[Football Analysis]01.이벤트 데이터 구성 파악

 

작성일자 : 2023-08-27
Ver 0.1.1

 

• 데이터 불러오기
  • pandas.read_csv: https://pandas.pydata.org/docs/reference/api/pandas.read_csv.html
  • pandas.read_pickle: https://pandas.pydata.org/docs/reference/api/pandas.read_pickle.html

• 중복 제거를 활용한 데이터 구성 파악
  • pandas.Series.unique: https://pandas.pydata.org/docs/reference/api/pandas.Series.unique.html
  • pandas.DataFrame.drop_duplicates: https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.drop_duplicates.html
  • pandas.DataFrame.sort_values: https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.sort_values.html
  • set: https://wikidocs.net/22262

데이터 불러오기¶

In [1]:

import pandas as pd, os

In [2]:

os.getcwd()

Out[2]:

'/Users/limjongjun/Desktop/JayJay/Growth/Python/soccer-analytics/Excercise'

In [3]:

# directory 변경
new_dir = '/Users/limjongjun/Desktop/JayJay/Growth/Python/soccer-analytics'
os.chdir(new_dir)

os.getcwd()

Out[3]:

'/Users/limjongjun/Desktop/JayJay/Growth/Python/soccer-analytics'

(1) 경기 정보 불러오기¶

In [5]:

dataset_name = 'England'
match_df = pd.read_csv(f'data/refined_events/{dataset_name}/matches.csv', index_col=0, encoding='utf-8-sig') #pkl 파일들 가장 밑에 csv 파일 존재
match_df

Out[5]:

	gameweek	datetime	venue	team1_id	team1_name	team1_goals	team2_id	team2_name	team2_goals	duration
match_id
2499719	1	2017-08-11 18:45:00	Emirates Stadium	1609	Arsenal	4	1631	Leicester City	3	Regular
2499727	1	2017-08-12 11:30:00	Vicarage Road Stadium	1644	Watford	3	1612	Liverpool	3	Regular
2499726	1	2017-08-12 14:00:00	St. Mary's Stadium	1619	Southampton	0	10531	Swansea City	0	Regular
2499721	1	2017-08-12 14:00:00	Stamford Bridge	1610	Chelsea	2	1646	Burnley	3	Regular
2499728	1	2017-08-12 14:00:00	The Hawthorns	1627	West Bromwich Albion	1	1659	AFC Bournemouth	0	Regular
...	...	...	...	...	...	...	...	...	...	...
2500092	38	2018-05-13 14:00:00	Anfield	1612	Liverpool	4	1651	Brighton & Hove Albion	0	Regular
2500091	38	2018-05-13 14:00:00	The John Smith's Stadium	1673	Huddersfield Town	0	1609	Arsenal	1	Regular
2500090	38	2018-05-13 14:00:00	Selhurst Park	1628	Crystal Palace	2	1627	West Bromwich Albion	0	Regular
2500098	38	2018-05-13 14:00:00	London Stadium	1633	West Ham United	3	1623	Everton	1	Regular
2500089	38	2018-05-13 14:00:00	Turf Moor	1646	Burnley	1	1659	AFC Bournemouth	2	Regular

380 rows × 10 columns

(2) 경기 정보 필터링¶

In [6]:

match_df[(match_df['team1_name'] == 'Manchester City') | (match_df['team2_name'] == 'Manchester City')] #Index

Out[6]:

