Something happened in the first week of October.
It's difficult to say what it was, but the Liberals have been increasing in the polls and are now challenging the Conservatives even in terms of seats.
Compared with where the Liberals were in early August, the rise is major.
We are talking of a good eight to 10 points nationwide and more in some provinces. The observed increase in the past week is especially visible in Ontario, where the Liberals have been ahead in every single poll, sometimes with a margin big enough to make them competitive nationally.
With that said, the Tories remain competitive as well and could benefit from winning most of the seats in the West. Also, the NDP's fall in Quebec has benefited Tories even more than the Grits, and gains in Quebec are definitely a possibility.
Below are the current projections. They are based on the most recent polls published in October. The calculations use past election results as well as the current polls (both national and riding ones) to predict the winner in the 338 ridings. They include regional and incumbency effects. The confidence intervals and the chance of winning are obtained through the use of 5,000 simulations that account for the uncertainty of the polls as well as for the distribution of the vote and the electoral system. In other words, these simulations try to include every possible scenario, given the information we currently have.
Our model isn't ready to call it a 50-50 race (as opposed to other projections that even have the Liberals ahead), but it's getting a lot closer. We still believe the Liberals aren't ahead by a big enough margin in Ontario. Moreover, polls have consistently underestimated incumbents in recent elections, including in 2011. The trend, however, is definitely in the Liberals' favour.
Is it because many NDP supporters are now switching to the Liberals in the hope of defeating Stephen Harper? Possibly. Explaining why people vote is a lot harder than simply projecting seats, and we don't want to conjecture too much.
To win a Canadian federal election, a party needs to dominate a couple of provinces. The table below shows the probabilities of winning the most seats in each province for each party.
As we can see, many provinces are currently projected to go to one party with certainty. Few will be surprised to see that Alberta would elect a majority of Conservative MPs, but the percentages in Quebec, Ontario, and B.C. are more interesting.
In Quebec, the NDP is still likely to win a majority of seats despite a drop of about 20 points. For once, the electoral system and vote splitting are working in this party's favour. The fact that the Bloc seems to be staying at 20 per cent or less is obviously helping considerably.
Conservatives and Liberals are in the fight to take the most votes, but their support is too concentrated to challenge the NDP everywhere. For all the talk of a NDP collapse in Quebec, it remains possible that the party could easily get 40 seats.
Ontario is currently the closest race between Liberals and Tories. The Grits are currently both polling and being projected as favourite, but it's really as tight as it gets if we look at seats. A look at the riding polls confirms the projections. As usual in Canadian politics, Ontario will most likely be the deciding province.
B.C. hasn't attracted the same attention as Quebec or Ontario, the first two but it's the only three-way race in the country. All three parties are polling about 30 per cent, and all three have a shot at winning the most seats. The province could ultimately decide if the Liberals will complete their comeback and defeat the Conservatives. Or, if Ontario decides to remain blue, it could decide who finishes second.
The last nine days of this campaign will be very interesting.
Justin Trudeau will have to decide if his chances are better in Ontario, B.C. or Quebec. So will Harper. For Mulcair, Ontario looks like a lost cause, and he'll most likely focus on Quebec and B.C. He's out of the race for first but not out of the one for second place — and finishing second could actually be important.
ALSO ON HUFFPOST:
Bryan Breguet has a B.Sc in economics of politics and a M.Sc in economics from the University of Montreal. He founded TooCloseToCall.ca in 2010 where he provides electoral analysis and projections. He has collaborated with the National Post, Journal de Montreal and l'Actualité.
He will provide analysis and updates for The Huffington Post Canada throughout the federal election campaign. For riding by riding projections, visit his interactive simulator.