By-election Breakdown

The spinners are out in full force when it comes to today's by-elections. The Cons are presenting a laughably skewed take in order to try to claim some momentous victory if they can eke out a win in territory which has voted for actual or expelled Cons all but once in the past 40 years; the Libs are understandably pointing out that the ridings aren't friendly terrain in order to justify a set of third-place finishes to come; and the NDP and Bloc are playing up their credentials in order to boost morale for election day. But let's put the current spin aside and take a look at the most likely outcomes in each riding.

New Westminster-Coquitlam

It's been well documented that the battle here is between the NDP's Fin Donnelly looking to take over the seat previously held by Dawn Black based on a combination of personal appeal and anger over the HST, while the Cons' Diana Dilworth tries to hide out and hope that the "Conservative" next to her name carries the day. Meanwhile, Lib Ken Beck Lee and Green Rebecca Helps figure to be fairly firmly entrenched in third and fourth place respectively - though it's worth wondering whether the flattening effect of a by-election can push Lee within striking distance of the top two contenders.

X-factor: The HST didn't manage to have much effect on a provincial by-election in Ontario, and today's vote may serve as the primary test of whether or not it'll sway votes in British Columbia.
Outcome Estimates: 65% chance of Donnelly win; 30% chance of Dilworth win; 5% chance of Lee win; 10% chance of Lee 2nd; 25% chance of Helps 3rd or higher

Hochelaga

Gilles Duceppe is now looking to downplay his party's all-out attack on the NDP. But all indications are that the NDP has put in place a couple of pieces from its winning formula in Outremont by winning the sign war and positioning itself as one of the two main alternatives in most commentary on the race - leaving the question of whether the final piece (shifting votes from other parties into the NDP's camp once that two-way race is set up) will push Jean-Claude Rocheleau ahead of the Bloc's Daniel Paille. The Libs normally run second in the riding but candidate Robert David hasn't been heard from much, while the Cons' Stephanie Cloutier looks to be even further back.

X-factor: It's been a common phenomenon in Quebec races involving the Bloc for federalist organizers and voters to unite behind one of the national parties. But for all of Rocheleau's strength through the campaign, I haven't heard much indication that he's managed to pull off the feat - and it's hard to see how he can overcome the Bloc's usual margin if not.
Outcome Estimates: 80% chance of Paille win; 15% chance of Rocheleau win; 5% chance of David or Cloutier win; 60% chance of Rocheleau 2nd; 20% chance of David 2nd; 40% chance of Cloutier 3rd or higher

Montmagny – L'Islet – Kamouraska – Rivière-du-Loup

I still have to figure it's a bad sign for the Cons if their star challenger Bernard Genereux is having to run against his own party. But there's at least a fairly plausible set of turnout assumptions which would nonetheless propel him to victory over the Bloc's Nancy Gagnon (particularly with the provincial Lib machine behind him), making this as likely a riding as any to change hands. Further back, while the NDP has done an effective job promoting Francois Lapointe, it'll have to be a pleasant surprise if he can make up a 10-point gap on the Libs to beat out Marcel Catellier for third place.

X-factor: Will the Cons' major highway announcement on Friday tip votes into their column based on Genereux' promise of patronage, or will it simply lead to voter fatigue?
Outcome Estimates: 60% chance of Gagnon win; 40% chance of Genereux win; 20% chance of Catellier 2nd; 30% chance of Lapointe 3rd or higher

Cumberland – Colchester – Musquodoboit Valley

While Bill Casey's old seat is being seen as an extremely close contest (with even the likes of Craig Oliver picking the NDP to win), I'll have to take a slightly less optimistic view of the likely outcomes. Yes, the NDP's Mark Austin is likely to finish a fairly strong second - but the Cons in one incarnation or another (PC, Con or Independent via Casey) have held the seat 36 of the last 40 years, and the safe bet has to be on Scott Armstrong continuing that pattern. With Lib Jim Burrows and Green Jason Blanch likely slotting into 3rd and 4th respectively, the other question to watch in the riding is whether the Christian Heritage Party's Jim Hnatiuk will develop enough traction to pass any of the national parties - but the likely answer there is "no".

X-factor: As much progress as the NDP has made within the riding provincially, the outcome will come down to whether the riding's voters are ready to forgive the Cons for their treatment of Casey now that one of his loyalists is carrying the party's banner. If so, then Armstrong should win in a walk based on the riding's historical tendencies; if not, then Austin figures to be the beneficiary of personal factors which easily trumped partisan identification in 2008.
Outcome Estimates: 70% chance of Armstrong win; 30% chance of Austin win; 25% chance of Burrows 2nd or higher; 20% chance of Hnatiuk 4th or higher.

As always, the above numbers reflect my own (hopefully reasonable educated) guesses rather than any mathematical formula. And it won't be long before we'll have some real results to mull over and analyze - hopefully with at least a couple more NDP MPs to show for the races.