U.S. athletes will claim 29 medals in Sochi, more than any other country, according to Tim and Dan Gaettinger. The brothers, who work for the analysis firm Discovery Corps in Canonsburg, Penn., designed a mathematical model to predict how many medals each country will win based on only four parameters — the geographical size of the country, the latitude of its capital city, the size of its economy relative to its population, and the value of its exports. The model is far from perfect, but it does explain about 55 percent of the variation in the number of medals won by different countries.
All of this year’s medals are shown in the chart above, allocated according to the model’s results. The brothers expect that countries not shown here will not win any medals. A bar of a different color represents the number of medals explained by each parameter, revealing the model’s workings in more detail. (These bars are not exact due to rounding.) Oslo, Norway, for example, is 60 degrees north of the equator, but that’s not primarily why the Gaettingers are optimistic about Norway’s chances this year. The country’s gross domestic product per capita is the reason for most of the medals Norway receives in their model. By contrast, Russia’s very large land area explains about half of the medals the Gaettingers expect the country to receive.
Why would land area be related to how well a country’s athletes do? The Gaettingers aren’t sure. So as not to allow their prior assumptions about what it takes to win in the Winter Olympics color their results, they tested all kinds of variables, and found that land area happened to be a strong predictor of success.
The extent of a country’s territory and the latitude of its capital city rarely change if ever, so this model relies on economic variables alone to explain the change in a country’s performance in the Winter Olympics over time.
Click below for more on the Gaettingers’ calculations.