Double Blind

I have been waking up every day for the last three days with 50 mails in my inbox saying there are comments available for moderation. I would love for this to happen, except for the fact that they’ve been 100% spam.

On the advice of Richard, I’ve decided to give Akismet, an automatic spam filter, a shot. We’ll see if it is effective, otherwise I might resort to using some sort of Bayesian filtration or even the dreaded “enter in the characters you see in this picture” form of self-moderation.

I have also been studying the spam that has been coming in. I’m guessing that a good majority of the spam is now designed to ‘nudge’ the filters into believing that future spam is legitimate. In other words, they are trying to pollute the filters with incorrect information. The cat and mouse game continues…

The 30-Year US treasury bond market, unlike the currency, precious metals, or index futures markets, behave in semi-regular ways that I can’t quite describe in print. This means that the markets have inefficiencies that can be exploited for profit. Whenever the market gets into certain types of “moods” (I use this word for the lack of any better term), I place trades to take advantage of short-term probabilistic movements in the bond market. This was my second bond trade of the year.

One full point on the chart (e.g. from 112 to 113) is US$1000 of movement. Realized profit is US$371.54 after commissions (US$1.73 per side). If anybody cares, in Canada these are taxed as capital gains, per the CRA Interpretation Bulletin IT-346R.

I always place my orders before I go to sleep – when I wake up I see the outcome of the trade. It is a system that has worked very well for me, as it completely takes the emotion away of looking at a chart. Either you wake up with a profit, a loss or you find your entry point was not reached. Whenever I look at the price quotations in real-time I tend to make very stupid decisions and thus I never trade in real-time.

In this trade, as the price target was not reached, I pulled out at the end of the day as there is no reason to hold the position – the trade was finished.

Using a ‘percentage return’ metric is irrelevant when it comes to trading futures. For example, had I traded two futures instead of one, I would have made double the return. However, the more leverage you apply to trades, the more dangerous trading becomes since you start running into the risk of ruin. That aside, if I can keep a 60% win rate, with each win providing a 3:2 gain over loss ratio, it amounts to an expectation of about US$250 per future per trade, providing I’m alert enough to discover these situations in the bond market when they occur.

This sort of thing is not for the light-hearted, however! You have been warned.

I entered into Calgary Grit’s election pool at the beginning of the election. My answers were here, while the results were here. While I placed well behind first place (tied for 10th with a score of 43 points while the leader had an amazing 70 out of 100 possible points), I managed to beat the only other guy I knew on the list, Declan, by a whole one point! Take that! :)

Thanks to the Calgary Grit for hosting the contest and I will be better prepared next time!

This is a cross-post with my former municipal election campaign site, SachaPeter.ca:

I was quoted in the Richmond Review by Matt Hoekstra. At around 7:00pm (just as the polls closed in BC) I headed over to the Conservative post-election celebration room in the Best Western Richmond hotel. This was the first event ever that had a gigantic platter of smoked salmon and capers (in addition to the other goodies, like sandwiches, cheese and crackers, spring rolls, etc.). The salmon was great and as usual, I stuffed myself silly as I usually do at these events.

A lot of the scruteneers started to trickle in around 8:30-9:00pm, but before then the room was sparsely populated. During that time, Eve Edmonds (of the News) and Matt (of the Review) questioned me about the results in general, and also what went right and wrong with the campaign. There was also a television crew from Fairchild TV that gave me a 3 minute interview and judging from his questions he was rather obsessed whether Reid’s social conservative history (concerning abortion and gay marriage) cost him the election in addition to what this tells about the people in Richmond. This was around the time that enough returns came in that it was quite apparent that Raymond Chan was going to win the seat.

I don’t know if I made it on television, but Matt Hoekstra did quote me in his filing to the Richmond Review, published on January 24, 2006:

Sacha Peter, a Conservative who ran for a seat on the school board in the last election, said some things worked in Reid’s campaign, others didn’t.

A fumble came when the campaign denied interview requests from some media outlets, he said.

Not only does that leave a bad impression on constituents, it also can hand the media control of the agenda. In Reid’s case, it was social conservative issues that the media dredged up.

“Certainly, when you’re not controlling the agenda, you’re not controlling the campaign.”

But Reid did a good job getting out in the community and conveying his thoughts in regular advertorials in the newspaper.

“He did a good job at getting out there. He is a very good communicator and he is intelligent,” he said. “The candidate spending the time in front of actual bona-fide voters as opposed to partisan hacks is a very good use of time.”

Sacha said Reid might have benefited from going on the attack as the Liberals did. He said negative campaigning can work, and the Liberals’ sponsorship scandal and questions over Chan’s promotion of his overseas schools would have been good attacks.

