[Your-Name-Here] - Fab Futures - Data Science

Home About

Football Player Valuation Analysis

Objectives

• Understand how performance correlates with market value.
• Identify undervalued/overvalued players based on transfer fees.
• Uncover hidden factors influencing valuation.

import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import numpy as np
import os

# Configuration
DATA_DIR = 'datasets/'
sns.set_theme(style="whitegrid")
pd.set_option('display.max_columns', None)

1. Data Loading and Preparation

Loading necessary datasets and merging them into a master dataframe for analysis.

def load_data():
    files = {
        'players': 'players.csv',
        'appearances': 'appearances.csv',
        'clubs': 'clubs.csv',
        'player_valuations': 'player_valuations.csv',
        'transfers': 'transfers.csv',
        'competitions': 'competitions.csv',
        'club_games': 'club_games.csv'
    }
    data = {}
    for name, file in files.items():
        path = os.path.join(DATA_DIR, file)
        if os.path.exists(path):
            print(f"Loading {name}...")
            # Parse dates where applicable
            parse_dates = []
            if name in ['appearances', 'player_valuations']:
                parse_dates = ['date']
            elif name == 'transfers':
                parse_dates = ['transfer_date']
            elif name == 'players':
                parse_dates = ['date_of_birth', 'contract_expiration_date']
            
            data[name] = pd.read_csv(path, parse_dates=parse_dates)
    return data

    print(f"Loaded {name}: {data[players].shape}")

data = load_data()

Loading players...
Loading appearances...
Loading clubs...
Loading player_valuations...
Loading transfers...
Loading competitions...
Loading club_games...

Feature Engineering & Merging

# Aggregate appearances (Total goals, assists, minutes, cards)
if 'appearances' in data:
    perf_agg = data['appearances'].groupby('player_id').agg({
        'goals': 'sum',
        'assists': 'sum',
        'minutes_played': 'sum',
        'yellow_cards': 'sum',
        'red_cards': 'sum',
        'game_id': 'count'  # Total games played
    }).rename(columns={'game_id': 'total_games'}).reset_index()
else:
    perf_agg = pd.DataFrame()

# Merge with Players
df = data['players'].copy()
if not perf_agg.empty:
    df = df.merge(perf_agg, on='player_id', how='left')

# Merge with Club info
if 'clubs' in data:
    clubs_subset = data['clubs'][['club_id', 'total_market_value', 'stadium_seats', 'coach_name', 'national_team_players']]
    df = df.merge(clubs_subset, left_on='current_club_id', right_on='club_id', how='left', suffixes=('', '_club'))

# Calculate Age
now = pd.Timestamp.now()
df['age'] = (now - df['date_of_birth']).dt.days / 365.25

# Filter for active players with market value
df_active = df[df['market_value_in_eur'] > 0].copy()

print(f"Active Players with Market Value: {len(df_active)}")
df_active.head()

