Wells asks;
…”anyone want to do the math on a swing of, say, six points from the Bloc to the Tories across Quebec and tell me how it changes the seat totals?”
Lance provides the answer..
Wells asks;
…”anyone want to do the math on a swing of, say, six points from the Bloc to the Tories across Quebec and tell me how it changes the seat totals?”
Lance provides the answer..
de yawn wit’ de backpack? hokiest thing I”ve seen since sixth grade. losing looks very good on the LPC and I hope they keep it up.
Although this is an interesting way to play with the numbers, a bi-election is never indicative of what will take place during a general election. I think the CPC has a chance to do very well in Quebec, but I’d be happy enough if they’d hold the line where they are and do a whole lot better in Ontario to be honest.
All of this number crunching does highlight why it is that the CPC has been moving left since the day they took the reigns of power. They have nowhere to go in Alberta, they own it. They have all the rural areas in Ontario they are likely to win, and they have a handful of ridings in Quebec. If they want a majority government they’re going to have to play to the urban fiscal conservative types that might utter the words global warming once in a while. Honestly, every time I hear Stephen Harper say climate change I just about wretch, but I make no difference to the equation anyhow, since I’m out of the country and not able to vote.
Of course the real reason that the libs lost the buy-elections is because they ran out of adscam money.
The calculations explain why the CP doesn’t want to go to the polls any time soon. Even with the tectonic shift of 18% they would only pick up 15 or so seats and that is nowhere near enough to comfortably achieve a majority.
With all due respect to Lance, I believe he’s done his math wrong. He appears to have interpreted a 15% swing as “subtract 15% of the Bloc vote and add 15% of the Conservative vote”.
But this isn’t the actual swing. The Bloc didn’t just lose 15% or 18% of their vote. They consistently lost about that percentage of total votes, and in terms of their own number of votes in 2006, they lost more than half of their supporters in each riding. In Outremont, in fact, the Bloc had their total vote reduced by about 80%.
Take Gaspesie–Iles-de-la-Madeleine for example, which apparently turns Tory with a 15% swing. Lance has essentially multiplied the Bloc vote by 0.85 and the Tory vote by 1.15, when he should have simply subtracted 15% from the Bloc percentage and added 15% to the Tory percentage. With equivalent voter turnout, this gives CPC a very comfortable margin of 8000 votes or so.
It turns out that Gaspesie is much close than a 15% swing. In the 2006 election, the Bloc had 42.7% of the vote, and the Tories 32.2%. A 6% pure swing from the Bloc to the Tories, therefore, would be enough to hand the Tories the riding.
I haven’t had the time to go through each riding yet, but it should be readily apparent from the most casual observation that Lance has significantly understated how much trouble the Bloc really is in.
Codswallops! You’re right Eugene!
I’ll just go fix that now. 🙂
Cheers,
lance
when all is said and done the libs ain’t got a pot to piss in. they’re broke with a really loseeeeeer leader
Right, I’ve corrected it now. Thanks Eugene.
Swings are:
– 6% off of the Bloc and added to the CPC. A gain of 5 seats.
– 10% off the Bloc and added to the CPC. A gain of 8 seats.
– 15%, a gain of 12.
– If they lose 18% as they did on Monday? A gain of 18 seats.
Cheers,
lance
Blogs are great . . . instant checking of data. Now if only NASA believed in that. 🙂
Cheers,
lance
Darn it, I guffed it again. Forgot to take the Lib wins into account on the correction. Maybe I should just go to bed. 🙂
Lance, sorry to make you process all that ugly data, but … you’re still understating the situation, actually. You don’t subtract 6% of the Bloc vote from their numbers; instead you should subtract 6% of the total vote in the riding from the Bloc vote (and then add it to the Tories, if you want a pure BQ->CPC swing).
Swing applies to total vote, not each party’s vote. As I was saying above, the Bloc didn’t lose 15% of their own votes — they lost 15% of the total percentage and more like about 50% of their own votes.
Sample calculation: in Saint-Maurice–Champlain, 48561 total ballots were cast (BQ 21532 CPC 16028). A 6% swing would send 2914 votes (6% of the total ballots) from the BQ to the CPC, so the new numbers would be CPC 18942 BQ 18618. In other words, a 6% swing should hand Saint-Maurice–Champlain to the CPC.
– 6% no gain,
– 10% 2 seats,
– 15%, a gain of 5,
– 18%, a gain of 11 seats.
Double checked this time.
Cheers,
lance
Oh. Hell.
Not that big a deal Eugene. It’s just sql. I have to change a view so that I can sum the total vote numbers for each poll and then riding. That’ll take the longest time to ensure its right. There’s 95k rows in Quebec alone.
Cheers,
lance
lance, which data set are you using? Table 11 gives candidates and votes/%s by riding, without having to sort through every individual poll.
Myself, I’m trying something in Java, currently with limited success.
I used the entire raw-data zip. I wanted to put it all in a db (postgres) for a project and this seemed like a good enough reason to get around to it.
It was a little bit of a pain. 🙂
I renamed all the CSV’s to have a provincial hint, cat’ed the all of each provinces csv’s into one file, tossed all the headers, converted all the accents, and then copied it in to a schema based on the headers. Thank god for emacs.
I wrote a query to total up the votes from each poll for the 4 or 5 major parties for each riding and dump the results into another table, one riding per row.
FWIW, I’ve updated the page now. I used both sets. The 6% off of Bloc onto CPC and 6% of total off Bloc and onto CPC. Yes, significant difference.
Second set:
6% – 15 seats
10% – 24 seats
15% – 37 seats
18% – 50 seats
If you’ve got postgres, I can send you a tar.bz2 dump if you’re interested, it’s only 1.7M (compressed from 40M)
Cheers,
lance
Gord,
15 more seats in Quebece alone.
Historically the Libs have benefitted in the rest of Canada for being the only real “federalist option” in Quebec. Such a collapse would most certainly translate into votes elsewhere.
Also, Dion bombed for non Quebec specific reasons as well. Dion is a dud, with very little voter appeal.
Finally, the Libs have little money.
The CPC doesn’t in any way fear a fall election.
Biff, try 40 more seats. Lance’s final iteration is the correct one. I guestimated 30 more in conversation with a friend yesterday. CPC makes significant inroads in to tradiional bleu territory, Toronto can go get stuffed. CPC majority. Is it bad form to call them truth and reconciliation commissions, since it has been used?
too many numbers guys. bottom line, bloc down, cpc up. an improvement.
like biff said the liberals got no money harper should engineer a fall election and bankrupt the leftards, worse case scenario would be another minority gov for the cons.
Juggling balloting stats is a fun hobby…but until we get the new liberal election system where winners lose and losers win well just have to put our vote with the best horse and see how the race goes.
>> “The calculations explain why the CP doesn’t want to go to the polls any time soon. Even with the tectonic shift of 18% they would only pick up 15 or so seats and that is nowhere near enough to comfortably achieve a majority.”
Well, on the plus side for the CPC, if there was such a swing in Quebec, you can rest assured that Ontario or enough of it would be swinging that way as well. An 18% swing – highly unlikely in a general election, but I suspect indicative of the direction the Bloc is going – is huge, means something has happened. If the Bloc is going down, and the CPC going up, there will be a momentum swing from Ontario undecideds and those who like to “back a winner”.
On the down side, and the flaw in this analysis of course, is assuming that all of that swing goes to the CPC. If the Bloc vote collapses during an election, then even a small Lib/NDP shaving off of that buffers the Bloc from a seat collapse.
But time can do wonders.
One anecdote: during the 1997 election, Chretien sent Dion around the province to fight for the Liberals and federalism. He refused to campaign in Jean Charest’s riding saying that he was not about to fight a federalist, and certainly not the federalist leader of another party in a close election. A swing of votes to the Liberals there could have given the Bloc the seat. All of which to say only that I’d rather have a few more Conservative seats than Bloc seats.
sounds like ted could be coming to his senses lol
Lance,
Your revisions are close to my calculations as well. I got similar results based on the various swings.
An interesting point is no con majority unless you get that significant swing. Cons nees 31 seats more. They will get between 5-10 more out of Ontario, unless there is a complete Liberal collapse or NDP supersurge, they may pick up a couple in BC and lose a couple down east, a wash, so they need 20 to 26 seats out of Quebec, inclremental.
It is going to be difficult. But a Bloc collapse would make it 1984 all over again.
I still say the unqunatifiable factor is the absolute disarray the Liberals are in. When their org meets the campaign it will fall apart within a week. That on its own gets you the 10 seats in Ontario, and if it is as bad as that add 5 more, from within the 416!
Turners campaign in 1984 was bad, Martins in 2006 was bad, Campbells was bad…..Dion’s will start as bad as Turners and Martins and end as bad as Campbells (she started out well)
It will be hard to watch
The other factor to watch out for, which hasn’t really been mentioned yet, is that Lib->NDP swing that happened in Outremont. Across the province, 10% or so combined with a substantial BQ->CPC shift, would gain another 5 to 10 extra Tory seats in the province. Simulating a swing of 20.2% BQ->CPC (as occurred in Roberval) and 18.3% LPC->NDP (as occurred in Outremont), while not being entirely realistic, is quite amazing and rather revealing too.
Stats are stats, particularly when you’re doing byelections. And names are just names. Not that I’m knocking what lance is doing – and thanks for doing it – by I’m more interested in two things: Are the Conservatives becoming the Patronage Party (or are the Liberals at least losing the ability to claim that title?) and are the Conservatives becoming the “Federalist” party (“Whatever that means” – Charley husband of Di.)
If that’s happening, you’re going to see Conservatives picking up seats not only in Quebec but in the Maritimes. Anyone who thinks Cape Breton votes for Brison because he has a cute butt rather than because he can (could) deliver the pork seriously underestimates the common sense of the common Maritimer.
Anyone got any polls on who’s becoming the Patronage Party? Or do the MSM try to avoid asking that question when the answers are inconvenient?
Interesting numbers, though I too would echo the sentiment that its pie-in-the-sky thinking that all the BQ vote would swing to the Cons. It would be interesting to see what your calculations would look like if you separated them out based on urban/rural ridings with the assumption that the BQ vote swings to CPC in the rural and to the NDP (as happened in Outremont) in the urban…..
So it’s 50 not 15…
That seems more in-line with what I felt would be the case, but lance’s first calculations and Kate’s stamp of approval in posting it caused me to take lance’s numbers as written.
That said, first, an 18% shift is a shakey number that could be reversed very quickly should the CP do something to offend franco-quebecer sensibilities. Second, a 10% figure has a much higher dgree of certainty and, if lance’s numbers are correct, we are talking 24 seats – Still too small a shift to give the CP 4 solid years to unveil the secret agenda in all of its paradigm-shifting glory.
Gord, there isn’t a hope in h e double l that there’s going to be an 18% total vote swing from the Bloc to the CPC. It _isn’t_ going to happen.
If anything, I’m thinking the first set of numbers is probably more correct because if 18% total vote does abandon the Bloc, then it’s going to split up between the other parties. Probably not 6/6/6 though. Sean has it correct. Maybe tonight I’ll do the urban NDP/rural CPC split based on 18% total vote and see what we get.
It was all just fun with numbers. You guys do that, right? Run what-if’s in the spreadsheet?
What? No?
Oh.
Cheers,
lance
It would be great if you could do that Lance I am interested to see what you get, and who doesn’t play around ‘what-ifs’ in spreadsheets?
The NDP will keep Ottawa honest. They at least speak out,no one else does.
Yes lance, I do run spreadsheets and in fact ran one (actually, several) on the last election with an emphasis on how many votes the CP needed to get a majority.
Your work confirms the dilemma that the CP has – that the best hunting ground for seats for the CP is indeed QC, unfortunately their aren’t enough to be had to get a majority and thus the need to tread a very balanced policy line so as to not alienate the ROC base too much.
I beg to differ with you on the possibility of there being a collapse of bloc vote to a separatist core (i.e. a loss of 18+ percent) – it could happen along the same lines as how Brian Mulroney achieved his que landslide in ’84.
And I also differ with you on the chances that the lost BQ support would split amoung all of the parties. The section of support that would fall off is essentially soft nationalist in character and thus the hard-line centralized canadian vision of both the NDP and the dion-led LP (but probably not an Iggy-led LP, BTW) is anathema to them. So, just as we witnessed on Monday, in a general election any support that the BQ lost would almost entirely move to the CP camp.
Having expressed those differences I want to reiterate that the chances of a massive abandonment of the BQ for the CP currently less than 50%. What we will soon see is a major PR/Policy campaign by the CP trumpeting the legal prevention of federal interference (i.e. spending money and overtaxing) in areas of provincial juristiction. This will win new and firm-up lots of support for the CP in both QC and AB and BC. Perhaps then ‘winning conditions’ will exist and the CP will engineer an election to capitalize on them.
“The NDP will keep Ottawa honest. They at least speak out,no one else does.”
Like the bunch of leftists with Tourettes that they are…
Those on the left have no monopoly on honesty – they are just as vain and corrupt as any other group.
Maybe the CPC could clawback more equalization from NL, NS and SK and spend it in Quebec on daycare or some other frivolous vote buying scheme. They got their Quebec bump in the polls that they were looking for from the last round of bribes. Remember, when it comes to the Quebecois Nation it’s all about “I know what you did last spring, but what have you done for me lately?”
As for the lost seats in Atlantic Canada being a wash with the ones the CPC will pick up in BC, dream on. The CPC isn’t endearing itself to BC’ers, they are feeling neglected as well. There will only be 2 Conservatives elected down east, Bill Casey (x-con) and Peter McKay. Remember, ABC.
For the cost of 3 Billion over 12 years, a pittance in today’s dollars, the CPC could have honoured AN ELECTION PROMISE and invested in an economic development plan that would have secured 25 out of 32 seats for generations to come.
If the CPC won every other seat outside of Atlantic Canada as they did in the last election they get 125 seats. Out of the 9 they won in AC in 2006, they might get 3 back, which leaves the CPC with 128 seats and needing 27 to win majority (155 seats).
If they win every other seat outside of AC, and had they HONOURED THEIR PROMISE, they could have easily, yes easily, have won 29 out of 32 seats bringing the CPC total to 154 seats, meaning they would have only needed 1 more seat from Manitoba, Quebec, Ontario, Sask., BC, NU, NWT and the Yukon, all things being equal.
Why don’t you run those numbers through your spreadsheets Lance and tell us what you find?
I bet it’s much easier to call me a welfare collecting inbred, right Lance? Or you could tell me to get off my @ss and get a job. Try to be creative this time.
Glenn said, “I bet it’s much easier to call me a welfare collecting inbred, right Lance? Or you could tell me to get off my @ss and get a job. Try to be creative this time.”
Did I say that, I’m must have had reason. Are you must be a Wpg Blue Bomber fan?
Cheers,
lance
4 billion to buy 1 seat. Money well spent Mr.Harper.