Wednesday, October 15, 2008

Final quick hitters & observations post-election















Winner: Me. They weren't dead on, but they were pretty darn close. To be honest, if the Liberals had of taken care of business in their stronghold of Ontario, I think my predictions would have been even closer. In the end, that's the difference between being the guy who gets a couple of pats on the back for his effort and being famous amongst his peers. lol

Anyway, here are the comparisons:

My national predictions (Seats, % Votes)
CPC 136 (35.7), Lib 87 (27.8), BQ 49 (10.8), NDP 34 (19.9) , Ind 2 (0.9), Grn 0 (4.8)

Actual '08 election results (Seats, % Votes)
CPC 143 (37.63), Lib 76 (26.24), BQ 50 (9.97), NDP 37 (18.20), Ind 2 (0.65) Grn 0 (6.88)

Winner
: Conservative leader Stephen Harper. Over these next few weeks and months, I'm certain Harper's leadership (and coming up short on a majority) will be cannon fodder for the press. However, unlike some pundits and Ottawa experts, I see another minority as a gift from heaven for the Tory leader. Why? Well, there are several reasons. Firstly, Harper has the full backing of his own caucus and party, unlike the Liberals who will certainly be re-entering unstable times with another leadership race. If Harper is smart (which I'm pretty sure he is), he will do what most Prime Minister's do in this situation. Pull up a chair and be a spectator to what will almost certainly be internal Liberal infighting. I mean, the negative "anonymous sources" in the Globe and Mail mid-election were pretty much evidence of that. Secondly, for someone who wants to change the country to a more conservative ethos via incrementalism, he has positioned himself perfectly. If you don't think so, just look at Pierre Trudeau's journey through governing. The man changed the face of this country while both losing to a weak leader and forming a minority during unsettling times. So there is hope yet for Harper and his mission. And finally, by not forming a large majority, he avoids doing what many of his conservative predecessors did. That being, coming in quickly and abruptly on a blue wave, scaring the political establishment, and then subsequently being shown the door as a result. So yes folks, there are advantages to being in a minority government after all.

Loser: Green party leader Elizabeth May. Not only did she do herself a disservice by running in Central Nova against political heavyweight Peter MacKay, she weakened her party's vision (and other candidates chances in the election) by getting too caught up in her local battle. Plus, it didn't help when she confused her own party supporters by calling for them to throw their support behind Dion in order to stop Harper. In the end, she ended up coming across as the leader of an interest group who could care less about the party she was running under and more about settling scores. It's hard to believe she wants to come back for more punishment in Central Nova, especially when she knows the Liberals will run a candidate.

Loser: NDP leader Jack Layton. Don't get me wrong, the dipper leader is well liked and universally trusted. Unfortunately, this just won't cut it for Jacko. Why? Well, at the same time last week, Mr. Layton was convincing Canadians that he was not only ready to replace the Liberals as Official Opposition, he was ready to be Prime Minister. Now that the big gains failed to materialize, I'm sure many Canadians will question his sincerity on such things, unless you're a die hard NDP supporter who drinks the orange Koolaid and doesn't consider missing official opposition status by more seats then you won altogether a failed mission.

Anyway, if the Liberals are smart, they will take full advantage of Layton's inability to sell himself as Official Opposition leader and, in turn, reclaim some of that territory on the left.

Loser: Liberal leader Stephane Dion. What more can be said then has already been said? Honestly, for a guy who wore an albatross around his neck for 3 months by trying to sell his Green Shift plan, he didn't do all that bad. And for that, people will probably question Harper's ability to win more then his opponents obvious political shortcomings.

Anyway, if there is one thing that Dion should be commended for, it's that he ran a respectful, policy oriented campaign under tough circumstances. I'm sure the next Liberal leader will find out very quickly that it's no picnic running against the Tory machine on a good day. A leader, if history repeats itself, will likely be an anglophone from outside Quebec. Which is probably not a good thing considering they just regained a pulse in Quebec while suffering their worst electoral defeat in Ontario in quite sometime.

Oh well, back to the drawing board for the Grits I suppose.

Winners: Gilles Duceppe and Danny Williams. The two regions (other then Alberta) where the Tories seat count declined. Let's just say, Duceppe's ability to beat back the conservatives from making addtional gains on his Bloc turf was the TSN turning point of this election. Had he not, Harper would certainly be governing from a clear majority this morning.

Losers: A disengaged and apathetic electorate. I don't often agree with Paul Zed on much of anything, but he put it perfectly last night about the disinterest in this election: "people are just electioned out." Well said, Paul.

For the record, NB turnout was above the national average of 2006 this time around. On the flip side, voter fatigue seems to be the worst in Newfoundland and Labrador. No surprise here since they had Danny-boy running around telling them who they could vote for.

