-
Notifications
You must be signed in to change notification settings - Fork 4
/
mlb-runline-dataset-production.py
208 lines (128 loc) · 9.21 KB
/
mlb-runline-dataset-production.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
# -*- coding: utf-8 -*-
"""
Created 2023
@author: Quant Galore
"""
from datetime import datetime, timedelta
import pandas as pd
import statsapi
import requests
import numpy as np
import sqlalchemy
import mysql.connector
Short_Long_Names = pd.read_csv("short_long_mlb_names.csv")
def name_converter(short_name):
long_name = Short_Long_Names[Short_Long_Names["short_name"] == short_name]["long_name"]
if len(long_name) < 1 :
return np.nan
else:
return long_name.iloc[0]
API_KEY = "your prop-odds.com api key"
# =============================================================================
# Start
# =============================================================================
# This is the production dataset, designed to only append new values, as oppposed to having to constantly re-build the dataset
# So, we set the start date to 7 days prior. This way, we check all of the games in that period to include any days that we missed
begin_date = (datetime.today() - timedelta(days = 7)).strftime("%Y-%m-%d")
ending_date = (datetime.today() - timedelta(days = 1)).strftime("%Y-%m-%d")
Schedule = statsapi.schedule(start_date = begin_date, end_date = ending_date)
Schedule_DataFrame = pd.json_normalize(Schedule)
date_range = pd.date_range(start = begin_date, end = ending_date)
odds_list = []
# The spread market represents the "runline" bet
market = "spread"
for date in date_range:
date = date.strftime("%Y-%m-%d")
url = f"https://api.prop-odds.com/beta/games/mlb?date={date}&tz=America/Chicago&api_key={API_KEY}"
games_url = f"https://api.prop-odds.com/beta/games/mlb?date={date}&api_key={API_KEY}"
games = pd.json_normalize(requests.get(games_url).json()["games"])
if len(games) < 1:
continue
for game_id in games["game_id"]:
Game = games[games["game_id"] == game_id]
sportsbook = []
odds_url = f"https://api.prop-odds.com/beta/odds/{game_id}/{market}?api_key={API_KEY}"
odds = requests.get(odds_url).json()
if len(odds) < 2:
continue
else:
# DraftKings generally offers the best odds, so for uniformity, we only include odds sourced from DraftKings
for book in odds["sportsbooks"]:
if book["bookie_key"] == "draftkings":
sportsbook = book
else:
continue
if len(sportsbook) < 1:
continue
odds_data = pd.json_normalize(sportsbook["market"]["outcomes"])
# The runline (-1.5) refers to the favorite winning by 2 or more points, so we have to first pull who the favorite is
moneyline_url = f"https://api.prop-odds.com/beta/odds/{game_id}/moneyline?api_key={API_KEY}"
moneyline_odds = requests.get(moneyline_url).json()
if len(moneyline_odds) < 2:
continue
else:
for moneyline_book in moneyline_odds["sportsbooks"]:
if moneyline_book["bookie_key"] == "draftkings":
moneyline_sportsbook = moneyline_book
else:
continue
if len(moneyline_sportsbook) < 1:
continue
moneyline_odds_data = pd.json_normalize(moneyline_sportsbook["market"]["outcomes"])
if moneyline_odds_data["odds"].max() < 0:
continue
moneyline_favorite = moneyline_odds_data[moneyline_odds_data["odds"] < 0].sort_values(by = "timestamp", ascending = True).head(1)["name"].iloc[0]
moneyline_underdog = moneyline_odds_data[moneyline_odds_data["odds"] > 0].sort_values(by = "timestamp", ascending = True).head(1)["name"].iloc[0]
# We sort by earliest available pre-game odds first, since the API may occasionally include odds that were set mid-game.
favorite = odds_data[(odds_data["handicap"] == -1.5) & (odds_data["name"] == moneyline_favorite)].sort_values(by = "timestamp", ascending = True).head(1)
underdog = odds_data[(odds_data["handicap"] == 1.5) & (odds_data["name"] == moneyline_underdog)].sort_values(by = "timestamp", ascending = True).head(1)
if len(favorite) < 1:
continue
elif len(underdog) < 1:
continue
team_1_favorite = favorite["name"].drop_duplicates().iloc[0]
team_2_underdog = underdog["name"].drop_duplicates().iloc[0]
team_1_favorite_odds = favorite["odds"].iloc[0]
team_2_underdog_odds = underdog["odds"].iloc[0]
odds_dataframe = pd.DataFrame([[team_1_favorite, team_1_favorite_odds, team_2_underdog, team_2_underdog_odds]],
columns = ["team_1", "team_1_spread_odds", "team_2", "team_2_spread_odds"])
full_odds_dataframe = pd.concat([Game.reset_index(drop = True), odds_dataframe], axis = 1)
if len(full_odds_dataframe) > 1:
continue
odds_list.append(full_odds_dataframe)
full_odds = pd.concat(odds_list).reset_index(drop = True).rename(columns = {"away_team":"away_name",
"home_team":"home_name",
"start_timestamp":"game_datetime"})
Merged_DataFrame = pd.merge(Schedule_DataFrame, full_odds, on = ["game_datetime", "away_name", "home_name"])
Merged_DataFrame["team_1"] = Merged_DataFrame["team_1"].apply(name_converter)
Merged_DataFrame["team_2"] = Merged_DataFrame["team_2"].apply(name_converter)
Featured_Merged_DataFrame = Merged_DataFrame[["game_datetime","away_name","home_name","away_score","home_score", "team_1", "team_1_spread_odds", "team_2", "team_2_spread_odds", "venue_name", "winning_team"]].copy().set_index("game_datetime")
# "team_1" always represents the favorite
# If the favorite was the home team, then home score - away score gives us the spread -- vice versa if the favorite is the away team
team_1_away_wins = Featured_Merged_DataFrame[Featured_Merged_DataFrame["away_name"] == Featured_Merged_DataFrame["team_1"]].copy()
team_1_away_wins["spread"] = team_1_away_wins["away_score"] - team_1_away_wins["home_score"]
team_1_home_wins = Featured_Merged_DataFrame[Featured_Merged_DataFrame["home_name"] == Featured_Merged_DataFrame["team_1"]].copy()
team_1_home_wins["spread"] = team_1_home_wins["home_score"] - team_1_home_wins["away_score"]
spread_dataframe = pd.concat([team_1_away_wins, team_1_home_wins], axis = 0)
def spread_converter(spread):
if spread >= 2:
return 1
else:
return 0
# If the favorite won the game by 2 or more points, we assign a 1
# If the favorite wins by less than 2 points, or if the underdog wins, we assign a 0
spread_dataframe["spread"] = spread_dataframe["spread"].apply(spread_converter)
Featured_Spread_DataFrame = spread_dataframe[["team_1", "team_1_spread_odds", "team_2", "team_2_spread_odds", "venue_name", "spread"]].copy().reset_index().set_index("game_datetime")
# To weed out any errors in the data set, we ensure to only include data where the odds are between -200 to +200
# The odds for these bets are almost never set outside of that range, so we exclude them.
Featured_Spread_DataFrame = Featured_Spread_DataFrame[(abs(Featured_Spread_DataFrame["team_1_spread_odds"]) < 200) & (abs(Featured_Spread_DataFrame["team_2_spread_odds"]) < 200)]
Featured_Spread_DataFrame = Featured_Spread_DataFrame[Featured_Spread_DataFrame["team_1"] != Featured_Spread_DataFrame["team_2"]]
Featured_Spread_DataFrame.index = pd.to_datetime(Featured_Spread_DataFrame.index).tz_convert("America/Chicago")
# We initialize our sqlalchemy engine, then submit the data to the database
engine = sqlalchemy.create_engine('mysql+mysqlconnector://username:password@database-host-name:3306/database-name')
existing_data = pd.read_sql("SELECT * FROM baseball_spread", con = engine)
existing_data["unique_indentifier"] = existing_data["game_datetime"].astype(str) + existing_data["team_1"].astype(str) + existing_data["team_2"].astype(str) + existing_data["venue_name"].astype(str)
Featured_Spread_DataFrame["unique_indentifier"] = Featured_Spread_DataFrame.index.strftime("%Y-%m-%d %H:%M:%S") + Featured_Spread_DataFrame["team_1"].astype(str) + Featured_Spread_DataFrame["team_2"].astype(str) + Featured_Spread_DataFrame["venue_name"].astype(str)
new_data = Featured_Spread_DataFrame[~Featured_Spread_DataFrame["unique_indentifier"].isin(existing_data["unique_indentifier"])]
new_data = new_data.drop("unique_indentifier", axis = 1)
new_data.to_sql("baseball_spread", con = engine, if_exists = "append")