Blog: Power & Principle

In Canada we are fortunate to have pretty good polling data on our leading political parties, even at the regional level.

Inspired, I imagine, at least in part by the success of the (somewhat mercurial) Nate Silver and his 538 project in the United States, bloggers like P.J. Fournier at 338Canada and Curtis Fric (supported by everyone's favourite Saskatchewan cartographer, Alex McPhee) at Polling Canada have done an excellent job of keeping tabs on the rise and fall of political popularity in Canada. 

We are also well served by Éric Grenier and his work predicting election outcomes, both at his old blog (ThreeHundredEight.com), the CBC, and now his Substack, The Writ.

The NBA season is about to start and so I thought I'd write a blog post detailing my adventure setting estimated win totals for the upcoming season.  

This is a fun project because it gets you thinking in a disciplined way about the prospects of your favourite team and others. 

An added bonus is addition of an element of competition, since the odds makers are happy and eager to publish their own results from their much more sophisticated (but not necessarily exceedingly more accurate) models. 

My basic approach is to sum a forecast of a team’s adjusted +/- into an aggregate total, then use Daryl Morey’s application of Bill James’ Pythagorean Wins estimate to determine how many wins each team will come away with over the season's 82 games. 

This approach isn’t perfect because +/- is built at the player-level rather than the team (unlike metrics like ORtg and DRtg), but it certainly serves my purpose of thinking more rigorously about basketball and so I figure the concept has some use.

To try and discipline my thinking about about the unfolding tragedy in Gaza, I have simulated a scenario involving an invasion of Gaza City by Israeli forces. The modest urban operations model employed here envisions a scenario where 100,000 Israeli troops, supported by air and armour, launch an assault against a resisting force of 50,000 Hamas and allied fighters entrenched in Gaza City.

This post models a potential Russian invasion of Eastern Ukraine. Relying on open-source data current to April 2021, the model results in a forecast of a successful Russian march to the Dnieper in 26 days, but at a cost of 10,500 dead Russian service members and 19,000 Ukrainian.

COVID grinding supply chains to a halt was one of the more remarkable aspects of the global pandemic. Unfortunately, restarting production--and safeguarding it going forward--is going to take a far different prescription than many currently offer.

With a swim meet coming up, I put together a list of some of the key things for new swimmers--both young and old--to think about. Racing isn't the most important thing for everyone, but I've never met an athlete who didn't want to perform their best once the whistle sounded. With a little bit of preparation, you can make the odds of that happening a lot more likely. 

The Swim Meet Prep Sheet covers the basic elements of an age group-level swim meet, and as such, is a useful primer for any young swimmers or new parents just entering the sport. 

I've broken the advice into five major sections:

• Lead-up to the meet 
• Gear checklist 
• Meet description (what to expect during your races) 
• Race rules (what you need to do so you don’t get disqualified) 
• Race tips 

Let me know what I've missed!

Remember, every race you compete in puts you to the test, making your body a little stronger and your mind a little wiser. By racing you’re also making it easier for your friends and competitors to push *themselves* harder as well. So even if your own race doesn’t go according to plan, just you being in the water can be a big help to someone else. 

In late January 2017 a new pool opened in my neighbourhood. This left me no excuse but to get back into the water after a six-year hiatus--the longest break since I started competitive swimming at age nine. 

During this lay-off I've been pretty good about making it to the gym for treadmill running and weights 3-or-so times a week. But I was in anything but water shape, so it was definitely time to go back.

I also wanted to take advantage of the new fitness tracker technology. I used a FitBit for a year or so a while back, and make sure I run with my phone to log all the steps. But now I wanted to track my progress in the water in a little closer detail. I figured I would give myself a year and then take a look. The data below is the product of that curiosity. 

With downloadable data comprising just a fraction of the information available online, learning how to use web extraction--or scraping--software has been on my to-do list for some time. 

A sophisticated person would take the time to learn a python package like Beautiful Soup or Scrapy. I thought about wandering down that path, but stumbled across a more user-friendly alternative. ParseHub is an intuitive, graphics-based web scraper that allows you to pull in significant amounts of data with very little preparatory learning. I spent a bit of time over the Christmas holidays playing around with it, and collected a few of the main lessons below.

Sports Reference.com has an excellent (though presumably soon-to-be-extinct) Olympics database. The data contained there is incredible--an even more detailed look at Olympic results than David Wallechensky's indispensable Summer and Winter Complete Book compendia. 

I've wanted to take a closer look at fatigue and average pace curves for swimming. This follows some running prediction work I did a while back based on Peter Riegel's 1981 "Athletic Records and Human Endurance" article in American Scientist. Plus, I always wanted to try out web scraping. So when I came across the easy-to-use scraper ParseHub, a good match of objectives was found.

