Moff-Thoughts

Predictions vs. Reality in London

Here are the results in London vs. my predictions. Short story: I missed how much the GPC vote would drop.

London-Fanshawe: Actual (Prediction) - Difference

NDP: 21 689 (20 000)  +1 689

CPC: 14 300 (12 000) + 2 300

LPC: 4 893 (6 000) -1 107

GPC 1 201 (3 500) - 2 299

CHP 535 (500) +35

I thought I was being cynical when I had the LPC all the way down to 6000.  I guess not.  And yeah, I really blew it with the GPC prediction.

London North Centre: Actual (Prediction) - Difference

CPC: 19 484 (17 500) +1 984

LPC: 17 765 (18 750) -985

NDP: 13 007 (13 750) -743

GPC: 2 235 (5 000) -2 765

AA: 236 (200) +36

I saw the NDP eating the LPC's lunch, and that's exactly what happened.  I didn't see the CPC increasing their vote as much as they did.  I gave the CPC a 30% chance of winning - I never saw them winning by this much. I also thought the turnout would be a fair bit higher than it was.

London West: Actual (Prediction) - Difference

CPC: 27 673 (23 000) +4 673

LPC: 16 652 (19 000) - 2 348

NDP: 16 109 (13 800) +2 309

GPC: 1 688 (4 000) -2 312

UP: 66 (200) -134

Yeah, I blew this one.  Badly. I knew Holder was well-liked in London but I underestimated how well liked he is. I thought the turnout would be up, but not this much! There has been a lot of construction in the northwest part of the riding, so maybe that has something to do with it.

Comparing London Predictions: DemocraticSpace vs. 308 vs. Me

I thought I'd look at 2 sites (http://www.democraticspace.com/ and http://www.threehundredeight.blogspot.com/) that post riding predictions and compare their results to my own.  DemocraticSpace posts ranges, so I took the mid-point of the range for their prediction.  Both sites use complicated, sophisticated models, whereas I just made up some numbers.

Here are the 3 urban London ridings:

London North Centre - DS | 308 | Me

CPC: 34.5 | 32.7 | 31.7

NDP: 22.5 | 22.5 | 24.9

LPC: 36.5 | 37.2 | 34.0

GPC: 7.5 | 7.3 | 9.1

If the CPC pull 34.5 in the riding then the Liberals are in a world of trouble.  I'm more bullish on the NDP and GPC than the two prediction models. This will be a trend.

London West - DS | 308 | Me

CPC: 40.5 | 41.4 | 38.3

NDP: 19.5 | 19.5 | 23.0

LPC: 32.5 | 31.8 | 31.6

GPC: 6.5 | 7.2 | 6.7

I expect increased interest in this election to increase voter turnout, which should depress the CPC vote share slightly.  As such, I have trouble seeing the CPC over 40. Again, I'm a little more bullish on the NDP.

London Fanshawe - DS | 308 | Me

CPC: 32.5 | 28.6 | 28.6

NDP: 47.5 | 50.9 | 47.6

LPC: 15.5 | 15.3 | 14.2

GPC: 3.5 | 4.3 | 8.3

Again, a landslide.  I'm far more bullish about the GPC.  I figure the lack of strategic voting should help the party here.

 

Thoughts on London-Fanshawe

NDP candidate Irene Mathyssen won by 5,000 votes over the CPC last election.

I grew up in this riding - I lived down the street from Tweedsmuir P.S. in the Hamilton/Gore area of east London. (In fact, I've lived in all 3 urban ridings in London, each for approx. 10 years)  The weekend before the debate I was driving through my old neighbourhood and it seemed that every 3rd house had an NDP sign.  And this is before the NDP were moving up in the polls.

The Green candidate, Matthew Peloza, has done a good job of stepping into Daniel O'Neail's shoes. He won the nomination quite early and has been working hard to gain supporters and financing.  Tons of GPC - Peloza signs in the riding. There's also a surprising number of CHP signs.

This one is going to be a landslide.  My guess:

NDP: 20 000 (47.6%)+/- 3000

CPC: 12 000  (28.6%) +/- 2000

LIB: 6 000 (14.2%) +/- 2500

GPC: 3 500 (8.3%) +/- 1500

CHP: 500 (1.2%) +/- 300

 

 

Thoughts on London West

Here are my thoughts on London West.

