Looking For Fucking Anything To Happen

This Campaign Is Slowly Driving Me Crazy

So, all of that panic was for nothing.

We’ve gotten another round of polls today - the trackers, Abacus, Leger, and Pallas - and nothing happened. That’s actually a lie - Mirimachi flipped back to the Liberals. Oh my. Oh heavens. Stop the fucking presses.

Remember the certainty that Quebec had fallen away from us after one (1) Angus Reid poll yesterday? Well Abacus joined the panic, but Pallas and Leger have the LPC-BQ gap at 15 and 16 respectively - so much for the certainty of a big surge in Quebec for the BQ, I guess. Everywhere else? The Ontario polling average is up to 9.5% again, BC seems to be stabilizing if not even trending back towards us, and the Prairies are bleeding votes that frankly we were never going to hold anyways so who gives a shit.

Now, I honestly could just run yesterday’s column again in this slot, but let’s be useful here and talk through what risks or not the government faces in its polling through Sunday. Mainstreet and EKOS both currently have weird, and it’s fair to say anomalous, weekend samples still in their sample that should flow out the next time they release, which would suggest they’ll break left. Abacus has a Quebec sample that’s too pro-Bloc, at least according to Leger, so replacing a L+3 with a L+7 would still help, if that’s what comes next.

Angus Reid’s probably too favourable in Ontario but too unfriendly in Quebec, and swapping those to shared L+8s would help the seat count. Pollara dropped some regional data on Curse Of Politics, and while it dropped the Ontario lead it’s now a less crazy lead and therefore less points to defend. And everything else was basically status quo of the previous poll anyways.

If there was a sustained amount of movement we’d know about it by now. If there was truly some latent momentum we’d have seen it by now. If there was something truly lurking around the corner by now we’d have seen it. There’s not.

On the question of whether there is some novel theory about how secretly the polls might be missing something, there’s not one. That's not a guarantee the polls will be perfect, though Canadian polls very often are, but if there is a broad based miss, it won’t be because Coletto and Graves and Leger and Maggi and and and didn’t think of something that PoilievreAssLicker1488 on Twitter came up with. There’s a huge financial risk to getting political polling wrong for these firms, because political polling is often an audition.

It’s rare in market research you get a definitive answer yes or no to your results, and therefore it’s hard to actually know who’s the best or not. One of the few things that has a verifiable record is political polling, which is why a lot of people who don’t have media deals do political polling occasionally. They want to audition and show off that they’re good at this. And if they tank? Their corporate clients, the ones that actually pay the bills, are gonna be pissed, and probably at least try to bring their business to whoever did get this right.

The idea that any of these people would let personal ideology get in the way of accuracy is for the birds. They are too focused on brand building to let petty nuisances like ideology matter, and this has always been true. The financial incentives of a bad poll, especially in a campaign as saturated by ones as this campaign has been, will be ruinous. If someone really takes a big risk by departing deeply from the status quo and it fails, they’re potentially fucked, especially without a high profile electoral contest to prove it was a one-off any time soon.

Now, the chances of a polling miss are non-zero, but there’s a difference between a polling miss and a polling miss that is big enough to meaningfully cost the Liberals majority government, let alone the election itself. More to the point, a good model should be able to properly identify the places where the polls are iffy, especially (in this context) right trending seats where straight swing would be too nice to the Liberals. I don’t know where the final numbers will end up but I do know there's going to be a handful of seats where it’s possible I tip over the right out of an abundance of caution simply because the trends terrify me. I’m not going to go into election night tipping every tied race Liberal for fun because my rep is on the line.

Most of the seats I’d have been worried about going into Election Day showing as red already flipped! There’s a couple of Northern Ontarios and I’m shitting bricks about a couple of Newfoundland seats but let’s be clear - the low hanging fruit is gone. The Liberals have a very stable majority government base under them unless something changes. I don’t see what possibly could, especially considering the clusterfuck of a Platform will cost the CPC time and energy.

1. So this week is going to suck. There’s little to talk about, my patience is already at zero, and there’s no momentum or tightening to speak of that isn’t just minor variation around a broadly similar story. The Liberals should win a comfortable majority government.