Olympic Post-mortem Part II: Who won the Sochi Olympics?

This is the second of two parts of an Olympic post-mortem (read the first part here), in which I arbitrarily assign ranks to the participating countries.

Following the 2012 Summer Games in London, I published a story in the Obiter declaring Grenada to be the Winner of the Olympics. I realize that this was an arbitrary designation, but it was not entirely unfounded. I developed a method of statistical evaluation to determine the relative performance of participating countries, accounting for their population and their wealth. Under this metric, a rich country with a high population (like the United States) is expected to win the most medals, while Grenada wildly exceeded expectations by winning a single medal. The results of my analysis aren’t terribly useful as a ranking tool, but they do pose interesting questions about why some countries perform the way they do at the Winter Games. Let’s take a look.

I’ll start by explaining my methodology. First, I assign each country a score based on the number of medals they win. A gold medal is worth three points, a silver two, and a bronze one. Canada’s medal count of 10 gold, 10 silver, and 5 bronze earns her a score of 55. This allows me to determine what proportion of the total available points (this year, there were 590 points available) each country successfully seized. This figure is the country’s “medal score”.

Second, I project each country’s “expected medal score” using key indicators. This is the controversial part. Obviously I have too much time on my hands, or I wouldn’t bother with this sort of thing. Nonetheless, I don’t have enough time to comprehensively evaluate and measure every factor that might contribute to a particular country’s success at the Games, so I have to use indicators that are readily available, reasonably reliable, and comparable across countries.

This year, I used the same indicators as I did for the 2012 London Games: population and GDP. Each country’s population and GDP are expressed as a proportion of the global total. In this case, “global” refers not to the entire world, but to a notional globe that includes only the countries who won medals at the Games. I set this threshold not to make my research and calculations easier, but on the principled basis that for the purposes of my invented competition, only countries with at least one person serious enough about winter sports to win an Olympic medal should be invited.

Population represents the talent pool in a particular country: presumably there are more people with a proclivity for skiing in a country of one billion than there are in a country of one million, whether that talent is discovered or not. A country’s expected medal score for population is the proportion of the global population that the country represents.

GDP represents a country’s individual ability to discover and support talented athletes with stipends, facilities, and training. Obviously, GDP is not a perfect indicator, since political conditions determine whether and how resources are allocated to athletics. Canada’s Own the Podium program has many funding partners, but its largest is the federal government, and not every nation contributes public funds (or any funds at all) to winter sports, regardless of the raw size of its economy. GDP thus represents a country’s theoretical ability to contribute financially to sport. A country’s expected medal score for GDP is the proportion of the global GDP that the country represents, and a country’s final expected medal score is the average of the population and GDP medal scores.

Finally, I calculate each country’s “performance”, which is the country’s actual medal score expressed as a percentage of its final expected medal score. If you follow any of the nonsense above, you’ve figured out that I am trying to determine whether a country exceeded or fell short of its theoretical potential to win medals, as determined by the size of its talent pool and ability to support its athletes. Thus, countries with a score over 100% can be said to have won x medals despite their small population, weak economy, or both, and countries with a score below 100% can be said to have won only y medals because they are failing to discover talented athletes, failing to support the ones they have, or both.

Before I get to the results, I want to acknowledge that I have failed to control for climate. Almost every country on earth contains areas in which facilities for training summer athletes can be built at relatively low cost. Training a long distance runner requires no equipment at all save a pair of decent shoes, and a nominally heated indoor track during winter, in places where it gets cold at all. There are almost no summer sports that demand an equipment cost greater than about $100 (with the exception of equestrian, the shooting sports, canoeing, kayaking, and now golf).

Winter sports are resource-intensive. Keeping arena ice frozen in Canada’s population centres is phenomenally expensive, so you can imagine how prohibitive the costs are in Mexico. In places without summer arenas, skiers and skaters practice on synthetic ice, a self-lubricating polymer that requires sophisticated equipment to manufacture and special expertise to install and maintain. They don’t have these things in Burundi. It doesn’t make economic sense, and no one cares about ice dancing there anyway.

I attempted to use annual snowfall data for each country to account for the natural advantages and disadvantages of colder and warmer countries, respectively. The problems with this were twofold. First, most countries have enough climatic variation to make it impossible to generalize. Vancouver gets 44 centimetres of snow each year; Toronto gets 122. Yet Vancouver was selected to hold the 2010 Winter Games because it had the resources to maintain indoor ice sheets, and because it is situated just hours from Whistler, which receives over 400 centimetres of snow each year. There is no “average Canadian snowfall”. Second, average snowfall data simply isn’t available for some places, even if it does occasionally snow. So, instead of properly controlling for climate, I will simply make particular note of countries that are relatively warm and/or receive little snow, yet still managed to score well by my estimation. There aren’t very many.

It should come as no surprise to anyone that Norway was the big winner here, as usual. With a population of just over 5 million and a GDP of US$282 billion, Norway’s medals earned her 53 points. This means that Norway earned 3086% of the medals it should have, ceteris paribus. No other country is even close. I suspect that this has something to do with how many medals are available in cross-country skiing and biathlon. Apparently there’s not much to do in Norway except to ski and shoot things.

