Justin Trudeau beware: the numbers are not necessarily your friends
By Paul Adams
Oct 9, 2012
Numbers are such wonderful bright shiny pointed things. If you work at a polling company, as I did for a time, there really is nothing quite so exciting as opening up the newest crop. They have the smell of fresh cut grass.
As your eyes race, tumbling through the data, you are at turns utterly surprised at the unexpected and delighted when your personal hunches are confirmed.
For a few minutes you feel as if you are in the sole possession of this magic thing. You are torn between the impulse to run out and tell your neighbour what you have and the desire to hoard it greedily to yourself.
Paul Adams is a veteran of the CBC, the Globe and Mail,
and EKOS Research. He has taught political science at
the University of Manitoba and journalism at Carleton
where he is an associate professor. His new book
Power Trap, on the dilemma of Canada’s opposition
parties will be published in September.
One of the weirdest parts about bonding with your dataset is how proud you are: your numbers are so tall and beautiful and smart – so filled with potential.
But love, as Shakespeare said, is blind.
And speaking of love, I come to the case of Justin Trudeau. A Canadian Press news story the other day based on a Harris/Decima poll said that 36% of Canadians “would be certain or likely to vote Liberal in the next election if Mr. Trudeau is at the party’s helm.”
That would be a fantastic turn for the Liberals who have mostly been running a poor third in the polls to the Conservatives and the NDP, as they did in the last election.
We know that polls like this early in a leadership race are often not much more than a measure of name recognition. Justin Trudeau has the most storied last name in Canadian politics. His possible opponents, even the former astronaut Marc Garneau, could only expect to become household names by winning the leadership.
Yet Harris/Decima’s chairman, Allan Gregg, one of the most eminent people in the polling business, who has been outspoken in warning about the dangers of reading too much significance into polls certainly did not understate this survey’s meaning to the future of Canadian politics.
“Justin Trudeau — more than any other prospective candidate we tested — holds the best prospect for a revival of the Liberal party,” Gregg told CP. “In fact he is the only candidate we tested that has the potential to broaden the Liberal vote beyond its current base.”
The headline on the CP story on the Globe and Mail website proclaimed that the poll had shown that “Trudeau could recreate [the] Liberals.”
Perhaps Mr. Gregg expressed more caveats than found themselves into the article, but he was even quoted as saying that the survey would, “debunk the myth that the Trudeau name is a liability in the province of Quebec or among francophones.”
That seems like overegging the cake just a little bit. Before we “debunk the myth” of the Trudeau name in Quebec, perhaps we should wait to see whether Justin Trudeau’s leadership rivals are able to re-stoke the fires of antagonism towards the Trudeau of the War Measures Act, the patriation of the constitution without the Quebec government’s consent, and opposition to Meech Lake. And how he handles it if they do.
Perhaps we should wait to see how this all looks when the PQ and the BQ, not to mention Tom Mulcair and Stephen Harper, are done with Justin Trudeau.
So far as I can tell, the Harris/Decima poll did not actually ask respondents to choose between Justin Trudeau and the leaders of the other parties. At least in its previous polling on this subject, the company simply asked people whether they were more or less likely to vote Liberal with Justin Trudeau at the helm, without mentioning the other parties or party leaders. Nothing wrong with such a question, so long as it is cautiously interpreted.
It is worth remembering as well that even the arch-federalist Stéphane Dion delivered the Liberals a large bounce in Quebec immediately after becoming leader, but that did not really “debunk” Dion’s unpopularity in French Quebec, as subsequent events proved.
As the American polling guru, Nate Silver, has pointed out, a candidate for the U.S. Senate with a five percentage point lead in the polls the day before the election has a 95% chance of winning. However, a similar lead a year before an election doesn’t mean much more than the flip of a coin.
If you love the sound of Newt Gingrich’s voice as I do — and evidently he does himself – you’ll love this clip from the Republican presidential nomination race last December when he was ahead in the polls.
“I’m going to be the nominee,” Gingrich told ABC news. “It’s very hard not to look at the recent polls and think that the odds are very high I’m going to be the nominee.”
Gingrich ended up carrying two states: one more than Ron Paul and 40 fewer than Mitt Romney.
I wish Justin Trudeau luck. As many commentators have noted, he is an uncertain quantity. Like so many polls, he is full of potential but a long way from being proven in the hard slog of a leadership race, or an election campaign.
Just one word of advice to him and to the Liberals considering his nomination: Beware of pollsters bearing numbers.