Skip to content


Switch branches/tags

Name already in use

A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?


Failed to load latest commit information.
Latest commit message
Commit time
Documentation Status



Scrape NHL data off the NHL API and website. This includes the Play by Play and Shift data for each game and the schedule information. It currently supports all preseason, regular season, and playoff games from the 2007-2008 season onwards.


You are going to need to have python installed for this. This should work for both python 2.7 and 3. I recommend having from at least version 3.6.0 but earlier versions should be fine.


To install all you need to do is open up your terminal and run:

pip install hockey_scraper

NHL Usage

The full documentation can be found here.

Standard Scrape Functions

Scrape data on a season by season level:

import hockey_scraper

# Scrapes the 2015 & 2016 season with shifts and stores the data in a Csv file
hockey_scraper.scrape_seasons([2015, 2016], True)

# Scrapes the 2008 season without shifts and returns a dictionary containing the pbp Pandas DataFrame
scraped_data = hockey_scraper.scrape_seasons([2008], False, data_format='Pandas')

Scrape a list of games:

import hockey_scraper

# Scrapes the first game of 2014, 2015, and 2016 seasons with shifts and stores the data in a Csv file
hockey_scraper.scrape_games([2014020001, 2015020001, 2016020001], True)

# Scrapes the first game of 2007, 2008, and 2009 seasons with shifts and returns a Dictionary with the Pandas DataFrames
scraped_data = hockey_scraper.scrape_games([2007020001, 2008020001, 2009020001], True, data_format='Pandas')

Scrape all games in a given date range:

import hockey_scraper

# Scrapes all games between 2016-10-10 and 2016-10-20 without shifts and stores the data in a Csv file
hockey_scraper.scrape_date_range('2016-10-10', '2016-10-20', False)

# Scrapes all games between 2015-1-1 and 2015-1-15 without shifts and returns a Dictionary with the pbp Pandas DataFrame
scraped_data = hockey_scraper.scrape_date_range('2015-1-1', '2015-1-15', False, data_format='Pandas')

The dictionary returned by setting the default argument "data_format" equal to "Pandas" is structured like:

  # Both of these are always included
  'pbp': pbp_df,

  # This is only included when the argument 'if_scrape_shifts' is set equal to True
  'shifts': shifts_df


The schedule for any past or future games can be scraped as follows:

import hockey_scraper

# As oppossed to the other calls the default format is 'Pandas' which returns a DataFrame
sched_df = hockey_scraper.scrape_schedule("2019-10-01", "2020-07-01")

The columns returned are: ['game_id', 'date', 'venue', 'home_team', 'away_team', 'start_time', 'home_score', 'away_score', 'status']

Persistent Data

All the raw game data files retrieved can also be saved to your disk. This allows for faster rescraping (we don't need to re-retrieve them) and the ability to parse the data yourself.

This is achieved by setting the keyword argument docs_dir=True. This will store the data in a directory called ~/hockey_scraper_data. You can provide your own directory where you want everything to be stored (it must exist beforehand). By default docs_dir=False.

For example, let's say we are scraping the JSON PBP data for game 2019020001. If docs_dir isn't False it will first check if the data is already in the directory. If so, it will load in the data from that file and not make a GET request to the NHL API. However if it doesn't exist, it will make a GET request and then save the output to the directory. This will ensure that next time you are requesting that data it can load it from a file.

Here are some examples.

The default saving location is ~/hockey_scraper_data.

# Create or try to refer to a directory in the home directory
# Will create a directory called 'hockey_scraper_data' in the home directory (if it doesn't exist)
hockey_scraper.scrape_seasons([2015, 2016], True, docs_dir=True)

User defined directory

USER_PATH = "/...."
hockey_scraper.scrape_seasons([2015, 2016], True, docs_dir=USER_PATH)

You can override the existing files by specifying rescrape=True. It will retrieve all the files from source and save the newer versions to docs_dir.

hockey_scraper.scrape_seasons([2015, 2016], True, docs_dir=USER_PATH, rescrape=True)

Live Scraping

Here is a simple example of a way to setup live scraping. I strongly suggest checking out this section of the docs if you plan on using this.

import hockey_scraper as hs

def to_csv(game):
    Store each game DataFrame in a file

    :param game: LiveGame object

    :return: None

    # If the game:
    # 1. Started - We recorded at least one event
    # 2. Not in Intermission
    # 3. Not Over
    if game.is_ongoing():
        # Print the description of the last event
        print(game.game_id, "->", game.pbp_df.iloc[-1]['Description'])

        # Store in CSV files
        game.pbp_df.to_csv(f"../hockey_scraper_data/{game.game_id}_pbp.csv", sep=',')
        game.shifts_df.to_csv(f"../hockey_scraper_data/{game.game_id}_shifts.csv", sep=',')

if __name__ == "__main__":
    # B4 we start set the directory to store the files
    # You don't have to do this but I recommend it

    # Scrape the info for all the games on 2018-11-15
    games = hs.ScrapeLiveGames("2018-11-15", if_scrape_shifts=True, pause=20)

    # While all the games aren't finished
    while not games.finished():
        # Update for all the games currently being played

        # Go through every LiveGame object and apply some function
        # You can of course do whatever you want here.
        for game in games.live_games:


Please contact me for any issues or suggestions. For any bugs or anything related to the code please open an issue. Otherwise you can email me at


Copyright (C) 2019-2022 Harry Shomer
This file is part of hockey_scraper

hockey_scraper is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with this program.  If not, see <>.