Ultimate Tennis Statistics and Tennis Crystal Ball

Ultimate Tennis Statistics is a Tennis Big Data Analysis tool with a nice web GUI. Tennis Crystal Ball is a set of Match Prediction and Tournament Forecasts algorithms powered by AI and machine learning.

Features

'GOAT' List - A.k.a. 'GOATometer' - Best players of Open Era ranked by 'GOAT' points with customizable weights (overall and by surface)
Player Profile - Player information, season summary, tournament results, matches, timeline, rivalries, ranking, performance indicators and statistics with charts, 'GOAT' points breakdown and records
Timelines - Dominance ('GOAT' points distribution among top players and seasons), Grand Slam, Tour Finals, Masters, Olympics, Davis Cup, Team Cups, Top Rankings and Surface timeline
Head-to-Head - Head-to-head between two players with H2H matches, season summary, performance and statistics comparison, ranking, performance and statistics charts, 'GOAT' points breakdown and Hypothetical Matchup prediction based on the Neural Network Match Prediction Algorithm
Heads-to-Heads - Heads-to-heads clusters among several players (i.e. among 'Big 4')
Greatest Rivalries - Explore greatest rivalries, overall or by season, tournament level, surface or round
Greatest Matches - Explore greatest matches ranked by special Match Greatness Score formula
Ranking Tables - Player weekly ranking tables
EloRatings - Player weekly Elo rating tables (overall, by surface, outdoor or indoor, set or game) using customized Elo rating formula
Ranking Charts - Player ranking and ranking point charts, compare players by constructing custom charts
Peak Elo Ratings - Peak Elo ratings list for comparing players in their peaks (overall, by surface, set, service/return game and tie-break)
Titles and Results - Titles and other results, filtered by various filters
Titles and Results Charts - Titles and results charts, filtered by level, surface and seasons
Top Performers - Find top performers in both performance and pressure situations categories
Performance Charts - Performance charts for various performance categories, filtered by seasons
Mental Toughness - Compare players by mental toughness using special mental toughness formula, filtered by various filters
Statistics Leaders - Find statistics leaders in different statistical categories (100+), including dominance and break points ratios
Top Match Stats - Top single match statistics figures filtered by various filters
Statistics Charts - Statistics charts for various statistical categories, including dominance and break points ratios
Seasons - Browse seasons and check season records, tournaments, rankings, performance, statistics and 'GOAT' points distribution among top players
Best Seasons - Find which are the players' best seasons of the Open Era based on 'GOAT' points (overall and by surface)
Tournaments - Browse tournaments, see players with most titles, historical tournament levels and surfaces and average participation
Tournaments Events - Browse all Open Era tournament events, see tournament event draw, performance, statistics, historical winners and records
Tournament Forecasts - Tournament Event Forecasts for in-progress tournaments driven by Neural Network Match Prediction Algorithm
Records Book - Various match, tournament result and ranking records, famous and infamous (the best player that never...)
Live Scores - Live Scores via Enetscores by Enetpulse

Web Site

https://www.ultimatetennisstatistics.com

Data Loaders

For Jeff Sackmann ATP CSV repository: https://github.com/JeffSackmann/tennis_atp (as of commit cf2201c on 2 dec 2019)
Database setup instructions: #232
Docker image mcekovic/uts-database with PostgreSQL database pre-populated with ATP tennis data as of season 2019: #337

Technology

PostgreSQL, Java, Spring Boot, Thymeleaf, JQuery, Bootstrap, Google Charts, Groovy...

Architecture

flowchart LR
  user((User)) --> tn
  subgraph uts[Tennis Stats]
    direction TB
    subgraph app[Spring Boot Java App]
      direction TB
      cc[Caffeine<br>Cache]
      mvc(Spring<br>MVC) --> tcb(Tennis<br>Crystal Ball<br>Predictor) & elo(Elo Ratings<br>Manager)
      mvc & tcb & elo --> cc
    end
    tn[[Tomcat<br>Native]] --> mvc
    app --> cb[Let's Encrypt<br>Certbot] -- SSL certificate ---> tn
  end
  subgraph dl[Data Load]
    direction TB
    load(Data Load<br>Groovy Script) --> gd[[Selenium<br>Geckodriver]] --> ff[Headless<br>Firefox]
  end
  app -- spawn ---> load
  app & load ---> db[(<br>PostgreSQL<br>DB)]
  load --> repo[Jeff Sackman<br>Tennis Repo]
  ff --> www([www])

  style uts fill:DimGray
  style app fill:CadetBlue
  style mvc fill:Indigo
  style elo fill:DarkOrchid
  style tcb fill:DarkMagenta
  style cc fill:Maroon
  style tn fill:ForestGreen
  style cb fill:Crimson
  style dl fill:DimGray
  style load fill:RoyalBlue
  style gd fill:IndianRed
  style ff fill:OrangeRed
  style db fill:SteelBlue
  style repo fill:DarkSlateGray

  click mvc href "https://github.com/mcekovic/tennis-crystal-ball/tree/master/tennis-stats/src/main/java/org/strangeforest/tcb/stats/controller" _blank
  click elo href "https://github.com/mcekovic/tennis-crystal-ball/tree/master/tennis-stats/src/main/java/org/strangeforest/tcb/stats/model/elo" _blank
  click tcb href "https://github.com/mcekovic/tennis-crystal-ball/tree/master/tennis-stats/src/main/java/org/strangeforest/tcb/stats/model/prediction" _blank
  click cc href "https://github.com/ben-manes/caffeine" _blank
  click tn href "https://tomcat.apache.org/native-doc" _blank
  click cb href "https://certbot.eff.org" _blank
  click load href "https://groovy-lang.org" _blank
  click gd href "https://github.com/mozilla/geckodriver" _blank
  click ff href "https://www.mozilla.org/en-US/firefox" _blank
  click db href "https://www.postgresql.org" _blank
  click repo href "https://github.com/JeffSackmann/tennis_atp" _blank

Screenshots

Home Page

GOAT List

Player Profile

Player Tournaments

Player Timeline

Player Rivalries

Player GOAT Points

Dominance Timeline

Surface Timeline

Greatest Rivalries

Ranking Charts

Ranking Points Chart

GOAT Points Chart

Peak Elo Ratings

Top Performers

Statistics Leaders

Seasons

Best Seasons

Tournaments

Records Book

License

Tennis Crystal Ball and Ultimate Tennis Statistics source code is licensed under Apache 2.0 License.

'GOAT' Formula, customizations of Elo Ratings for tennis, Match Prediction, Tournament Forecasts and other algorithms by Ultimate Tennis Statistics are licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

In short: Attribution is required. Non-commercial use only.

mcekovic / tennis-crystal-ball

readme