In [1]:

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

from footyscripts import footyviz

#plotting settings
%matplotlib inline

In [2]:

matches = ['2013-08-3266-Barcelona_VS_Levante',
           '2013-08-3278-Málaga_VS_Barcelona',
           '2014-08-8360-Burnley_VS_Chelsea',
           '2014-08-8370-Chelsea_VS_Leicester']

df = pd.read_csv("../datasets/_opta/_opta_data_.csv", encoding='utf-8', index_col=0)
df = df[df['match'].isin(matches)].sort(['match', '@minsec'])
df = df[df['@player_id']==306]
df.fillna(False, inplace=True)

In [3]:

#add to footyviz !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
x_size = 105.0
y_size = 68.0
df['x']=df['x']/100*x_size 
df['y']=df['y']/100*y_size
df['to_x']=df['to_x']/100*x_size
df['to_y']=df['to_y']/100*y_size

-c:4: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_index,col_indexer] = value instead
-c:5: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_index,col_indexer] = value instead
-c:6: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_index,col_indexer] = value instead
-c:7: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_index,col_indexer] = value instead

In [4]:

fig, ax = footyviz.draw_pitch()
footyviz.draw_events(df[(df.type=='all_passes') & (df.team_name=='Barcelona')], base_color='red', alpha=0.8)
footyviz.draw_events(df[(df.type=='all_passes') & (df.team_name=='Chelsea')], base_color='#235c96', alpha=0.8)

fig.savefig('fabregas.jpg')

In [5]:

df['success_pct'] = df['@type']=='completed'
df['count'] = 1

df[df.type=='all_passes'].groupby('team_name').agg({'x': [np.mean, np.std],
                                                    'assists': np.sum,
                                                    'long_ball': np.sum,
                                                    'count': np.sum,
                                                    'success_pct': np.mean})

Out[5]:

	count	x		success_pct	assists	long_ball
	sum	mean	std	mean	sum	sum
team_name
Barcelona	86	67.527209	16.035325	0.825581	6	6
Chelsea	166	55.232530	16.986273	0.903614	5	22

2 rows × 6 columns

In [6]:

df['fifth'] = np.digitize(df['x'], np.arange(0, 105, 105.0/5))

df[df.type=='all_passes'].groupby(['team_name', 'fifth']).agg({'assists': np.sum,
                                                               'long_ball': np.sum,
                                                               'count': np.sum})

Out[6]:

		count	assists	long_ball
team_name	fifth
Barcelona	1	1	0	0
	2	3	1	1
	3	30	0	3
	4	40	1	2
	5	12	4	0
Chelsea	1	4	0	1
	2	37	0	5
	3	69	1	15
	4	49	2	1
	5	7	2	0

10 rows × 3 columns