	gameweek	datetime	venue	team1_id	team1_name	team1_goals	team2_id	team2_name	team2_goals	duration
match_id
2499720	1	2017-08-12 16:30:00	The American Express Community Stadium	1651	Brighton & Hove Albion	0	1625	Manchester City	2	Regular
2499734	2	2017-08-21 19:00:00	Etihad Stadium	1625	Manchester City	1	1623	Everton	1	Regular
2499739	3	2017-08-26 11:30:00	Vitality Stadium	1659	AFC Bournemouth	1	1625	Manchester City	2	Regular
2499754	4	2017-09-09 11:30:00	Etihad Stadium	1625	Manchester City	5	1612	Liverpool	0	Regular
2499767	5	2017-09-16 14:00:00	Vicarage Road Stadium	1644	Watford	0	1625	Manchester City	6	Regular
2499774	6	2017-09-23 14:00:00	Etihad Stadium	1625	Manchester City	5	1628	Crystal Palace	0	Regular
2499781	7	2017-09-30 16:30:00	Stamford Bridge	1610	Chelsea	0	1625	Manchester City	1	Regular
2499794	8	2017-10-14 14:00:00	Etihad Stadium	1625	Manchester City	7	1639	Stoke City	2	Regular
2499802	9	2017-10-21 14:00:00	Etihad Stadium	1625	Manchester City	3	1646	Burnley	0	Regular
2499818	10	2017-10-28 14:00:00	The Hawthorns	1627	West Bromwich Albion	2	1625	Manchester City	3	Regular
2499822	11	2017-11-05 14:15:00	Etihad Stadium	1625	Manchester City	3	1609	Arsenal	1	Regular
2499834	12	2017-11-18 15:00:00	King Power Stadium	1631	Leicester City	0	1625	Manchester City	2	Regular
2499841	13	2017-11-26 16:00:00	The John Smith's Stadium	1673	Huddersfield Town	1	1625	Manchester City	2	Regular
2499857	14	2017-11-29 20:00:00	Etihad Stadium	1625	Manchester City	2	1619	Southampton	1	Regular
2499865	15	2017-12-03 16:00:00	Etihad Stadium	1625	Manchester City	2	1633	West Ham United	1	Regular
2499873	16	2017-12-10 16:30:00	Old Trafford	1611	Manchester United	1	1625	Manchester City	2	Regular
2499881	17	2017-12-13 19:45:00	Liberty Stadium	10531	Swansea City	0	1625	Manchester City	4	Regular
2499895	18	2017-12-16 17:30:00	Etihad Stadium	1625	Manchester City	4	1624	Tottenham Hotspur	1	Regular
2499904	19	2017-12-23 15:00:00	Etihad Stadium	1625	Manchester City	4	1659	AFC Bournemouth	0	Regular
2499915	20	2017-12-27 19:45:00	St. James' Park	1613	Newcastle United	0	1625	Manchester City	1	Regular
2499921	21	2017-12-31 12:00:00	Selhurst Park	1628	Crystal Palace	0	1625	Manchester City	0	Regular
2499934	22	2018-01-02 20:00:00	Etihad Stadium	1625	Manchester City	3	1644	Watford	1	Regular
2499943	23	2018-01-14 16:00:00	Anfield	1612	Liverpool	4	1625	Manchester City	3	Regular
2499954	24	2018-01-20 17:30:00	Etihad Stadium	1625	Manchester City	3	1613	Newcastle United	1	Regular
2499966	25	2018-01-31 20:00:00	Etihad Stadium	1625	Manchester City	3	1627	West Bromwich Albion	0	Regular
2499972	26	2018-02-03 12:30:00	Turf Moor	1646	Burnley	1	1625	Manchester City	1	Regular
2499982	27	2018-02-10 17:30:00	Etihad Stadium	1625	Manchester City	5	1631	Leicester City	1	Regular
2499990	28	2018-03-01 19:45:00	Emirates Stadium	1609	Arsenal	0	1625	Manchester City	3	Regular
2500004	29	2018-03-04 16:00:00	Etihad Stadium	1625	Manchester City	1	1610	Chelsea	0	Regular
2500016	30	2018-03-12 20:00:00	Bet365 Stadium	1639	Stoke City	0	1625	Manchester City	2	Regular
2500033	32	2018-03-31 16:30:00	Goodison Park	1623	Everton	1	1625	Manchester City	3	Regular
2500045	33	2018-04-07 16:30:00	Etihad Stadium	1625	Manchester City	2	1611	Manchester United	3	Regular
2500057	34	2018-04-14 18:45:00	Wembley Stadium	1624	Tottenham Hotspur	1	1625	Manchester City	3	Regular
2500065	35	2018-04-22 15:30:00	Etihad Stadium	1625	Manchester City	5	10531	Swansea City	0	Regular
2500078	36	2018-04-29 13:15:00	London Stadium	1633	West Ham United	1	1625	Manchester City	4	Regular
2500085	37	2018-05-06 12:30:00	Etihad Stadium	1625	Manchester City	0	1673	Huddersfield Town	0	Regular
2500024	31	2018-05-09 19:00:00	Etihad Stadium	1625	Manchester City	3	1651	Brighton & Hove Albion	1	Regular
2500095	38	2018-05-13 14:00:00	St. Mary's Stadium	1619	Southampton	0	1625	Manchester City	1	Regular

(3) 경기 이벤트 데이터 불러오기¶

• 17/18 시즌 18Round Mancity vs Tottenham

In [7]:

match_id = 2499895
match_events = pd.read_pickle(f'data/refined_events/{dataset_name}/{match_id}.pkl') #pkl 파일을 Dataframe으로 불러오기
match_events

