On polls and seat projections

It didn’t take long after QMI published a poll they commissioned through Abacus Data for someone to take the numbers and plug them into a seat projection model.

First, some notes on the poll. I’ve seen some non-Conservatives question the validity of the poll, citing that it was commissioned by QMI, which operates the Sun chain of newspapers. I’ll leave the discussion of media bias in select outlets for others to take up, but critics should note that Abacus, while new, is a reputable firm, and their CEO did his PhD in Alberta, so he knows a thing or two about Alberta politics.

I’ve never been a big fan of seat projection models because of the simple reality that, for various reasons, popular vote percentages do not easily translate into seat percentages in single-member plurality voting systems.  Moreover, while there are significant regional variations in regional support patterns, regional sub-samples are so small that they have high margin of error, and are less reliable than the province-wide sample as a whole. I’m not saying that sub-samples are not useful, but that seat is projection is difficult enough already, without the added complication of smaller regional sub-samples. Finally, in a province that has historically been either a “two party plus” or “one party plus” system, local constituency effects (star candidates, concentrated campaigning, etc.) matter even more to Alberta opposition MLAs, who almost always have to run ahead of their parties.

However, this election is also different in that it will be the most competitive Alberta election since 1993, and it is the first multiparty election Alberta will have experienced in decades. This election is also being contested with new electoral district boundaries, which means several ridings have changed dramatically. In 2008, there were several ridings which were won by margins in the low hundreds and a few with margins of less than one hundred, so stiff multiparty competition and widespread vote-splitting will make this election even more competitive, and competitive elections with low margins between winners and losers make seat projections even more difficult.

As for what the poll says, my take is that this isn’t anything new, and that these numbers look similar to numbers from February 2011. To me, these numbers show that Redford’s honeymoon period with the electorate has ended; the PCs have fallen from their post-leadership high of about 50%, and Redford’s leadership rating is a net negative. Given the centrist policy stances of Alison Redford, it is possible that at least part of the drop in PC support is attributable to some Liberals (and possibly New Democrats) who supported Redford returning back to their party. The big winner in this round of polling is the Wildrose Alliance, who have succeefully adapted to a post-Stelmach Alberta and, who are within striking distance again. They also have the only leader with a net positive rating, which is especially noteworthy, given that Smith is running slightly ahead of her party (her ‘favorable’ rating is 31%; the Wildrose is at 29%), whereas Redford is running behind her party (Redford has a ‘favorable’ rating of 29%; the PCs are at 34%). (Yes, I know these aren’t ‘approval’ numbers, per se, but the comparison is meant to be illustrative, rather than scientific.)

All of this is subject to change, as a lot can happen over the course of a campaign. The Wildrose may be within striking distance but only one of their MLAs was elected under a Wildrose banner, and, while many of their campaign team are veterans from either the federal Conservative Party or the Alberta PCs of yore, they are up against the best political machine in Canada. Similarly, the Liberals may have regained some of the support they lost to the Tories, but they will still have to fight for a share of earned media in what has largely been billed as a two-horse race. The New Democrats have the benefit of already being mobilized for their federal leadership race and of being able to focus on a handful of targeted ridings. As indicated by their initial attack ad campaign, the PCs will continue with their ‘devil you know over the devil you don’t’ narrative. I still think Redford has deliberately ceded the right wing in order to court soft-progressive support and ‘unaligned moderates’, which pretty much describes the political leanings of most Albertans.

I do not believe that Redford wants to win a majority anywhere near the size of the last PC government caucus. From a purely political standpoint, a ridiculously large caucus (anything above 70 seats) actually hurts the governing party in the long-run because government MLAs, lacking a strong opposition to fight off, instead become preoccupied with fighting each other. A 60-seat government caucus with a 20-seat Wildrose Official Opposition and eight seats split between the Liberals and the NDP would allow Redford to continue with her centrist policy agenda. She said herself that she’s pragmatic, and this House configuration would make it easy for her to shift right or left as the circumstances dictate.

Regardless of what happens, this will be the most interesting election Alberta has seen in a long time.


About jbsantos
Polling, politics, PR and outdoor pursuits.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: