How to Analyze Football Data Using Python and StatsBomb: FIFA World Cup 2022 Final Example

Introduction

The FIFA World Cup 2022 Final between Argentina and France will be remembered as one of the most thrilling matches in football history. With Argentina clinching victory in a penalty shootout after a 3-3 draw, this match provided a wealth of tactical, individual, and team insights. This analysis working with only Argentina’s player passing data to find out key patterns, tactics and insight that shaped this epic encounter.

Libreries and Data Import

To conduct this analysis, we utilized the StatsBomb API for detailed event data. Key Python libraries included:

pandas for data manipulation,
mplsoccer for visualization,
matplotlib for plotting, and
scipy for data smoothing.
The dataset included detailed events such as passes, shots, fouls, and defensive actions from the final matc

1
2
3
4
5
6
7


# Import Libreries & Packeges
import pandas as pd
import matplotlib.pyplot as plt
from mplsoccer import Pitch, VerticalPitch, FontManager
from matplotlib import rcParams
from statsbombpy import sb
from scipy.ndimage import gaussian_filter

1
2
3


# Call Statsbomb API to import data
matches = sb.matches(competition_id=43, season_id=106)
matches[matches['competition_stage']=='Final']

	match_id	match_date	kick_off	competition	season	home_team	away_team	home_score	away_score	match_status	...	last_updated_360	match_week	competition_stage	stadium	referee	home_managers	away_managers	data_version	shot_fidelity_version	xy_fidelity_version
9	3869685	2022-12-18	17:00:00.000	International - FIFA World Cup	2022	Argentina	France	3	3	available	...	2023-08-17T15:55:15.164685	7	Final	Lusail Stadium	Szymon Marciniak	Lionel Sebastián Scaloni	Didier Deschamps	1.1.0	2	2

1 rows × 22 columns

1
2


# Select Fifa Worldcup 2022 "Argentina vs France" final match
events = sb.events(match_id=3869685)

Data Preprocessing

Before diving into analysis, preprocessing steps involved filtering and structuring the data:

Extracting key events (passes, shot,, goals, player positions),
Filtering player-specific data (Messi, Argentina),
Data Normalization

1
2


# Check the data
events.tail()

	50_50	bad_behaviour_card	ball_receipt_outcome	ball_recovery_offensive	ball_recovery_recovery_failure	block_deflection	block_offensive	carry_end_location	clearance_aerial_won	clearance_body_part	...	substitution_outcome	substitution_outcome_id	substitution_replacement	substitution_replacement_id	tactics	team	team_id	timestamp	type	under_pressure
4402	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	Argentina	779	00:19:07.472	Half End	NaN
4403	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	France	771	00:05:58.866	Half End	NaN
4404	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	Argentina	779	00:05:58.866	Half End	NaN
4405	NaN	Yellow Card	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	France	771	00:49:35.193	Bad Behaviour	NaN
4406	NaN	Yellow Card	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	Argentina	779	00:04:57.514	Bad Behaviour	NaN

5 rows × 94 columns

1
2


# Check the available colomns
events.columns

Index(['50_50', 'bad_behaviour_card', 'ball_receipt_outcome',
       'ball_recovery_offensive', 'ball_recovery_recovery_failure',
       'block_deflection', 'block_offensive', 'carry_end_location',
       'clearance_aerial_won', 'clearance_body_part', 'clearance_head',
       'clearance_left_foot', 'clearance_other', 'clearance_right_foot',
       'counterpress', 'dribble_nutmeg', 'dribble_outcome', 'dribble_overrun',
       'duel_outcome', 'duel_type', 'duration', 'foul_committed_advantage',
       'foul_committed_card', 'foul_committed_offensive',
       'foul_committed_penalty', 'foul_committed_type', 'foul_won_advantage',
       'foul_won_defensive', 'foul_won_penalty', 'goalkeeper_body_part',
       'goalkeeper_end_location', 'goalkeeper_outcome', 'goalkeeper_position',
       'goalkeeper_technique', 'goalkeeper_type', 'id', 'index',
       'interception_outcome', 'location', 'match_id', 'minute', 'off_camera',
       'out', 'pass_aerial_won', 'pass_angle', 'pass_assisted_shot_id',
       'pass_body_part', 'pass_cross', 'pass_deflected', 'pass_end_location',
       'pass_goal_assist', 'pass_height', 'pass_inswinging', 'pass_length',
       'pass_outcome', 'pass_outswinging', 'pass_recipient',
       'pass_recipient_id', 'pass_shot_assist', 'pass_switch',
       'pass_technique', 'pass_through_ball', 'pass_type', 'period',
       'play_pattern', 'player', 'player_id', 'position', 'possession',
       'possession_team', 'possession_team_id', 'related_events', 'second',
       'shot_aerial_won', 'shot_body_part', 'shot_end_location',
       'shot_first_time', 'shot_freeze_frame', 'shot_key_pass_id',
       'shot_one_on_one', 'shot_outcome', 'shot_statsbomb_xg',
       'shot_technique', 'shot_type', 'substitution_outcome',
       'substitution_outcome_id', 'substitution_replacement',
       'substitution_replacement_id', 'tactics', 'team', 'team_id',
       'timestamp', 'type', 'under_pressure'],
      dtype='object')