Out[7]:

	match_id	event_id	period	time	team_id	team_name	player_id	player_name	event_type	sub_event_type	tags	start_x	start_y	end_x	end_y
0	2499895	215108367	1H	1.784	1625	Manchester City	8325	S. Agüero	Pass	Simple pass	[Accurate]	52.00	34.68	40.56	34.68
1	2499895	215108368	1H	3.324	1625	Manchester City	105339	Fernandinho	Pass	Simple pass	[Accurate]	40.56	34.68	29.12	8.16
2	2499895	215108369	1H	6.406	1625	Manchester City	8277	K. Walker	Pass	Simple pass	[Accurate]	29.12	8.16	44.72	10.20
3	2499895	215108370	1H	7.124	1625	Manchester City	38021	K. De Bruyne	Pass	Simple pass	[Accurate]	44.72	10.20	69.68	6.12
4	2499895	215108371	1H	8.676	1625	Manchester City	11066	R. Sterling	Pass	Simple pass	[Accurate]	69.68	6.12	58.24	12.24
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
1607	2499895	215110122	2H	2875.703	1624	Tottenham Hotspur	210044	E. Dier	Pass	Simple pass	[Accurate]	36.40	26.52	48.88	29.92
1608	2499895	215110123	2H	2876.142	1624	Tottenham Hotspur	240070	H. Winks	Duel	Ground attacking duel	[Anticipation, Lost, Not accurate]	48.88	29.92	43.68	24.48
1609	2499895	215109959	2H	2876.768	1625	Manchester City	447205	P. Foden	Duel	Ground defending duel	[Anticipated, Won, Accurate]	55.12	38.08	60.32	43.52
1610	2499895	215109960	2H	2878.046	1625	Manchester City	11066	R. Sterling	Duel	Ground attacking duel	[Free space right, Lost, Not accurate]	60.32	43.52	60.32	43.52
1611	2499895	215110124	2H	2878.216	1624	Tottenham Hotspur	240070	H. Winks	Duel	Ground defending duel	[Free space left, Lost, Not accurate]	43.68	24.48	NaN	NaN

1612 rows × 15 columns

이벤트 데이터 인덱싱(indexing), 슬라이싱(slicing), 필터링(filtering)¶

(1) 열 인덱싱(column indexing)¶

In [8]:

match_events['player_name']

Out[8]:

0          S. Agüero
1        Fernandinho
2          K. Walker
3       K. De Bruyne
4        R. Sterling
            ...     
1607         E. Dier
1608        H. Winks
1609        P. Foden
1610     R. Sterling
1611        H. Winks
Name: player_name, Length: 1612, dtype: object

(2) 행 인덱싱(row indexing)¶

In [9]:

match_events.loc[0]

Out[9]:

match_id                  2499895
event_id                215108367
period                         1H
time                        1.784
team_id                      1625
team_name         Manchester City
player_id                    8325
player_name             S. Agüero
event_type                   Pass
sub_event_type        Simple pass
tags                   [Accurate]
start_x                      52.0
start_y                     34.68
end_x                       40.56
end_y                       34.68
Name: 0, dtype: object

(3) 슬라이싱¶

In [10]:

match_events[0:10]

Out[10]:

	match_id	event_id	period	time	team_id	team_name	player_id	player_name	event_type	sub_event_type	tags	start_x	start_y	end_x	end_y
0	2499895	215108367	1H	1.784	1625	Manchester City	8325	S. Agüero	Pass	Simple pass	[Accurate]	52.00	34.68	40.56	34.68
1	2499895	215108368	1H	3.324	1625	Manchester City	105339	Fernandinho	Pass	Simple pass	[Accurate]	40.56	34.68	29.12	8.16
2	2499895	215108369	1H	6.406	1625	Manchester City	8277	K. Walker	Pass	Simple pass	[Accurate]	29.12	8.16	44.72	10.20
3	2499895	215108370	1H	7.124	1625	Manchester City	38021	K. De Bruyne	Pass	Simple pass	[Accurate]	44.72	10.20	69.68	6.12
4	2499895	215108371	1H	8.676	1625	Manchester City	11066	R. Sterling	Pass	Simple pass	[Accurate]	69.68	6.12	58.24	12.24
5	2499895	215108382	1H	9.282	1624	Tottenham Hotspur	240070	H. Winks	Duel	Ground loose ball duel	[Won, Accurate]	45.76	55.76	47.84	61.20
6	2499895	215108372	1H	9.415	1625	Manchester City	38021	K. De Bruyne	Duel	Ground loose ball duel	[Lost, Not accurate]	58.24	12.24	56.16	6.80
7	2499895	215108383	1H	11.223	1624	Tottenham Hotspur	8292	D. Rose	Pass	Simple pass	[Not accurate]	47.84	61.20	55.12	47.60
8	2499895	215108373	1H	12.536	1625	Manchester City	38021	K. De Bruyne	Others on the ball	Touch	[Interception]	48.88	20.40	49.92	0.00
9	2499895	215108384	1H	15.649	1624	Tottenham Hotspur	0	NaN	Interruption	Ball out of the field	[]	54.08	68.00	NaN	NaN

