If some pundits are to be believed, Quebec is crucial to Liberal hopes. Quebec public opinion exists in a world apart from the rest of the country. In national polls, however, it is but one region. So it makes sense to pay close attention to the Quebec polling companies, which conduct large sample surveys just in Quebec. The two most prominent are Léger and CROP, both of whom have had recent polls, Leger in September, CROP in August .
I have developed a new seat model that takes into account language preference in Quebec. However, one tends only to get a linguistic breakdown in large sample Quebec polls. I have combined the two polls (I had to infer the Anglophone vote in the Léger poll which has not put details in the web yet) and used my new linguistic model to get the following outcome:
Lysiane Gagnon argues in the column linked to above that the Liberals think they can win as many as 12 more seats than their current 14. This calculation would bring them up 10 seats.
However, the Bloc would remain the dominant party and Liberal gains here would not be decisive in determining the overall outcome.
The Conservatives at the moment have just enough support to hold most but not all their Quebec seats, while the NDP appears to be slipping under the radar and at the moment would do better than expected here.