Getting Started

First…

  • Step 1:

    • Fork/Clone this GitHub repository to your local machine

    • Make sure you have Python >= 3.10

  • Step 2:

    • pip install -r requirements.txt

  • Step 3:
    • Set the following environment variables (this can be done via terminal or in your IDE):

      • api_key

      • secret_key

      • log_path (defaults to ./)

      • log_level (defaults to 10 (debug), 20=info, 30=warning, 40=error, 50=critical) (recommended: 20)

    • Setting env vars with terminal:
      • If using bash:

        • export var_name=value

      • In python:

        • os.environ[var_name] = value

    • You can also just set these values directly in the StrategyConfig, but it is recommended to use env vars to store api keys safely.

  • Step 4 (TRADE!):

    • open main.py

if __name__ == '__main__':
    config = StrategyConfig()
    config.event_id = 'evt_xxx'
    #  adjust other parameters as you would like. See details below.
    strat = Strategy(config)
    trade_strategy(strat)

Note

Replace the strategy import above with one of the prebuilt strategies, or your own!

See Strategies: for more information on strategies (custom and prebuilt)

See StrategyConfig Class: for more information on setting up your config.

  • Setting your StrategyConfig

    • event_id: the event id for the event in which you want to trade

    • order_size: optional, the number of shares per order. Some included strategies will allow you to

    • array_len: length of numpy array storing historical websocket data. Useful if you are using moving averages, etc.

    • always_update: whether you want to update array when there is no new information

    • message_consumers: number of async message consumers you want. Defaults to 12, more may be necessary for large events.

    • phase: a list containing one or both of ‘live’, ‘ipo’

    • log_level: logging levels in accordance with python logging package.

    • log_path: where the user wants to store logs in /log folder. Defaults to current directory

    • max_spend: maximum dollar amount the user wants to invest in this event

    • max_position: maximum shares of a player the user wants to own. The jockbot will never buy more than this number of any one player.

    • print_to_console: whether or not to print log messages to console.

  • Step 5: