Applying Weighting to the Goals Models#
[1]:
import sys
sys.path.append("../../")
import penaltyblog as pb
Get data from football-data.co.uk#
[2]:
fb = pb.scrapers.FootballData("ENG Premier League", "2019-2020")
df = fb.get_fixtures()
df.head()
[2]:
| date | datetime | season | competition | div | time | team_home | team_away | fthg | ftag | ... | b365_cahh | b365_caha | pcahh | pcaha | max_cahh | max_caha | avg_cahh | avg_caha | goals_home | goals_away | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| id | |||||||||||||||||||||
| 1565308800---liverpool---norwich | 2019-08-09 | 2019-08-09 20:00:00 | 2019-2020 | ENG Premier League | E0 | 20:00 | Liverpool | Norwich | 4 | 1 | ... | 1.91 | 1.99 | 1.94 | 1.98 | 1.99 | 2.07 | 1.90 | 1.99 | 4 | 1 |
| 1565395200---bournemouth---sheffield_united | 2019-08-10 | 2019-08-10 15:00:00 | 2019-2020 | ENG Premier League | E0 | 15:00 | Bournemouth | Sheffield United | 1 | 1 | ... | 1.95 | 1.95 | 1.98 | 1.95 | 2.00 | 1.96 | 1.96 | 1.92 | 1 | 1 |
| 1565395200---burnley---southampton | 2019-08-10 | 2019-08-10 15:00:00 | 2019-2020 | ENG Premier League | E0 | 15:00 | Burnley | Southampton | 3 | 0 | ... | 1.87 | 2.03 | 1.89 | 2.03 | 1.90 | 2.07 | 1.86 | 2.02 | 3 | 0 |
| 1565395200---crystal_palace---everton | 2019-08-10 | 2019-08-10 15:00:00 | 2019-2020 | ENG Premier League | E0 | 15:00 | Crystal Palace | Everton | 0 | 0 | ... | 1.82 | 2.08 | 1.97 | 1.96 | 2.03 | 2.08 | 1.96 | 1.93 | 0 | 0 |
| 1565395200---tottenham---aston_villa | 2019-08-10 | 2019-08-10 17:30:00 | 2019-2020 | ENG Premier League | E0 | 17:30 | Tottenham | Aston Villa | 3 | 1 | ... | 2.10 | 1.70 | 2.18 | 1.77 | 2.21 | 1.87 | 2.08 | 1.80 | 3 | 1 |
5 rows × 111 columns
Create the Weights#
[4]:
xi = 0.001
weights = pb.models.dixon_coles_weights(df["date"], xi)
Train the model#
[5]:
clf = pb.models.DixonColesGoalModel(
df["goals_home"], df["goals_away"], df["team_home"], df["team_away"], weights
)
clf.fit()
The model’s parameters#
[6]:
clf
[6]:
Module: Penaltyblog
Model: Dixon and Coles
Number of parameters: 42
Log Likelihood: -879.671
AIC: 1843.343
Team Attack Defence
------------------------------------------------------------
Arsenal 1.145 -0.96
Aston Villa 0.821 -0.642
Bournemouth 0.824 -0.649
Brighton 0.773 -0.836
Burnley 0.869 -0.941
Chelsea 1.348 -0.812
Crystal Palace 0.531 -0.912
Everton 0.898 -0.813
Leicester 1.283 -1.055
Liverpool 1.537 -1.264
Man City 1.727 -1.255
Man United 1.31 -1.242
Newcastle 0.781 -0.768
Norwich 0.32 -0.527
Sheffield United 0.756 -1.158
Southampton 1.069 -0.761
Tottenham 1.218 -0.972
Watford 0.733 -0.666
West Ham 1.026 -0.708
Wolves 1.031 -1.148
------------------------------------------------------------
Home Advantage: 0.239
Rho: -0.078
[7]:
clf.get_params()
[7]:
{'attack_Arsenal': np.float64(1.144949905929318),
'attack_Aston Villa': np.float64(0.8211833050535601),
'attack_Bournemouth': np.float64(0.8238000396326658),
'attack_Brighton': np.float64(0.7730705244669612),
'attack_Burnley': np.float64(0.8689343441725552),
'attack_Chelsea': np.float64(1.3475855486891566),
'attack_Crystal Palace': np.float64(0.5309831147050881),
'attack_Everton': np.float64(0.898409705429681),
'attack_Leicester': np.float64(1.282663505849936),
'attack_Liverpool': np.float64(1.5372296425830276),
'attack_Man City': np.float64(1.7267149046878532),
'attack_Man United': np.float64(1.310318077147705),
'attack_Newcastle': np.float64(0.7810821985130814),
'attack_Norwich': np.float64(0.32031537569664587),
'attack_Sheffield United': np.float64(0.756397077539036),
'attack_Southampton': np.float64(1.068666804409662),
'attack_Tottenham': np.float64(1.2184558429326078),
'attack_Watford': np.float64(0.7325661771016169),
'attack_West Ham': np.float64(1.0260467035815914),
'attack_Wolves': np.float64(1.0306272018767515),
'defence_Arsenal': np.float64(-0.9597520519607475),
'defence_Aston Villa': np.float64(-0.6424612731039607),
'defence_Bournemouth': np.float64(-0.6485814937773848),
'defence_Brighton': np.float64(-0.8360247260429235),
'defence_Burnley': np.float64(-0.9412626602561426),
'defence_Chelsea': np.float64(-0.8121173879097644),
'defence_Crystal Palace': np.float64(-0.911934038315484),
'defence_Everton': np.float64(-0.8128562445182332),
'defence_Leicester': np.float64(-1.0548116769925087),
'defence_Liverpool': np.float64(-1.2639681889819574),
'defence_Man City': np.float64(-1.2550748290155953),
'defence_Man United': np.float64(-1.2420940177658861),
'defence_Newcastle': np.float64(-0.7679483380308282),
'defence_Norwich': np.float64(-0.5269116672528777),
'defence_Sheffield United': np.float64(-1.1579048379110184),
'defence_Southampton': np.float64(-0.7609806462356308),
'defence_Tottenham': np.float64(-0.9715679787843047),
'defence_Watford': np.float64(-0.6656257049737819),
'defence_West Ham': np.float64(-0.7077751518777993),
'defence_Wolves': np.float64(-1.148290961379967),
'home_advantage': np.float64(0.23856467457373232),
'rho': np.float64(-0.07846937040455315)}
Predict Match Outcomes#
[8]:
probs = clf.predict("Liverpool", "Wolves")
probs
[8]:
Module: Penaltyblog
Class: FootballProbabilityGrid
Home Goal Expectation: [1.8729287]
Away Goal Expectation: [0.79188351]
Home Win: 0.6215177796020613
Draw: 0.23427193067485133
Away Win: 0.14421028809547223
1x2 Probabilities#
[9]:
probs.home_draw_away
[9]:
[np.float64(0.6215177796020613),
np.float64(0.23427193067485133),
np.float64(0.14421028809547223)]
[10]:
probs.home_win
[10]:
np.float64(0.6215177796020613)
[11]:
probs.draw
[11]:
np.float64(0.23427193067485133)
[12]:
probs.away_win
[12]:
np.float64(0.14421028809547223)
Probablity of Total Goals >1.5#
[13]:
probs.total_goals("over", 1.5)
[13]:
np.float64(0.7529851084800863)
Probability of Asian Handicap 1.5#
[14]:
probs.asian_handicap("home", 1.5)
[14]:
np.float64(0.37547110035717224)
Probability of both teams scoring#
[15]:
probs.both_teams_to_score
[15]:
np.float64(0.4710502787522432)