Breaking... Compas poll sees Tory majority, NDP official opposition.

The latest Compas poll, released just now (about 9:40 a.m.), will be among the last before the pre-vote polling blackout.

Compas has national party support thus: Tories, 46%; NDP, 26%; Grits, 17%; Bloc, 7%; Greens, 4%.

This is outcome is hardly a sure thing, Compas says, given that 61% of its latest poll respondents say they might change their minds in the next 48 hours. On that score, the NDP has the least committed voters, with 79% saying they could switch before casting their ballot; 71% of Greens saying that; 65% of Grits switchable; 51% of Tories; and 45% of Bloc voters.

The NDP wildcard is that, as in other pollsters' reports in recent weeks, the NDP leads as the second choice among potential switchers. Among voters who might switch, 34% say they'd vote NDP instead, 29% say they'd go Grit instead, 20% would switch to the Greens, 11% to the Tories, and 6% to the Bloc.

If you're rooting for the sort of transformational change the past two weeks has promised, this latest poll confirms it: the NDP will eclipse the Liberals for the first time in history.

Compas says the big story of the election is the NDP, which has the most to gain and the most to lose. If the planets align for the Dippers, they could win 36% of the vote. Or, if things fall apart, NDP support could crater to 20%. I'll just note that "the most to lose" doesn't have you at 20%, which actually would be an improvement from where the Dippers were when the election was called. It would a huge loss - or, more accurately, disappointment - against high expectations that developed during the second half of the campaign.

The NDP downside is that its vote is "soft," which is inevitable given that so much of is Canadians who've never or seldom voted NDP. I'm no psychologist, but if you've changed your voter sentiments - from Liberal to NDP (the Liberals have evolved into a Tory clone), or from NDP to Conservative (you're a Grit whose afraid of the socialist hordes), it's tougher to switch back. That means you'll changed your sentiments twice now, and most of us don't feel great about the instability of intellect that suggests.

The NDP upside is that it has the most room to grow, as the second choice of more voters telling Compas they could switch than any party.

I'm at odds with two conclusions here of Compas pollster Conrad Winn.

Apart from having never seen the Tory support placed so high, at 46%, and only twice (out of two dozen polls I've studied in the campaign) placed so low, at 17%,

I'd say that at 51% stability for the Tories, against 79% for the NDP, the Tories have their own problem with switchability. "Only" 51% of Tory Compas respondents saying they might switch in the next 48 hours compares well only with the other parties. For that matter, the 65% "switchability" of only semi-committed Grit voters wouldn't seem to augur well for the Grits, either.

The second point is Winn's investment in voters who don't know their local candidate's names. Compas is on about this at considerable length in its latest, 15-page report. The fact NDP voters are least likely to know their local NDP candidates' name doesn't say as much to me as to Wynn, who sees that as a bad omen for Dippers.

It follows that in a time of great flux - which we appear to be in - first-time NDP voters are going to be even more likely to vote leader and party than local candidate, a phenomenon as old as voting itself. If you're determined to oust Harper, or thwart the socialist hordes, or rescue the Grits, or come home to the Bloc - a lot of folks doing that will give diddly-squat about the name on the ballot, only the party affiliation. (I must confess here that I do not know the name of my MPP. I voted McGuinty in last election, as I will in the next.)

 

 

Comments

[Chuck Deshaies]

Those numbers can't be right. A 20 point lead for the Conservatives? It is either a rogue poll - or every other pollster has been way off for a very long time.

And people who still can't make up their minds at this late stage shouldn't be allowed anywhere near a polling station tomorrow.

[Darwin O'Connor]

Yes, it seems like it a 1 out of 20 polls that is outside the normal range of error. I assume the other pollsters will be realizing new numbers today and then we'll see.

[Wascally Wabbit]

Looks like that is a definite outlier- here are others from today
niknanos Pollster Nik Nanos
CTV/Globe/ #Nanos Apr 28-30 C 37.0, N 30.6, L 22.7, B 5.5, G 3.2 bit.ly/nanos Sat 1day tight | polling tday, more #s @9ET
#
Tweeted by kady O'Malley
cc @kady RT @Colettod: Abacus Poll: CPC 37, NDP 32, LPC 18 n=1,007, April 28-29 http://bit.ly/jDVwyS
11 minutes ago · LikeUnlike
#
Paul Darlaston davidakin David Akin
Abacus poll: 58% say CLSH should quit if he fails to get majority. 73% say MI should quit if #LPC fail to win >77 seats. #elxn41

[winter]

very small sample size. their last poll was way out of whack as well.

[Kevin Dornwinkle]

This is obviously a rogue poll. Every other single poll I've seen shows the CPC in the mid to high 30s, not anywhere close to 46%

[albertaclipper]

Ekos has been out to lunch for 5 months now so I think we can take Compas Research poll as pretty much as gospel. (no I don't go to church. It's a phrase)

[D. Peturk]

I think the Compas poll is right on.

The NDP "surge is definately soft everywhere except Quebec, and in Quebec they have no ground force.

Frank Graves and the rest of the Liberal polsters can't differentiate between real results and wishful thinking.

[Mark]

Every poll had Ford and Smitherman tied and we know how that turned out. We will have to wait until tomorrow to find the true results.

[Frib]

That's a poll much more to my liking. I firmly believe that all this NDP talk is just that, talk. I don't think it will be reflected in the results.

[whatever]

See that Conrad Winn is up to his old tricks again. Surprised that he didn't find that Canadians want Harper as PM for life.

[Paul]

If any of you NDP'ers actually read the poll methodology which I doubt since you never bother reading the budget,

These numbers are based on "Distribution of the Vote if Voters Do Not Change Their Minds and If NDP Voters Turn Out to Vote in the Same Proportion as Other
Voters."

[joomla development services]

It is really nice to read this post. I learnt so many things here. I love your post and expect more post from you. Thanks for sharing it.

[Supra Shoes]

You bet!

[Girin Jackson]

A great article. Well stated.