(4) 필터링 또는 불린 인덱싱(Boolean indexing)¶

In [11]:

match_events[match_events['event_type'] == 'Shot'].head() #Shooting에 대한 Action만 필터링

Out[11]:

	match_id	event_id	period	time	team_id	team_name	player_id	player_name	event_type	sub_event_type	tags	start_x	start_y	end_x	end_y
204	2499895	215108558	1H	549.772	1625	Manchester City	8325	S. Agüero	Shot	Shot	[Head/body, Opportunity, Position: Out low rig...	95.68	25.84	104.0	34.0
299	2499895	215108640	1H	822.817	1625	Manchester City	14808	İ. Gündoğan	Shot	Shot	[Goal, Head/body, Opportunity, Position: Goal ...	95.68	38.76	104.0	34.0
452	2499895	215108755	1H	1380.188	1625	Manchester City	8325	S. Agüero	Shot	Shot	[Left foot, Opportunity, Position: Goal low ce...	88.40	49.64	104.0	34.0
456	2499895	215108760	1H	1385.615	1625	Manchester City	11066	R. Sterling	Shot	Shot	[Left foot, Opportunity, Position: Out high le...	91.52	42.84	104.0	34.0
480	2499895	215108786	1H	1474.472	1625	Manchester City	38021	K. De Bruyne	Shot	Shot	[Left foot, Opportunity, Position: Goal low le...	80.08	36.72	104.0	34.0

In [12]:

match_events[match_events['player_name'] == 'K. De Bruyne'].tail(6)

Out[12]:

	match_id	event_id	period	time	team_id	team_name	player_id	player_name	event_type	sub_event_type	tags	start_x	start_y	end_x	end_y
1498	2499895	215109844	2H	2448.209	1625	Manchester City	38021	K. De Bruyne	Pass	Simple pass	[Accurate]	73.84	60.52	80.08	62.56
1502	2499895	215109848	2H	2455.843	1625	Manchester City	38021	K. De Bruyne	Pass	Simple pass	[Accurate]	67.60	61.20	85.28	64.60
1513	2499895	215109860	2H	2483.958	1625	Manchester City	38021	K. De Bruyne	Pass	Simple pass	[Accurate]	71.76	18.36	93.60	13.60
1516	2499895	215109864	2H	2489.462	1625	Manchester City	38021	K. De Bruyne	Pass	Simple pass	[Accurate]	83.20	18.36	94.64	15.64
1536	2499895	215109885	2H	2522.511	1625	Manchester City	38021	K. De Bruyne	Pass	Simple pass	[Accurate]	73.84	8.16	81.12	36.72
1549	2499895	215109899	2H	2577.333	1625	Manchester City	38021	K. De Bruyne	Pass	Simple pass	[Accurate]	44.72	18.36	29.12	26.52

중복 제거를 활용한 데이터 구성 파악¶

(1) Series 원소 중복 제거¶

• 전후반 구분

In [13]:

match_events['period'].unique() #Unique 값으로 변환

Out[13]:

array(['1H', '2H'], dtype=object)

• 팀 구분

In [14]:

match_events['team_name'].unique()

Out[14]:

array(['Manchester City', 'Tottenham Hotspur'], dtype=object)

• 경기 출전 선수

In [15]:

match_events['player_name'].unique()

Out[15]:

array(['S. Agüero', 'Fernandinho', 'K. Walker', 'K. De Bruyne',
       'R. Sterling', 'H. Winks', 'D. Rose', nan, 'H. Kane',
       'N. Otamendi', 'M. Dembélé', 'E. Dier', 'H. Lloris',
       'J. Vertonghen', 'F. Delph', 'Ederson', 'K. Trippier',
       'Son Heung-Min', 'C. Eriksen', 'L. Sané', 'E. Mangala',
       'İ. Gündoğan', 'D. Alli', 'Gabriel Jesus', 'E. Lamela', 'P. Foden',
       'M. Sissoko', 'Bernardo Silva'], dtype=object)

(2) DataFrame 행 중복 제거¶

• 팀과 선수의 ID 및 이름

In [16]:

match_events[['team_id', 'team_name', 'player_id', 'player_name']].drop_duplicates() #이벤트 id 기준으로 정렬됨

