Importing the Libraries¶

In [1]:
# basic operations
import numpy as np

# for dataframe manipulations
import pandas as pd 

# for data visualizations
import matplotlib.pyplot as plt
import seaborn as sns

# for missing values
import missingno as mno

# for date time manipulation
import datetime

# for interactivity
import ipywidgets as widgets
from ipywidgets import interact
from ipywidgets import interact_manual

# setting up the background style for the plots
plt.style.use('fivethirtyeight')

Reading the Data

In [2]:
# reading the data and also checking the computation time
%time data = pd.read_csv('data.csv')

# lets also check the shape of the dataset
print(data.shape)

Wall time: 540 ms
(18207, 89)
In [3]:
# lets check the column names present in the data
data.columns
Out[3]:
Index(['Unnamed: 0', 'ID', 'Name', 'Age', 'Photo', 'Nationality', 'Flag',
       'Overall', 'Potential', 'Club', 'Club Logo', 'Value', 'Wage', 'Special',
       'Preferred Foot', 'International Reputation', 'Weak Foot',
       'Skill Moves', 'Work Rate', 'Body Type', 'Real Face', 'Position',
       'Jersey Number', 'Joined', 'Loaned From', 'Contract Valid Until',
       'Height', 'Weight', 'LS', 'ST', 'RS', 'LW', 'LF', 'CF', 'RF', 'RW',
       'LAM', 'CAM', 'RAM', 'LM', 'LCM', 'CM', 'RCM', 'RM', 'LWB', 'LDM',
       'CDM', 'RDM', 'RWB', 'LB', 'LCB', 'CB', 'RCB', 'RB', 'Crossing',
       'Finishing', 'HeadingAccuracy', 'ShortPassing', 'Volleys', 'Dribbling',
       'Curve', 'FKAccuracy', 'LongPassing', 'BallControl', 'Acceleration',
       'SprintSpeed', 'Agility', 'Reactions', 'Balance', 'ShotPower',
       'Jumping', 'Stamina', 'Strength', 'LongShots', 'Aggression',
       'Interceptions', 'Positioning', 'Vision', 'Penalties', 'Composure',
       'Marking', 'StandingTackle', 'SlidingTackle', 'GKDiving', 'GKHandling',
       'GKKicking', 'GKPositioning', 'GKReflexes', 'Release Clause'],
      dtype='object')

Cleaning Data¶

In [4]:
# checking if the data contains any NULL value
# Visualize missing values as a matrix 
mno.bar(data.iloc[:, :40],
        color = 'orange',
        sort = 'ascending') 
plt.title('Checking Missing Values Heat Map for first half of the data', fontsize = 15)
plt.show()
In [5]:
# Visualize missing values as a matrix 
mno.bar(data.iloc[:, 40:]) 
plt.title('Checking Missing Values Heat Map for second half of the data')
plt.show()

Missing Values Imputation¶

In [6]:
# filling the missing value for the continous variables for proper data visualization

data['ShortPassing'].fillna(data['ShortPassing'].mean(), inplace = True)
data['Volleys'].fillna(data['Volleys'].mean(), inplace = True)
data['Dribbling'].fillna(data['Dribbling'].mean(), inplace = True)
data['Curve'].fillna(data['Curve'].mean(), inplace = True)
data['FKAccuracy'].fillna(data['FKAccuracy'], inplace = True)
data['LongPassing'].fillna(data['LongPassing'].mean(), inplace = True)
data['BallControl'].fillna(data['BallControl'].mean(), inplace = True)
data['HeadingAccuracy'].fillna(data['HeadingAccuracy'].mean(), inplace = True)
data['Finishing'].fillna(data['Finishing'].mean(), inplace = True)
data['Crossing'].fillna(data['Crossing'].mean(), inplace = True)

