Friday, February 8, 2013

Forum Federal Poll: 32% Con, 30% Liberal, 26% NDP

In a first, and in heavy opposition to the recent Abacus poll, Forum Research has released their usual monthly federal political issues poll and it shows the Liberals moving into second place in the popular vote, ahead of the NDP - the only non-Nanos poll to do so, at least for their regular topline numbers.

Forum Research (Federal - February 6, 2013)
Conservative: 32% (-4%) - 136 seats (-26 seats)
Liberal Party: 30% (+5%) - 103 seats (+26 seats)
New Democratic: 26% (-2%) - 84 seats (+13 seats)
Bloc Québécois (QC Only): 24% (-1%) - 14 seats (-13 seats)
Green Party: 4% (=) - 1 seat (=)

Compared to the last Forum poll out in January, the Liberals for the first time since probably long before the May 2011 election, have reached above 100 seats in a projection done by myself. The Liberals gain most of their seats in Ontario and Quebec, where they sit at 34% (up from 27%) and 31% (up from 29%) respectively. That's good for a strong 70+ seats in central Canada alone.

The Liberals also lead in Atlantic Canada (34% to the Con's 31%, and the NDP's 29%) and post strong numbers in the Prairies (35% to the Con's 36%, but I put this down to a small sample size), and even hitting above 20% in Alberta - good for two seats, one in Calgary and one in Edmonton.

The Conservatives and NDP battle it out in BC, just line in Abacus, with the NDP leading 37% to the Con's 34%. In Quebec, the NDP rise to 29% support (up from 26%), which gives them 32 seats but they're all extremely close, especially with those stableish Bloc numbers. Sure, the separatists lost over a dozen seats, but they're definitely not far behind the pack.

As always, Forum did some Trudeau polling, showing a hypothetical Trudeau-led LPC mopping the floor with the other two parties.

Forum Research (Federal - Trudeau as Liberal Leader)
Liberal Party: 41% (+6%) - 169 seats (+36 seats)
Conservative: 30% (-3%) - 110 seats (-27 seats)
New Democratic: 20% (-1%) - 48 seats (+9 seats)
Bloc Québécois (QC Only): 22% (-3%) - 11 seats (-17 seats)
Green Party: 2% (-1%) - 0 seats (-1 seat)

The hypothetical Trudeau Liberals have seen a jump in support since the last Forum polling on this, landing us just short of a majority government in a 338-seat House of Commons (you need 170 seats). With strong results in Ontario (45% support to the Con's 33%) and Quebec (41% to the NDP's 24% and the Bloc's 22%), the Liberals would dominate under these conditions.

Even Alberta, where the Liberals sit at 29% in this hypothetical polling, give us five seats - five!!! That's absolutely freaking amazing.

Why exactly the standard topline numbers have seen such an increase for the Liberals is unknown, though it isn't hard to draw a line between Trudeau's fairly obvious route towards the leadership, and an increase in support for us.

Of course, if that is true, then it could easily deflate once people realize more about Trudeau. That does happen often to popular messiahs. But this poll speaks a lot to the weakness of the NDP - the Official Opposition, with its "strong" leadership and upbeat attitude, is being shoved to the side simply on the idea of Justin Trudeau leading the Liberal Party. That doesn't bode well for Canada's social democrats.


  1. Before you get too excited, read the detailed breakdown of the past party support of those polled in this survey. In the last election, the Cons were first with 39.6%, the NDP second at 30.6% and Libs third at 18.9. Based on 1095 respondents, the breakdown of past election support should be 434 Cons, 335 NDP and 207 Lib. Instead those responding broke down as 313 Cons, 260 NDP and 297 Lib. In short, the poll respondents are more pro-Liberal than the electorate from the last vote. This would explain the much higher Liberal numbers and the lower numbers for both Cons and NDP. The devil is certainly in the details.

  2. Hey Brian,

    I've looked at the data and came to a bit of a different conclusion. While the numbers are distorted at first glance, remember that these are decided voters/respondents that Forum has put into the table - *not* the full base which includes undecideds. If you were to see the table that includes undecideds, it would make much more sense. So the sample itself is not purposely biased; I'd also assume that Forum puts in leaners and so on to counteract the partisan bias in the decideds sample.

    While there is some bias in the numbers, Forum is an experienced pollster and they will work to correct it to the best of their abilities. After all, that is why the margin of error does exist. So I think these numbers are reliable enough to get somewhat excited over.