Hackathon Prompts

Submission PowerPoint template: Finalist_Presentation_Template

 

Instructions: Please select one of the prompts below to work on. If picking the open prompt, you must acquire approval from the NBA Basketball Analytics team (Jason Rosenfeld, Philip Chang, Max Horowitz, Patrick Harrel) before beginning work and before 10:00am. Failure to do so may result in disqualification.

  1. Develop a new method or tool for evaluation of defensive performance in the NBA.
    1. Some ideas include evaluation of players by age (e.g. older players’ tendencies vs. younger players’ tendencies), a visualization / video tool to aid in tracking the associated data, and / or a new and tested metric (e.g. “Defensive PER”)
  2. Develop a new method or tool for evaluation of the effectiveness of timeouts as an offensive or defensive strategy.
    1. For example, explore getting rid of certain timeouts and the residual effects, or the tendencies of offenses or defenses after timeouts depending on which team initiated the timeout (including coaching strategy and play results)
  3. Build a tool and / or model to predict the outcomes of shots attempted.
    1. Some ideas include predicting the make / miss outcome of each shot, the potential success of different types of shots that are generated from live game play, or the rebounds that occur as a result of missed shots
  4. Open Topic: Participants are allowed to pursue a creative, original topic.
    1. Guidelines: Solutions / tools must have realistic and practical application for the basketball community.
    2. Please remember that no prepared work may be utilized during the Hackathon; all code and other components must be created at the event.
    3. All topics must be approved by NBA League Office personnel at the event in advance of beginning any work on the proposed topic. Failure to receive approval will result in disqualification from the competition.
    4. In addition to the defense, timeout, and player lineup subjects outlined in the prompts, possible subjects include scheduling and methods to quantify performance outside of traditional statistics (e.g. with Hustle Stats).

Judging Criteria

Five (5) finalists will be selected based on the following criteria:

  1. Originality and creativity of analysis – 25%
  2. Strategic and practical implementation – 25%
  3. Replicability of analysis and organization of code – 25%
  4. Effective communication and summary of results – 15%
  5. Cohesive writing that clearly demonstrates results – 10%

Three (3) award winners will be judged from the finalists based on the following criteria:

  1. Quality of presentation – 25%
  2. Quality of slides – 15%
  3. Results and methods – 30%
  4. Accuracy of results – 30%