data['Weight'].fillna('200lbs', inplace = True)
data['Contract Valid Until'].fillna(2019, inplace = True)
data['Height'].fillna("5'11", inplace = True)
data['Loaned From'].fillna('None', inplace = True)
data['Joined'].fillna('Jul 1, 2018', inplace = True)
data['Jersey Number'].fillna(8, inplace = True)
data['Body Type'].fillna('Normal', inplace = True)
data['Position'].fillna('ST', inplace = True)
data['Club'].fillna('No Club', inplace = True)
data['Work Rate'].fillna('Medium/ Medium', inplace = True)
data['Skill Moves'].fillna(data['Skill Moves'].median(), inplace = True)
data['Weak Foot'].fillna(3, inplace = True)
data['Preferred Foot'].fillna('Right', inplace = True)
data['International Reputation'].fillna(1, inplace = True)
data['Wage'].fillna('€200K', inplace = True)
In [7]:
pd.set_option('max_rows', 100)
data.isnull().sum()
Out[7]:
Unnamed: 0                     0
ID                             0
Name                           0
Age                            0
Photo                          0
Nationality                    0
Flag                           0
Overall                        0
Potential                      0
Club                           0
Club Logo                      0
Value                          0
Wage                           0
Special                        0
Preferred Foot                 0
International Reputation       0
Weak Foot                      0
Skill Moves                    0
Work Rate                      0
Body Type                      0
Real Face                     48
Position                       0
Jersey Number                  0
Joined                         0
Loaned From                    0
Contract Valid Until           0
Height                         0
Weight                         0
LS                          2085
ST                          2085
RS                          2085
LW                          2085
LF                          2085
CF                          2085
RF                          2085
RW                          2085
LAM                         2085
CAM                         2085
RAM                         2085
LM                          2085
LCM                         2085
CM                          2085
RCM                         2085
RM                          2085
LWB                         2085
LDM                         2085
CDM                         2085
RDM                         2085
RWB                         2085
LB                          2085
LCB                         2085
CB                          2085
RCB                         2085
RB                          2085
Crossing                       0
Finishing                      0
HeadingAccuracy                0
ShortPassing                   0
Volleys                        0
Dribbling                      0
Curve                          0
FKAccuracy                    48
LongPassing                    0
BallControl                    0
Acceleration                  48
SprintSpeed                   48
Agility                       48
Reactions                     48
Balance                       48
ShotPower                     48
Jumping                       48
Stamina                       48
Strength                      48
LongShots                     48
Aggression                    48
Interceptions                 48
Positioning                   48
Vision                        48
Penalties                     48
Composure                     48
Marking                       48
StandingTackle                48
SlidingTackle                 48
GKDiving                      48
GKHandling                    48
GKKicking                     48
GKPositioning                 48
GKReflexes                    48
Release Clause              1564
dtype: int64
In [8]:
# impute with 0 for rest of the columns
data.fillna(0, inplace = True)

# lets check whether the data still has any missing values
data.isnull().sum().sum()
Out[8]:
0

Feature Engineering¶

In [9]:
# creating new features by aggregating the features

def defending(data):
    return int(round((data[['Marking', 'StandingTackle', 
                               'SlidingTackle']].mean()).mean()))

def general(data):
    return int(round((data[['HeadingAccuracy', 'Dribbling', 'Curve', 
                               'BallControl']].mean()).mean()))

def mental(data):
    return int(round((data[['Aggression', 'Interceptions', 'Positioning', 
                               'Vision','Composure']].mean()).mean()))

def passing(data):
    return int(round((data[['Crossing', 'ShortPassing', 
                               'LongPassing']].mean()).mean()))

def mobility(data):
    return int(round((data[['Acceleration', 'SprintSpeed', 
                               'Agility','Reactions']].mean()).mean()))
def power(data):
    return int(round((data[['Balance', 'Jumping', 'Stamina', 
                               'Strength']].mean()).mean()))

def rating(data):
    return int(round((data[['Potential', 'Overall']].mean()).mean()))

def shooting(data):
    return int(round((data[['Finishing', 'Volleys', 'FKAccuracy', 
                               'ShotPower','LongShots', 'Penalties']].mean()).mean()))
In [10]:
# adding these categories to the data

data['Defending'] = data.apply(defending, axis = 1)
data['General'] = data.apply(general, axis = 1)
data['Mental'] = data.apply(mental, axis = 1)
data['Passing'] = data.apply(passing, axis = 1)
data['Mobility'] = data.apply(mobility, axis = 1)
data['Power'] = data.apply(power, axis = 1)
data['Rating'] = data.apply(rating, axis = 1)
data['Shooting'] = data.apply(shooting, axis = 1)

# lets check the column names in the data after adding new features
data.columns
Out[10]:
Index(['Unnamed: 0', 'ID', 'Name', 'Age', 'Photo', 'Nationality', 'Flag',
       'Overall', 'Potential', 'Club', 'Club Logo', 'Value', 'Wage', 'Special',
       'Preferred Foot', 'International Reputation', 'Weak Foot',
       'Skill Moves', 'Work Rate', 'Body Type', 'Real Face', 'Position',
       'Jersey Number', 'Joined', 'Loaned From', 'Contract Valid Until',
       'Height', 'Weight', 'LS', 'ST', 'RS', 'LW', 'LF', 'CF', 'RF', 'RW',
       'LAM', 'CAM', 'RAM', 'LM', 'LCM', 'CM', 'RCM', 'RM', 'LWB', 'LDM',
       'CDM', 'RDM', 'RWB', 'LB', 'LCB', 'CB', 'RCB', 'RB', 'Crossing',
       'Finishing', 'HeadingAccuracy', 'ShortPassing', 'Volleys', 'Dribbling',
       'Curve', 'FKAccuracy', 'LongPassing', 'BallControl', 'Acceleration',
       'SprintSpeed', 'Agility', 'Reactions', 'Balance', 'ShotPower',
       'Jumping', 'Stamina', 'Strength', 'LongShots', 'Aggression',
       'Interceptions', 'Positioning', 'Vision', 'Penalties', 'Composure',
       'Marking', 'StandingTackle', 'SlidingTackle', 'GKDiving', 'GKHandling',
       'GKKicking', 'GKPositioning', 'GKReflexes', 'Release Clause',
       'defending', 'general', 'mental', 'Defending', 'General', 'Mental',
       'Passing', 'Mobility', 'Power', 'Rating', 'Shooting'],
      dtype='object')

Data Visualization¶

In [11]:
# lets check the Distribution of Scores of Different Skills

plt.rcParams['figure.figsize'] = (18, 12)
plt.subplot(2, 4, 1)
sns.distplot(data['Defending'], color = 'red')
plt.grid()

plt.subplot(2, 4, 2)
sns.distplot(data['General'], color = 'black')
plt.grid()

plt.subplot(2, 4, 3)
sns.distplot(data['Mental'], color = 'red')
plt.grid()

plt.subplot(2, 4, 4)
sns.distplot(data['Passing'], color = 'black')
plt.grid()

plt.subplot(2, 4, 5)
sns.distplot(data['Mobility'], color = 'red')
plt.grid()

plt.subplot(2, 4, 6)
sns.distplot(data['Power'], color = 'black')
plt.grid()

plt.subplot(2, 4, 7)
sns.distplot(data['Shooting'], color = 'red')
plt.grid()

plt.subplot(2, 4, 8)
sns.distplot(data['Rating'], color = 'black')
plt.grid()

plt.suptitle('Score Distributions for Different Abilities')
plt.show()
In [12]:
# comparison of preferred foot over the different players

plt.rcParams['figure.figsize'] = (8, 3)
sns.countplot(data['Preferred Foot'], palette = 'pink')
plt.title('Most Preferred Foot of the Players', fontsize = 20)
plt.show()
In [13]:
# plotting a pie chart to represent share of international repuatation

labels = ['1', '2', '3', '4', '5']  #data['International Reputation'].index
sizes = data['International Reputation'].value_counts()
colors = plt.cm.copper(np.linspace(0, 1, 5))
explode = [0.1, 0.1, 0.2, 0.5, 0.9]

plt.rcParams['figure.figsize'] = (9, 9)
plt.pie(sizes, labels = labels, colors = colors, explode = explode, shadow = True,)
plt.title('International Repuatation for the Football Players', fontsize = 20)
plt.legend()
plt.show()