Active Players with Market Value: 31078

	player_id	first_name	last_name	name	last_season	current_club_id	player_code	country_of_birth	city_of_birth	country_of_citizenship	date_of_birth	sub_position	position	foot	height_in_cm	contract_expiration_date	agent_name	image_url	url	current_club_domestic_competition_id	current_club_name	market_value_in_eur	highest_market_value_in_eur	goals	assists	minutes_played	yellow_cards	red_cards	total_games	club_id	total_market_value	stadium_seats	coach_name	national_team_players	age
0	10	Miroslav	Klose	Miroslav Klose	2015	398	miroslav-klose	Poland	Opole	Germany	1978-06-09	Centre-Forward	Attack	right	184.0	NaT	ASBW Sport Marketing	https://img.a.transfermarkt.technology/portrai...	https://www.transfermarkt.co.uk/miroslav-klose...	IT1	Società Sportiva Lazio S.p.A.	1000000.0	30000000.0	48.0	25.0	8808.0	19.0	0.0	136.0	398	NaN	70634	NaN	13	47.520876
1	26	Roman	Weidenfeller	Roman Weidenfeller	2017	16	roman-weidenfeller	Germany	Diez	Germany	1980-08-06	Goalkeeper	Goalkeeper	left	190.0	NaT	Neubauer 13 GmbH	https://img.a.transfermarkt.technology/portrai...	https://www.transfermarkt.co.uk/roman-weidenfe...	L1	Borussia Dortmund	750000.0	8000000.0	0.0	0.0	13508.0	4.0	2.0	152.0	16	NaN	81365	NaN	18	45.360712
2	65	Dimitar	Berbatov	Dimitar Berbatov	2015	1091	dimitar-berbatov	Bulgaria	Blagoevgrad	Bulgaria	1981-01-30	Centre-Forward	Attack	NaN	NaN	NaT	CSKA-AS-23 Ltd.	https://img.a.transfermarkt.technology/portrai...	https://www.transfermarkt.co.uk/dimitar-berbat...	GR1	Panthessalonikios Athlitikos Omilos Konstantin...	1000000.0	34500000.0	38.0	13.0	8788.0	11.0	1.0	122.0	1091	NaN	28703	NaN	11	44.876112
3	77	NaN	Lúcio	Lúcio	2012	506	lucio	Brazil	Brasília	Brazil	1978-05-08	Centre-Back	Defender	NaN	NaN	NaT	NaN	https://img.a.transfermarkt.technology/portrai...	https://www.transfermarkt.co.uk/lucio/profil/s...	IT1	Juventus Football Club	200000.0	24500000.0	0.0	0.0	307.0	0.0	0.0	4.0	506	NaN	41507	NaN	19	47.608487
4	80	Tom	Starke	Tom Starke	2017	27	tom-starke	East Germany (GDR)	Freital	Germany	1981-03-18	Goalkeeper	Goalkeeper	right	194.0	NaT	IFM	https://img.a.transfermarkt.technology/portrai...	https://www.transfermarkt.co.uk/tom-starke/pro...	L1	FC Bayern München	100000.0	3000000.0	0.0	0.0	1080.0	0.0	0.0	12.0	27	NaN	75000	NaN	17	44.747433

2. Core Valuation Analysis

Performance Metrics vs Market Value

How do goals, assists, and playing time correlate with market value?

plt.figure(figsize=(15, 5))

plt.subplot(1, 3, 1)
sns.scatterplot(data=df_active, x='goals', y='market_value_in_eur', alpha=0.5)
plt.title('Goals vs Market Value')
plt.xscale('log')
plt.yscale('log')

plt.subplot(1, 3, 2)
sns.scatterplot(data=df_active, x='assists', y='market_value_in_eur', alpha=0.5)
plt.title('Assists vs Market Value')
plt.xscale('log')
plt.yscale('log')

plt.subplot(1, 3, 3)
sns.scatterplot(data=df_active, x='minutes_played', y='market_value_in_eur', alpha=0.5)
plt.title('Minutes Played vs Market Value')
plt.yscale('log')

plt.tight_layout()
plt.show()

# Correlation Matrix
cols = ['market_value_in_eur', 'goals', 'assists', 'minutes_played', 'total_games', 'age']
corr = df_active[cols].corr()
plt.figure(figsize=(8, 6))
sns.heatmap(corr, annot=True, cmap='coolwarm', fmt='.2f')
plt.title('Correlation: Performance vs Value')
plt.show()

Market Dynamics: Age Curves

How does valuation change with age?

plt.figure(figsize=(10, 6))
sns.lineplot(data=df_active, x='age', y='market_value_in_eur', estimator='median')
plt.title('Median Market Value by Age')
plt.show()

Transfer Fee vs Market Value

Are players generally over or undervalued compared to their transfer fees?

*** Not happy with the results... to be continued

if 'transfers' in data:
    # Merge transfers with current market value (approximate comparison)
    # Note: Ideally we should compare transfer fee with market value AT THE TIME of transfer.
    # Here we look at the 'market_value_in_eur' column in the transfers table itself which records value at transfer time.
    
    tf = data['transfers'].dropna(subset=['transfer_fee', 'market_value_in_eur'])
    tf = tf[tf['transfer_fee'] > 0]  # Filter out free transfers for this plot
    
    plt.figure(figsize=(10, 6))
    sns.scatterplot(data=tf, x='market_value_in_eur', y='transfer_fee', alpha=0.3)
    
    # Add y=x line
    max_val = max(tf['market_value_in_eur'].max(), tf['transfer_fee'].max())
    plt.plot([0, max_val], [0, max_val], 'r--', label='Fair Value (Fee = MV)')
        
    plt.title('Transfer Fee vs Market Value at Time of Transfer')
    plt.xscale('log')
    plt.yscale('log')
    plt.legend()
    plt.show()