Out[16]:

	team_id	team_name	player_id	player_name
0	1625	Manchester City	8325	S. Agüero
1	1625	Manchester City	105339	Fernandinho
2	1625	Manchester City	8277	K. Walker
3	1625	Manchester City	38021	K. De Bruyne
4	1625	Manchester City	11066	R. Sterling
5	1624	Tottenham Hotspur	240070	H. Winks
7	1624	Tottenham Hotspur	8292	D. Rose
9	1624	Tottenham Hotspur	0	NaN
12	1624	Tottenham Hotspur	8717	H. Kane
14	1625	Manchester City	70086	N. Otamendi
19	1624	Tottenham Hotspur	11152	M. Dembélé
20	1624	Tottenham Hotspur	210044	E. Dier
21	1624	Tottenham Hotspur	25381	H. Lloris
22	1624	Tottenham Hotspur	48	J. Vertonghen
33	1625	Manchester City	8464	F. Delph
35	1625	Manchester City	71654	Ederson
54	1624	Tottenham Hotspur	8945	K. Trippier
59	1624	Tottenham Hotspur	14911	Son Heung-Min
64	1624	Tottenham Hotspur	54	C. Eriksen
83	1625	Manchester City	245364	L. Sané
88	1625	Manchester City	70085	E. Mangala
89	1625	Manchester City	14808	İ. Gündoğan
107	1625	Manchester City	0	NaN
139	1624	Tottenham Hotspur	13484	D. Alli
1047	1625	Manchester City	340386	Gabriel Jesus
1362	1624	Tottenham Hotspur	20441	E. Lamela
1452	1625	Manchester City	447205	P. Foden
1457	1624	Tottenham Hotspur	25804	M. Sissoko
1476	1625	Manchester City	265673	Bernardo Silva

In [17]:

match_events[['team_id', 'team_name', 'player_id', 'player_name']].drop_duplicates().sort_values('team_id') #team_id 기준으로 정렬

Out[17]:

	team_id	team_name	player_id	player_name
59	1624	Tottenham Hotspur	14911	Son Heung-Min
1362	1624	Tottenham Hotspur	20441	E. Lamela
54	1624	Tottenham Hotspur	8945	K. Trippier
139	1624	Tottenham Hotspur	13484	D. Alli
5	1624	Tottenham Hotspur	240070	H. Winks
7	1624	Tottenham Hotspur	8292	D. Rose
9	1624	Tottenham Hotspur	0	NaN
12	1624	Tottenham Hotspur	8717	H. Kane
64	1624	Tottenham Hotspur	54	C. Eriksen
19	1624	Tottenham Hotspur	11152	M. Dembélé
20	1624	Tottenham Hotspur	210044	E. Dier
21	1624	Tottenham Hotspur	25381	H. Lloris
22	1624	Tottenham Hotspur	48	J. Vertonghen
1457	1624	Tottenham Hotspur	25804	M. Sissoko
0	1625	Manchester City	8325	S. Agüero
107	1625	Manchester City	0	NaN
1047	1625	Manchester City	340386	Gabriel Jesus
1452	1625	Manchester City	447205	P. Foden
89	1625	Manchester City	14808	İ. Gündoğan
88	1625	Manchester City	70085	E. Mangala
33	1625	Manchester City	8464	F. Delph
35	1625	Manchester City	71654	Ederson
14	1625	Manchester City	70086	N. Otamendi
4	1625	Manchester City	11066	R. Sterling
3	1625	Manchester City	38021	K. De Bruyne
2	1625	Manchester City	8277	K. Walker
1	1625	Manchester City	105339	Fernandinho
83	1625	Manchester City	245364	L. Sané
1476	1625	Manchester City	265673	Bernardo Silva

• 이벤트 유형의 종류

In [18]:

match_events[['event_type', 'sub_event_type']].drop_duplicates().sort_values(['event_type', 'sub_event_type'])

Out[18]:

	event_type	sub_event_type
16	Duel	Air duel
18	Duel	Ground attacking duel
19	Duel	Ground defending duel
5	Duel	Ground loose ball duel
34	Foul	Foul
650	Foul	Hand foul
298	Free kick	Corner
35	Free kick	Free kick
203	Free kick	Free kick cross
28	Free kick	Goal kick
1346	Free kick	Penalty
10	Free kick	Throw in
954	Goalkeeper leaving line	Goalkeeper leaving line
9	Interruption	Ball out of the field
555	Interruption	Whistle
214	Offside
353	Others on the ball	Acceleration
113	Others on the ball	Clearance
8	Others on the ball	Touch
189	Pass	Cross
145	Pass	Hand pass
13	Pass	Head pass
15	Pass	High pass
313	Pass	Launch
0	Pass	Simple pass
160	Pass	Smart pass
300	Save attempt	Reflexes
481	Save attempt	Save attempt
204	Shot	Shot
1047	Substitution	Player in
1048	Substitution	Player out

(3) List 원소 중복 제거¶

• 추가 태그의 종류

In [19]:

match_events['tags'].sum() #리스트 형태로 변환

Out[19]:

['Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Not accurate',
 'Interception',
 'Accurate',
 'Lost',
 'Not accurate',
 'Won',
 'Accurate',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Lost',
 'Not accurate',
 'Won',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Feint',
 'Interception',
 'Not accurate',
 'Interception',
 'Not accurate',
 'Take on left',
 'Won',
 'Accurate',
 'Take on right',
 'Lost',
 'Not accurate',
 'Accurate',
 'Lost',
 'Not accurate',
 'Won',
 'Accurate',
 'Accurate',
 'Accurate',
 'Not accurate',
 'Interception',
 'Accurate',
 'Lost',
 'Not accurate',
 'Won',
 'Accurate',
 'Not accurate',
 'Interception',
 'Accurate',
 'Accurate',
 'Accurate',
 'Anticipation',
 'Lost',
 'Not accurate',
 'Anticipated',
 'Won',
 'Accurate',
 'Interception',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Neutral',
 'Accurate',
 'Neutral',
 'Accurate',
 'Not accurate',
 'Interception',
 'Accurate',
 'Accurate',
 'Neutral',
 'Accurate',
 'Neutral',
 'Accurate',
 'Not accurate',
 'Not accurate',
 'Interception',
 'Accurate',
 'Neutral',
 'Accurate',
 'Neutral',
 'Accurate',
 'Free space right',
 'Lost',
 'Not accurate',
 'Free space left',
 'Won',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Missed ball',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Through',
 'Not accurate',
 'Interception',
 'Neutral',
 'Accurate',
 'Neutral',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Lost',
 'Not accurate',
 'Won',
 'Accurate',
 'Interception',
 'Accurate',
 'Accurate',
 'Lost',
 'Not accurate',
 'Sliding tackle',
 'Won',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Free space right',
 'Lost',
 'Not accurate',
 'Free space left',
 'Won',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Lost',
 'Not accurate',
 'Won',
 'Accurate',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Take on left',
 'Won',
 'Accurate',
 'Take on right',
 'Lost',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Not accurate',
 'Anticipated',
 'Won',
 'Accurate',
 'Anticipation',
 'Neutral',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Lost',
 'Not accurate',
 'Won',
 'Accurate',
 'Accurate',
 'Accurate',
 'Neutral',
 'Accurate',
 'Neutral',
 'Accurate',
 'Accurate',
 'Accurate',
 'Neutral',
 'Accurate',
 'Neutral',
 'Accurate',
 'Accurate',
 'Accurate',
 'Through',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Anticipated',
 'Won',
 'Accurate',
 'Anticipation',
 'Lost',
 'Not accurate',
 'Accurate',
 'Not accurate',
 'Interception',
 'Neutral',
 'Accurate',
 'Neutral',
 'Accurate',
 'Not accurate',
 'Interception',
 'Accurate',
 'Through',
 'Not accurate',
 'Interception',
 'Not accurate',
 'Accurate',
 'Not accurate',
 'Accurate',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Take on left',
 'Won',
 'Accurate',
 'Take on right',
 'Lost',
 'Not accurate',
 'Left foot',
 'Not accurate',
 'Interception',
 'Accurate',
 'Not accurate',
 'Accurate',
 'Not accurate',
 'Interception',
 'Not accurate',
 'Interception',
 'Accurate',
 'Accurate',
 'Accurate',
 'Not accurate',
 'Interception',
 'Free space left',
 'Sliding tackle',
 'Lost',
 'Not accurate',
 'Free space right',
 'Won',
 'Accurate',
 'Key pass',
 'High',
 'Accurate',
 'Head/body',
 'Opportunity',
 'Position: Out low right',
 'Not accurate',
 'Accurate',
 'Lost',
 'Not accurate',
 'Won',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Accurate',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Not accurate',
 'Interception',
 'Missed ball',
 'Through',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Take on right',
 'Won',
 'Accurate',
 'Take on left',
 'Lost',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Free space left',
 'Lost',
 'Not accurate',
 'Free space right',
 'Won',
 'Accurate',
 'Right foot',
 'Accurate',
 'Interception',
 'Accurate',
 'Free space right',
 'Won',
 'Accurate',
 'Free space left',
 'Lost',
 'Not accurate',
 'Accurate',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Lost',
 'Not accurate',
 'Won',
 'Accurate',
 'Take on left',
 'Lost',
 'Not accurate',
 'Take on right',
 'Won',
 'Accurate',
 'Accurate',
 'Accurate',
 'Free space right',
 'Won',
 'Accurate',
 'Free space left',
 'Lost',
 'Not accurate',
 'Accurate',
 'Right foot',
 'High',
 'Accurate',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Not accurate',
 'Interception',
 'Neutral',
 'Accurate',
 'Neutral',
 'Accurate',
 'Accurate',
 'Accurate',
 'Lost',
 'Not accurate',
 'Won',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Accurate',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Not accurate',
 'Interception',
 'Accurate',
 'Accurate',
 'Take on right',
 'Lost',
 'Not accurate',
 'Take on right',
 'Lost',
 'Not accurate',
 'Take on left',
 'Won',
 'Accurate',
 'Take on left',
 'Won',
 'Accurate',
 'Take on right',
 'Won',
 'Accurate',
 'Take on left',
 'Lost',
 'Not accurate',
 'Not accurate',
 'Assist',
 'High',
 'Accurate',
 'Goal',
 'Head/body',
 'Opportunity',
 'Position: Goal center left',
 'Accurate',
 'Position: Goal center left',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Lost',
 'Not accurate',
 'Won',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Neutral',
 'Accurate',
 'Neutral',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Through',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Lost',
 'Not accurate',
 'Won',
 'Accurate',
 'Not accurate',
 'Interception',
 'Accurate',
 'Not accurate',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Free space left',
 'Won',
 'Accurate',
 'Free space right',
 'Lost',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Not accurate',
 'Interception',
 'Accurate',
 'Accurate',
 'Take on left',
 'Won',
 'Accurate',
 'Take on right',
 'Lost',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Free space left',
 'Lost',
 'Not accurate',
 'Free space right',
 'Won',
 'Accurate',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Take on left',
 'Won',
 'Accurate',
 'Take on right',
 'Lost',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Lost',
 'Not accurate',
 'Won',
 'Accurate',
 'Accurate',
 'Not accurate',
 'Accurate',
 'Right foot',
 'High',
 'Not accurate',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Free space right',
 'Lost',
 'Not accurate',
 'Free space left',
 'Won',
 'Accurate',
 'Left foot',
 'Not accurate',
 'Interception',
 'Not accurate',
 'High',
 'Not accurate',
 'Interception',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Right foot',
 'High',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Neutral',
 'Accurate',
 'Neutral',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Through',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Through',
 'Not accurate',
 'Accurate',
 'Key pass',
 'Through',
 'Accurate',
 'Left foot',
 'Opportunity',
 'Position: Goal low center',
 'Accurate',
 'Position: Goal low center',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Left foot',
 'Opportunity',
 'Position: Out high left',
 'Not accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Not accurate',
 'Interception',
 'Accurate',
 'Feint',
 'Not accurate',
 'Interception',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Not accurate',
 'Free space right',
 'Lost',
 'Not accurate',
 'Free space left',
 'Won',
 'Accurate',
 'Left foot',
 'Opportunity',
 'Position: Goal low left',
 'Accurate',
 'Position: Goal low left',
 'Accurate',
 'Interception',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Not accurate',
 'Interception',
 'Not accurate',
 'Accurate',
 'Lost',
 'Not accurate',
 'Won',
 'Accurate',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'High',
 'Not accurate',
 'Accurate',
 'Take on right',
 'Won',
 'Accurate',
 'Take on left',
 'Lost',
 'Not accurate',
 'Counter attack',
 'Accurate',
 'Counter attack',
 'Accurate',
 'Counter attack',
 'Accurate',
 'Counter attack',
 'Accurate',
 'Counter attack',
 'Accurate',
 'Counter attack',
 'Right foot',
 'Opportunity',
 'Position: Goal low center',
 'Accurate',
 'Position: Goal low center',
 'Accurate',
 'Accurate',
 'Free space left',
 'Lost',
 'Not accurate',
 'Free space right',
 'Won',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Free space left',
 'Won',
 'Accurate',
 'Free space right',
 'Lost',
 'Not accurate',
 'Take on left',
 'Won',
 'Accurate',
 'Take on right',
 'Lost',
 'Not accurate',
 'Take on left',
 'Lost',
 'Not accurate',
 'Take on right',
 'Won',
 'Accurate',
 'Not accurate',
 'Interception',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Dangerous ball lost',
 'Not accurate',
 'Interception',
 'Accurate',
 'Left foot',
 'Accurate',
 'Right foot',
 'Blocked',
 'Not accurate',
 'Interception',
 'Accurate',
 'Anticipated',
 'Won',
 'Accurate',
 'Anticipation',
 'Lost',
 'Not accurate',
 'Accurate',
 'Neutral',
 'Accurate',
 'Neutral',
 'Accurate',
 'Not accurate',
 'Fairplay',
 'Not accurate',
 'Fairplay',
 'Not accurate',
 'Accurate',
 'Not accurate',
 'Interception',
 'Not accurate',
 'Interception',
 'Accurate',
 'Not accurate',
 'Interception',
 'Free space right',
 'Won',
 'Accurate',
 'Free space left',
 'Lost',
 'Not accurate',
 'Lost',
 'Not accurate',
 'Won',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Counter attack',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Counter attack',
 'Accurate',
 'Counter attack',
 'Counter attack',
 'Free space left',
 'Neutral',
 'Accurate',
 'Free space right',
 'Neutral',
 'Accurate',
 'Counter attack',
 'Lost',
 'Not accurate',
 'Won',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Take on right',
 'Lost',
 'Not accurate',
 'Take on left',
 'Won',
 'Accurate',
 'Accurate',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Accurate',
 'Missed ball',
 'Accurate',
 'Lost',
 'Not accurate',
 'Won',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Not accurate',
 'Interception',
 'Accurate',
 'Accurate',
 'Left foot',
 'Opportunity',
 'Position: Out low right',
 'Not accurate',
 'Not accurate',
 'Interception',
 'Not accurate',
 'Interception',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Not accurate',
 'Interception',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Neutral',
 'Accurate',
 'Neutral',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Right foot',
 'Opportunity',
 'Position: Out low right',
 'Not accurate',
 'Not accurate',
 'Interception',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Neutral',
 'Accurate',
 'Neutral',
 'Accurate',
 'Accurate',
 'Not accurate',
 'Interception',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Yellow card',
 'Not accurate',
 'Interception',
 'Not accurate',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Take on left',
 'Neutral',
 'Accurate',
 'Take on right',
 'Neutral',
 'Accurate',
 'Accurate',
 'Not accurate',
 'Accurate',
 'Free space left',
 'Won',
 'Accurate',
 'Free space right',
 'Lost',
 'Not accurate',
 'Neutral',
 'Accurate',
 'Neutral',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'High',
 'Not accurate',
 'Interception',
 'Not accurate',
 'Right foot',
 'Blocked',
 'Not accurate',
 'Interception',
 'Accurate',
 'Won',
 'Accurate',
 'Lost',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Lost',
 'Not accurate',
 'Won',
 'Accurate',
 'Not accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 'Accurate',
 ...]

