jemorriso / pysbr

PySBR

A Python client for accessing the Sportsbook Review GraphQL endpoint.

This library allows you to get odds and other information from Sportsbook Review (SBR) for any of the leagues that are supported on the website, including NFL, NCAAB, UFC, ATP, EPL and many others. For a given event, you can get the current, best, opening lines and line history for any combination of betting markets and sportsbooks available on the website.

The SBR GraphQL endpoint is undocumented and subject to change, so use at your own risk. Try and avoid getting rate limited so we can keep this available for everyone to use!

Project documentation found on ReadTheDocs

There is also a discord channel set up for this project that may be helpful.

Installation

pip install python-sbr

Examples

from pysbr import *
from datetime import datetime

Current lines for Sunday's NFL games

dt = datetime.strptime('2020-12-06', '%Y-%m-%d')

nfl = NFL()
sb = Sportsbook()

NFL and Sportsbook are examples of config classes that are provided so you don't need to remember various interal IDs, such as team, league, sport, sportsbook, or betting market ids. The full list of config classes can be found in the documentation.

Using the NFL config class to get the NFL's league id, the EventsByDate query can then be used to retrieve all the events on some date for a given league or sport.

e = EventsByDate(nfl.league_id, dt)

All queries have list and dataframe as available instance methods, producing a readable and formatted list or dataframe of the server response. You can also view the raw response from the server by accessing the _raw instance variable.

e.list()[0]

    {'scores': [],
     'sport id': 4,
     'league id': 16,
     'season id': 8582,
     'event id': 4143414,
     'description': 'Jacksonville Jaguars@Minnesota Vikings',
     'location': 'Minneapolis',
     'country': 'USA',
     'event status': 'scheduled',
     'datetime': '2020-12-06T10:00:00-08:00',
     'stadium type': 'artificial',
     'participants': [{'participant id': 1529,
       'is home': False,
       'source': {'name': 'Jacksonville',
        'nickname': 'Jaguars',
        'short name': 'Jacksonville',
        'abbreviation': 'JAC',
        'location': 'Jacksonville'}},
      {'participant id': 1541,
       'is home': True,
       'source': {'name': 'Minnesota',
        'nickname': 'Vikings',
        'short name': 'Minnesota',
        'abbreviation': 'MIN',
        'location': 'Minnesota'}}],
     'event group': {'event group id': 22,
      'name': 'REG Week 13',
      'alias': 'Week 13'}}

Then you can get the current lines for any betting markets and sportsbooks of interest by using CurrentLines.

cl = CurrentLines(e.ids(), nfl.market_ids(['1Qou', 'ps', '1st half moneyline']), sb.ids(['Pinnacle', 'Bovada']))

Notice how you can use the NFL config class to find the betting market ids for 1st quarter over/under, full game point spread, and 1st half moneyline. See the documentation for accepted formats. Similarly, you can use Sportsbook to find sportsbook ids.

cl.list(e)[0]

    {'market id': 91,
     'event id': 4143414,
     'sportsbook id': 20,
     'datetime': '2020-12-05T07:26:01-08:00',
     'participant id': 1529,
     'spread / total': 0,
     'decimal odds': 3.56,
     'american odds': 256,
     'event': 'Jacksonville Jaguars@Minnesota Vikings',
     'market': '1st Half - Draw No Bet',
     'sportsbook': 'Pinnacle',
     'participant': 'JAC'}

In order to get the 'event', 'market', 'sportsbook' and 'participant' fields, pass the event object into list or dataframe for any lines-related query.

---

If an event is already completed, 'scores' will be populated in the response. The score data can then be used to calculate the result of the bet.

dt = datetime.strptime('2020-11-30', '%Y-%m-%d')

e = EventsByDate(nfl.league_id, dt)
cl = CurrentLines(e.ids(), nfl.market_ids(['1Qou', 'ps', '1st half moneyline']), sb.ids(['Pinnacle', 'Bovada']))

cl.list(e)[:2]

    [{'market id': 91,
      'event id': 4143550,
      'sportsbook id': 20,
      'datetime': '2020-11-30T17:12:25-08:00',
      'participant id': 1536,
      'spread / total': 0,
      'decimal odds': 2.7,
      'american odds': 170,
      'event': 'Seattle Seahawks@Philadelphia Eagles',
      'market': '1st Half - Draw No Bet',
      'result': 'L',
      'profit': -100.0,
      'participant score': 17,
      'sportsbook': 'Pinnacle',
      'participant': 'PHI'},
     {'market id': 91,
      'event id': 4143550,
      'sportsbook id': 20,
      'datetime': '2020-11-30T17:12:25-08:00',
      'participant id': 1548,
      'spread / total': 0,
      'decimal odds': 1.5128,
      'american odds': -195,
      'event': 'Seattle Seahawks@Philadelphia Eagles',
      'market': '1st Half - Draw No Bet',
      'result': 'W',
      'profit': 51.28,
      'participant score': 23,
      'sportsbook': 'Pinnacle',
      'participant': 'SEA'}]

You can see that Seahawks won the 1st half by a score of 23-17, netting a profit of $51.28 on a $100 bet on Seahawks 1st half ML if you were to bet Pinnacle's closing line.

