Ontario Reset: Ford In Majority

Meditation On Polls, Models, And Movement

There are two parts to people’s memories of a campaign – how did the polls do during the whole campaign, and how’d they do at the end? Last-poll accuracy is a skill, and a very important one, but it’s hard for me to argue that a pollster who was off in la la land the whole campaign who then gets close to the pin at the end had a good cycle (see Selzer, Ann in the US, whose entire 2020 cycle was showing great Democratic strength and then her final poll shifted right hard at the end – and was right).

You could say the same thing about forecasters, in that I don’t think PJ Fournier had a great 2021 despite the fact that his final forecast was basically just as fine as everyone else’s. His column calling O’Toole the favourite midway through the campaign has been widely mocked in these pages, but he at least got to the right answer by the end. Whether the final outcome or the process matters more is a matter of taste.

Why am I being esoteric on the nature of forecasts right now? Well, anyone who follows me on Twitter will know – I no longer have Ford outside majority government, and right now there’s a 4 seat gap in mine and 2CloseToCall’s forecast – hardly the kind of huge gap there has been at times. Right now, I think Doug Ford is in majority, and while the path to that no longer being true is very obvious, it being obvious doesn’t mean it will happen.

There’s two questions to handle – the question of the modelling, and of the punditry. I’ll handle the modelling first – the Liberal vote share has come down more in the last few days than the PC vote has, and because of the way my model works, when the Liberal vote falls province wide, their vote falls more in the suburban target seats in the GTA than it does in the rurals, because when their vote goes up, it goes up more in the suburban targets and less in the rurals. It’s a function of math, and when I finally got around to updating the model again Monday, the Liberals had fallen down to the wrong side of a lot of seats in Scarborough, Brampton, and Mississauga.

Do I think that that basic insight – that the Liberal vote will be more efficient in the suburbs than Fournier or Bryan think, for a given vote share – is still correct? Sure, and the results in Australia on Saturday give me even more ammo for that belief. Doug Ford managing to avoid bigger losses in the suburbs than he gets across the board would be the biggest surprise possible on June 2nd, win lose or draw. It’s just a fanciful notion that DoFo will be the one to overcome the Global Fucking Realignment and limit his suburban damage compared to his province wide vote.

None of that means he is a lock to lose – but if you tell me that Fournier has the PCs at 81 seats, as he does as of the time of this writing, then I’m going to tell you he has lost his fucking mind. Fundamentally, for the same final vote shares, I’ll probably have the Liberals a handful of seats higher than Bryan and a handful more higher than Fournier because I think the Liberals vote is more efficient than they do. If I am wrong, c’est la vie. In terms of process, I’d much rather go down swinging with this.

In terms of the punditry, yeah, I deserve to take some shit – I thought the left would consolidate by now, and they haven’t. Maybe there will be a last week sprint to one or the other – I think that would have to be the Liberals, given Horwath’s spent the whole campaign bouncing around in the low to mid 20s and seemingly has no exit velocity for upward trajectory – but it hasn’t happened yet, and there’s some saying about the definition of insanity that sounds compelling here.

If you want to think I’m a nonce for thinking Del Duca would be Premier, go ahead I guess, but the theory Del Duca would be Premier was based on a series of three parts – that the PC vote would decline (it has), that the Liberal vote will be more efficient than consensus (discussed above, but to be determined), and that the left would consolidate (nope). That’s the big error in my punditry – I was banking on an amount of consolidation on the left and the Ontario Liberals being able to slingshot themselves up on the back of the NDP, in the same way that Justin Trudeau did to Tom Mulcair in 2015. You want to mock me for the punditry? This is my error, and I own it.

I have made a killing – my best take, across multiple elections and multiple countries, is assuming that minor parties will underperform their polls. Yes, it occasionally bites me – we’ll see how much it did so in Australia this week, although it seems like much of the Green surge in terms of winning seats is being undone on late counts – but trusting the NDP to shit the bed, and the Lib Dems in the UK, has been so consistent and profitable as to make me just blind to the idea that the NDP wouldn’t fade here down the stretch. I really didn’t see a situation where Ford was down to 36% and the opposition was so uselessly divided it couldn’t run him down.

Where are we here? If the left doesn’t consolidate, Ford wins. If Ford raises his vote share much from here, then the amount of left consolidation needed increases. I think a PC lead of 7% is where the majority government line is, Bryan thinks it’s 5%. This is a valid model disagreement, and one that I find mostly uninteresting until we see the results.

If you want to be a Liberal optimist, a decent amount of the left consolidation in 2015 happened after Thanksgiving weekend, where family got together, sorted out that it was the Liberals, and then moved to make that happen. There was a holiday weekend just now (not that much of Ottawa got to enjoy the holiday, the guy without power groaned. Seriously, it’s been 72 hours without it now, I’m done being patient.)

The problem with that is waiting for the left consolidation is like, well, me waiting for my power to come back. It’s a great theory, but right now, there’s no evidence it’s going to come. If you had told me my polling average would have Ford at 36.3%, I’d be telling you he’s cooked. And now he isn’t, because he’s up 9%. Unless that fundamental truth changes, it’s game over, Ford’s getting a majority, and I’m gonna have to pray that most people care more about final outcomes than the road to getting there.