A Taste of the Danforth By Election

The by election in Toronto Danforth (or, as it must be referred to in every article on the subject, "Jack Layton's old seat") has been called for March 12 March 19. If you're looking for political excitement, by elections may be as good as it gets over the next three years of majority government politics.

By elections are notoriously hard to handicap, but a glance at last year's results makes the favourite obvious:

Layton (NDP) 61%
Lang (Lib) 18%
Koenig (CPC) 14%
Mugnatto-Hamu (Green) 6%

A 43-point margin is daunting, but not insurmountable. After all, three of the seventeen Harper era by elections have seen one party pick up at least 25 points - the NDP in Outremont (+30), the Liberals in Winnipeg North (+37), and the Tories in Cumberland Colchester (+37). There were unique circumstances in play for all three, but the death of the most popular politician in Canada is fairly unique, so a win by the Grits isn't impossible.

But it won't be easy.

While the Liberals held the riding from 1988 to 2004, local demographics favour the NDP. On my nifty regression spreadsheet that projects vote based on demographics, Toronto Danforth is the 16th best NDP riding in the country and the 60th best Liberal riding. The riding has been orange provincially for over a decade, and the Ontario Dippers took it by 23 points in the fall. In short, voters on the Danforth are the type of people who vote NDP and they've tended to vote NDP of late.

That means the NDP will have a lot of votes ID'd on the ground, and a good organization in place - even if many volunteers are distracted by the leadership race. The Grits should be a bit more focused and there are still a lot of Liberals left in Toronto to pitch in, so they should be able to at least match the NDP on the ground. Or at least they would have, had they not given the NDP a one-month head start by waiting so long to nominate their candidate.

That candidate will be Grant Gordon, an ad executive who made a stir online with a humorous nomination pamphlet. Gordon is a fine candidate, but he's not the "star" George Smitherman and other insiders not-too-wisely speculated about last month.

Gordon will be up against an equally solid-yet-unexciting NDP candidate - law professor Craig Scott. Communications consultant Andrew Keyes will carry the banner for the Tories, but they haven't been above 15% on the Danforth since the merger. Adriana Mugnatto-Hamu will once again run for the Greens.

As unpredictable as by elections are, all signs point to the NDP defeating the Liberals, albeit with a reduced margin.


Be sure to check out Pundits Guide for an update on the campaign-to-date.

Taste of Victory on the Danforth

There's a lot of talk today about the eventual by election for Jack Layton's Toronto-Danforth seat. Many Liberals are salivating at the prospect of a victory there which would pull them within a mere 67 seats of official opposition status:

Liberals hope to paint Layton’s Toronto riding red again

Liberal Party president Alfred Apps is counting out Stephen Harper’s Conservatives, predicting a two-way fight with the NDP for Jack Layton’s Toronto seat.

Mr. Apps, a Bay Street lawyer who knows the Toronto political landscape well, told The Globe Wednesday that Mr. Layton won the seat because of his “terrific municipal background and powerful personality.”

He added: “Jack actually won it on the strength of his personal reputation and charisma.”

Given that by elections are all about expectations, I certainly wouldn't be advising Alf Apps, or anyone else, to raise expectations of a Liberal win on the Danforth. But it's still likely worth examining just how probable a Liberal pick-up actually is.

For starters, we need to accept that anything can happen in a by election. In Kevin Lamoureux's Winnipeg North by election upset last year, NDP support fell from 62% to 41% while Liberal support jumped from 12% to 46%. So at least in theory, Jack Layton's 43-point victory in May is not an insurmountable hill to climb.

But that's still a mighty tall hill, which only looks taller when you take a proper scan of the landscape. As the map at the top of this post shows, the two ridings to the East of it fell to the NDP in May. A relative no-name, Matthew Kellway, beat a 6-term incumbent in the Beaches by 11 points. That's nothing to sneeze at.

The riding has gone NDP provincially the previous four elections - most recently by 6,500 votes. Yes, Peter Tabuns is a popular incumbent, but there are a lot of people in the riding who are used to voting orange, even when Jack's name isn't on the ballot.

Finally, it's a riding that, demographically, "should" be an NDP seat. When I put the Census demographics into the regression meat grinder, it comes out as the 16th strongest NDP seat in the country.

Add it all up, and the odds of an upset on the Danforth are exceedingly small. Raising hopes to the contrary is only setting Liberals up for disappointment.