Opening lines for teams over date range

ncaab = NCAAB()
sb = Sportsbook()

start = datetime.strptime('2020-11-25', '%Y-%m-%d')
end = datetime.strptime('2020-12-03', '%Y-%m-%d')

For NCAA basketball all the teams have been added to the config class, so here you can use the team_ids method.

gonzaga = ncaab.team_id("GONZ")
baylor = ncaab.team_id("baylor")

EventsByParticipants includes any games that Gonzaga or Baylor played in over the given date range.

e = EventsByParticipants([gonzaga, baylor], start, end, ncaab.league_id)

Use the OpeningLines query to get the opening lines on Bovada for all the matching games:

ol = OpeningLines(e.ids(), ncaab.market_id('ps'), sb.id('Bovada'))

ol.list(e)[1]

    {'market id': 401,
     'event id': 4285569,
     'sportsbook id': 9,
     'datetime': '2020-11-23T17:55:49-08:00',
     'participant id': 651,
     'spread / total': -4,
     'decimal odds': 1.9091,
     'american odds': -110,
     'event': 'Gonzaga Bulldogs@Kansas Jayhawks',
     'market': 'Point Spread (Including OT)',
     'result': 'W',
     'profit': 90.91,
     'participant score': 102,
     'sportsbook': 'Bovada',
     'participant': 'GONZ'}

ol.list(e)[5]

    {'market id': 401,
     'event id': 4286230,
     'sportsbook id': 9,
     'datetime': '2020-12-01T15:10:19-08:00',
     'participant id': 687,
     'spread / total': -5.5,
     'decimal odds': 1.9091,
     'american odds': -110,
     'event': 'Illinois Fighting Illini@Baylor Bears',
     'market': 'Point Spread (Including OT)',
     'result': 'W',
     'profit': 90.91,
     'participant score': 82,
     'sportsbook': 'Bovada',
     'participant': 'BAY'}

Gonzaga opened as a 4 point favorite against Kansas, and Baylor opened as a 5.5 point favorite against Illinois.

Search for event and get line history

ufc = UFC()
sb = Sportsbook()

For some leagues there is no configuration class, or it doesn't have teams or participants. In this case SearchEvents comes in handy. You can search for an upcoming event by participant like so:

s = SearchEvents('vettori')

s.list()[0]

    {'description': 'Jack Hermansson@Marvin Vettori',
     'datetime': '2020-12-05T20:59:00-08:00',
     'league': 'UFC',
     'sport': 'fighting',
     'sport id': 9,
     'event id': 4293126,
     'participants': [{'psid': 66013,
       'rot': 24149,
       'event id': 4293126,
       'participant id': 66013,
       'is home': False,
       'source': {'team id': 66013,
        'league id': 26,
        'name': 'Jack Hermansson',
        'nickname': None,
        'short name': None,
        'abbreviation': None},
       'name': 'Jack Hermansson'},
      {'psid': 66261,
       'rot': 24150,
       'event id': 4293126,
       'participant id': 66261,
       'is home': True,
       'source': {'team id': 66261,
        'league id': 26,
        'name': 'Marvin Vettori',
        'nickname': None,
        'short name': None,
        'abbreviation': None},
       'name': 'Marvin Vettori'}],
     'league id': 26}

You can get the complete line movement history for an event for a given market and sportsbook. Here you can see the line moving in Vettori's favor.

vettori = s.list()[0]['participants'][1]['participant id']

lh = LineHistory(s.id(), ufc.default_market_id, sb.id('Pinnacle'), vettori)

lh.list(s)[0]

    {'market id': 126,
     'event id': 4293126,
     'sportsbook id': 20,
     'datetime': '2020-12-05T20:27:03-08:00',
     'participant id': 66261,
     'spread / total': 0,
     'decimal odds': 1.6944,
     'american odds': -144,
     'event': 'Jack Hermansson@Marvin Vettori',
     'market': '2way',
     'sportsbook': 'Pinnacle',
     'participant': 'Marvin Vettori'}

lh.list(s)[1]

    {'market id': 126,
     'event id': 4293126,
     'sportsbook id': 20,
     'datetime': '2020-12-05T20:20:20-08:00',
     'participant id': 66261,
     'spread / total': 0,
     'decimal odds': 1.7092,
     'american odds': -141,
     'event': 'Jack Hermansson@Marvin Vettori',
     'market': '2way',
     'sportsbook': 'Pinnacle',
     'participant': 'Marvin Vettori'}

