the CPC she wouldn’t be preemptively (and at considerable length) blame shifting about how the budget is a poison pill. The thing about confidence is it’s easy, and generally concise.

The Liberals are going to be in government in 4 months. Whether it’s in majority or minority is unclear, but if the Opposition is stupid enough to do this damn silly thing, they’ll lose. The Liberals are already at the majority line, and an election gets them there easily. The polls aren’t moving against Carney, and the Conservatives aren’t about to waltz into office.

(With Budget Week here, consider a paid subscription to ScrimshawUnscripted. All my work will remain available for free, but with the political fallout of the budget and an NDP Leadership race coming - plus ramping up to Quebec, Alberta, and likely BC elections on the horizon and the shitshow that is the fight against Ford - paid subscriptions are a greatly appreciated way of saying thanks.)