Projection post mortem

Over the past few weeks, I engaged in one of my favourite pastimes - predictions. To aid me, I developed a bit of an aid in the form a projection model. Here I try to evaluate how it did and what its strengths and weaknesses were.

First, its biggest flaw is that it is almost wholly dependant on outside factors. It is a poll-driven machine and, if the polls are not accurate, then it is not accurate.

Here is how the polling worked out vs. the actual result (both contrasted against the 2006 results):

 CPCLibBQNDPGrn
Poll AvgActualPoll AvgActualPoll AvgActualPoll AvgActualPoll AvgActual
Atl-9.2%-4.9%-2.0%-4.5%--+3.8%+3.4%+7.2%+3.7%
QC-4.4%-2.9%-0.1%+3.0%-0.8%-4.0%+3.9%+4.7%+1.8%-0.5%
ON-0.7%+4.1%-5.1%-6.1%--+1.3%-1.2%+4.9%+3.3%
Pra-0.6%+5.4%-6.4%-7.2%--+3.4%+0.1%+5.0%+2.7%
AB-7.1%-0.4%+3.5%-3.9%--+2.9%+1.0%+1.8%+2.3%
BC+1.5%+7.1%-5.6%-8.3%---2.7%-2.5%+7.5%+4.1%
Natl*-1.9%+1.3%-3.6%-4.0%--+1.6%+0.7%+4.7%+2.3%
*National numbers excluding the BQ were used to project the 3 northern seats

As you can see, the pollsters were a bit off (or at least my average of their polls was). No pollster got it exactly right, though Angus-Reid came very close in both national and regional numbers.

Using the actual regional results, my model gives the following:
CPC 145
Lib 77
BQ 48
NDP 35
Ind 3

Using only Angus-Reid's poll, my model gives the following:
CPC 141
Lib 76
BQ 49
NDP 39
Oth 3

So, I think I am relatively pleased with the model, even with the slightly off polling results, I seem to have done better than most prognosticators for a change.

Here is a list of the gross error for all the predictions listed on Calgary Grit's round-up and myself:
  1. Ekos Predictions (18)

  2. Calgary Grit's model (23.6)
  3. Barry Kay Seat Projections (24)
  4. Andrew Coyne (26)
  5. nbpolitico (28)
  6. UBC Stock Market (30)
  7. David Akin (30)

  8. Kady O'Malley (35)
  9. Democratic Space (36)
  10. Election Prediction Project (38)
  11. Scott Reid (40)
  12. Andrew Steele (44)
If I had had a crystal ball or a better gut, using the real results would have projected a gross error of 8, while using the best poll (Angus-Reid) would have given an error of 6 - obviously showing that the model needs a bit of work!!

But of course I am forgetting my blogging colleague NBT. Who needs fancy projection models when you've got him? His gross error: 22, good enough for second place! As the prominently displayed link on my other site says, a better prognosticator than I!

Projection history: Final :: Eighth :: Seventh :: Sixth :: Fifth :: Fourth :: Third :: Second :: First :: Methodology

Tuesday, October 14, 2008

Live blog (sort of)

(all times Atlantic, will go live at 11 when west coast polls close)

9:17 - So much for my model in the Atlantic. As I had feared, Atlantic polls were not localized enough to be of much use. A *huge* CPC drop in Newfoundland and Labrador poisoned the sample. Tories could gain as many as four seats in NB, two in PEI and hold their own or gain one in NS.

9:52 - Huge upset in Miramichi where Charlie Hubbard is defeated handily. Saint John and Moncton are too close to call and my gut would have had them fall to the Tories before Miramichi. Go figure.

10:48 - Tory vote seems to be up in Ontario considerably over the polls, if this holds things could be a lot different than most folks predicted.

10:59 - Current vote totals have the CPC up 9% over the Libs in Ontario, leading in places like Ottawa Centre... if this holds it would be almost certainly a majority government. But it is pretty early.

11:16 - It continues to look good for the Conservatives - I think that CTV might want to retract calling Florida for Dion a Harper minority.

11:38 - Despite a steep Liberal drop in Ontario, Gerard Kennedy is pulling ahead in his race. He could be the only Liberal gain in that province.

11:59 - Kennedy declared the winner by CBC, ahead by nearly 2000 votes. If there is a Liberal leadership race which is largely a rematch of the last one, expect a major Kennedy speaking point to be that he ran against an incumbent and picked up the only Liberal gain in Ontario while Ignatieff, Rae and Hall Findlay all ran in ultra safe Liberal strongholds.

12:10 - Peter Kent pulls ahead in Thornhill? Would be very interesting if that holds.

12:15 - Who is this independent James Ford in Edmonton--Sherwood Park? He keeps edging in and out of the lead.