The Conservatives took this riding from the Liberals in 2008 by just over 2 100 votes.  Before the campaign, though, I would have called this one a coin-flip even if both parties were at the same popularity levels in Ontario.  The Liberal candidate, Doug Ferguson, had a few things in his favour:

• The former Liberal incumbent Sue Barnes, seemed to be going through the motions over the last few years.  A lot of Liberals in the riding stayed home in 2008, as shown by a 3 000 drop in turnout.  This is more dramatic than it sounds, since the population of the riding has been on the rise.
• The Green candidate in the 2008 election, Monica Jarabek ran a very strong, very well financed (by Green standards) campaign.  The Greens just missed receiving 10% in the polls, which many on the Green team blamed on May's apparent call for strategic voting.  The riding association blew up spectacularly after the campaign, with Jarabek leaving the Greens for the Liberals.  If you ever meet me in person, ask me about it sometime.  Anyhow, the current Green candidate Brad Corbett is excellent, but he does not have access to the funding or the organization Monica built up.

Working in Holder's favour:

• He's the incumbent in the riding, he's much better known than he was in 2008, and like Glen Pearson, he's an unbelievably nice guy.
• Peter Ferguson, the NDP candidate, is running again and has more name recognition than he had in 2008.  Since he'd pull more votes away from the Liberals than the Conservatives, this helps Holder.

To me that balances out the 2 100 votes from last election, so I would have called this one a coin flip.

Not any more. Although Holder badly bungled the Diamond Aircraft file in the early part of the campaign, the province wide shift from the LPC to NDP is going to give this one to Holder.

My best guesses:

CPC: 23 000 (38.3%) +/- 1500

LPC: 19 000 (31.6%) +/- 3000

NDP: 13 800 (23.0%) +/- 3500

GPC: 4 000 (6.7%) +/- 1500

UP: 200 (0.3%) +/- 150

Here's where my uncertainty is stemming from:

1. I'm uncertain what the voter turn-out will be, due to a combination of population increase in the riding and interest in this election. I have the turnout to be about the same as in 2006 and up 2 500 or so from 2008.
2. I'm uncertain how much of the overall NDP surge will make it to this riding, and what proportion of those people will be drawn from the LPC and GPC.
3. I'm uncertain how a less close race than 2008 will pull fewer people away from the NDP and GPC and towards the LPC.

I give the Tories a 90% chance of taking this one, the Liberals a 10% chance and the NDP no chance.

 

Election Day Wagers

Election day has arrived which means one thing - an excuse to gamble.   Here are some bets that have been made on Twitter.  I hope these encourage you to make your own.  If you do make a bet, please let me know at @MikePMoffatt (note: Two f's, Two t's in last name).

Bets seem to be taking two forms: How many seats a particular party will win and the outcome of particular ridings.

Proposition 1: Which candidate will win London West. @jesse_helmer picks Ferguson - LPC, @MikePMoffatt picks Holder - CPC.

Stakes: If Helmer wins, Moffatt is to post picture of himself holding a Ferguson - LPC sign and post on Twitter.  If Moffatt wins, Helmer is to post picture of himself holding a Caroline Rioux - GPC Ottawa--Vanier sign.

Proposition 2: Which candidate will win Etobicoke-Lakeshore. @scottfeschuk picks Ignatieff - LPC, @colbycosh takes the field (a.k.a. anyone else).

Stakes: $50.

Proposition 3: How many seats will the BQ win overall. @MikePMoffatt picks over 10, @scottfeschuk picks under 10.

Stakes: Winner picks loser's Twitter avatar for a day.

Proposition 4: How many seats will the CPC win overall. @MikePMoffatt picks over 130, @emilytroy picks under 130.

Stakes: Loser must take a pic of themselves holding a sign of winner's choosing.

Proposition 5: How many seats will the NDP win overall. @scottfeschuk picks over 90, [unknown] picks under 90.

Stakes: Winner picks loser's Twitter avatar for a day.

Proposition 6a: Conservative seats vs. Liberal and NDP ones. @MikePMoffatt picks CPC > LPC + NDP, @pkbwood picks CPC < LPC + NDP.

Stakes: Loser buys the first round.

Proposition 6b: Conservative seats vs. Liberal and NDP ones. @MikePMoffatt picks CPC > LPC + NDP, @rockbandprogram picks CPC < LPC + NDP.

Stakes: One pint.

 

Proposition 7: How many seats will the GPC win overall. @MikePMoffatt picks 1 or more, Rebecca Young (not on Twitter) picks 0.

