The following projection is based upon a series of polls released between Sept. 8 to 16 from Ipsos, Nanos, DART, Abacus and Leger, producing a blended and weighted sample of more than 8000 respondents. The Liberal Party leads with 165 seats, followed by the Conservatives with 133, the New Democrats with 20, the Bloc Quebecois with 13, the Greens with 5, and one each for the People's Party and an Independent. This is not to be interpreted as a prediction of the future, but rather is an estimate of what the parliamentary seat distribution during the opening week of the election campaign.
As has been the pattern for almost two months, change in public opinion at the national level has appeared minimal and has moved at a glacial pace. Even now, aggregations of national polls suggest a virtual dead heat between the Liberals and Conservatives. However, there has been movement at the regional level. The surprising aspect of the growing seat lead by the Liberals despite the near tie in popular vote estimates is largely attributable to Ontario, where there is a substantial number of swing ridings, coloured grey in the accompanying LISPOP map. The Liberal lead over the Conservatives is now estimated at seven percentage points, producing more seat changes than occur elsewhere. Since the previous projection there has been a somewhat compensating increase for the Conservatives in BC, where they now register a five-point lead over the Liberals.
Table 1: Federal Seat projections - September 19 2019 (2015 election results in brackets)
|Canada||165 (184)||133 (99)||20 (44)||13 (10)||5 (1)||1||1|
|Atlantic||25 (32)||7 (0)||-||-||-||-||-|
|Quebec||51 (40)||12 (12)||1 (16)||13 (10)||-||1||-|
|Ontario||71 (80)||39 (33)||11 (8)||-||-||-||-|
|Prairies / North||8 (8)||22 (18)||1 (5)||-||-||-||-|
|Alberta||- (4)||33 (29)||1 (1)||-||-||-||-|
|British Columbia||10 (17)||20 (10)||6 (14)||-||5 (1)||-||1|
Note: The "regional swing model" is more fully explained in a paper presented by Dr. Barry Kay to the 2009 annual meeting of the American Political Science Association, entitled "A Regional Swing Model for Converting Canadian Popular Vote into Parliamentary Seats 1963-2008." It should be noted that the application of the model above does not make use of the "incumbency effect" described in that paper. In tests for past elections, using late campaign polls to project electoral outcomes, the model has proved to be accurate within an average of four seats per party since 1963. Readers interested in post-dictions for past federal elections dating back to 1963, for projections using pre-election polls dating back to the 1980 federal election and for three Ontario provincial elections, may contact me at [email protected].