Elopy
This is a module to allow you to maintain state for teams as you progress them through games played against opponents. This uses the the constants set out from 538's NBA Elo. But you could inherit and overload the functions to work for any sport you wanted.
To Install:
pip install elopy
Code Examples:
from elopy.elo import Elo
team_a = Elo(start_elo=1600, k=20, hca = 100)
team_b = Elo(start_elo=1400, k=20, hca = 100)
#To get the win probability of team_a beating team_b at home
team_a_probs = team_a.win_probs(team_b, is_home=True)
#To get the win probability of team_b beating team_a at home
team_b_probs = team_b.win_probs(team_a, is_home=True)
#If you want to get away probabilities then set is_home to False
#To get the point spread of team_a vs. team_b. Positive represents an underdog
#negative represents a favorite
#at home
point_spread_home = team_a.point_spread(team_b, is_home=True)
#not at home. Will return a value
point_spread_away = team_a.point_spread(team_b, is_home=False)
#update elo's after a team has played the other. Let's say team a beat team b
#by 15 points as a visitor
team_a.play_game(team_b, 15, is_home=False)
#This will update both team a and team b's Elo ratings. If you run the same line
as above again it will be as if both teams are playing a second game
team_a.play_game(team_b, 15, is_home=False)
#This means that team a played team b again and beat them by 15 points again as
the visitor
#Then to get each team's Elo rating just access the elo attribute
print(team_a.elo)
print(team_b.elo)
