ATP Tennis Record Analysis

Use data from https://github.com/JeffSackmann/tennis_atp to analyze ATP players’ records.

Perform all data analysis using pandas.

Tasks

1. Using the GitHub Repository

• Clone the repository from GitHub.
• Carefully read the documentation in the repository to understand the meaning of each dataset and field.

2. Analyze atp_players.csv

• What columns (fields) are included in this file?
• How many total players are listed?
• Among all players, how many have incomplete dob (date of birth) information?
• Group players by ioc (nationality). Which countries have the most players?
• How many players are from TPE and TWN respectively?
• Among the players from TPE, what surnames are included?

3. Analyze atp_matches_202[1-4].csv

• How many matches have TPE players won and lost?
• How many distinct tournament levels (tourney_level) are there?
• In how many matches is the winner’s height greater than / equal to / less than the loser’s height?
  • For each tourney_level, analyze the relationship between height and match outcome.
• Round player ages (.round()) and:
  • Use value_counts() to count the ages of winners.
  • Compute win rates by player age.
  • List the top 10 players with the most wins.
  • List the top 10 players with the highest win rates.
  • Create a table showing the number of matches between players of different ages:
    • The index should be the winner’s age, and the columns should be the loser’s age.
    • Each cell should represent the number of matches played.
    • Only consider players aged 19–28.
  • Create another table showing the win rate between players of different ages:
    • Index = player’s age; columns = opponent’s age; values = win rate.
  • Identify which age groups have win rates above 49% against every age group (within ages 19–28).

4. Your Own Questions