Let's check the Players with International Reputation as 5

In [14]:
data[data['International Reputation'] == 5][['Name','Nationality',
                            'Overall']].sort_values(by = 'Overall',
                                        ascending = False).style.background_gradient(cmap = 'magma')
Out[14]:
Name Nationality Overall
0 L. Messi Argentina 94
1 Cristiano Ronaldo Portugal 94
2 Neymar Jr Brazil 92
7 L. Suárez Uruguay 91
22 M. Neuer Germany 89
109 Z. Ibrahimović Sweden 85
In [15]:
# plotting a pie chart to represent the share of week foot players

labels = ['5', '4', '3', '2', '1'] 
size = data['Weak Foot'].value_counts()
colors = plt.cm.Wistia(np.linspace(0, 1, 5))
explode = [0, 0, 0, 0, 0.1]

plt.pie(size, labels = labels, colors = colors, explode = explode, shadow = True, startangle = 90)
plt.title('Distribution of Week Foot among Players', fontsize = 25)
plt.legend()
plt.show()
In [16]:
# different positions acquired by the players 

plt.figure(figsize = (13, 15))
plt.style.use('fivethirtyeight')
ax = sns.countplot(y = 'Position', data = data, palette = 'bone')
ax.set_xlabel(xlabel = 'Different Positions in Football', fontsize = 16)
ax.set_ylabel(ylabel = 'Count of Players', fontsize = 16)
ax.set_title(label = 'Comparison of Positions and Players', fontsize = 20)
plt.show()
In [17]:
# defining a function for cleaning the Weight data

def extract_value_from(value):
  out = value.replace('lbs', '')
  return float(out)

# applying the function to weight column
#data['value'] = data['value'].apply(lambda x: extract_value_from(x))
data['Weight'] = data['Weight'].apply(lambda x : extract_value_from(x))

# plotting the distribution of weight of the players
sns.distplot(data['Weight'], color = 'black')
plt.title("Distribution of Players Weight", fontsize = 15)
plt.show()
In [18]:
# defining a function for cleaning the wage column

def extract_value_from(column):
    out = column.replace('€', '')
    if 'M' in out:
        out = float(out.replace('M', ''))*1000000
    elif 'K' in column:
        out = float(out.replace('K', ''))*1000
    return float(out)
In [19]:
# applying the function to the wage and value column
data['Value'] = data['Value'].apply(lambda x: extract_value_from(x))
data['Wage'] = data['Wage'].apply(lambda x: extract_value_from(x))

# visualizing the data
plt.rcParams['figure.figsize'] = (16, 5)
plt.subplot(1, 2, 1)
sns.distplot(data['Value'], color = 'violet')
plt.title('Distribution of Value of the Players', fontsize = 15)

plt.subplot(1, 2, 2)
sns.distplot(data['Wage'], color = 'purple')
plt.title('Distribution of Wages of the Players', fontsize = 15)
plt.show()
In [20]:
# Skill Moves of Players

plt.figure(figsize = (10, 6))
ax = sns.countplot(x = 'Skill Moves', data = data, palette = 'pastel')
ax.set_title(label = 'Count of players on Basis of their skill moves', fontsize = 20)
ax.set_xlabel(xlabel = 'Number of Skill Moves', fontsize = 16)
ax.set_ylabel(ylabel = 'Count', fontsize = 16)
plt.show()
In [21]:
data[(data['Skill Moves'] == 5.0) & (data['Age'] < 20)][['Name','Age']]
Out[21]:
Name Age
25 K. Mbappé 19
1004 J. Sancho 18
1143 Vinícius Júnior 17
2495 M. Ødegaard 19
In [22]:
# To show Different Work rate of the players participating in the FIFA 2019