Stakes: If Young wins, Moffatt is to post picture of himself holding an NDP sign while on a bicycle and post on Twitter.  If Moffatt wins, Young is to post picture of herself holding a GPC sign and post it on Facebook.

Proposition 8: Which candidate will win Saanich—Gulf Islands. Jana Seijts [Facebook] picks Lunn - CPC, @MikePMoffatt picks May - GPC.

Stakes: The loser has to have their face painted green and a post a picture on Facebook.

Proposition 9: How many seats will the LPC win overall. Jana Seijts [Facebook] picks over 45, @MikePMoffatt picks under 45.

Stakes: Loser must take a pic of themselves holding a sign of winner's choosing and post on Facebook.

Thoughts on London North Centre - T-1 Day to Election

A few thoughts on my home riding.

Last election Glen Pearson (LPC) defeated Paul Van Meerbergen (CPC) by 3 300 votes.  Going into this campaign there were 3 things working in Glen's favour:

• Van Meerbergen was (and is) a very well-known, albeit controversial, city councillor, whereas the current CPC candidate Truppe is not very well known in the riding. I figure this should add about 400-600 votes to the difference, between CPCers staying home and CPCers switching to LPC (which counts double).
• The Green Party recorded 5 600 votes last campaign, thanks to an excellent campaign by Mary Anne Hodge.  Hodge is running again, but the GPC is down somewhat in the polls from '08. I would guess that all else being equal, this drops the GPC vote by 600-800, with 150-250 going to Pearson (the rest staying home or going to the NDP, so no boost to the CPC).
• Although the NDP candidate last election was not all that well known (Steve Holmes), the current candidate is less so (German Gutierrez). Guiterrez ran in the last municipal election and received approximately 4 votes (actually it was around 150, but who is counting?) I figure this would add another 50-150 votes to Pearson's total without boosting the CPC.

So instead of a 3 300 vote gap, we would be looking at a 3 800 - 4 200 vote gap. A fairly comfortable victory.  Before the campaign I would have put the odds of the Liberals winning at around 95-98%.

With the Liberals falling in the polls, now it is a lot less certain.  I still think Pearson and the Liberals will take the riding.  Between the Greens and Animal Alliance (the only small party in the riding), there should be around 89-92% of the vote left to split between the NDP, Liberals and Tories.  I have trouble seeing the CPC doing better than the 33% they got last time.  A lot has to break right for a party to win with at most 33% when 90% is up for grabs. (33/29/28 gets you there but 33/34/23 doesn't)  And I have trouble seeing the NDP at more than 25-28% in this riding.

My prediction is that Pearson takes the riding by 1 000 - 1 500 votes over the CPC.  Approx 18.5K-19K for the LPC, 17.5K for the CPC, 13.5-14K for the NDP and 5K for the GPC.

Although I voted GPC and although I had a Twitter spat with Glen Pearson yesterday, I would be happy with such a result. Glen has been a very good MP and is an even better person.

Updated to add: I was curious to see what those numbers would mean as far as percentage of votes.  Assuming 200 votes for the Animal Alliance candidate and using the midpoint of the ranges, it breaks down as follows:

LPC: 18 750 (34.0%)

CPC: 17 500 (31.7%)

NDP: 13 750 (24.9%)

GPC: 5 000 (9.1%)

AA: 200 (0.4%)

I have the turnout up about 1 500 from last election, hence the fall in the CPC percentage with the approx. the same number of votes.

On a personal level, I wouldn't be too thrilled with this outcome, as the GPC are about 500-600 votes shy of receiving a campaign refund.

These look pretty reasonable to me.  Although historically the NDP have performed average-to-slightly-above in this riding relative to the rest of the province, I don't see that happening here due to Pearson's personal popularity and NDP-LPC fence sitters believing the LPC is the best chance to prevent a CPC victory.

Update 2: Unbelievably hypocritical of me to only have point estimates.  So here are confidence intervals:

LPC: 18 750 +/- 2000

CPC: 17 500  +/- 1200

NDP: 13 750 +/- 3000

GPC: 5 000 +/- 1000

AA: 200 +/- 150

Here's where my uncertainty is stemming from:

1. I'm uncertain what the voter turn-out will be, due to a combination of population increase in the riding and interest in this election.
2. I'm uncertain how much of the overall NDP surge will make it to this riding, and what proportion of those people will be drawn from the LPC and GPC.
3. I'm uncertain how a close race will pull people away from the NDP and GPC and towards the LPC.

