One month into the federal election campaign, the three major parties remain right where they were at the start: locked in a three-way race.

Here are the latest weekly ballot tracking numbers from Nanos Research, with the percentage-point change from the previous week in brackets:

  • NDP: 30.8 per cent (+1.7)
  • Conservatives: 28.8 per cent (-1.3)
  • Liberals: 29.7 per cent (-0.2)

The new numbers are very similar to Nanos polling taken up to and including July 31, two days before the writ dropped on Aug. 2.

In the July poll, the NDP had 30.1 per cent support, the Conservatives had 31.5 per cent and the Liberals were at 29.3 per cent.

The latest polling, which was completed in the four weeks up to and including Aug. 28, suggests Ontario continues to be a battleground, perhaps explaining why leaders are giving so many speeches there.

The Conservatives are sitting at 38 per cent support in Ontario, while the Liberals aren’t far behind at 35 per cent. The NDP is in third place with 23 per cent.

British Columbia, meanwhile, is another close race, with the NDP at 32 per cent support, the Liberals at 30 per cent and the Conservatives at 26 per cent.

B.C. is also where the Green Party is polling most strongly, with 12 per cent support, compared to five per cent nationally.

The New Democrats have their strongest lead in Quebec, where they enjoy a 15-point lead, sitting at 41 per cent support. The Liberals have 26 per cent support, while the Bloc Quebecois have 17 per cent and the Conservatives are at 11 per cent.

The Liberals are polling strongest in Atlantic Canada, where they have 49 per cent support, compared to 36 per cent for the NDP and 14 per cent for the Conservatives.

The Conservatives have a commanding lead in the Prairies, with 45 per cent support versus 27 per cent for the NDP and 18 per cent for the Liberals.

Election day is Oct. 19.


A random telephone survey of 1,000 respondents is considered accurate within plus or minus 3.1 percentage points, 19 times out of 20. The weekly ballot tracking is determined using random landline and cellphone interviews with 1,000 adult Canadians, and may be weighted by age and gender. The current report is based on a four-week rolling average up to and including Aug. 28, 2015.

The regional numbers have sample sizes ranging from 78 respondents in Atlantic Canada to 261 in Ontario, which means they have a wider margin of error than the survey overall.

Click here for full methodology and all the charts.