Alberta NDP Leader Rachel Notley on the campaign trail.
Alberta NDP Leader Rachel Notley on the campaign trail. Credit: David J. Climenhaga Credit: David J. Climenhaga

A new poll by Abacus Data released Saturday suggests the Alberta NDP, led by Rachel Notley, is breaking ahead of the United Conservative Party, led by Danielle Smith, and heading into winning territory. 

“Among all eligible voters, the NDP has gained seven points from our last survey before the election started,” said Abacus Chairman and CEO David Coletto in his analysis of the polling data culled by the firm from questions asked of 885 eligible voters between Tuesday and Friday. 

“This is the first time in our tracking that the NDP support is above 40 per cent and is the largest lead we’ve measured for either party in the five waves of research we completed,” Coletto said. 

“Among eligible voters likely to vote, the NDP lead is slightly smaller (five points) with the NDP at 46 per cent, the UCP at 41 per cent, and undecided at 9 per cent,” he said. “Among decided likely voters, the ballot is closer with the NDP ahead by six, 51 per cent to 45 per cent.”

Importantly, the polling shows the NDP leading in Calgary – where the consensus among pundits holds the election will be won or lost. 

“The NDP is slightly ahead in Calgary (42 per cent to 36 per cent),” Coletto said. When the undecideds are removed, he added, that becomes 49 per cent for the NDP and 41 per cent for the UCP, with the Alberta Party at five per cent. 

This new data shows the NDP in a winning position, with undecided voters finally moving – and breaking toward the NDP. 

Still, the Abacus results do run counter to the findings of another recent poll by Mainstreet Research, which said Saturday it was in the field on the same dates talking to 1,338 adults. 

Mainstreet’s poll saw the UCP ahead by six points, at 45 per cent compared with 39 per cent for the NDP.

Regular readers will know that I’m not a fan of Mainstreet, whose polls are sometimes accurate, sometimes not so accurate, and at least once, in President and CEO Quito Maggi’s own words, “a catastrophic failure.”

That failure was the 2017 Mainstreet prediction that former Calgary mayor Naheed Nenshi would lose to his Conservative challenger by a wide margin. Nenshi, as we all know now, won. 

I’m sure Mainstreet has fine-tuned its methodology since then, so we should view all opinion surveys, including Mainstreet’s, thoughtfully and with an appropriate level of caution.  

In a Substack post, the always entertaining Evan Scrimshaw, who seems incapable of pulling his punches, expressed his doubts about the surge of UCP support in Edmonton reported by Mainstreet and called the Calgary results from Abacus “genuinely frightening for the UCP.”

While you never really know for sure with polls – they are, after all, just a snapshot in time – Abacus’s results do have a believable feel.

Among the reasons I say that: What many of us are observing of the sign wars in our own communities; the apparent impact of serial revelations of the outrageous things said by Smith over the past few months that are suddenly contradicted by her when they’re reported; the growing public awareness of the troubling role played in Smith’s UCP by the extremist Take Back Alberta faction; the increasingly panicky rhetoric from UCP candidates; and the ridiculous hysterical and defamatory social media attacks on the quality of Abacus polling by UCP trolls.

Conservative parties, though, even ones as badly led as Smith’s, enjoy an enormous structural advantage on the Alberta electoral map. So it’s prudent for all observers to continue to treat this election as too close to call.

My guess is there will very soon be some more polls from other respectable pollsters and we’ll have a stronger sense of which way voters are heading. 

So brace yourselves, if voter sentiment appears to be clearly breaking toward the NDP, the reaction from the UCP will not be pretty. 

The election is scheduled to take place on May 29. 

A note on margins of error

It is metronomically argued by commentators who pay too much attention to political polling that the margin of error is critically important and must always be reported. Frankly, I have my doubts about this, and think it’s prudent to be skeptical of all polls, even ones by reputable pollsters with good track records and believable margins of error. 

Since few understand the intricacies of polling and the many ways it can be manipulated, I increasingly suspect reporting margins of error just muddies the water further for most readers. They are usually reported as plus or minus a small percentage, 19 times out of 20, or something like that. If you’re not a pollster, what does that really mean when there are so many other factors to consider, including the hysterical lies of nuts trying to create a stop-the-steal narrative?  

For what it’s worth, the Abacus poll reports a margin of error of plus or minus 3.4 per cent; Mainstreet reports one of plus or minus 2.7 per cent. 

David J. Climenhaga

David J. Climenhaga

David Climenhaga is a journalist and trade union communicator who has worked in senior writing and editing positions with the Globe and Mail and the Calgary Herald. He left journalism after the strike...