Given what I've written above I believe I've probably over-estimated the GPC. 4500 +/- 1000 is a better estimate.

I'd give the odds of each party winning as LPC: 65%, CPC: 30%, NDP: 5%. And 5% is probably pushing it - I just can't see the NDP winning here, but it's been a really weird election.

Space Disco Track of the Day: Slick - Space Bass (1979)

Amazing track by one of my all-time favourite disco groups, Slick:

 

Discography from Discogs:

Slick were the instrumental section of Fat Larry's Band. Scored great disco successes in 1979 with jazz-funk classics "Space Bass" and "Sexy Cream" (banned from radio play in the UK). If you want pure disco, though, listen for "Put Your Pants On"...

This gives me an excuse to post a bonus track - the non-Space Disco 'Sexy Cream', which is absolutely fantastic as well:

 

Space Disco Track of the Day: Capricorn - Capricorn (1980)

Been way too long since I've posted something. So here goes - a largely forgotten Italian piece of SpaceDisco:

 

From DiscoMusic.com:

"Capricorn" by Capricorn was another often forgotten Disco classic with Italian roots from Claudio Simonetti of Easy Going and Vivien Vee fame.

"Capricorn" is a great electronic space Disco track and worth searching for. Highly recommended.

Discogs discographies: Capricorn, Claudio Simonetti.

George Carlin on Voting

I have solved this political dilemma in a very direct way: I don't vote. On Election Day, I stay home. I firmly believe that if you vote, you have no right to complain. Now, some people like to twist that around. They say, 'If you don't vote, you have no right to complain,' but where's the logic in that? If you vote, and you elect dishonest, incompetent politicians, and they get into office and screw everything up, you are responsible for what they have done. You voted them in. You caused the problem. You have no right to complain. I, on the other hand, who did not vote - who did not even leave the house on Election Day - am in no way responsible for that these politicians have done and have every right to complain about the mess that you created.

Idelogy, Economics and the Attack on Science

It was not a good week for the science of economics in Canada.  The Globe and Mail published an analysis (their words) that 'Corporate tax cuts don't spur growth' (their words again), based wholly on the fallacy that correlation is causation. Around the same time, the Canadian Centre for Policy Alternatives published a study (their words)  using the same fallacious method.  The 6-page 'study' is a breathtaking piece of ideological anti-science from a group with good people, who often does good work, and quite frankly should know better.

Sadly both the Globe piece and the CCPA study garnered a great deal of press.  Stephen Gordon  quickly published a rebuttal that contained 2 major points, which I'm paraphrasing:

1. The Globe piece and the study got the correlations wrong.
2. Correlation is not causation.  You need to control for a great deal many other factors, which the advanced attempt to do.  (He then links to those studies)

In response, Armine Yalnizyan attempts to defend against Gordon's criticism by stating:

As to the second point, that corporate tax cuts are associated with economic growth: a correlation does not a cause make.

I wish she had told that to the CCPA before they published their study. It seems a strange point to make in defense of the CCPA study.  Armine goes on to make a number of excellent points and later in the week published a terrific piece in Economy Lab on the TFSA. She's a wonderful economist, so I'm a little puzzled why she would decide to attempt to defend the indefensible.

And their methodology is indefensible. This sort of correlation-is-causation anti-science is used by every group with an ideological axe to grind, from right-wing arguments against anthropogenic climate change to Stormfront-esque anti-science on differences between the races. I'm sure we've all heard the arguments of the latter: Group X is correlated with social bad Y, without accounting for little things like poverty levels or education.

That is not to suggest theories such as anthropogenic climate change or the effects of corporate income tax rates should be immune to criticism.  The right way to do it is through careful study and taking into account those confounding factors, such as McKittrick and Essex have been doing with climate change.

The sad thing is that there's a lot we don't know about open economy macroeconomics and changes to corporate income tax rates.  It may even turn out that the CCPA is right, which would fundamentally change the way we think about this issue, the way Card and Krueger fundamentally changed the way we think about the minimum wage.  I believe Card and Krueger will win the Econ Nobel someday for their work.  Economists who changed the way we think about open economy macroeconomics would receive similar praise.

The CCPA could be those economists and advance science. They have instead chosen to be another group that uses anti-science to advance their ideology. I find that sad, since I know they're capable of so much more.