Queen's Gazette | Queen's University

The Magazine Of Queen's University

Search form

The man who's never wrong

The man who's never wrong

[photo of Nik Nanos]
Photo supplied by Nanos Research

As the polls closed on election night this October, Canadians were shocked to see just how much of the country had turned red. After nearly 10 years of Conservative leadership, it was a surprise to see the Liberals surge from third place into a commanding victory.

One man wasn’t surprised though, because it went just as he’d predicted.

For Nik Nanos, Artsci’88, Artsci’89, and EMBA’10, forecasting the Liberal majority with pinpoint accuracy was business as usual. Mr. Nanos has been calling federal and provincial elections for nearly three decades now, and he has yet to get it wrong.

His first public poll was in 1988 while he was still a Queen’s student, and Mr. Nanos projected Peter Milliken would beat out Kingston and the Islands’ Conservative incumbent Flora MacDonald. In the years since that prediction, Nanos Research, which is led by Nik Nanos and his brother John Nanos, Artsci’92, has moved on to serving as the official pollster for CTV, The Globe and Mail, and Bloomberg News.

Though he was right on the mark with his final numbers, Mr. Nanos says the most recent election was a difficult one to call because so much changed during the two-month campaign period.

“From the get-go, the outcome of the election was uncertain. The Conservatives, Liberals and New Democrats were all competitive, and it’s very unusual to have a three-way competitive race,” he says. “It took 66 days to sort out the first decision point, which was who would be the main challenger to Stephen Harper.”

In quick succession, national debate about wearing the niqab at citizenship ceremonies, Tom Mulcair’s lukewarm performance at the first French debate, and the release of the NDP fiscal platform drove support away from the NDP.

“The election started with an unprecedentedly low number of undecided voters — about 10 percent, when it’s usually between 15 and 20 — and an exceptionally high number of Canadians who were willing to vote Liberal or NDP,” Mr. Nanos says.  “As soon as the numbers started breaking towards the Liberals, the ‘anti-Harper forces’ coalesced around Justin Trudeau.“

That anything resembling an “anti-Harper” contingent existed was a particularly bad sign for the Conservatives.

“Anecdotally, my perfectly predictable but non-statistical model of who will win an election is based on the single question of, ‘who is this election about?’” Mr. Nanos says. “Usually, whoever it’s about is the person who loses. This election was a referendum on Stephen Harper and he lost, the previous federal election was about Michael Ignatieff, before that it was Stéphane Dion — even in the Ontario provincial election, it was about Tim Hudak, and he lost.”

For much of the campaign period though, it was far from obvious that Harper and the Conservatives weren’t going to win at least a minority. In fact, it wasn’t until after Thanksgiving weekend, just a week before voting day, that Mr. Nanos felt confident calling a Liberal win.

It’s pretty clear the Liberals won the Battle of Thanksgiving.

“Going into that weekend, it was a two-way race between Liberals and Conservatives, but the Liberals came out of it with a six-point lead. We know that holidays in an election are a key decision point for a lot of people: they’re gathering with friends and family, breaking bread, and discussing what’s going on locally and nationally. It’s pretty clear the Liberals won the Battle of Thanksgiving, and in the last three days that Nanos Research did tracking before the election, they picked up the extra points that moved them into majority territory.”

While the piece of wisdom that predicts a bad outcome for whoever the focal point of an election is proved true, there was another bit of political popular wisdom that proved incorrect.

According to Nanos, it’s generally agreed that to win an election, it’s essential to run a negative campaign. Despite the Conservatives’ many ads suggesting that Trudeau “just isn’t ready”, it wasn’t enough to clinch the win, and instead the Liberals’ positive campaign proved more compelling.

“The Liberals wanted to attract, not only New Democrats, but also progressive-minded Conservatives. They didn’t do anything to repel voters, their platform passed scrutiny, they didn’t demonize their opponents, and they were constantly reaching to make a winning coalition,” Mr. Nanos says. “This is something that’s attracting the interest of politicos around the world. They’re looking at Canada and wondering if the triumph of positivity is indicative of a new global trend, or just a one-off.”

While a political climate defined by its positivity might make for a less cynical nation, it’s too soon tell if it’s the new normal. Mr. Nanos says it will take at least a decade to know whether it’s an enduring change.

To hear him speak, it’s easy to forget that Mr. Nanos is just a predictor of public opinion and not a soothsayer. Despite his perfect record and his strong sense of what Canadians are thinking, he says there are no secrets to his success.

“There are no proprietary algorithms or statistical models that we use — there are no trade secrets here. We use rigorous methods, enforce high standards and are the only research company that was doing random telephone surveys, had live agents asking questions, and validated our results. It’s the most expensive way to do research, but it’s the most robust in terms of consistency and reliability. We still rely on those tried-and-true methods learned while I was an undergraduate at Queen’s.“

Much like he did back in 1988, Mr. Nanos says there’s nothing stopping an upstart from doing polling of their own and getting more accurate numbers than Nanos Research. But with a 30-year track record behind him, Mr. Nanos isn’t too worried about the prospect.

“If they can do it, more power to them,” he says with a laugh.