Toronto Centre By-Election Prediction

Toronto Centre, looking towards Yonge & Bloor from 30 Gloc, July 2011. Kenneth M. Kambara

I'm skeptical of polling results that aren't commissioned by the parties in Canada, in the wake of the recent Alberta and BC debacles. I understand it's hard to get good data without a huge budget. The latest Forum poll has the Liberal-NDP race in Toronto Centre heating up to 47 to 39%, respectively for Chrystia Freeland and Linda McQuaig. I have concerns with the poll, as it's skewed older (62% of the sample are 55 or older), overrepresents Liberals (58% of the sample compared to 41% of the vote in 2013), and I do believe it used landlines. I'm sure it was weighted, but I'm leery of putting too much stock into it.

That said, I thought it might be interesting to examine the data, looking at how the 2011 vote mapped onto current preferences. The 2011 vote distribution was:

Rae (LPC) 41.01%, Wallace (NDP) 30.21%, Moore (CPC) 22.64, Michelson (GPC) 5.02%

Using back of the napkin calculations from the latest Forum poll, Freeland is holding on to only 64% of Rae's vote. She's getting 25% of the Conservative vote, 12% of the NDP vote, and 45% of the Green vote. That puts her roughly at 38%.

McQuaig is holding on to 86% of the NDP vote, getting 31% of the Liberal vote, 12% of the Conservative vote, and 36% of the Green vote. That puts McQuaig at around 43%.

What should be troubling to Freeland is that she's only managing to hold on to 64% of Rae's vote and only getting 25% of the Conservative vote. My sense is that Freeland and Trudeau have been pitching a fairly centrist platform and the fact that more defections from the Conservatives in an urban riding should be a cause for concern. McQuaig is holding on to 86% of Susan Wallace's total and siphoning off almost a third of the Rae vote. Progressives may be moving towards the NDP and rewarding Mulcair's tough stance against Stephen Harper.

By elections are strange animals. Lower turnout is to be expected and the ground game tends to matter more. Turnout will be the key, but if the NDP wins, the data at the polling district level will be telling. If the NDP is able to win neighborhoods in St. James Town and Regents Park, they have a shot at making inroads into communities with new Canadians. If the NDP can win Cabbagetown, the Liberals have a lot to be worried about, as they're losing their grip on the upper middle class. I'm not quite sure progressive Liberals will be that motivated to vote for Freeland, so their turnout may be lower. The NDP seems energized with a youthful energy and has tighter messaging. If the NDP wins or is within 5% of Freeland, it's a huge accomplishment for the party and will cause some soul searching in the Liberal camp.

I think this will be a nailbiter and predict Linda McQuaig (NDP) will pull it off, just north of 40%. 

Update: I was wrong. See update on the new blog site.