Ridings to watch

Here are nbpolitico's ridings to watch for the 2008 federal election. Polls are closed in the Atlantic and I am watching the returns. I'll follow the Election Act and not let you in on the secrets, but in the mean time, here are the seats that my projections showed within 5 points or "the ridings to watch" (parties listed in the order the projection model shows them placing):
  1. St. John's East: NDP vs. Lib
  2. Central Nova: NDP vs. Grn
  3. Dartmouth--Cole Harbour: Lib vs. NDP
  4. Honoré-Mercier: CPC vs. BQ
  5. LaSalle--Émard: BQ vs. Lib
  6. Outremont: Lib vs. NDP
  7. Rimouski-Neigette--Témiscouata--Les Basques: Ind vs. BQ vs. Lib
  8. Algoma--Manitoulin--Kapuskasing: NDP vs. Lib
  9. Beaches--East York: NDP vs. Lib
  10. Brampton West: Lib vs. CPC
  11. Brant: CPC vs. Lib
  12. Etobicoke--Lakeshore: Lib vs. CPC
  13. Guelph: CPC vs. Lib
  14. Kenora: Lib vs. NDP vs. CPC
  15. London West: CPC vs. Lib
  16. Mississauga South: CPC vs. Lib
  17. Mississauga--Erindale: Lib vs. CPC
  18. Newmarket--Aurora: CPC vs. Lib
  19. Oak Ridges--Markham: Lib vs. CPC
  20. Oakville: CPC vs. Lib
  21. Oshawa: CPC vs. NDP
  22. Ottawa South: Lib vs. CPC
  23. Sudbury: Lib vs. NDP
  24. Thunder Bay--Rainy River: NDP vs. Lib
  25. Welland: NDP vs. Lib vs. CPC
  26. Churchill: NDP vs. Lib
  27. Saint-Boniface: CPC vs. Lib
  28. Winnipeg South Centre: Lib vs. CPC
  29. Palliser: CPC vs. NDP
  30. Regina--Qu'Appelle: CPC vs. NDP
  31. Saskatoon--Rosetown--Biggar: NDP vs. CPC
  32. Edmonton--Strathcona: NDP vs. CPC
  33. Burnaby--Douglas: NDP vs. CPC
  34. Esquimalt--Juan de Fuca: Lib vs. CPC vs. NDP
  35. Newton--North Delta: Lib vs. CPC vs. NDP
  36. North Vancouver: CPC vs. Lib
  37. Richmond: CPC vs. Lib
  38. Vancouver Island North: NDP vs. CPC
  39. Vancouver Kingsway: Lib vs. NDP
  40. Vancouver Quadra: Lib vs. CPC
  41. Nunavut: Lib vs. CPC
Enjoy!

Monday, October 13, 2008

Final projection

After a little break for Thanksgiving weekend, here is the final update based on 7 new polls.

Angus Reid - Oct 8-10 (0.71 weight)
EKOS - Oct 10-12 (1.00 weight)
Ipsos-Reid - Oct 7-9 (0.57 weight)
Nanos Research - Oct 10-12 [ON, QC, Atl only] (1.00 weight)
Strategic Counsel - Oct 10-11 [ON, QC only] (0.93 weight)
Harris-Decima -Oct 9-12 [ON, QC, Atl, BC only] (0.93 weight)
Segma - Oct 5-9 [ON, QC, Atl, BC only] (0.43 weight)


National results (change vs. 2006, change vs. last projection)
CPC 129 (+5, -4), Lib 85 (-18, +3), BQ 50 (-1, +1), NDP 41 (+12, n/c), Ind 3 (+2, n/c), Grn 0 (n/c, n/c)

Atlantic (change vs. 2006, change vs. last projection)
Lib 22 (+2, n/c), NDP 6 (+3, n/c), CPC 3 (-6, n/c), Ind 1 (+1, n/c), Grn 0 (n/c, n/c)

Quebec (change vs. 2006, change vs. last projection)
BQ 50 (-1, +1), Lib 12 (+1, n/c), CPC 11 (+1, n/c), Ind 2 (+1, n/c), NDP 0 (n/c, -1)

Ontario (change vs. 2006, change vs. last projection)
CPC 47 (+7, +1), Lib 41 (-13, n/c), NDP 18 (+6, -1)

Saskatchewan and Manitoba (change vs. 2006, change vs. last projection)
CPC 21 (+1, n/c), NDP 5 (+2, n/c), Lib 2 (-3, n/c)

Alberta (change vs. 2006, change vs. last projection)
CPC 27 (-1, -1), NDP 1 (+1, +1)

British Columbia (change vs. 2006, change vs. last projection)
CPC 20 (+3, -4), NDP 10 (n/c, +1), Lib 6 (-3, +3)

North (change vs. 2006, change vs. last projection)
Lib 2 (n/c, n/c), NDP 1 (n/c, n/c)

A few remarks...

