Alberta: The 15 Seats To Decide Government

On The Battleground In Alberta

There are two ways to frame a discussion of the polling in Alberta right now – the first is that, if strictly accurate, an average of the current polling from Mainstreet and Leger would show a NDP lead over the UCP in seats, and the second is that the polls right now tell us nothing. The first is true – the NDP right now, strictly taking an average of the polls, would win 46 seats, per my model. There’s a reason I haven’t tweeted this fact out, which is that I don’t think that insight is in any way meaningful.

The second of those claims is also not very true, which is that this tells us nothing. It’s true, in a sense – we have no idea how big of a lead the NDP or UCP will have on polling day, and therefore any seat projection today is merely a snapshot of the current situation – but it does tell us quite a bit that can inform the broader conversation, which is why, after Mainstreet and Leger last week, I decided to reframe the ScrimshawScoreboard, if just for once.

Evan Scrimshaw @EScrimshaw

Alberta #ScrimshawScoreboard, done a bit differently: UCP: 40 NDP: 32 In-Play: 15 Here's the basic truth about the Alberta election: the NDP have 32 seats they *should* win fairly easily. They need 12 more. If the NDP win 12 of 15, they win If not, the UCP's back in Government

/photo/1

An NDP performance under 32 seats next year should be considered a resignation worthy offence for Rachel Notley – they really should be able to hold their 24 seats and win the 8 (lighter) Orange seats highlighted on their side of the ledger. On the other side, those 40 UCP seats really should be safe for them, even in the (realistically) worst case scenarios, and in the middle, are the 15 seats that I think will decide the election.

Now, some in the comments of that post have suggested minor quibbles with that list, and I’ll be the first to admit it is not a hard and fast list – we don’t have a UCP leader yet, and the identity of the leader will inform the map. If it’s Travis Toews, then Calgary will be better for the UCP than the current polls and there will be a resurgent Wildrose Independence. If it’s Smith, Wildrose Independence will collapse but the NDP will do better in Calgary.

But what we do have is a general sense of the map, and it’s an interesting map of seats. It’s 1 Edmonton-area seat, three rural ones, and 11 Calgary seats. The NDP need to win 12 of those seats, and the thing is, they probably need to go 10/11 in Calgary at a minimum. (The advantage of the Alberta electorate being so geographically stratified is that we can pretty much pencil in the results of north of 70 seats without even really trying, and hone in on the truly competitive map easily.)

Sherwood Park should go NDP, so if you’re counting seats for them, it’s 33. I have them as a decent shot in Lesser Slave Lake, but it’s also rural Alberta and the chances that the NDP are being overestimated in the polls there, as they always are, is decent, so let’s keep that in the “in play” section. Leduc and Fort Sask are seats that the model says are competitive but I think the UCP win, especially if Smith is the leader. More importantly, I don’t think the NDP are winning either of them without winning Lesser Slave Lake, so they’re clearly a tier back in terms of non-urban targets.

That leaves the NDP with an optimistic 34 before getting into Calgary, where there are 11 seats that would fall to the NDP if the NDP wins Calgary by 8%, which is what a current polling average says they are right now. (Before you yell at me about believing the polls, wait.) If they win Calgary by an amount in that range, they’ll have a chance to win majority government, but a very slim one – they’ll need a proper Calgary Strategy to pull that off. (Will I keep harping about this fact? Till the day they come up with one or I die, whichever comes first.)

We’ll start the “Problems For The Alberta NDP” section with the fact that Leger has their lead in Calgary at 3%, not 8%, and that sized lead is not enough to flip enough seats to win, and that the reason the NDP have a big enough lead in Calgary is because Mainstreet has even bigger leads there. The Leger lead certainly passes more of a smell test, I’ll just put it that way. The other part of this is that Alberta polling is notoriously bad, almost always substantially understating the support of the right. Hell, even in 2015, the NDP won a lower share of the vote than expected, they just won the size of majority they did because of efficient vote splits.

But even if the polls are right and the NDP gets and holds a sizable Calgary lead, the NDP are probably screwed, for one simple reason – they have no margin for error. This model assumes that there is likely to be an even incumbency benefit, but that won’t be true. Some UCP members will, I’m sure, be utterly useless and underperform what they “should” do, but so will some be better than expected. The chance that the UCP won’t be able to steal one or two seats they “shouldn’t” win based on provincial swing is very low, because somewhere, some incumbent is going to have done a very good job of showing up to every event and helped a lot of constituents with the basic, non-partisan work of being an MLA. And given that the NDP need to basically run the table, it’s really hard to get all the coinflips to go.

It can happen, but the NDP are playing poker and are down to a situation where there are maybe two cards that can come that will beat the UCP’s hand. It’s not impossible, but it is the case that the NDP are heavy underdogs at this point, because of a sheer function of the math.

The NDP’s path to victory is narrow, but it exists – which is better than many would have thought after the 2019 election and worse than a lot of people hoped after the years of bad Kenney polls. Either way, the map is clear at this point – all roads go through Calgary.