What a team concedes most from corners

defensive_routine_summary and defensive_zone_summary flip restart_routines’s attacking-team perspective to the defending side: which routine types, and which destination zones, is this team’s defending most exposed to. Same per-match, two-contestant requirement as defensive_set_piece_summary.

from pathlib import Path

from wa_setpieces import load_matches
from wa_setpieces.core.defending import defensive_routine_summary, defensive_zone_summary
from wa_setpieces.viz.plots import plot_defensive_routine_bars

try:
    _here = Path(__file__).resolve().parent
except NameError:
    _here = Path.cwd()
DATA = _here.parent / "tests" / "data" / "sample_match.json"

events = load_matches(DATA)

By routine type:

conceded_routines = defensive_routine_summary(events, "corner")
conceded_routines

	contestantId	routine_type	attempts_faced	shots_conceded	goals_conceded	shot_rate_conceded
0	cxb4hqite921i36gwrezdts7c	central_six_yard	6	0	0	0.0
1	cxb4hqite921i36gwrezdts7c	short	1	0	0	0.0
2	f2yd0yzt0om6qhks9gbowu1d6	penalty_area	2	0	0	0.0

team_id = conceded_routines["contestantId"].iloc[0]
fig, ax = plot_defensive_routine_bars(conceded_routines, team_id=team_id)

By destination zone – where the ball ended up, not how it got there:

conceded_zones = defensive_zone_summary(events, "corner")
conceded_zones

	contestantId	destination_zone	attempts_faced	shots_conceded	goals_conceded	shot_rate_conceded
0	cxb4hqite921i36gwrezdts7c	six_yard_box	5	0	0	0.0
1	cxb4hqite921i36gwrezdts7c	penalty_area	1	0	0	0.0
2	cxb4hqite921i36gwrezdts7c	wide_final_third	1	0	0	0.0
3	f2yd0yzt0om6qhks9gbowu1d6	penalty_area	2	0	0	0.0

fig, ax = plot_defensive_routine_bars(
    conceded_zones, team_id=team_id, metric="shots_conceded",
    title="Shots conceded by destination zone",
)