“The Conservatives are scary mantra is really frightening a lot of people. It’s been quite effective.”

I will elaborate on these statements and will say with having no connection to the strategists or impact on the campaign strategy. The only involvement I had in the campaign was compiling together a list of emails received by the campaign in response to various local issues as a courtesy to one of the members, more deeply connected to the party, that asked me. I was originally going to do absolutely nothing with respect to this election since I was on post-election detox from the November 2005 school trustee election, but I agreed anyway as the task wasn’t too onerous for a computer-literate person like me to perform.

I remember watching Global TV where they were doing riding profiles and interviewing all the candidates in the riding. With Richmond, all the three other candidates were interviewed, except when it came time for Reid, the reporter said something to the tune of “Despite the fact we attempted to contact their office multiple times over multiple days, he could not be reached for an interview because they claimed he was busy working with constituents.” In replacement of what should have been a nice 30 second clip of him talking about what he could do for Richmond (like they did for all the other candidates), Global brought up some speech he did in 1999 about talking against abortion and the reporter mentioning his past with Focus on the Family.

This looks really bad, but it could have been an isolated incident. Maybe he was truly busy shaking hands with voters and getting out there. But in subsequent interviews, he was unavailable. He didn’t have a quote in the Vancouver Sun’s riding-by-riding brief before election day. He didn’t have a quote in the Vancouver Province. It just said that he was “unavailable to be reached.” Now it doesn’t matter the reason, but for anybody paying attention they should realize that something is not right. If he’s this unavailable from the media, what makes people think he will be available for him? This was a huge mistake which probably cost him at least 1000-1500 votes in the campaign, possibly enough to cover the 1800 votes he lost the seat by.

I was speculating off the record with both Matt and Eve what could possibly justify such an apparently sacrificial strategy. Little did I know that Eve already had already ’scooped’ the answer in the Richmond News’ January 24, 2006 publication:

… Reid’s campaign manager, Robbie Robertson, blamed a Jewish-owned, liberal biased media for some of the challenges of the campaign.

“The CanWest Global media empire is controlled by a Jewish family (the Aspers) and they have been the most aggressive family to attack Christians, especially Conservative Christians,” Robertson said.

The Richmond News, the Vancouver Sun and Province, the National Post and BCTV News on Global are owned by CanWest.

This would have been the last reason I would have thought that his campaign had for refusing interviews. I was floored when I read this. At least Darrel managed to disown him at the last second and say that Robertson doesn’t speak for the campaign, but I suspect that this will sabotage any chance remaining for Reid to obtain a nomination in a future election in Richmond. The fact that Reid didn’t use the excuse of “he didn’t tell me about any of these interview requests” shows that he implicitly agreed with the decision to stay away from the media.

Certainly if Robertson spoke up earlier about this, it would have cost Reid another 2500 votes in the election. It’s patently obvious that Reid made a huge mistake hiring Robertson as a campaign manager. That type of behaviour completely justifies the “scary” logo that the Liberals have been trying to attach to the Conservatives.

Instead, Reid should have not been so defensive about his past and be publicly proud of what he stands for. It will then cease to be a campaign issue that will dog him throughout the campaign. I personally don’t share many of the views of the pro-life lobby, but I don’t shy away from talking about issues where I know I am in the minority with (e.g. I don’t think that low voter turnout is particularly bad). As long as you keep your chin up, and be able to justify your beliefs with something morally consistent and not off in the radical fringe, I will respect your opinion and not think anything else of it. I think Reid’s handlers bought into the conspiracy theory that all media were out to get him. The correct strategy is to state what you believe in, and mention it early in the campaign to get it over with. Harper mentioned the gay marriage/free vote issue on day 1 of the election and managed to kill that issue right there. Reid should have done the same thing and do it with his chin up. By avoiding the dialogue, Reid was not setting the agenda for the campaign and justifiably lost on that front when the media did the work for the Chan campaign.

I have talked to Darrel on two occasions and he is articulate and intelligent (other than apparently choosing which people to handle his campaign). Unfortunately I think he got caught up with “Paul Martin syndrome” and was campaigning based on what he (or his campaign manager) wanted his public image to be versus what the person he actually is.

Richmond was a winnable riding for the Conservatives, but the Conservatives fell on their swords yet again.

There are ramifications with respect to the results in this election.

First, the public didn’t want to give absolute control to the Conservatives. They will have a relatively free reign to pass moderate legislation through this year since nobody wants an election in 2006. Whether the Conservatives will try abusing this in a parliamentary game of ‘chicken’ or not remains to be seen, but I would consider it to be a strategic error to introduce anything too radical. Instead, their goal should be to learn the ropes of operating a government and introducing rules to make sure that government corruption is limited in the future, specifically with contracts to “consulting” firms. As long as the Conservatives can be a good caretaker in the next couple years (including not brining the fiscal balance into deficit which they stand a good chance of doing since the Liberals have squandered the surplus over the past 5 years), the public should be able to reward them with another minority. I think Steven Harper will keep a tight lid on things and shake the “scary” brush that the Liberals have been tar-and-feathering him with for the past couple years.

The Conservatives, in order to win the next election, have to appeal to urban voters in Toronto, Montreal and Vancouver (in that order) that they can run a country and are not going to ram a social agenda on the public even when given a minority. They have to do this in such a fashion that they will not alienate their supporters in the parts of the country that voted for them, although they have a lot of political capital to spend since their victories tended to be overwhelming in places other than Ontario. Unfortunately for us in BC, this means that the Conservatives are going to have to appeal to central Canada and Quebec, which will cause some more resentment in this province and Alberta. The Conservative goal should be to aim for an election in the fall of 2007.

The Liberals didn’t do as badly as some were predicting (including myself), and in order for them to win the next election, they have to make the Conservatives look as incompetent as possible in government. The first few months for the Conservatives will be brutal (look for some entertaining question periods), but after that it will be progressively more difficult for the Liberals to pick up dirt. Compounding the Liberals’ problems is that the party is still internally divided between Martin and Chretien supporters and they need to pick a leader that can unify the party and get to battle in an election quickly. The longer the Conservatives stay in power, the greater in trouble the Liberals are. The Liberals also have a massive debt to pay off from their election expenses and are constrained with time and resources. Ideally they would want to force the government down on the spring budget vote of 2007 if I see things correctly.

The big loser in this election was the Bloc. They were looking forward to this election and making boastful claims that they were aiming for 50% of the popular vote in the province. They got 42%. The Conservatives foiled their plans, something the rest of the country didn’t see and the Bloc will be prepared for them the next time. This also means that the Bloc would be unlikely to entertain some sort of informal coalition with the Conservatives, lest they be giving up their opportunity for more seats in the following election. The Bloc’s correct strategy is to continue gunning after the Liberals as there isn’t much they can do about the Conservatives without helping them. I’m not sure how Duceppe is going to worm his way out of this one other than taking a “wait and see” approach with the Conservatives.

The Bloc also cannot ally with the Liberals and their combined seat count is 154. They could, in theory, team up with the independent MP in Quebec to form a majority, but this is an unlikely coalition. In the future, it might be enough to bring down the government.

The NDP won 10 more seats this election than the previous one, but this gets them no closer to holding the “balance of power” in the House of Commons. The Conservatives have 124 and the NDP have 29, which is a combined 153 seats. They need 155 to pass legislation. Although the Conservatives can listen to the NDP (and it would be good press for them if they could co-operate with the NDP and get some joint resolutions passed), the NDP really can’t wield as much power as they could in the previous parliament because they are just 2 seats short. The Conservatives will still need the consent of either the Bloc or Liberals.

So what I see in the future is a very carefully guided parliament with little in terms of anything radical. Longer term, the parties will be jockeying for an election in 2007 (whether the spring or fall), but this will pale in comparison for something potentially a little more important for the country: Quebec Referendum 2009.

The Canadian election is over. I doubt there will be another election until at least the fall of 2007 (the next focus of my electoral research will be on select states in the 2006 USA midterm elections for the US Senate and Governorships). In terms of how I predicted, I am quite happy with my projection models. There were significant deficiencies in my models in terms of how the candidates performed in Quebec (I obviously, along with the rest of the country, thought the Bloc would do better at the expense of the actual Conservative performance). I also thought the Liberals would fare worse in British Columbia and Quebec.

That said, my Conservative seat projection (123) was very close to the actual result (124), while the Liberal result was off by 11 (103 actual vs. 92 predicted). The BQ was off by 9 and the NDP was off by 3. My provincial predictions were pretty much correct except for BC (where the Liberals got more seats where I had previously predicted NDP/Conservative pickups) and Quebec (where the Conservatives surprised). I nailed Ontario and the rest of the country, plus or minus a few seats (Reg Alcock losing by 100 votes in Winnipeg was certainly not on my radar).

The prediction puts me ranked 4th in terms of the pundits out there (summarized on the UBC ESM webpage) as my absolute error in seat count was 24. Milton Chan of Election Prediction did one hell of a job calling this election with an absolute error of 12 seats – I’m sure we are all looking at the Quebec numbers and asking ourselves how we could have missed it. I saw the 32% polling numbers outside of Montreal and just assumed that these would be sucked up from the Liberal and Bloc candidates, but the Bloc candidates would still be getting narrow victories in most of the seats, so I just credited the Conservatives with two extra seats. Not so.

In terms of my provincial predictions in BC, I got 28 out of 36 seats correct, which was a woefully inadequate performance. Considering I live in this province and also that I consider my projections of the 2005 provincial election a failure, I’ll have to re-examine my numbers and hopefully come with a more accurate projection in the next election.

In terms of the money game, I am very happy with my performance. I don’t think this could have gone much better than it has. I recall that six months ago, when people were openly questioning Steven Harper’s leadership of the Conservative party that he was trading at about 10% on Tradesports to get more seats than the Liberals. When the election was called, I thought the Conservatives were going to be running a campaign strictly on Gomery and corruption – they didn’t. They had been spending the last six months preparing and researching a campaign and absolutely slaughtered Martin and the Liberals, who spent most of their ammunition trying to keep their government alive.

When the Conservatives were trading at around 30% in mid-December, I knew that they were fighting the right campaign – on policy. That’s when I starting buying them in volume. This was during the time when the Conservatives were still at 30% popularity in the polls and media pundits were claiming they couldn’t break out of the 30% threshold. Little did they know that the strategic focus of the campaign was to present a policy-based election, to give something for Canadians to vote for. When the Conservatives were trading at 50% after the RCMP investigation on income trusts broke out during Christmas time, I knew it was over and placed a second and larger stake. The rest, they say, was history.

It’s too bad election trading doesn’t scale up by a factor of 100 as I could probably make a career out of it assuming the markets were as inefficient as they were this time around. I don’t think they will be nearly as inefficient the next time, which means I have to tighten the screws on my modelling and be that much more prepared than the rest of the market.

The Liberals are going to be spending some time in opposition and they will be credible, considering a lot of their MPs know the ins and outs of government operations which they were running just months earlier. They’ll be able to ask pointed questions on material weaknesses of government departments they left over the past 12 years and will try to embarrass the Conservatives. The Conservatives need to climb up the learning curve on operating a government very quickly before their time expires in this minority government. It will be an interesting game to watch.

This election has been fairly textbook, and mostly to my expectations. I will analyze the results compared to my projections in a future message.

Tradesports:
Conservative Victory (won): +US$662.86
NDP getting more than 22 seats (won): +US$296.34
Bloc getting more than 58 seats (lost): -US$25.08
Total: US$934.12

UBC ESM:
Minority government (won): +CAD$376

Total gain on election: CAD$1462, assuming a 0.86 exchange ratio.

Here are my final seat projections for the 2006 Canadian Election, province-by-province, with comments:

British Columbia – Conservative 19, NDP 14, Liberal 3

Previously discussed – the NDP are the left wing alternative this election. You can thank Carole James and Jack Layton for this, as BC traditionally hates the NDP Federally when they’re in provincial government. Right now the BC Liberals (a party that typically has a membership consisting of federal Liberals and Conservatives) are in control, so this will bode well for the NDP.

Alberta – Conservative 28

The only real competition is in Edmonton Center, where Anne McLellan is toast.

Saskatchewan – Conservative 12, Liberal 1, NDP 1

Nothing to report here.

Manitoba – Conservative 6, Liberal 4, NDP 3

Churchill’s NDP support is being split which highly suggests a Liberal victory.

Ontario – Liberal 53, Conservative 43, NDP 10

You have to take a look at the different areas of the province. The Greater Toronto region is divided into two area codes – 416 and 905. The 416 area code is closer to the city core, while the 905 area code is on the outskirts of the 416 area. The 905 area outside Toronto is heavily Liberal, while the 905 area not near Toronto is either NDP or Conservative leaning. The 416 area is heavily Liberal with a sprinkling of NDP support (thanks to Jack Layton and Olivia Chow, who will both win their seats). The eastern part of the province (dominated by Ottawa) is Conservative leaning, with moderate Liberal and NDP support, the latter two concentrated in the Ottawa metro area. The Northern part of the province is Liberal and NDP supporting, while the southwest is Conservative leaning. When you weight the seat distribution (there is a heavy weight on the urban areas) and account for “special seats” that have deviant expectations, you end up with the result above.