1
2
3
4
5


# Create x and y lables from "location" data
events[['x','y']]= events['location'].apply(pd.Series)

# Create end_x and end_y Lables from "pass_end_location" Data
events[['end_x','end_y']]= events['pass_end_location'].apply(pd.Series)

Ploting & Visualization

Aregntina’s Pass Analysis

Analysing Argentina’s passes and player data will give us key insights into the team’s build-up play. Here we will try to understand:

Completed vs Missed/Other passes
Player individual passes
Messi’s passes pattern
Passes leads to goal

1
2
3
4
5


# Create new data frame for all the passe's by Aregentine players
ag_pass_df = events[(events['type']=='Pass') & (events['team']=='Argentina')]

# Filter sucessfull passes
mask_pass = ag_pass_df['pass_outcome'].isnull()

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22


# Set up the pitch
pitch = Pitch(pitch_type='statsbomb', pitch_color='#22312b', line_color='#c7d5cc')
fig, axs = pitch.draw(figsize=(15, 13), constrained_layout=True, tight_layout=False)
fig.set_facecolor('#22312b')

# Plot the completed passes
pitch.lines(ag_pass_df[mask_pass].x, ag_pass_df[mask_pass].y,
            ag_pass_df[mask_pass].end_x, ag_pass_df[mask_pass].end_y,
            lw=5, transparent=True, comet=True, color='#3cada2', ax=axs, label='completed passes')
pitch.scatter(ag_pass_df[mask_pass].x, ag_pass_df[mask_pass].y, alpha = 0.2, color = "#3cada2", ax=axs)

# Plot the other passes
pitch.lines(ag_pass_df[~mask_pass].x, ag_pass_df[~mask_pass].y,
            ag_pass_df[~mask_pass].end_x, ag_pass_df[~mask_pass].end_y,
            lw=5, transparent=True, comet=True, color='#ba4f45', ax=axs, label='other passes')
pitch.scatter(ag_pass_df[~mask_pass].x, ag_pass_df[~mask_pass].y, alpha = 0.2, color = "#ba4f45", ax=axs)

# Set up the legend
axs.legend(facecolor='#22312b', handlelength=5, edgecolor='None', fontsize=15, loc='upper left', labelcolor='white')

# Set the title
ax_title = axs.set_title('Argentina Passes Against France', fontsize=20, color='white')

1
2


# Create player name list
players = ag_pass_df['player'].unique()

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25


# Set up the pitch
pitch = Pitch(pitch_type='statsbomb', pitch_color='#22312b', line_color='#c7d5cc')
fig, axs = pitch.draw(nrows=5, ncols=4, figsize=(20,15),constrained_layout=True, tight_layout=False)
fig.set_facecolor('#22312b')

# Convert the axs into one dimension
axs = axs.flatten()

# Plot each player's passes
for i, player in enumerate(players):
    player_passes = ag_pass_df[ag_pass_df['player'] == player]
    
    pitch.lines(player_passes[mask_pass].x, player_passes[mask_pass].y,
             player_passes[mask_pass].end_x, player_passes[mask_pass].end_y,
             lw=2, transparent=True, comet=True, color='#3cada2', ax=axs[i], label='completed passes')
    pitch.scatter(player_passes[mask_pass].x, player_passes[mask_pass].y, alpha = 0.2, color = "#3cada2", ax=axs[i])
    axs[i].set_title(f"Passes by {player}", fontsize=10, color='white')

# Remove extra pitches
for i in range(17,20):
    axs[i].remove()

# Set the title
fig.suptitle("Argentine Player's Individual Passes", fontsize = 25, color='white')
plt.show()

1
2


# Filter all the passes from Messi
messi_pass = ag_pass_df[ag_pass_df['player']=='Lionel Andrés Messi Cuccittini']

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12


# Set up the pitch
pitch = Pitch(pitch_type='statsbomb', pitch_color='#22312b', line_color='#c7d5cc')
fig, axs = pitch.draw(figsize=(10, 8), constrained_layout=True, tight_layout=False)
fig.set_facecolor('#22312b')

