More on the Desnethe-Missinippi-Churchill allegations

A few days ago, I noted the story about Conservative allegations of electoral impropriety in the Saskatchewan riding of Desnethé-Missinippi Churchill River, pointing out that the last poll to report (Ahtahkakoop), which Harrison had complained had put the Liberals over the top, had in the last election given Harrison only one vote. And only one vote when PC in 2000 (no one voted Allliance.) (The data is all downloadable at elections Canada.)

What I didn't realize was that the poll Ahtahkakoop used to be called Shell Lake, which is where my mother grew up and all my cousins live (or used to live before they grew up and moved away; I'm hoping to visit my uncles there two weeks hence.) Anyway, here's a map (within which, by the way, sits both my grand-father's farm and my great-grand-father's!):The Conservative Candidate complains that Ahtahkakoop went overwhelmingly Liberal. But this, as I mentioned, is consistent with their recent voting pattern, and consistent with other native communities in the area. Note that other reserves also appear in the map: Sturgeon lake (no CPC votes in 2004), Big River 118 (3 CPC votes in 2004 from 391 eligible voters), Mistawasis 103 (3 votes from 439 eligible voters), Sturgeon Lake (124) (no votes from 211), Sturgeon Lake (125) (3 votes from 300), Montreal Lake 106 (10 votes from 343). So Ahtahkakoop merely followed its normal voting pattern.

And in this it seems to have been consistent with how natives voted in 2004. In that election, there were 176 polling stations. Of these there were seven in which the CPC got no votes (Descharme Lake, Garson Lake, Kinoosao, La Loche, Pelican Narrows, Sturgeon lake); sixteen where they only got one vote (Shoal Lake--Ruby Lake, Ahtahkakoop 104, Sandy Bay, Cumberland House, Stanley Mission, Turnor Lake, La Loche Reserve--Clearwater River, Cumberland 20, Missinipe--Grandmother's Bay, Pelican Narrows, Pelican Narrows, Hall Lake, La Loche, Sturgeon Landing, Camsell Portage, Stony Rapids), eight with 2 votes; nine with 3 votes; etc. (If you want you to check my figures, download the data from Elections Canada yourself.)

What's the pattern here? I don't know all of these communities. But their names suggest that they are native. And in this riding, not many natives vote CPC.

The local Conservatives surely know this and concentrate their efforts elsewhere. What's their complaint? That the turn-out was higher than they expected, or, rather, hoped.

Is there any reasons to explain why it might be higher? Three come to mind:

1. it is winter, rather than summer (cf. the turnout in Nov. 2000 to June 2004 here)
2. the Liberal candidate was a high-profile native leader from near-by Prince Albert;
3. the Conservatives had said they would ditch the Kelowna accord

Yes, the Conservative Harrison has repeated allegations of Liberal threats and inappropriately placed campaign literature. But allegations like those swirl in every campaign office about every party. If any are true, they should be punished.

But the data before us can be explained without widespread fraud. The native vote, which everyone expected to go overwhelmingly against the Conservatives, did, and their turn-out was high. And that should be a good thing, no?

Update CBC: Merasty's victory has shrunk to 73 votes and Harrison has until tomorrow to apply for a recount.