Quebec – Bloc 60, Liberal 10, Conservatives 4, Independent 1

You can divide Quebec into two areas – Montreal and the rest of the province. The Liberals and Conservatives have roughly equal support in Montreal, which spells big trouble for the Liberals. Outside Montreal, the Conservatives have been polling respectably well which should give them a few seats to work with. The lone independent is Andre Arthur, in the riding of Portneuf-Jacques-Cartier. In Montreal, the Liberal support, where they have it, is very concentrated and this will translate into a few seats. Outside of Montreal, they are absolutely toast except in Hull, where this riding is virtually an extension of Ottawa. Despite the fact that the Bloc’s popular support is going to be approximately the same as the last election, vote splitting is going to give them more seats this time around.

Atlantic Canada – Liberal 18, Conservative 11, NDP 3

Nothing is changing here from the previous election except for a single seat in Newfoundland and a couple seats in New Brunswick.

Territories – Liberal 2, NDP 1

The Northwest Territories should go NDP, although there may be surprises here simply because of the lower number of voters and historical volatility.

Final projection:
Conservatives 123
Liberals 92
Bloc Quebecois 60
NDP 32
Independent 1

Traditionally, when a party gets about 40% of the popular vote they’re considered to be in the running for a majority government. The only problem with this assumption is that the last three elections have had the Bloc Quebecois, which make it difficult to obtain a majority without an overwhelming majority of seats in the rest of the country. The Conservatives have the bulk of their support in Western Canada, the Liberals have their support in Toronto, Montreal and Vancouver, the NDP have their support in select areas of urban and rural areas of the country with certain characteristics (union influences), while the Bloc obviously have their support in Quebec.

Because of this regionalism, when the parties win their seats, they tend to win them with overwhelming majorities. Look at your average Alberta riding – for example, the riding of Medicine Hat (the home of soon-to-be finance minister Monte Solberg) is going to vote 75-80% Conservative, which is some 40% more than he actually needs to get elected. The Liberal riding of Etobicoke North (in metro Toronto) is going to be won with way more than 50% of the vote.

Because there are no major vote splitting mechanisms (the right-wing vote is consolidated in one major party, of which this did not exist when Jean Chretien was in power) and with the Bloc Quebecois continually taking nearly 50% of the popular vote in Quebec means that in order for a party to form a majority government, they will need closer to 45% of the popular vote in the rest of the country in order to form a majority government assuming that they continue to achieve massive victories in the seats in their ‘core’ region.

The other question I will already be asking is – When will the next election be? My guess will be the fall of 2007.

As you may know, I like to discuss the results of elections. What separates me from most pundits is that I stick money where my mouth is. Some people considering it gambling, but I consider it informed speculation – my track record in the past is pretty good in terms of making money from these particular events.

Here is my existing risk profile coming into the Federal Election:

Betting profile on Tradesports:

Conservatives to get more seats than the Liberals. If they do, then I’m up US$667. If not, then I’m down US$1123. I made these bets in three batches – the first batch was when the Conservatives were favoured 30%, the second batch when it was 50%, and the final batch was at around 70%. I’ve put in some “risk-free capital” at higher odds which represents the higher average existing odds I have on the contracts (63%).

NDP to get more than 22.5 seats. If they do, I’m up US$298. If not, then I’m down US$362. I made this bet approximately in mid-December.

Bloc to get more than 57.5 seats. If they do, I’m up US$29. If not, then I’m down US$21. I made this trade today at 42% odds. I originally made a stronger bet on the Bloc earlier in the election (a $250 bet at approximately 51%), but I reversed it for a $30 loss a couple weeks after when things were looking more uncertain. I still think it will be really close for the Bloc – there will be a lot of vote splitting between the Liberals and the Conservatives which should get the Bloc some seats due to vote splitting.

Betting profile on the UBC Election Stock Market:

A minority government will be formed. If this happens, I’m up $376. If not, I’m down $1000. I liquidated all of my seats and popular vote contracts, so I’m hoping to make a 38% return on the ESM this election.

A few scenarios (assuming US exchange rate is 0.86):
1. Conservative majority = $224 loss.
2. Conservative minority = $1152 gain.
3. Liberal minority = $930 loss.
4. Liberal majority = $2306 loss.

Factoring in the NDP and Bloc contracts, the best-case and worst-case scenarios are:
1. Best case: Conservative minority, NDP > 22 seats, Bloc > 57 seats: $1532 gain
2. Worst case: Liberal majority, NDP < 23 seats, Bloc < 58 seats: $2751 loss. (I’m sure some of you out there are cheering for this outcome!)

We’ll see if the accumulated research works out. I’ve kind of hedged myself in the event of a Conservative majority government and I don’t think the Liberals have much of a chance, so I think January 23rd will be quite profitable for me.

Declan Dunne over at Crawl Across the Ocean went through the effort of predicting 308 ridings. This includes predicting 21 Conservatives, 11 NDP and 4 Liberals in BC. Good luck tomorrow!