Update! - 29OCT2013
I've added the results and a link to a recent Ipsos poll showing the Liberals and NDP tied for first, and the Tories in 3rd place. (It is in the comments)
Simple answer: Ayeup! In fact if current trends continue, I'd expect it. How is this possible? For that answer we need to look at some past examples of a government slipping from 1st to third.
The most recent happened only a few weeks ago, when the Nova Scotia NDP did this. What's important to remember is that part (though not all) of what caused the party to slip was a scandal mid-term (expenses) and accusations of poor governance. Mid-term polls had the party solidly in second.
Go back to 2009 in the province and you see the same story but for the PC Party. Again, an expenses scandal, again, solidly in second during the mid-term, and again, ended up in third 2 years later.
Well maybe this is just a Nova Scotia thing?
Lets look at some other provinces. Lets jump on over to BC, where we have a counter example. The last change of government was 2001. Polls (wikipedia does not have this, but I have records from mustel) had the NDP dropping from 30 to 20 in the same period (mid-term to election) if not for the fact that there was no other party to pick up the slack (greens remained stable) we likely would have seen the same. Manitoba in 1988 does not have easily accessible poll records, but I do know that when Doer was picked to lead the NDP the party was neck and neck with the Liberals. The final easy example of a government dropping to 3rd was Ontario in 1995. Again, I have records Wikipedia does not (this time from Environics - I'm willing to share the excel file if anyone doubts me) that had a 3-way race in the mid-term.
We are actually thus able to estimate how far the CPC would drop. Using the recent poll showing the parties at a 32-32-23 split, we can use these last elections to make the following assumptions.
1 - There is about a 50% chance, give or take, that the CPC will suffer the same fate, and thus, end in 3rd. This happening depends not on what the NDP does, but, on what the Liberals do. The more clear and obvious a Liberal victory, the more willing anti-Liberal voters will be to switch to a party they agree with more than they are to vote for a party with a better chance of winning. This means a 4 point swing to the Liberals.
2 - The NDP can't muck this up. It's always possible for an opposition to scare off voters. We've seen this in 3 of the last 4 provincial elections to varying degrees. If the NDP can hold it together - and they seem to be more in danger of being boring than they do of being scary - they could then get about 4 points themselves.
But this is only popular vote. How would this translate into seats?
That's much harder to presume due to the new ridings, and the fact that a provincial breakdown will skew these numbers. Here, however, are some estimates.
BC and the Territories
The Tories will retain some strength here, and so will the NDP, despite a strong Liberal showing.
I fully expect gains due to the new smaller ridings in Alberta and the newfound distaste for the Tories. The Liberals will take Calgary Northeast.
Huge Liberal sweep. Both the NDP and Tories have done things unpopular in the region, and, Trudeau is popular here. Add to this the trends backing this up, and we will see the entire region go bright red.
A subtotal of the above. The Liberal strength will come from Ontario and Quebec, while the NDP will make up for lost ground in Quebec.
Easier to estimate than Quebec because of the huge swings expected. Ontario has been known to swing in the last week of the campaign based on how well or poorly parties are doing elsewhere. In fact the entire final-week swing for the NDP in 2011 came from Ontario, as their swing had stopped in every other province including Quebec.
The NDP will need a lot of seats from Quebec to overtake the Tories. Are there enough seats here?
The new ridings will make it more difficult for the Tories to hold on, and, have hurt the Bloc. The province will turn into a horserace between the Liberals and the NDP, and with leaders from the province, this will suck up votes from all sides. Poll averaging tells us, however, that due to vote efficiencies, we are in for a long night.
It's too close to call right now.
The final answer may be determined by 100 votes in ridings like Saint Maurice, Avignon-Matane-etc, and Compton, as well as if the Bloc can manage to take even a single seat.
The Tories have an advantage in that the NDP will need a few points in the polls to beat them in seats, but, it's certainly possible. As we get closer to 2015 and polls in Quebec settle down* we will have a much better picture of what will happen.
*Current polls can't agree on who will sweep Montreal, the Corona, or the outer Regions of Quebec.