Reality check. And a third poll confirming the NDP surge.
That's a reality check for me, not you. Judging from your sober reaction to my exuberance over yesterday's Ekos poll, an ER self-administered reality check is in order.
But first, I neglected to assay both big polls today. Fortunately for ER, the latest Nanos Research poll for CTV-Globe and Mail confirms the Toronto Star-La Presse-Angus Reid survey I did discuss earlier this morning. Namely that the NDP leads in Quebec, is running second nationally (every time I type that I can't believe I'm doing so), and the Grits are mired in the low 20s nationally.
The Globe-CTV's latest Nanos Research poll, also released this morning, shows roughly the same results, with the Tories a little stronger than Angus Reid, at 37.8% public approval, though still a bit short of majority. Nanos, too, has the NDP placing a strong second, at 27.8% (again, close to that magic 29% the NDP needs to solidify official-opposition status). Nanos has the Liberals way back at 22.9%, the Bloc at 5.8%, and the Greens at 4.7%.
Now the reality check, which is seat projections.
Everything I've written about everybody's chances of doing whatever means squat depending on how their popular vote, no matter if it's surging or plummeting, translates into seats. And here's where we enter fantasyland.
Here's the average of 10 outfits that do seat projections, at pixel time:
Tories, 148; Liberals, 67; NDP, 56; Bloc 36; Greens, 0, Independents, 1.
An anti-Tory coalition in that scenario is unlikely, since the gap between the Tories and the second-place Grits is just too big to be easily justifiable to either the public or the G-G. Plus, it would require the participation of the Bloc, and even then the tripartite coalition would have just 4 seats over the 155 required for a majority. Harper would scream bloody murder, and the G-G would blanch at the instability of that 4-seat margin.
But here's the thing - that average masks wildly different projections among the 10 prognosticators.
Here's the range for each party about the seat guesstimators:
Tories, 161-131; Liberals, 78-57; NDP, 100-42; BQ, 42-14.
In the historic 1990 Ontario election, the landmark status of that outcome wasn't so much the NDP forming an Ontario government for the first time as the downright bizarre way that scores of riding contests broke the NDP's way. The NDP won at least 30 seats in deeply NDP-hostile territory, notably in southwestern Ontario, due to the weirdest, least predictable vote splits imaginable.
How those same splits play out May 2-3 defies prediction utterly, as the ranges above show. They show the NDP, for instance, either picking up 64 net new seats or just 6. They have the Grits picking up one seat, or losing 20. The BQ drops a mere 5 seats or 33, practically wiping the Bloc from existence.
As an isolated example: every other report I've seen of the Tories' 10 Quebec City region seats has the Tories easily holding most of them, or losing most of them.
So it's a mug's game writ large, if you'll pardon the pun. And I'll try to be more restrained about where the chips will fall.
Or, as Paul Wells said Monday the tail of his convincing Tories-headed-for-comfortable-minority-or-narrow-majority blog post, "UPDATE, Monday evening: Ekos has numbers so different from Nanos‘ that you can argue all night over who’s right!"
And the next day's Kinsella: "The polls are all over the map, and have been for weeks. The seat projections are based on contradictory polls, and are therefore a joke. I accordingly give up. At this point, the only poll that matters - as the old saying goes - is the one May 2.
Fair enough. Extreme caution is indicated.
Except that the parties are at this critical juncture applying their limited resources to those seats they consider winnable, which gets us back to the black art of seat projections. And which has all of us - those not focused on "The Wedding of the Century," and that would be Kate and Bill, not Jack and Iggy -cogitating on an historic realignment on the left; an opposition-led minority government; a renewed Tory minority still constrained from reshaping the country to Harper's lifelong design; or a Harper majority whose outcome, Harper once said, would result in a Canada we won't recognize when he's done with it.