Getting odds on Wimbledon futures

sb = Sportsbook()
atp = ATP()

Sometimes it may not be obvious how to find a given market id. In this case you can call sport_config on a league configuration class and search for the market you are looking for. See the documentation for other config methods.

print(atp.sport_config())

    {'default market id': 126, 'consensus market ids': [126, 395, 396], 'markets': [{'periods': 0, 'alias': '', 'market types': [{'alias': 'Money Lines', 'market id': 126, 'name': '2way', 'url': 'money-line'}, {'alias': 'Point Spreads', 'market id': 395, 'name': 'Point Spread', 'url': 'pointspread'}, {'alias': 'Total Games', 'market id': 396, 'name': 'American Total', 'url': 'totals'}], 'market group id': 217, 'name': 'Full Game', 'url': 'full-game'}, {'alias': 'Futures', 'market types': [{'alias': 'French Open Winner', 'market id': 719, 'name': 'French Open Winner', 'url': 'french-open-winner'}, {'alias': 'Wimbledon Winner', 'market id': 720, 'name': 'Wimbledon Winner', 'url': 'davis-cup-winner'}, {'alias': 'US Open Winner', 'market id': 721, 'name': 'US Open Winner', 'url': 'us-open-winner'}, {'alias': 'Australian Open Winner', 'market id': 723, 'name': 'Australian Open Winner', 'url': 'australian-open-winner'}], 'market group id': 224, 'name': 'Futures', 'url': 'futures'}], 'sport id': 8}

wimbledon_futures_id = 720

start = datetime.strptime('2021-06-28', '%Y-%m-%d')
end = datetime.strptime('2021-07-12', '%Y-%m-%d')

You can use EventsByDateRange to find Wimbledon's event id.

e = EventsByDateRange(atp.league_id, start, end)

wimbledon_eid = e.list()[0]['event id']

With Wimbledon's event id and the futures market id, you can get the current odds on Bet365 for each competitor to win Wimbledon.

cl = CurrentLines(wimbledon_eid, wimbledon_futures_id, sb.id('bet365'))

[o for o in cl.list(e) if o['participant'] in ['Dimitrov', 'Djokovic', 'Nadal']]

    [{'market id': 720,
      'event id': 4138561,
      'sportsbook id': 5,
      'datetime': '2020-04-04T10:46:32-07:00',
      'participant id': 5371,
      'spread / total': 0,
      'decimal odds': 67,
      'american odds': 6600,
      'event': '2021 Wimbledon Winner ATP',
      'market': 'Wimbledon Winner',
      'sportsbook': 'Bet365',
      'participant': 'Dimitrov'},
     {'market id': 720,
      'event id': 4138561,
      'sportsbook id': 5,
      'datetime': '2020-04-04T10:46:32-07:00',
      'participant id': 5373,
      'spread / total': 0,
      'decimal odds': 2.5,
      'american odds': 150,
      'event': '2021 Wimbledon Winner ATP',
      'market': 'Wimbledon Winner',
      'sportsbook': 'Bet365',
      'participant': 'Djokovic'},
     {'market id': 720,
      'event id': 4138561,
      'sportsbook id': 5,
      'datetime': '2020-04-04T10:46:32-07:00',
      'participant id': 5684,
      'spread / total': 0,
      'decimal odds': 8,
      'american odds': 700,
      'event': '2021 Wimbledon Winner ATP',
      'market': 'Wimbledon Winner',
      'sportsbook': 'Bet365',
      'participant': 'Nadal'}]

Development setup

Use Pipenv. Clone this repo, and then run pipenv install --dev --pre black to create a virtual environment with dev dependencies installed.

To run the test suite:

pipenv run pytest --cov=pysbr tests/

Inside conftest.py there are 3 global variables, QUERY_SERVER, WAIT_MEAN and WAIT_DEVIATION that you can change to actually query the SBR server when testing, otherwise the test suite will not query the server, it will use the saved cassettes.

Release History

• 0.3.2
  • FIX: Add fake-useragent to install-requires
• 0.3.1
  • FIX: HTTP error 463
• 0.3.0
  • ADD: 'participant full name' and 'sportsbook alias' keys on lines lists / dataframes
  • ADD: New league config classes
• 0.2.1
  • CHANGE: Update README
• 0.2.0
  • FIX: LineHistory bug
  • ADD: id methods for Query and Config classes
• 0.1.3
  • CHANGE: Update docs
• 0.1.0
  • Initial release

Meta

Jeremy Morrison – jeremymorrison.ca - [email protected]

Distributed under the MIT license. See LICENSE for more information.

github.com/JeMorriso/PySBR

Contributing

1. Fork it (https://github.com/JeMorriso/PySBR/fork)
2. Create your feature branch (git checkout -b feature/fooBar)
3. Commit your changes (git commit -am 'Add some fooBar')
4. Push to the branch (git push origin feature/fooBar)
5. Create a new Pull Request

pysbr's Issues

UConn Huskies missing?

