How the Olympic Medal Table Should Be Presented--and Read

In most media reports about the Olympics, summary Medal Tables are used as measures of "national performance," if not "worth". This is particularly visible in terms of the great Hollywood-esque nationalist / jingoistic Kitch of the putative "competition" between the US and China.  We learn that the "U.S. Edges China in Medal Race," pointing at a particular sport, volleyball, as the tool that does that "edging" by virtue of the fact that the US gained its last gold medal in that sport. And those two examples are from supposedly respectable newspapers, one in the US, the other in the UK. Not tabloids. Chinese official media did the same, with a message in the opposite direction, of course. Russian media: ditto. Germany, the Netherlands as well. The "medal competition" is presented as if it were a great struggle between the new geopolitical-geoeconomic "enemies," the US and China, with a sprinkle of cold war references to Russia, the real "enemy" (lurking, ever so suspiciously, under the name "Russian Olympic Committee"). The typical journalistic language praises the "two giants" for a nail-biting competition for the top position. Which of course goes to . . . the United States, of course. Who else.

The validity--or, for that matter, sanity--of using sports as basis for national pride is highly dubious. In so many ways this whole exercise is a giant waste of money and an exercise in jingoism, vanity and futility. 

But let us assume, for a moment, that someone actually wants to find out about the relative "performance" of "nations" based on the Olympics. Someone with a genuine interest in the relative significance of elite sports in various countries, and the efficiency of the organization of sports on all levels, from the streets to the elite professional circles.

Viewed that way, there is a great degree of inconsistency in the ways the results tend to be presented. The countries of the world are ranked in such tables either by the sheer number of gold medals "their" athletes had achieved. Such accounts ignore silver and bronze medals as if those signaled "defeat." The alternative involves adding up the number of gold, silver and bronze medals for each country. That method is also problematic: It gives equal weight to all three medals, whereas we all feel a bronze, while very nice, is not the same is the silver, and neither of those is the same as a gold medal. 

Worse yet, the presentation of the medal table in the way it is done by almost all of the world's media creates a false equivalency between the chances for victory for athletes from, say, the People's Republic of China (the world's largest society and the world's largest economy) and, say, Eritrea. That cannot possibly be right, anybody with a reasonable sense of fairness might think.

Here I propose a simple way to present olympic medals that considers all three medals and gives them different weight. Namely, 3 for gold, 2 for silver, and 1 for bronze. The data come from the IOC's "official" medal count table. To control for population size and economic power, I plot the medal scores by the total GDP of each country (obtained from the World Bank's World Development Indicators dataset.) Syria and Taiwan are missing from the GDP data, so they had to be omitted.

Here is the raw distribution of country-to-country medal scores by GDP.

We can make three observations from this distribution.

• The economic might of a country is a fantastic predictor of olympic success. Here, approximately 73% of the medal results is explained by GDP (R-square=0,7328).
• There is not much to talk about concerning the "giants" competition between the US and China. The two top scorers are almost exactly on the linear regression line. 
• By and large, the most interesting part in terms of olympic medals, is the middle in terms of GDP.

Here is the same graph, with my indication--see the oval shape laid over the graph--of that part of the field which contains the most successful "nation states," i.e., those states which find themselves farthest above the regression line. They are those states which perform "better" in terms of olympic medal scores than what would be expected solely on the basis of their economic power.  

For easier identification, I have marked, in the next graph, (with a different color, and abbreviated country names) those cases are on the top "border" of the patch that contains the high achievers--read: the "nation" states that are most outstanding in terms of the Olympic "performance" of their athletes.

Proceeding from the left to right, here is the list of the countries that had the best "national" performance in terms of olympic medal scores this time around:  Jamaica, Georgia, Uganda, Uzbekistan, Azerbaijian, Slovenia, Serbia, Croatia, Belarus, Bulgaria, Kenya, Cuba, Ukraine, Hungary, New Zealand, Czech Republic, Poland, the Netherlands, Australia, the ever so nebulous "Russian Olympic Committee," the United Kingdom, and Japan. As it turns out, the People's Republic of China is right on the regression line and the United States is very slightly below it (i.e., based on its economic might, it would have been expected to score a tiny bit better). For orientation, I have also inserted the global mean in GDP and medal scores (see the black square--it looks like this: █ -- you'll find it in the middle of the graph). Most interesting are the countries higher and to the left of that square (i.e., those with better-than-average medal scores, with less than average GDP figures).

So, how much better is each "nation" is doing than it would be expected, based on its GDP? That is indicated in the vertical distance between the dot that represents the country and the regression line. Cuba, for instance, has a medal score of 32, its vertical line crosses the regression line at around 5,5, so Cuban athletes performed 32/5,5=5,8 "better" than could have been expected, based on their GDP alone. And so on.   

I see three noteworthy empirical regularities in this distribution. (There might be more--I'd be happy to hear if anyone finds any of them worth mentioning.)

1. Many erstwhile state socialist societies of east-central Europe, and a few of those of the former USSR, are still doing better than other countries in their GDP bracket, almost irrespective of their relative wealth, ranging from Georgia to the "Russian Olympic Committee." The same is true for still-socialist Cuba. Some European Union member former-socialist states--here: the Baltic states and perhaps the Czech Republic and perhaps Poland--have begun losing their traditional advantages, but they are still doing quite above the expected value.
2. In each summer Olympics, there are a few states from the Global South that break into the ranks of the best performer erstwhile-state-socialist states. This time around Jamaica, Kenya and Uganda did remarkably well, a fantastic achievement, given their limited economic resources.
3. A former colonial power--the UK--and two of its erstwhile white-settler colonies--New Zealand and Australia--also did well, and so did Japan, using a certain "home country advantage." 

The "negative" cases I will not elaborate on, everyone can think of their expectation--if it is not named in the last graph, it did not do particularly well. 

So much about what reality shows us, beyond the Hollywood-esque story of "the US Edging Out China."