Skip to main content
Menu
How We’re Projecting Each Country’s Medal Count In Tokyo

This article is part of our Tokyo Olympics series.

The 2021 2020 Olympics are already underway in Tokyo, and that means it’s nearly time to start counting up all the hardware won by each country’s athletes. Of course, there are plenty of ways to track the gold, silver and bronze medals hauled in by every Olympic team, but what’s often missed by the standard tracker is context: How much should we expect each country to be winning, based on how well it tends to do in the sports that have — and haven’t — already handed out their medals?

For the 2018 Winter Olympics, we wrote a few stories along these lines, attempting to measure performance relative to expectations. But for Tokyo, we wanted to expand that concept into a full interactive that tracks a country’s medals above or below baseline as the games progress. (We did this with data provided graciously by Bill Mallon of Olympedia.org, an essential resource for anyone with an interest in Olympics data and history.) Keeping in mind that this is not a forecast in the same sense as, say, our NFL predictions — but rather a baseline accounting of how countries have tended to do in each event — here are the basics of how it works:

Each country’s expected gold, silver and bronze medals are based on its performance in each sport over the past three Summer Olympics, weighted by recency. (Why three? My research found that number of Olympic Games was most correlated with a country’s medal count in a given Olympics.) Our “sport” categories are relatively broad — we grouped the 339 events that award medals into 40 categories, such as archery, cycling, table tennis and wrestling — and we tracked the historical share of each medal type won by the country in those sports. To form our baseline expectations, we then made sure the rates added up to a 100 percent total share1 among countries actually participating in each of those sports this year.

An illustration of five Olympic medals that are each a color of an Olympic ring, arranged in the order of the Olympic rings

related: Which Countries Are Doing Better — Or Worse — Than Expected At The Tokyo Olympics? Read more. »

We also made adjustments to handle a few special cases. For new sports with no Olympic history, we made the extremely naive assumption that each participating country has an equal probability of winning each medal. For sports with some prior history (but not complete over each of the past three Olympics), we blended the observed shares with the naive all-countries-equal approach. Since there’s also a lot of evidence that home countries get a big bounce during the games they host, we added a boost to Japan’s expected medals for the Tokyo Games, and we reduced the medal rates of Brazil in 2016, the U.K. in 2012 and China in 2008 by the same amount.2 Finally, just like in the 2018 Winter Olympics, we needed to adjust the expected medals for Russia — err, the “ROC” — over the doping scandal that has affected how the team can present itself and who is able to participate in the games.3

Using this basic system, our interactive will allow you to track the medal count in a new way — comparing results with how each country should generally expect to be doing based on its recent history of success. Because while the Olympics aren’t only about the glory of making the podium, earning a medal (mostly) lasts a lifetime and immortalizes an entire generation of athletes in the memories of fans worldwide.


How to watch skateboarding like an Olympic judge | FiveThirtyEight


How an armless archer trains his brain to win Olympic medals| FiveThirtyEight

Footnotes

  1. Or more, in the case of events that award multiple bronze medals.

  2. Specifically, we used Mallon’s research suggesting that the home country sees its medal total inflated by about 50 percent while hosting, relative to its performance in surrounding Olympics.

  3. We used the same 25 percent penalty as we did for the Pyeongchang Games.

Neil Paine is a senior writer for FiveThirtyEight.

Comments