Olympic Medals by Country

Introduction

The dataset is available on Kaggle but was originally created by scraping data from www.sports-reference.com. It lists all Winter and Summer Games from 1896 to 2016 and includes athlete names, physical characteristics (sex, age, height, weight), country/team, host city of each Games, sport/event, and medal won (if any).

In this analysis, I will be analyzing the medals won for each country and determining overall which country has been most successful throughout the history of the Olympic Games.

To see the R code I used to generate these visualizations, click here.

The basics: an overview of medals by country

Let's first take stock of the overall trend if we look at the number of athletes who have competed for each country as compared to how many medals they have won.

Looking at this scatterplot, we might conclude that the United States is incredibly dominant at the Olympics, as it is the country that has sent the most athletes and won the most medals, by far.

plot of chunk unnamed-chunk-5

Before moving on, let's take a look at how this splits out by gender. We can see that American men make up a large percentage of the athletes and medals for the United States (for more details, please read 120 Years of Olympics Summer Games History). We also see that the overall trends of athletes and medals is fairly similar.

plot of chunk unnamed-chunk-6

Data SPIn 1: 1896 to 1988

One of the main issues that jumps out of the data is the prominence of the Soviet Union. Although the last year the Soviet Union competed in the Olympics was 1988, they have the second highest number of medals in the history of the Olympic Games. What would the data look like if we only considered Olympic medals before the fall of the Soviet Empire?

We see here that the Soviet Union does come a bit closer to the United States. This view is problematic for a few important reasons:

  1. There are other countries besides the Soviet Union (Yugoslavia, Germany, the Czech Republic, etc) which have changed politically over the years. This method of looking only at data defined by the Soviet Union may adversely affect the results of other countries.
  2. Women have gained much more prominence in the Olympic Games in recent years. In the United States, this is because of the creation of Title IX in 1972, which began to filter through sports and ramped up Olympic participation in the 1990s. Restricting the data to the years before the Soviet Union fell would exclude most of the accomplishments of female athletes.

plot of chunk unnamed-chunk-7

Data spin 2: average medals per year

In order to address the issue of some countries participating in fewer Olympic Games, we could look at the average number of medals per year. This helps to normalize the data so we can compare, for example, the United States (who competed in 35 Games) to the Soviet Union (who competed in 10).

This view of the data shows us that the Soviet Union did, in fact, have a higher average number of medals per year than the United States has had. We also see East Germany jump up to the top of the pile as well. Neither of these are terribly surprising, as both countries poured quite a lot of state-sponsored support into winning Olympic competitions.

A note of caution here. We see that the Unified Team has the highest number of average medals per year in this plot. This was theh team that many Soviet athletes competed under in the Summer and Winter Games in 1992, before the resulting countries had finalized their affiliation with the International Olympic Committee. Since this team was only in place for one year, it is interesting to note how well they did, but countries will a longer history may want to be given more weight in the analysis. In order to cut down on some of the extreme examples of this, I've restricted the data to only countries with at least 100 athletes.

plot of chunk unnamed-chunk-9

We can also see similar trends if we split each country by gender. The top four countries (The Soviet Union, The United States, the Germanies, and the Unified Team) all show the standard pattern of a higher men's average for medals per year. Where it starts to get interesting is when you look at the top of the dense grouping. Russian and Chinese women both have a higher average than men. I will dig more deeply into the reasons behind these differences in a later analysis.

plot of chunk unnamed-chunk-10

Data spin 3: average medals per athlete

We've seen that it's not always effective to look at overall data or a small segment of data (restricted by year). The fact that some countries are significantly larger and can therefore compete on a larger level (and have more talent to choose from) has a huge impact on their performance at the Games.

When we looked at the average by year, this helped to normalize some of these issues. However, large countries often have the ability to send more athletes to the Games than others (some countries may send 2 athletes per Games, some may send 100). In order to account for this, we should look at the average number of medals per athlete.

We can see here that while the United States has by far the most medals, the Soviet Union and East Germany have a higher average number of medals per athlete.

plot of chunk unnamed-chunk-26

Looking at the average medals per athlete when split by gender, we see that American women have a higher average than American men. The same holds true for the Soviet Union and for East Germany.

plot of chunk unnamed-chunk-11

Conclusion

There's no question that the United States has historically been the powerhouse of the Olympic Games. We have sent almost 55% more athetes (12,146) than the next closest competitor (Great Britain, 7,860) and we have won more than twice as many medals (5,219) as the next closest competitor (the Soviet Union, 2,451). If we measure on pure quantity, the United States comes out on top.

However, if we want to add in the quality of that competition, it's a different story. Do we most often send the BEST competitors? Many of our athletes (like Michael Phelps) have dominated their events and brought home multiple medals. Others have struggled a bit more to make an impact. Combining both quality and quantity into determining which country has been most successful, we should qualify that the United States is the most successful country that is still sending athletes to the Olympics. The Soviet Union and East Germany both had better overall averages per athlete, but are no longer in existence.