The big story at these Games is 2^nd place Slovenia. Standing out in their fluorescent green and blue uniforms, alpine athletes from this former Yugoslavian country earned two gold, two silver, and four bronze medals. This gives them a score of 2598%. Nothing to sniff at.

A few other countries outperformed expectations by an order of magnitude, with Latvia (3^rd), Austria (4^th), and Switzerland (5^th) all posting scores over 1000%.

Host country Russia is naturally disadvantaged by its sizeable population of over 150 million. At the same time, much of the country is a vast frozen wasteland, which explains why the Russians swept the podium in the men’s 50km “skiathlon” on the final day of competition. Lots of room to practice. I would be remiss, however, if I did not acknowledge the role of defectors in Russia’s success. South Korean short track skater Victor Ahn was cut from his home team in 2011. Russia offered him citizenship, which he accepted, and he won three golds (one as part of the relay team) and a bronze in Sochi. No Korean finished ahead of him during the entire competition. Oops. In any case, warts and all, Russia’s score was a respectable 232%, putting them in 12^th.

Korea, with a score of 121%, scored the closest to 100%, meaning that they performed almost exactly as expected, coming in 18^th place. Incidentally, they would have scored 193% and come in 15^th if they’d kept Victor Ahn.

The bottom of the barrel includes both the United States, placing 22^nd out of 26 countries with a score of 50%, and China, dead last with a score of 10%. Obviously the Summer Games are a bigger focus for the Chinese team’s exceptionally large talent pool and rigorous training program.

As promised, there are two countries particularly worthy of praise due to their high performance despite the climatic odds. All but two of the countries that won medals at all have at least one corner of homeland suitable for training winter Olympians. This makes the performance of these two countries remarkable. Australia scored 71% and placed 20^th, winning two silvers and a bronze, all in freestyle ski and snowboard events. All three athletes train overseas, sometimes in Canada, since domestic facilities are virtually non-existent. A scoring metric that accounted for climate would have functioned so as to bump Australia’s score well over 100%.

The second country is, of course, the Netherlands. I’m not suggesting it never gets cold there. However, the Elfstedentocht, an annual 200km outdoor speed skating race on the low country’s canals, hasn’t been held since 1997 because it hasn’t been cold enough for long enough for the ice to reach the requisite 15cm thickness. The Dutch are absolutely crackers for speed skating. In 2012, when it looked like the Elfstedentocht (Dutch for “eleven cities tour”) might happen but was eventually called off, Prime Minister Mark Rutte suggested that the Dutch assuage their disappointment and sorrow by – get this – going to local indoor rinks and skating their cares away. The podium at the Olympic oval is the temporary February residence of the House of Oranje every four years. It will probably always be that way. With a score of 899% and a 6^th place finish under this admittedly imperfect system, there is a strong case that the Netherlands may be as much a winner as Norway or Slovenia.

You’re probably wondering how our beloved home team fared this year. With a very respectable medal count, Canada scored 506%, more than quintupling statistical expectations, though expectations were certainly much higher in reality. This score is in no small part thanks to the first ever “double double” – twin gold medals in both hockey and curling. No country has ever done that before, and I don’t suspect that Slovenia will be challenging us in that respect.

Lest we become prideful, however, I wish to provide some food for thought. Even with a score as high as Canada’s, some areas of the country do outperform others. I simply didn’t have time to slice and dice the entire Olympic team by province of origin, but I did take the time to gather stats for a few provinces, and confirmed two hypotheses. First, that bobsled brakewoman Heather Moyse, even if you only assign her half a gold medal for being half of the winning team, single-handedly puts Prince Edward Island, with its tiny population and economy, at the very top of Canadian provinces at these Games. As against the rest of Canada, PEI scores 2135%, and scores an even more impressive 10614% as against the rest of the world, tripling Norway’s score.

The second hypothesis is the one that everyone’s favourite Première Ministre has been hinting at all week: that Québeckers are outperforming the rest of Canada. Athletes from la belle province won 3 gold, 3 silver, and 2 bronze medals of their own, and won 0.16 of a gold medal as part of the men’s hockey team, 0.32 of a gold medal as part of the women’s hockey team, and 0.75 of a silver medal in the women’s short track relay. As part of Canada, Québec won 156% of its fair share of medals (the rest of Canada scored just 82%), and scored 796% as against the rest of the world (which would put them in the top ten), compared to the remaining provinces’ 419%, excellent but comparatively dismal.

Everyone knew from the 1950s to the 1970s that one of the major reasons why the Montréal Canadiens kept winning the Stanley Cup was that they had the right of first refusal for players in their extensive Québec-based farm system and within 50 miles of Montréal itself, a gift that kept on giving well after the introduction of the draft in 1963. French Canadians love winter. I’ve always believed that if a majority of residents agree in a referendum (which, of course, gets less likely as the years march on), the federal government has no right to interfere with Québec’s right to self-determination. Looking at the figures above, I suddenly see the argument for keeping them on Canada’s Olympic team against their will.

I hope that my Olympic fever and reading week writing projects have kept you entertained. It was a fun time to be Canadian, and I know I’ll get excited all over again in four years. The difference is that, in four years, I will appreciate the little bronze medal successes of seemingly insignificant countries in an entirely new way, since those triumphs are perhaps more significant in the circumstances than another double double. Here’s to Monaco’s bobsled bronze in Pyeongchang.