handycapper

Releases the data from horse racing charts so that it may be explored in fun and creative ways.

Includes:

• a native desktop app that parses PDF charts, saves the data within to an embedded database, and provides a friendly experience for viewing and searching the data
• exporting the data to Microsoft Excel
• exposing the data via a JSON-based REST API
• a Java SDK for programmatic access to the chart data and/or database
• an SQL editor to execute queries against the database

Version 1.0.1 released on May 24th 2018 to account for a long owner name at SA May 20th 2018 Race 3.

TL;DR

An Equibase PDF chart results file e.g.

...loaded via the handycapper app:

...allows the chart to be consumed in a simpler manner:

...to present new insights:

...be viewed from different perspectives:

...and more easily discovered:

Should you wish to go deeper, export the data and analyze with familiar tools like Excel:

Got a creative streak? Experiment and innovate by integrating the data into new tools and visualizations with the built-in REST API:

User interfaces not your thing? No problem - use the SDK to create software that can work with the chart data and/or the database directly:

And conduct research at a granular level with the full power of SQL:

Quickstart

1. Download either Handycapper.app.zip (MacOS) or Handycapper-1.0.1.exe.zip (Windows) from Releases, unzip, install and start the application.
2. Load an Equibase PDF chart file (for either a single race or even a full raceday) e.g. 2017 Breeders' Cup Day 2
3. Double-click on a row to open the race view:

Note: the MacOS .app has been signed with my Apple Developer certificate. The .exe is not signed and may prompt warnings from your Windows OS. handycapper does not require an external network or internet connection to function.

Technical documentation

• Database (include schema)
• REST API endpoints
• JSON API design

FAQ

• Who created this?

  Robin Howlett

• Why?

  When I moved to the US from Ireland, I was very familiar with horse racing but found the US racing system's size, scope, nomenclature and organization difficult to comprehend. I wanted to see how it all worked - what were the racing circults, how healthy was the industry, how did trainers operate etc. I felt the source data would tell me the story, but the racing data industry's business model is geared towards handicapping software that requires a good deal of prior knowledge to operate effectively and primarily for wagering purposes. I just wanted to explore the raw data and see where it led me.

  After connecting with some racing folks on Twitter, I felt a gap existed for regular racing fans to interact with racing data in a highly-flexible, multi-faceted manner:

  • to provide the simplest possible solution for someone to get started; users should be able to explore racing data with a one-click installation
  • viewing the chart with an easy-to-use interface that presented the data in a less intimidating manner for less experienced fans
  • using Excel to perform some ad-hoc calculations of their own
  • exposing an API so that new racing experiences, experiments, and media could be created by those perhaps outside of the racing industry today
  • an SDK to provide a reference data model to encourage software engineers to consider horse racing as an accessible
  • a database for fine-grained data-driven research and to highlight the rich, data- and wagering-centric aspects of the sport

  Also, I enjoy learning new technologies using racing as a domain, and I wanted to demonstrate my product and technical prowess.

• How do you get the data from the chart/PDF?

  I wrote Chart Parser. It basically figures out the meaning of every character in the PDF by its location, value and/or context. By grouping the related characters together, a data model can be built. handycapper takes that data model and saves it to a relational database. Once its in a DB, the world's your oyster.

• How much does it cost?

  Nothing. It is open-source and MIT licensed.

• What can I do with the data?

  Both handycapper and Chart Parser are designed for internal personal non-commercial use only. The data that Equibase provides in the PDF charts belongs to them and you should be aware of and respect their terms of use.

  This project contains and references a single sample Equibase PDF chart included for testing, educational and demonstration purposes only.

• What technologies does it use?

  Lots, but primarily Java 8, Spring Boot, JavaFX & ControlsFX, H2, jOOQ, Apache PDFBox, Apache POI, Flyway, Jetty, Jackson, Bouncy Castle

• Can I use a different database engine?

  Yes - both MySQL and PostgreSQL have support out of the box, though to use them you must build from source via Maven and the mysql and postgresql Maven Profiles declared in the pom.xml file respectively. Usernames and password configuration in application-mysql.properties and application-postgresql.properties will need to be modified. The default Maven and Spring profile is h2.

• Should I use this for handicapping and wagering decisions?

  If you wish but I would strongly discourage it. The professional products out there (Equibase, OptixEQ, STATS Race Lens, DRF Formulator etc.) are designed for this and you should buy and use those instead.

• Can I pay you?

  Nah. Donate to a charity instead.

Feedback

Feedback, requests, bugs etc. can be sent to me via:

• robin.howlett at gmail.com
• Twitter: @robinhowlett