(Photo is of the Canadian women's 4x100m freestyle relay celebration, after their bronze medal finish--one of my favourite races at Rio. By the CP's Frank Gunn, published in Maclean's.)

Ticket to Ride is a simple yet surprisingly enjoyable board game, easily picked up by pre-teen family members—and your friends with wine glass firmly in hand. I’m not much of a gamer myself, so I can’t speak to the game’s relative merits or strategic weaknesses. But I’ve played it a few times now and a) scoring is by far the most complicated part of the whole exercise (so many trains, so many cards!*); and b) this problem is a good excuse to practice a little spreadsheet coding. 

I haven’t read much of the self-help / effective habits genre, but it was recommended I take Tim Ferriss’ Tools of the Titans for a spin. Though I can’t say the in-depth discussion of psychedelics or quirky workout routines held much interest for me, there is a tonne if interesting stuff in there. I can’t suggest it enough. 

What follows are a few nuggets I found worth writing down. I’ve denoted ‘TF’ for the suggestions from Ferriss directly, and wrote out the names of his interviewees and their references for the others. 

1. TF: idols, icons, titans, billionaires “are nearly all walking flaws who’ve maximised 1 or 2 strengths.” (pxxiii).
2. TF: key traits of the successful: I can think. I can wait. I can fast (through difficulties and disasters). (pxxviii)
3. Justin Mager: remember that blood tests, etc are just snapshots.  Human body is a process, not a fixed number.  So need context.  (p73)
4. Chris Sacca: go to as many high-level meetings as possible—even if you’re not invited. Just figure out how to be helpful.  If anyone asks why you’re there, say you’re taking notes for them. (p166)
5. Steve Martin: Key to success is to be so good they can’t ignore you. (p173)
6. TF: “It’s not what you know, it’s what you do consistently.” (p185)
7. Derek Sivers: never turn down a gig. Big break came from working a pig show. “When you’re earlier in your career, I think the best strategy is just to say ‘yes’ to everything. Every little gig. You just never know what are the lottery tickets.” (p187)   
8. TF: busy = lack of control. “Lack of time I lack of priorities. If I’m ‘busy’, it is because I’ve made choices that put me in that position.” (p189)
9. TF: block out 2-3 hours each day just to focus on ONE item. (p200)
10. Nelson Mandela: asked how did you survive all those years in prison? “I didn’t survive. I prepared.” (p211)
11. Scott Adams: everyone has at least a few areas they can be in top 25% with some effort. Key is to combine two streams in some interesting and useful combination—that’s what makes your work rare and valuable. (p270)
12. Noah Kagan: block out time each week in your calendar just for learning.  (p327)
13. TF: Cut wifi during writing time and write ‘TK’ as placeholder for things you need to research later. Don’t let that get you away from getting words out on the page. (p348)
14. Brad Feld: creativity not likely in random 30-45min blocks of time. Need concerted chunks, uninterrupted, 3-5 hours minimum. (p387)
15. Shay Carl: “You can tell the true character of a man by how his dog and his kids react to him.” (p441)
16. Will Macaskill: “If you earn 68k per year, then globally speaking, you are the 1%.” (p446)
17. Jack Dorsey: best investment ever made is walking to work every day (5 miles, 1 hour 15min). (p510)
18. Mark Twain: “Whenever you find yourself on the side of the majority, it’s time to pause and reflect.” (p524)
19. Ludwig Wittgenstein: “The limits of my language mean the limits of my world.” (p524) 
20. TF: 10% of people will find a way to take anything personally. Expect it and treat it as math. (p535)
21. Naval Ravikant: “The first rule of handling conflict is don’t hang around people who are constantly engaging in conflict.” (p547)                  
22. Josh Waitzkin: don’t learn chess by memorising opening moves, learn it by using just three pieces in an endgame scenario; learn the broader principles. Reduce complexity to master the central rules of the game.  (p578)
23. Glengarry Glen Ross: Always tell the truth.  It’s the easiest thing to remember. (p592)
24. Robert Rodriguez: when starting a project, what ask yourself what assets do you have?  Is how El Mariachi included bar, bus, and pitbull—all things he could borrow from friends and family.  = movie shot for $7k.  (p629)  
25. Jocko Willink: deal with setbacks by thinking ‘good’. “Oh, mission got cancelled? Good.  We can focus on another one. Didn’t get the new high-speed gear we wanted? Good. We can keep it simple. Didn’t get promoted? Good.  More time to get better. Got injured? Good.  Needed a break from training. Got beat?  Good. We learned. Unexpected problems? Good. We have the opportunity to figure out a solution.” (p640-1)

Cleveland’s signing of the parrot-toting slugger Edwin Encarnacion was more than a savvy grab by a World Series runner-up looking to bulk up on home runs. It was also demonstrative of a Blue Jays front office that both comically misread the declining long-ball market and revealed the new regime's noncommittal strategic attitude that we all feared.

