The following projection is based on a series of polls released between Sept. 20 to 24, since the embarrassing photos of Justin Trudeau were made public. It draws from a blended and weighted sample of 11,000 respondents from surveys administered by Ipsos, DART, Nanos, Angus Reid, Ekos, Abacus and Forum. The Liberal Party continues to lead with 150 seats, a decline of 15 seats from last week’s projection. The Conservatives edge closer behind with 142 seats. The NDP gained two seats to bring it up to 22, the Bloc Québécois gained four to currently stand at 17. The Green party and People’s party remain unchanged with five seats and one seat, respectively. One seat is projected for an independent candidate. This is not to be interpreted as a prediction of the future, but rather is an estimate of what the parliamentary seat distribution might have looked like in the days since Sept. 18.
One generalization to note is that the Liberals have declined slightly in almost every region, but by numbers within the margin of sampling error of 1-to-3 percentage points. In some places, it has cost the Liberals seats, but not in other places. The beneficiaries of the Liberal decline have also varied. In Quebec, the Conservatives declined as much as the Liberals, and it is the Bloc that has gained. In Ontario, the improvement was registered by the Conservatives, who moved from a seven-point to a three-point overall support deficit. The NDP also gained a couple of seats there. Changes in public opinion in the west had hardly any impact on seat movement.
Table 1: Federal Seat projections - September 25 2019 (2015 election results in brackets)
|Canada||150 (184)||142 (99)||22 (44)||17 (10)||5 (1)||1||1|
|Atlantic||25 (32)||7 (0)||-||-||-||-||-|
|Quebec||47 (40)||12 (12)||1 (16)||17 (10)||-||1||-|
|Ontario||60 (80)||48 (33)||13 (8)||-||-||-||-|
|Prairies / North||7 (8)||23 (18)||1 (5)||-||-||-||-|
|Alberta||- (4)||33 (29)||1 (1)||-||-||-||-|
|British Columbia||11 (17)||19 (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].