# Plot the pass convex hull
hull = pitch.convexhull(messi_pass.x, messi_pass.y)
poly = pitch.polygon(hull, ax=axs, edgecolor='cornflowerblue', facecolor='cornflowerblue', alpha=0.3)
scatter = pitch.scatter(messi_pass.x, messi_pass.y, ax=axs, edgecolor='black', facecolor='cornflowerblue')

# Set the title
axs_title = axs.set_title("Messi's Pass Convex Hull", fontsize=20, color='white')

1
2
3
4


shot_pass_df = events.loc[(events['pass_assisted_shot_id'].notnull()) & (events['team']=='Argentina'), ['x', 'y', 'end_x', 'end_y', 'pass_assisted_shot_id']]
shot_df = events.loc[(events['type']=='Shot') & (events['team']=='Argentina'), ['id', 'shot_outcome', 'shot_statsbomb_xg', 'minute']].rename({'id':'pass_assisted_shot_id'}, axis=1)
shot_pass_df = shot_pass_df.merge(shot_df, how='left')
mask_goal = shot_pass_df['shot_outcome'] == 'Goal'

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31


# Setup the pitch
pitch = VerticalPitch(pitch_type='statsbomb', pitch_color='#22312b', line_color='#c7d5cc',
                      half=True, pad_top=2)
fig, axs = pitch.grid(endnote_height=0.03, endnote_space=0, figheight=7,
                      title_height=0.08, title_space=0, axis=False,
                      grid_height=0.82)
fig.set_facecolor('#22312b')

# Plot the completed passes
pitch.lines(shot_pass_df.x, shot_pass_df.y, shot_pass_df.end_x, shot_pass_df.end_y,
            lw=6, transparent=True, comet=True, cmap='jet',
            label='pass leading to shot', ax=axs['pitch'])

# Plot the goals
pitch.scatter(shot_pass_df[mask_goal].end_x, shot_pass_df[mask_goal].end_y, s=300,
              marker='football', edgecolors='black', c='white', zorder=2,
              label='goal', ax=axs['pitch'])
pitch.scatter(shot_pass_df[~mask_goal].end_x, shot_pass_df[~mask_goal].end_y,
              edgecolors='white', c='#22312b', s=300, zorder=2,
              label='shot', ax=axs['pitch'])

# Set the title
axs['title'].text(0.5, 0.5, f'Argentina Passes Leading to Shots', color='white',
                  va='center', ha='center', fontsize=20)

# Set legend
legend = axs['pitch'].legend(facecolor='#22312b', edgecolor='None',
                             loc='lower center', handlelength=4)
for text in legend.get_texts():
    text.set_fontsize(15)
    text.set_color('white')

Argentina’s Pressure Analysis

In this section we will try to understand at what positions argentine players dominated the most and at what level

1
2


# Filter pressure data
mask_presure = (events['type'] == 'Pressure') & (events['team'] == 'Argentina')

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19


# Setup pitch
pitch = Pitch(pitch_type='statsbomb', line_zorder=2,
              pitch_color='#22312b', line_color='#efefef')

# Draw
fig, axs = pitch.draw(figsize=(9, 7))
fig.set_facecolor('#22312b')
bin_statistic = pitch.bin_statistic(events[mask_presure].x, events[mask_presure].y, statistic='count', bins=(25, 25))
bin_statistic['statistic'] = gaussian_filter(bin_statistic['statistic'], 1)
pcm = pitch.heatmap(bin_statistic, ax=axs, cmap='hot', edgecolors='#22312b')

# Add the colorbar and format off-white
cbar = fig.colorbar(pcm, ax=axs, shrink=0.6)
cbar.outline.set_edgecolor('#efefef')
cbar.ax.yaxis.set_tick_params(color='#efefef')
ticks = plt.setp(plt.getp(cbar.ax.axes, 'yticklabels'), color='#efefef')

# Set the title
axs_title = axs.set_title('Pressure Applied by Argentina', fontsize=20, color='white')

Conclusion

Argentina’s performance in the FIFA World Cup 2022 Final showcased their tactical discipline and cohesive play, particularly evident in their passing structure and defensive pressure.

Passing Insights: Argentina’s passing network revealed a well-balanced approach, effectively utilizing both central and wide areas. The team prioritized ball retention and progressive play, with Lionel Messi orchestrating attacks through key passes. Their passing patterns created numerical superiority in crucial zones, allowing them to control the game’s tempo and exploit spaces in France’s defense.

Defensive Pressure: Argentina applied structured defensive pressure, particularly in the first half. Their pressing actions were concentrated in the midfield, aimed at disrupting France’s build-up play and limiting their access to dangerous areas. By maintaining compact defensive lines, Argentina forced turnovers and quickly transitioned into attack, a strategy that kept France under constant pressure.

This combination of controlled passing and well-coordinated pressing was instrumental in Argentina’s dominance for large parts of the match, ultimately contributing to their historic victory.