In [20]:

set(match_events['tags'].sum())

Out[20]:

{13484,
 14808,
 14911,
 20441,
 245364,
 25804,
 265673,
 340386,
 447205,
 8325,
 'Accurate',
 'Anticipated',
 'Anticipation',
 'Assist',
 'Blocked',
 'Counter attack',
 'Dangerous ball lost',
 'Fairplay',
 'Feint',
 'Free space left',
 'Free space right',
 'Goal',
 'Head/body',
 'High',
 'Interception',
 'Key pass',
 'Left foot',
 'Lost',
 'Missed ball',
 'Neutral',
 'Not accurate',
 'Opportunity',
 'Position: Goal center',
 'Position: Goal center left',
 'Position: Goal center right',
 'Position: Goal high left',
 'Position: Goal low center',
 'Position: Goal low left',
 'Position: Goal low right',
 'Position: Out high center',
 'Position: Out high left',
 'Position: Out high right',
 'Position: Out low right',
 'Position: Post low left',
 'Right foot',
 'Sliding tackle',
 'Take on left',
 'Take on right',
 'Through',
 'Won',
 'Yellow card'}

In [21]:

match_events[match_events['event_type'] == 'Substitution'] #tags에 교체 대상 선수의 player id가 입력되어 있음

Out[21]:

	match_id	event_id	period	time	team_id	team_name	player_id	player_name	event_type	sub_event_type	tags	start_x	start_y	end_x	end_y
1047	2499895	0	2H	660.0	1625	Manchester City	340386	Gabriel Jesus	Substitution	Player in	[8325]	NaN	NaN	NaN	NaN
1048	2499895	0	2H	660.0	1625	Manchester City	8325	S. Agüero	Substitution	Player out	[340386]	NaN	NaN	NaN	NaN
1362	2499895	0	2H	1860.0	1624	Tottenham Hotspur	20441	E. Lamela	Substitution	Player in	[14911]	NaN	NaN	NaN	NaN
1363	2499895	0	2H	1860.0	1624	Tottenham Hotspur	14911	Son Heung-Min	Substitution	Player out	[20441]	NaN	NaN	NaN	NaN
1452	2499895	0	2H	2220.0	1625	Manchester City	447205	P. Foden	Substitution	Player in	[14808]	NaN	NaN	NaN	NaN
1453	2499895	0	2H	2220.0	1625	Manchester City	14808	İ. Gündoğan	Substitution	Player out	[447205]	NaN	NaN	NaN	NaN
1457	2499895	0	2H	2280.0	1624	Tottenham Hotspur	25804	M. Sissoko	Substitution	Player in	[13484]	NaN	NaN	NaN	NaN
1458	2499895	0	2H	2280.0	1624	Tottenham Hotspur	13484	D. Alli	Substitution	Player out	[25804]	NaN	NaN	NaN	NaN
1476	2499895	0	2H	2340.0	1625	Manchester City	265673	Bernardo Silva	Substitution	Player in	[245364]	NaN	NaN	NaN	NaN
1477	2499895	0	2H	2340.0	1625	Manchester City	245364	L. Sané	Substitution	Player out	[265673]	NaN	NaN	NaN	NaN