Now I get that a professional sports franchise is a business, and that I as a fan have no right to demand the firm hold a commitment to winning above maximizing revenues. To this end, while Encarnacion leaving will not make the team better, I can see how it could provide some relief to the bottom line. And so the decision to let EE walk and sign Kendrys-freaking-Morales was rational, and therefore at least somewhat defensible.

Noted curmudgeon and Rogers-apologist Andrew Stoeten certainly thinks so. Before Christmas he argued on Twitter that fans will pack SkyDome* regardless of whether or not the team wins. Loyalty to the brand, in other words, will outweigh any of this off-season's stumbles.

I've had a few weeks to think about this and I'm still not convinced. Is it not true that people like winners? If so, shouldn't this fact be salient to revenue?

The idea is intriguing and, it seems to me, should be a pretty clear matter of the empirical record. So let’s fire up the Stata and off to Baseball Reference.com we go!

I recently wrote a paper looking at Canadian defence spending in historical context.  The frame of the paper was to look at the Harper era, but I used it as an excuse to go all the way back to Sir John A.  I've included here an assortment of a few graphs and the main dataset, saved in Stata .dta format. Data comes from splicing Stats Canada's incomparable 'Historical Statistics' collection with a newer CANSIM series.  The dataset also includes PM and government party, by year.

 In the early hours of Wednesday morning, two small boats carrying 23 refugees from the besieged Syrian town of Kobane set off from Turkey's Bodrum peninsula, bound for the Greek island of Kos.  Presumably, their aim was to enter the EU in search of asylum and a new life.  

The boats, however, sank.  At least 12 of the passengers drowned, including five children.  One of these was a three-year old boy named Aylan Kurdi, whose body was recovered by the Turkish coast guard laying face-down in the sand.  A Turkish news agency recorded these images and within hours they were on the cover of newspapers worldwide.

I spent a bit of time today running Peacekeeping.org's massive dataset through the hopper.  My Stata coding was pretty brutish, but it did come up with the accompanying graph pretty quickly.  The dataset covers all UN 'chapter 6-and-a-half' missions from 1991 to the present, for all countries and presented in monthly totals.  I've just reproduced the relative Canadian contribution here.  

In my puttering around the UN DPKO website all I can find are pdf stat releases with a cutoff of about November 1990.  I gather the International Peace Institute (IPI), which runs Peacekeeping.org and built the Peacekeeping Database, gleaned their dataset from these documents.  With a little luck the DPKO will one day issue data on older missions.  In the meantime, I've yet to see a better collection of data on the topic.

This spring I wrote a paper for the the Department of Foreign Affairs and International Trade Canada's (DFAIT), under their International Security Research and Outreach Programme (ISROP).  It basically was a chance for me to extend my earlier thinking about China's economy and environment, and take a look at China's domestic security and international position.  

What follows are the executive summary, a couple of quick thoughts, and a link to the full text.

For anyone interested in the program--and it is a great one--the current ISROP research themes can be found at: 

http://www.international.gc.ca/isrop-prisi/index.aspx?lang=eng

I've been putting together a Canadian Foreign Policy dataset, relying heavily on the early 1980s Statistics Canada Historical Data Series publication (partly because I like navigating a book format, partly because the CANSIM database is such a labryrinth of inpenetrable search returns and a graveyard of terminated longitudinal series).  

Some of the series I've constructed are total defence spending and total federal spending.  The following graphs are quick sketches of the expenditure trend, both at the macro scale and over the last few decades.

I like using defence spending in terms of the total federal budget because it gives a better sense of relative government priorities

than the more common GDP denominator.  A government cutting EVERYTHING isn't signalling it wants to get away from military spending, it's signalling it just wants to spend less.  Falling defence budgets are just a byproduct.  Budget shares, however, signal a movement in some other direction than military force.

Note that it is harder to get government financial data than just GDP, so sometimes this metric simply isn't available.  And of course, if the variable of interest is power potential rather than just foreign policy concern, then GDP is a more suitable measure.  

Just finished up a chapter on politics in Africa, this time focusing on communications technology, political competition, and a bit of economic development.  As usual, there were a lot more graphs than could fit in the final product.  

What follows are a bit of a random collection, but hopefully useful to the reader nonethless.  They were created on Stata and all the associated data is up on the 'Academic Details - Datasets' page.  Sources follow in the captions below. 

As part of a book chapter I'm writing on ICT and African governance I've been going through some World Bank data on the Sub-Saharan region.  Part of the process is just plugging in some variables and looking at the graphs that come out.  There are always some interesting stories that arise, but this one is by far my favourite.

Seems like Naomi Klein--who when I was an undergraduate seemingly sold books like Dan Brown--couldn't have been more wrong.