Small thing:

ncaab = NCAAB()
ncaab.team_id("uconn") # Not there
ncaab.team_id("huskies") # Not there
ncaab._team_ids

Can't find UConn Huskies for the life of me... is it just missing?? If not, what pattern can I use to find their id?

Great job with this by the way!!

OpeningLines for MLB require indexing sb.ids()

Grabbed the CurrentLines for Pinnacle (MLB, sometime in June 2019) and it worked fine with sb.ids('Pinnacle'):
closemoneylines = CurrentLines(e.ids(),mlb.market_ids(['moneyline']),sb.ids('Pinnacle'))

Tried the same thing with 'OpeningLines' instead of 'CurrentLines' and I got the following error:
TypeError: Expected <class 'int'>, got [20]

I checked and saw Pinnacle's ID was 20, so I just indexed sb.ids('Pinnacle') as shown below:
openmoneylines = OpeningLines(e.ids(),mlb.market_ids(['moneyline']),sb.ids('Pinnacle')[0])

Not a big deal but just wanted to point it out in case anyone runs into this issue! Maybe I'm doing something wrong too haha

HTTPError: 463 Client Error

I've used this package in the past with success. But now I try running EventsByDateRange and get the following error:

HTTPError: 463 Client Error: for url: https://www.sportsbookreview.com/ms-odds-v2/odds-v2-service

Any idea what happened here? Did I get my IP banned...

What's the rate limit?

Maybe we can implement a method to avoid hitting it.

SBR Service is no longer working

Hello,

As of today, SBR odds service appears to no longer be working. Is anyone else having issues?

Thanks for all the work you have done on this project! It is much appreciated.

Line History for NBA Totals not returning

Hey Jeremy - great library. I was trying to pull line history for totals using the LineHistory function and it works for every other market except totals markets.

lh = LineHistory(4295653, 402, sb.id('BetOnline'), [1161, 1167])
lh.list()

lh = LineHistory(4295653, 401, sb.id('BetOnline'), [1161, 1167])
lh.list()

An example for you if you want to try.

Sportbook Names

Is there a reverse look up for Sportsbook Names being used? From the documentation, the examples are Bovada and Pinnacle. Using BestLines, the SportBook Ids can be obtained. How can one obtain the Sports Book names?

Thanks.

Dataframe scores

In Events queries, period / quarter scores are out of order. NaN where there should be scores has also been observed.

Finding missing IDs (especially sportsbook IDs)

Great stuff!

How did you go about getting sbr team ids? If you point me the way I can help fill in the missing data.

Issue with LineHistory() and multiple participants

When I run:
ol = LineHistory(max(e.ids()), ncaab.market_ids('ps')[0], sb.ids('Bovada')[0],[baylor[0],gonzaga[0]])
print(ol.dataframe())

I only get the dataframe for Gonzaga line history, and not both. Should the function be able to accept a list of Team IDs and return a full line history with both teams?

Filter by bookie functionality not working?(We found just bookie name field incorrect due to sister sites/old names)

Based on this snippit of code:

   # Example options found here: https://github.com/JeMorriso/PySBR/blob/main/pysbr/config/sportsbooks.yaml
    bookie_list = ['BetOnline', 'Bovada']
    game_type = ['moneyline']

    ### END OF Configurable variables ###

    dt = datetime.strptime(datetime.today().strftime('%Y-%m-%d'), '%Y-%m-%d')
    nfl = NFL()
    nba = NBA()
    sb = Sportsbook()
    e_nfl = EventsByDate(nfl.league_id, dt)
    e_nba = EventsByDate(nba.league_id, dt)
    cl_nfl = CurrentLines(e_nfl.ids(), nfl.market_ids(game_type), sb.ids(bookie_list))
    cl_nba = CurrentLines(e_nba.ids(), nba.market_ids(game_type), sb.ids(bookie_list))

I should not get getting match results like so from NBA:

        elif game.league == "NBA":
            print("NBA LEAGUE GAME PROCESSING")
            for match in cl_nba.list(e_nba):
                if match['event'].lower().find(str(game.team_chosen).lower()):  # match found
                    if match['american odds'] >= int(game.american_odds):
                        if isinstance(game.bookie_odds, int):  # If another bookie has better value replace again
                            if match['american odds'] > game.bookie_odds:
                                game.bookie = match['sportsbook']
                                game.bookie_odds = match['american odds']
                        else:
                            game.bookie = match['sportsbook']
                            game.bookie_odds = match['american odds']

Where SportsBetting is still in the list when I specifically filter for only ['BetOnline', 'Bovada']

Version Used: 0.2.1
Python Version: 3.9