Sportec¶

• Load local event files
• Load local tracking files
• Load Open Data
• Load remote files

Load local event files¶

In [1]:

from kloppy import sportec

dataset = sportec.load_event(
    event_data="../../kloppy/tests/files/sportec_events.xml",
    meta_data="../../kloppy/tests/files/sportec_meta.xml",
    
    # Optional arguments
    coordinates="sportec",
    event_types=["pass", "shot"]
)

dataset.to_df().head()

from kloppy import sportec dataset = sportec.load_event( event_data="../../kloppy/tests/files/sportec_events.xml", meta_data="../../kloppy/tests/files/sportec_meta.xml", # Optional arguments coordinates="sportec", event_types=["pass", "shot"] ) dataset.to_df().head()

Out[1]:

	event_id	event_type	result	success	period_id	timestamp	end_timestamp	ball_state	ball_owning_team	team_id	player_id	coordinates_x	coordinates_y	end_coordinates_x	end_coordinates_y	receiver_player_id	set_piece_type	body_part_type
0	17364900000006	PASS	COMPLETE	True	1	0.000	None	alive	None	DFL-CLU-000004	DFL-OBJ-0000SP	56.41	68.00	77.75	38.71	DFL-OBJ-0000ZS	KICK_OFF	None
1	17364900000007	PASS	COMPLETE	True	1	3.123	None	alive	None	DFL-CLU-000004	DFL-OBJ-0000ZS	73.94	37.21	NaN	NaN	DFL-OBJ-002G3I	None	None
2	17364900000014	PASS	COMPLETE	True	1	31.797	None	alive	None	DFL-CLU-00000A	DFL-OBJ-00017V	35.57	68.00	21.24	28.58	DFL-OBJ-0027B9	THROW_IN	None
3	17364900000031	SHOT	BLOCKED	False	1	79.480	None	alive	None	DFL-CLU-000004	DFL-OBJ-002706	21.24	28.58	NaN	NaN	None	None	RIGHT_FOOT
4	17364900000036	PASS	INCOMPLETE	False	1	95.173	None	alive	None	DFL-CLU-000004	DFL-OBJ-002G3I	8.72	4.21	NaN	NaN	None	None	None

Load local tracking files¶

In [3]:

from kloppy import sportec


dataset = sportec.load_tracking(
    raw_data="../../kloppy/tests/files/sportec_positional.xml",
    meta_data="../../kloppy/tests/files/sportec_meta.xml",
    
    # Optional arguments
    sample_rate=1,
    limit=10,
    coordinates="sportec",
    only_alive=False
)

dataset.to_df().head()

from kloppy import sportec dataset = sportec.load_tracking( raw_data="../../kloppy/tests/files/sportec_positional.xml", meta_data="../../kloppy/tests/files/sportec_meta.xml", # Optional arguments sample_rate=1, limit=10, coordinates="sportec", only_alive=False ) dataset.to_df().head()

Out[3]:

	period_id	timestamp	frame_id	ball_state	ball_owning_team_id	ball_x	ball_y	ball_z	ball_speed	DFL-OBJ-002G3I_x	...	DFL-OBJ-002G3I_d	DFL-OBJ-002G3I_s	DFL-OBJ-002G5S_x	DFL-OBJ-002G5S_y	DFL-OBJ-002G5S_d	DFL-OBJ-002G5S_s	DFL-OBJ-002FVJ_x	DFL-OBJ-002FVJ_y	DFL-OBJ-002FVJ_d	DFL-OBJ-002FVJ_s
0	1	0.00	10000	dead	DFL-CLU-000004	2.69	0.26	0.06	0.00	0.35	...	None	0.00	NaN	NaN	None	NaN	NaN	NaN	None	NaN
1	1	0.04	10001	alive	DFL-CLU-00000A	3.41	0.26	0.08	65.59	0.34	...	None	1.74	NaN	NaN	None	NaN	NaN	NaN	None	NaN
2	1	0.08	10002	alive	DFL-CLU-000004	4.22	0.33	0.09	65.16	0.32	...	None	1.76	NaN	NaN	None	NaN	NaN	NaN	None	NaN
3	1	0.12	10003	alive	DFL-CLU-000004	5.02	0.38	0.09	74.34	0.31	...	None	1.78	NaN	NaN	None	NaN	NaN	NaN	None	NaN
4	1	0.16	10004	alive	DFL-CLU-000004	5.79	0.44	0.08	73.58	0.29	...	None	1.80	NaN	NaN	None	NaN	NaN	NaN	None	NaN

5 rows × 21 columns

Load open data¶

There are 7 games of open Sportec Open DFL Tracking and Event Data (Bassek et al. 2025).

You can use the following match ids to load these games automatically with kloppy.

match_id	home	away
J03WMX	1. FC Köln	FC Bayern München
J03WN1	VfL Bochum 1848	Bayer 04 Leverkusen
J03WPY	Fortuna Düsseldorf	1. FC Nürnberg
J03WOH	Fortuna Düsseldorf	SSV Jahn Regensburg
J03WQQ	Fortuna Düsseldorf	FC St. Pauli
J03WOY	Fortuna Düsseldorf	F.C. Hansa Rostock
J03WR9	Fortuna Düsseldorf	1. FC Kaiserslautern

In [ ]:

from kloppy import sportec

match_id = "J03WMX"
event_dataset = sportec.load_open_event_data(match_id=match_id)
tracking_dataset = sportec.load_open_tracking_data(match_id=match_id)

from kloppy import sportec match_id = "J03WMX" event_dataset = sportec.load_open_event_data(match_id=match_id) tracking_dataset = sportec.load_open_tracking_data(match_id=match_id)

Load remote files¶

Kloppy supports remote files through fsspec FileSystem under the hood. This allows you to work with files in AWS S3, Google Cloud, Azure Blob, HDFS, FTP, and SFTP without extra tools. For example you can pass:

• Individual s3 file paths: (e.g raw_data=s3://.../sportec_positional.xml)

Note: Kloppy might throw an the first time to help you identify missing cloud specific dependencies like s3fs.

In [ ]:

from kloppy import sportec

dataset = sportec.load_tracking(
    raw_data="s3://.../sportec_positional.xml",
    meta_data="s3://.../sportec_meta.xml",
    
    # Optional arguments
    sample_rate=1,
    limit=10,
    coordinates="sportec",
    only_alive=False
)

from kloppy import sportec dataset = sportec.load_tracking( raw_data="s3://.../sportec_positional.xml", meta_data="s3://.../sportec_meta.xml", # Optional arguments sample_rate=1, limit=10, coordinates="sportec", only_alive=False )