NDP surge

Based on the spate of recent polls, one cannot deny something is happening and possibly something quite big.  Until we see an established pattern, I don't plan to update my projections, but lets just say it looks like the NDP will win more than 2 seats in Quebec.  Outside Quebec, the NDP maybe doing well, but there is nothing really unusual about their numbers and I don't totally buy the idea that since Quebec has swung towards the NDP, so will English Canada.  If we look at each region we can see the following.

Atlantic Canada

The polls have high margins of error, but the NDP certainly has gained here, but still sits in third and considering there are strongest in Nova Scotia where they have an unpopular provincial NDP government, I don't expect them to gain a lot here.  Maybe a seat or two, but that is about it.  As for vote splitting, it is tough to say, however there are at least 5 ridingsif not more when incumbency is factored in where the Tories are over 45% meaning they will probably hold those ridings either way.  At this point I see no reason to alter the predictions here.

Ontario

Some polls show the NDP in the low 20s, but their polling in Ontario is no higher than they were 10 days before the election in 2004, 2006, and 2008 so I am quite skeptical about any gains here.  With the economy being a bigger issue in Ontario than other provinces and the NDP being seen as weak here, I expect they will finish in third.  In fact if an election were held today, I suspect the results in Ontario would be very similiar to 2008 with only a few close ridings changing hands.  Also, Ontario had an NDP government from 1990-1995 and considering voter turnout is much higher amongst those over 40 than under 40, many voters still remember this and will be very reluctant to vote for them.  Also in most ridings even with the changing polls it is still clear that the Liberals have a far better chance of defeating Tories so those who vote strategically will probably stick with the Liberals unless the NDP polls well ahead nationally

Western Canada

In the Prairies, the Tories got over 50% in 7 of the 9 ridings they won in Manitoba, and 10 of the 13 ridings they won in Saskatchewan and considering the NDP rise is having almost no impact on the Tories, I doubt it will result in too many seat changes here.  Also the NDP in Saskitoba is only around 25% which is close to what they got last time around.  In the case of British Columbia, this could cost the Tories seats, but most polls only show the NDP in the high 20s like last time around and considering the bad experience BC had with the NDP in power provincially during the 90s, I don't think they will get much above 30% and I doubt they will surpass the Tories as the Tories have a pretty strong base in BC to begin with.

Quebec

Quebec is a whole different story and no doubt if those numbers hold we could see a massive shift in many seats.  This would not be the first time this has happened in Quebec, after all in 1984, the PCs went from 1 to 58 seats, in the 2007 provincial election, the ADQ nearly formed government after only having 4 seats in the previous election so a huge NDP gain in Quebec is quite possible.  In fact considering Quebec is more left wing than any other province in Canada, I have often wondered why the NDP didn't do better here.  Most of their support comes from the Bloc Quebecois, so that is the party that is most likely to suffer.  It may actually save the Tories some seats due to vote splitting as there aren't a whole lot of Tory-NDP switchers in Quebec.  I also don't think it will hurt the Liberals as badly as some think since Liberal support in the Montreal area is mostly on the West Island which is fairly affluent and the NDP has never done well amongst the affluent.  If anything, their gains will more likely be on the East Island which is working class and also Bloc Quebecois.  Either way, I suspect my predictions are probably fairly off but until the polls stabilize and we get a clearer picture, I don't plan to change anything.  One thing I have found is Ontario is usually the easiest province to predict in terms of seats while Quebec is the most difficult as it is not uncommon for Quebecers to move en masse towards a certain party.  In the case of Ontario, any party that won a riding by more than 10% will usually hold it and I suspect most if not all ridings won by more than 10% last time around will back the same party.  I will update my predictions after the Easter weekend when we have a clearer idea of what is going on.  The one thing that is clear is this election could be a game changer despite all the predictions that this would be a Seinfield election and we would get an almost identical parliament.