Olympics Data Analysis

Goal

Use Olympics medals/athletes data to:

• Rank countries by medals
• See trends over years
• Identify top sports

Analysis pipeline

  flowchart LR
  A["Raw results"] --> B["Count
(medals per country)"]
  B --> C["Rank
(sort_values, top 10)"]
  C --> D["Visualize
(bar chart)"]
  D --> E["Trend
(medals by year)"]

Step 1: Load

import pandas as pd
 
df = pd.read_csv("data/olympics.csv")
print(df.head())

Step 2: Basic counts

medals = df.groupby("country").size().sort_values(ascending=False).head(10)
print(medals)

Step 3: Plot top countries

import matplotlib.pyplot as plt
 
plt.figure(figsize=(10, 4))
plt.bar(medals.index, medals.values)
plt.title("Top 10 countries by medals")
plt.xticks(rotation=25)
plt.tight_layout()
plt.show()

Visualize it

Deliverable

• Top 10 countries
• Change over years
• Sports that contribute most

What one row means

Before you count anything, find out what a row represents. In a medals file, one row might be one athlete's medal, one team's medal, or one country's total for a year. The answer changes the numbers. In a team event such as relay, each team member may have their own row, so counting rows would give a country four medals for one race. Look at df.head() and df.columns, then decide what "one medal" means and remove duplicates if needed (for example with drop_duplicates on event, year and country).

Ranking countries

The one-line ranking from the lesson works in three steps: groupby("country") splits rows by country, .size() counts the rows in each group, and sort_values(ascending=False).head(10) orders them and keeps the top ten. Here is the same idea on a tiny table you can run:

import pandas as pd
 
df = pd.DataFrame({
    "country": ["USA", "USA", "CHN", "CHN", "CHN", "GBR", "USA", "GBR"],
    "year": [2016, 2020, 2016, 2020, 2020, 2016, 2020, 2020],
    "medal": ["Gold", "Silver", "Gold", "Gold", "Bronze", "Silver", "Gold", "Gold"],
})
 
top = df.groupby("country").size().sort_values(ascending=False)
print(top)
# USA    3
# CHN    3
# GBR    2

Two countries can tie, and the order between tied countries is not guaranteed unless you add a second sort key. For a fair ranking, sort by gold medals first and total second.

Trends over the years

To see change over time, group by two columns and reshape with unstack, so each country becomes a line:

by_year = df.groupby(["year", "country"]).size().unstack(fill_value=0)
print(by_year)
# country  CHN  GBR  USA
# year
# 2016       1    1    1
# 2020       2    1    2

fill_value=0 matters: a country with no medals in a year would otherwise show NaN. Plotting by_year.plot() then draws one line per country.

Common mistakes

• Counting rows when rows are athletes. Team medals get counted many times.
• Comparing raw totals across eras. Games have added events over time, so also look at the share of all medals.
• Mixing country names and codes, such as "United States" and "USA", which splits one country into two groups.
• Missing values in the medal column. Rows for athletes who did not medal may be included; filter them out with dropna(subset=["medal"]).
• Plotting too many bars. Show the top 10 so labels stay readable.

Turning it into findings

State each result as a plain sentence: which countries lead, how their share changed, and which sports drive it. Then check the finding by looking at one country and one sport by hand, to make sure the pipeline agrees with common sense.

Check your understanding

🧪 Try It Yourself

Exercise 1 – Medal counts per country

Press Run to see what your code prints.

Exercise 2 – Top N ranking

Press Run to see what your code prints.

Exercise 3 – Medals by year (trend)

Press Run to see what your code prints.