The final projection echoes the buzz that the NDP will pick up a seat in Alberta and the conventional wisdom that Elizabeth May will lose in Central Nova - however it shows both of these things happening by less than a percentage point. In other words, it says these things with no certainty whatsoever.

We also show a Conservatve collapse in Atlantic Canada offset by gains in Ontario and British Columbia, allowing them to essentially break even, while the NDP grows at the expense of Liberal totals.

I've posted a chart of which seats are projected for each party and the whole projection system for your interest and in the interests of accountability.

No matter who for, don't forget to vote tomorrow. I hope you enjoyed these projection and the musings of NBT and I over the past few weeks. Thanks for stopping by.

Today's chart:


NOTE: These projections do not reflect the wishes, nor predictions, of the author. They are based on simple math using the criteria outlined here and here.

NBT's 2008 Election Projections











National results (Seats, % Votes)
CPC 136 (35.7), Lib 87 (27.8), BQ 49 (10.8), NDP 34 (19.9) , Ind 2 (0.9), Grn 0 (4.8)

Atlantic Canada (Seats, % Votes)
Lib 22 (35.9) , CPC 6 (29.7), NDP 3 (18.3), Ind 1 (0.9), Grn 0 (5.6)

Quebec (Seats, % Votes)
BQ 49 (41.0), Lib 15 (27.5) , CPC 10 (21.1) , Ind 1 (1.2) , NDP 0 (13.6)

Ontario (Seats, % Votes)
CPC 49 (36.9) , Lib 40 (32.3), NDP 17 (19.5) Grn 0 (5.4)

Saskatchewan and Manitoba (Seats, % Votes)
CPC 25 (42.3), NDP 2 (21.5), Lib 1 (22.3) Grn 0 (4.1)

Alberta (Seats, % Votes)
CPC 28 (54.4), Lib 0 (20.5), NDP 0 (18.7) Grn 0 (3.9)

British Columbia (Seats, % Votes)
CPC 18 (41.8), NDP 11 (17.2), Lib 7 (28.8) Grn (6.7)

North (Seats, % Votes)
Lib 2 (38.5), NDP 1 (30.9) CPC 0 (21.6)

A few quick speculations...

There was much chatter around Jack Layton and whether or not he and his NDP party would replace the Liberals as Her Majesty's Loyal Opposition. Several polls for much of the campaign showed that Layton was one of the most popular party leaders in Canada ahead of Dion, May and Duceppe. Nonetheless, he will not be successful in translating this into electoral success, much like Broadbent in '88, since I believe the Liberals will reap most of the benefits from an impending Tory majority. In other words, a fair size chunk of last minute, undecided voters will jump over to the Liberals in order to stop Harper.

You could say that New Democrat Thomas Mulclair's by-election win in Outremont was anything but routine, but one thing that may surprise the dippers once again on this one is how short lived his victory will end up. Liberals regain their stronghold tonight and put an end to the NDP's dreams of forming a modest caucus in La belle province.

Liberals make it a clean sweep of Atlantic Canadian islands (Newfoundland, Cape Breton Island and Prince Edward Island). First time this has happened since confederation.

Nova Scotia and New Brunswick keep the status quo, with the exception of electing Bill Casey as an independent and electing one more Tory to the Atlantic NB caucus. Oh yeah, and another thing, for those that thought Pictou-Antigonish-Guysborough (now Central Nova) would be close, you are wrong. Peter in a cakewalk (1000+ vote margin).

Aside from Central Nova, here are 15 ridings to keep your eye on tonight. Anyway, for all those that dropped by Paint the Political Picture, nbpolitico and I can't say how much we appreciate all your comments, opinions and discussion.

Oh, and one last thing, I second this statement by Tim Powers: "... if you do one thing today, make it out to vote. I am just walking over to do it myself now. There is much at stake in this election. There are clear choices. Get some skin in the game and vote.A big congratulations to everyone from every party who put their name on the ballot. It is a courageous and commendable thing to do. Most of us have no clue about the sacrifices you have made to try to serve Canada. Good on you all. Politics matters and you have made the effort to make a difference." That said, have a great election night folks!!

Morer

Not only are Gerry and I close in policy thought, it would seem that we have the same predictions and gut feelings as well. So if we're wrong, we'll still end up being right.

Don't expect voter turnout to break any records after tonight's count comes in. As I see it, it will most likely decline from the improved 2006 levels. Probably around sixty per cent or so...maybe even lower. I mean, honestly, when there's that many people out there who aren't motivated by any of the current political leaders, it's bound to happen.