/tennis-vis

A project for visualizing tennis players statistics.

Primary LanguageJupyter NotebookMIT LicenseMIT

TennisVis

  • A project for visualizing tennis players statistics.
  • Published Python Dash App can be found at: https://tennis-vis.herokuapp.com/
  • Raw data source: https://tennisabstract.com, https://www.atptour.com/
  • Current Version: 2.4
  • Last data/ATP Rank update: 2024-09-09/2024-09-09
  • Current Functionality:
    1. Home
      • A Network for All ATP Grand Slam Champions in Tennis History
    2. Dynamics
      • ATP Dynamic Ranking from 2000-01-10 to 2024-09-09
      • GS Titles Accumulation of Important Players from 1990 to 2024
    3. GeoTennis
      • Geographic Distribution of ATP Top100 Players from 2000-01-10 to 2024-09-09
    4. Records Search
      • A comprehensive match records search interface, criterions includes:
        • Dates
        • Opponents, e.g., Novak Djokovic, Top10 Players, Big3, Big4
        • Surfaces, e.g., Hard, Grass
        • Tournemants, e.g., Wimbledon, ATP1000, Olympics
        • Rounds, e.g., R128, QF, F
        • Results, e.g., win, tie-break lose
        • Streak, e.g., return the longest winning streak among matches of all previous seletions
        • Layout, e.g., lite or display all match statistics
    5. Stats Visualization
      • A graphic interface for multiple players stats comparison within a period:
        • Titles, e.g., all titles, grand slams
        • Head 2 Head, e.g., h2h on grass
        • Moveing Average Winning Ratio, e.g. MA winning ratio per 100 matches
        • Serve Stats, e.g., ace percent, second serve win percent
        • Points Overview, e.g., the distribution of Dominance Rate
        • Break Points Overview, e.g., average break points conversion rate
        • Win/Loss Counts, e.g., big heart matches vs crystal heart matches
        • Winning ratio by surface, e.g., on Clay
        • Winning ratio by tournament, e.g., finals of Australian Open
        • Winning ratio at rounds of Grand Slam, e.g., semi-finals on US Open
    6. DSTennis
      • A currently simple interface for visualizing regression results of tennis players' stats:
        • Variables are at the player level, e.g., height, weight, career winning ratio, career average dominant rate, left or right handed.
        • Simple linear regression displays the scatter plots and estimated regression line.
        • Multiple linear regression displays the estimated coefficients.