Parsing custom events¶

In [1]:

from kloppy import statsbomb
from kloppy.domain import ShotEvent, EventFactory, create_event, ShotEvent
from dataclasses import dataclass
from typing import Optional

import polars as pl

from kloppy import statsbomb from kloppy.domain import ShotEvent, EventFactory, create_event, ShotEvent from dataclasses import dataclass from typing import Optional import polars as pl

Define a custom shot event class¶

• Extends ShotEvent to include xg (expected goals).
• Adds inverted_xg property to compute 1 - xG.
• Overrides __str__ to provide a readable string representation.

In [2]:

@dataclass(repr=False)
class StatsbombShotEvent(ShotEvent):
    xg: Optional[float] = None

    @property
    def inverted_xg(self) -> Optional[float]:
        if self.xg is not None:
            return 1 - self.xg
        else:
            return None

    def __str__(self):
        return f"<Shot event_id='{self.event_id}' time='{self.time}' player='{self.player}' xg={self.xg} result='{self.result}'>"

@dataclass(repr=False) class StatsbombShotEvent(ShotEvent): xg: Optional[float] = None @property def inverted_xg(self) -> Optional[float]: if self.xg is not None: return 1 - self.xg else: return None def __str__(self): return f""

Implement a custom event factory¶

• Extracts xg from raw_event if available.
• Converts xG value to float.
• Uses create_event to construct a StatsbombShotEvent instance

In [3]:

class StatsbombEventFactory(EventFactory):
    def build_shot(self, raw_event, **kwargs) -> ShotEvent:
        xg = (
            raw_event["shot"]["statsbomb_xg"]
            if "shot" in raw_event and raw_event["shot"]
            else None
        )
        if xg is not None:
            xg = float(xg)

        return create_event(
            StatsbombShotEvent,
            xg=xg,
            raw_event=raw_event,
            **kwargs,
        )

class StatsbombEventFactory(EventFactory): def build_shot(self, raw_event, **kwargs) -> ShotEvent: xg = ( raw_event["shot"]["statsbomb_xg"] if "shot" in raw_event and raw_event["shot"] else None ) if xg is not None: xg = float(xg) return create_event( StatsbombShotEvent, xg=xg, raw_event=raw_event, **kwargs, )

For the sake of this demonstration, we will use Statsbomb Open Event Data and apply our custom EventFactory

In [4]:

dataset = statsbomb.load_open_data(
    match_id=15946,
    # Optional arguments
    coordinates="statsbomb",
    event_factory=StatsbombEventFactory(),  # Use our custom EventFactory
)

dataset = statsbomb.load_open_data( match_id=15946, # Optional arguments coordinates="statsbomb", event_factory=StatsbombEventFactory(), # Use our custom EventFactory )

/cw/dtaijupiter/NoCsBack/dtai/pieterr/Projects/kloppy/kloppy/_providers/statsbomb.py:83: UserWarning: 

You are about to use StatsBomb public data.
By using this data, you are agreeing to the user agreement. 
The user agreement can be found here: https://github.com/statsbomb/open-data/blob/master/LICENSE.pdf

  warnings.warn(

Filter and print shots¶

In [5]:

shots = dataset.filter("shot")

for shot in shots[:5]:
    print(shot)

shots = dataset.filter("shot") for shot in shots[:5]: print(shot)

<Shot event_id='becd7956-ce44-479e-8fc9-16a2d1f1f349' time='P1T02:29' player='Lionel Andrés Messi Cuccittini' xg=0.07699243 result='OFF_TARGET'>
<Shot event_id='9107d374-2942-4876-a14f-1b9f86901c15' time='P1T05:40' player='Jordi Alba Ramos' xg=0.05166811 result='OFF_TARGET'>
<Shot event_id='ddd194ca-08fb-43d0-87c2-33647f975f9f' time='P1T15:29' player='Lionel Andrés Messi Cuccittini' xg=0.016932096 result='SAVED'>
<Shot event_id='86596ddb-d824-4e5e-b18c-b4442e9ce7cf' time='P1T16:20' player='Rubén Sobrino Pozuelo' xg=0.1226044 result='OFF_TARGET'>
<Shot event_id='3ed2b107-be17-42d5-9d1b-25006a0e55cb' time='P1T18:16' player='Luis Alberto Suárez Díaz' xg=0.041750744 result='OFF_TARGET'>

Use the inverted_xg property¶

In [6]:

for shot in shots[:5]:
    print(shot.inverted_xg)

for shot in shots[:5]: print(shot.inverted_xg)

0.92300757
0.94833189
0.983067904
0.8773956
0.958249256

Convert shots to a Polars DataFrame¶

In [7]:

shots.to_df("player_id", "xg", engine="polars")

shots.to_df("player_id", "xg", engine="polars")

Out[7]:

shape: (28, 2)

player_id	xg
str	f64
"5503"	0.076992
"5211"	0.051668
"5503"	0.016932
"6613"	0.1226044
"5246"	0.041751
…	…
"6935"	0.140808
"3501"	0.334535
"3501"	0.105343
"5503"	0.09483
"5503"	0.420624

In [ ]: