Opinion polls can't be counted on to reliably predict election results, that is for certain, but that doesn't mean they should be thrown out entirely.
Defenders of political polls like to talk about them as snapshots of the electorate. When they're right, the polling firms puff out their chests and when they're wrong, they blame a rapidly changing political landscape. Instead of looking at polls as snapshots, sharply in focus with those percentages broken down to the tenths of a point, it's better to look at them more broadly, more like tomorrow's weather forecast.
The Nanos survey, commissioned for the Globe and Mail newspaper and released Tuesday, should be seen that way, as well.
The precise numbers and the colourful bar graphs and pie charts are pretty to look at but are of limited value. The most important information in the polls, especially this far out from an October election, are the trends.
That can be hard to take in a society that values quantitative information far more than qualitative, even when the numbers aren't robust and don't hold up to tests for validity and reliability. With that in mind, forget the Nanos data and read no further than the executive summary on page 2, specifically the first two sentences:
"Canadians are more likely to think that a majority government of any kind, regardless if it is a majority Conservative government or a majority NDP government, will have a more positive effect on the Canadian economy than a minority government would when asked about outcomes in the next federal election. However, Canadians are more likely to think that an NDP government will have a more positive impact on the economy (47%), versus a Conservative government (32%)."
Ignore the numbers and just read that second sentence as a more-or-less statement. More likely, more positive, that's the language that pollsters should speak, not specific numbers and outcomes.
Taken at face value, those sentences point to two fascinating trends that could shape the outcome of the election.
First, the efforts by the Conservatives to demonize Liberal leader Trudeau are working but not in the way the Conservatives expected. For many voters, the attack ad message "Justin - just not ready" is being interpreted as "I wonder if that Mulcair fellow is ready." Second, campaigning on a soundly managed and healthy economy is proving to be a castle
made of sand for the Harper Conservatives, reinforcing the idea of a closer examination of Mulcair.
Pollsters and political operatives are well aware that voters tend to move quickly and en masse in an unstoppable wave.
If the summer love affair Canadians seem to be having with Mulcair and the NDP takes root, the ridiculous notion of an NDP majority government in Ottawa will be no more ridiculous than an NDP majority in Edmonton. Momentum is critical and if a growing number of voters see Mulcair as a prime minister in waiting, it will be difficult (but not impossible) for Harper to fight back. Mulcair would benefit with a chat with Adrian Dix, as would Harper talking to Christy Clark, to hear how easily a monumental lead can get blown away. It's a long time until voting day. There is no lead that is safe and there is no obstacle that can't be overcome.
Along with the feeling of Canadians regarding majority and minority governments, the most interesting finding of in the Nanos survey was regarding if Canadians feel their personal finances are better off, worse off or the same since 2011. More Canadians feel they are worse off than better or the same and those numbers hold across the country, except for Quebec, and across gender and age. That's a worrisome outcome for Harper but it gets worse. The demographic that reported the lowest replies of better off was the 60-plus category. If the Conservatives are losing the retirees, they are truly in for a world of hurt come election day.
The polls are pointing to serious black clouds on the horizon for the Conservatives and sunny skies for the NDP but it's still too soon to say who's going to get caught out in the rain.