novig-liquidity 1.0.3

A Python wrapper for the Novig API with built-in filtering and liquidity validation.

Novig Liquidity

A Python wrapper around the Novig API with built-in filtering and validation.
This package helps you fetch sports market data and filter it based on liquidity and stat types.

Installation

pip install novig-liquidity

Requirements

Dependencies are listed in requirements.txt and are installed automatically. Core ones include:

• aiohttp
• redis
• pydantic
• requests

Usage

Create a filters.json file defining which stat types you care about. Example:

{
  "WNBA": [
    {
      "raw_name": "points_rebounds_assists",
      "display_name": "Points, Rebounds, and Assists",
      "active": true
    },
    {
      "raw_name": "points",
      "display_name": "Points",
      "active": true
    },
    {
      "raw_name": "rebounds_assists",
      "display_name": "Rebounds and Assists",
      "active": true
    },
    {
      "raw_name": "points_assists",
      "display_name": "Points and Assists",
      "active": true
    },
    {
      "raw_name": "points_rebounds",
      "display_name": "Points and Rebounds",
      "active": true
    },
    {
      "raw_name": "three_pointers_made",
      "display_name": "Three Pointers Made",
      "active": true
    },
    {
      "raw_name": "double_double",
      "display_name": "Double-Double",
      "active": true
    }
  ],
  "NFL": [
    {
      "raw_name": "rushing_yards",
      "display_name": "Rushing Yards",
      "active": true
    },
    {
      "raw_name": "rushing_attempts",
      "display_name": "Rushing Attempts",
      "active": true
    },
    {
      "raw_name": "longest_rush",
      "display_name": "Longest Rush",
      "active": true
    },
    {
      "raw_name": "passing_yards",
      "display_name": "Passing Yards",
      "active": true
    },
    {
      "raw_name": "receiving_yards",
      "display_name": "Receiving Yards",
      "active": true
    },
    {
      "raw_name": "receptions",
      "display_name": "Receptions",
      "active": true
    },
    {
      "raw_name": "longest_reception",
      "display_name": "Longest Receptions",
      "active": true
    },
    {
      "raw_name": "passing_completions",
      "display_name": "Passing Completions",
      "active": true
    }
  ]
}

Example script

import asyncio
import json
from Config import Novig

if __name__ == "__main__":
    # Load filters from file
    with open("filters.json", "r") as f:
        filters = json.load(f)

    # Choose your filter type and amounts
    total_and_difference_filter = {
        # "filter_type": "total_difference",
        "filter_type": "total_and_difference",
        "difference_amount": 3000,
        "highest_order_amount": 1500
    }
    
    # OR
    
    total_difference_filter = {
        "filter_type": "total_difference",
        "difference_amount": 3000,
    }
        
    

    # Create Novig instance
    novig = Novig(filters=filters, filter_amount_dict=total_and_difference_filter)

    # Run it
    results = asyncio.run(novig.run())
    print(results)

if __name__ == "__main__":
   raw = asyncio.run(Novig.get_raw_data(["NFL"]))
   import json
   with open("raw.json", "w") as f:
        json.dump(raw, f, indent=4)

Filter Types

Two filter types are currently supported:

1. total_difference
   Keeps markets where the liquidity difference between over/under is at least difference_amount.

2. total_and_difference
   Keeps markets where:

   • Liquidity difference ≥ difference_amount, and
   • Either over/under side has a highest order with liquidity_left ≥ highest_order_amount.

Output

novig.run() returns a dict keyed by league name (e.g., "WNBA", "NFL"), each with a list of filtered market entries. Each entry includes:

• key_name (market description)
• liquidity (grouped by over/under, includes highest_order, totals, etc.)
• additional_data (player name, stat type, line, game title, start time)

Features

• Async API calls via aiohttp
• Filtering on liquidity differences and totals
• Stat-type validation via your filters.json
• Clean models (Pydantic + dataclasses where applicable)