plt.figure(figsize = (15, 5))
plt.style.use('fivethirtyeight')

sns.countplot(x = 'Work Rate', data = data, palette = 'hls')
plt.title('Different work rates of the Players Participating in the FIFA 2019', fontsize = 20)
plt.xlabel('Work rates associated with the players', fontsize = 16)
plt.ylabel('count of Players', fontsize = 16)
plt.xticks(rotation = 90)
plt.show()
In [23]:
# To show Different potential scores of the players participating in the FIFA 2019

plt.figure(figsize=(16, 4))
plt.style.use('seaborn-paper')

plt.subplot(1, 2, 1)
x = data.Potential
ax = sns.distplot(x, bins = 58, kde = False, color = 'y')
ax.set_xlabel(xlabel = "Player's Potential Scores", fontsize = 10)
ax.set_ylabel(ylabel = 'Number of players', fontsize = 10)
ax.set_title(label = 'Histogram of players Potential Scores', fontsize = 15)

plt.subplot(1, 2, 2)
y = data.Overall
ax = sns.distplot(y, bins = 58, kde = False, color = 'y')
ax.set_xlabel(xlabel = "Player's Overall Scores", fontsize = 10)
ax.set_ylabel(ylabel = 'Number of players', fontsize = 10)
ax.set_title(label = 'Histogram of players Overall Scores', fontsize = 15)
plt.show()
In [24]:
# violin plot 

plt.rcParams['figure.figsize'] = (20, 7)
plt.style.use('seaborn-dark-palette')

sns.boxplot(data['Overall'], data['Age'], hue = data['Preferred Foot'], palette = 'Greys')
plt.title('Comparison of Overall Scores and age wrt Preferred foot', fontsize = 20)
plt.show()

Countries with Most Players

In [25]:
# picking up the countries with highest number of players to compare their overall scores

data['Nationality'].value_counts().head(10).plot(kind = 'pie', cmap = 'inferno',
                                        startangle = 90, explode = [0, 0, 0, 0, 0, 0, 0, 0, 0.1, 0])
plt.title('Countries having Highest Number of players', fontsize = 15)
plt.axis('off')
plt.show()
In [26]:
# Every Nations' Player and their Weights

some_countries = ('England', 'Germany', 'Spain', 'Argentina', 'France', 'Brazil', 'Italy', 'Columbia')
data_countries = data.loc[data['Nationality'].isin(some_countries) & data['Weight']]

plt.rcParams['figure.figsize'] = (15, 7)
ax = sns.violinplot(x = data_countries['Nationality'], y = data_countries['Weight'], palette = 'Reds')
ax.set_xlabel(xlabel = 'Countries', fontsize = 9)
ax.set_ylabel(ylabel = 'Weight in lbs', fontsize = 9)
ax.set_title(label = 'Distribution of Weight of players from different countries', fontsize = 20)
plt.show()
In [27]:
# Every Nations' Player and their overall scores

some_countries = ('England', 'Germany', 'Spain', 'Argentina', 'France', 'Brazil', 'Italy', 'Columbia')
data_countries = data.loc[data['Nationality'].isin(some_countries) & data['Overall']]

plt.rcParams['figure.figsize'] = (15, 7)
ax = sns.barplot(x = data_countries['Nationality'], y = data_countries['Overall'], palette = 'spring')
ax.set_xlabel(xlabel = 'Countries', fontsize = 9)
ax.set_ylabel(ylabel = 'Overall Scores', fontsize = 9)
ax.set_title(label = 'Distribution of overall scores of players from different countries', fontsize = 20)
plt.show()
In [28]:
# Every Nations' Player and their wages

some_countries = ('England', 'Germany', 'Spain', 'Argentina', 'France', 'Brazil', 'Italy', 'Columbia')
data_countries = data.loc[data['Nationality'].isin(some_countries) & data['Wage']]

plt.rcParams['figure.figsize'] = (15, 7)
ax = sns.barplot(x = data_countries['Nationality'], y = data_countries['Wage'], palette = 'Purples')
ax.set_xlabel(xlabel = 'Countries', fontsize = 9)
ax.set_ylabel(ylabel = 'Wage', fontsize = 9)
ax.set_title(label = 'Distribution of Wages of players from different countries', fontsize = 15)
plt.grid()
plt.show()
In [29]:
# Every Nations' Player and their International Reputation

some_countries = ('England', 'Germany', 'Spain', 'Argentina', 'France', 'Brazil', 'Italy', 'Columbia')
data_countries = data.loc[data['Nationality'].isin(some_countries) & data['International Reputation']]

plt.rcParams['figure.figsize'] = (15, 7)
ax = sns.boxenplot(x = data_countries['Nationality'], y = data_countries['International Reputation'], palette = 'autumn')
ax.set_xlabel(xlabel = 'Countries', fontsize = 9)
ax.set_ylabel(ylabel = 'Distribution of reputation', fontsize = 9)
ax.set_title(label = 'Distribution of International Repuatation of players from different countries', fontsize = 15)
plt.grid()
plt.show()
In [30]:
some_clubs = ('CD Leganés', 'Southampton', 'RC Celta', 'Empoli', 'Fortuna Düsseldorf', 'Manchestar City',
             'Tottenham Hotspur', 'FC Barcelona', 'Valencia CF', 'Chelsea', 'Real Madrid')

data_clubs = data.loc[data['Club'].isin(some_clubs) & data['Overall']]

plt.rcParams['figure.figsize'] = (15, 8)
ax = sns.boxplot(x = data_clubs['Club'], y = data_clubs['Overall'], palette = 'inferno')
ax.set_xlabel(xlabel = 'Some Popular Clubs', fontsize = 9)
ax.set_ylabel(ylabel = 'Overall Score', fontsize = 9)
ax.set_title(label = 'Distribution of Overall Score in Different popular Clubs', fontsize = 20)
plt.xticks(rotation = 90)
plt.grid()
plt.show()
In [31]:
# Distribution of Ages in some Popular clubs

some_clubs = ('CD Leganés', 'Southampton', 'RC Celta', 'Empoli', 'Fortuna Düsseldorf', 'Manchestar City',
             'Tottenham Hotspur', 'FC Barcelona', 'Valencia CF', 'Chelsea', 'Real Madrid')

data_club = data.loc[data['Club'].isin(some_clubs) & data['Wage']]

plt.rcParams['figure.figsize'] = (15, 8)
ax = sns.boxenplot(x = 'Club', y = 'Age', data = data_club, palette = 'magma')
ax.set_xlabel(xlabel = 'Names of some popular Clubs', fontsize = 10)
ax.set_ylabel(ylabel = 'Distribution', fontsize = 10)
ax.set_title(label = 'Disstribution of Ages in some Popular Clubs', fontsize = 20)
plt.xticks(rotation = 90)
plt.grid()
plt.show()
In [32]:
# Distribution of Wages in some Popular clubs

some_clubs = ('CD Leganés', 'Southampton', 'RC Celta', 'Empoli', 'Fortuna Düsseldorf', 'Manchestar City',
             'Tottenham Hotspur', 'FC Barcelona', 'Valencia CF', 'Chelsea', 'Real Madrid')

data_club = data.loc[data['Club'].isin(some_clubs) & data['Wage']]

plt.rcParams['figure.figsize'] = (16, 8)
ax = sns.boxplot(x = 'Club', y = 'Wage', data = data_club, palette = 'magma')
ax.set_xlabel(xlabel = 'Names of some popular Clubs', fontsize = 10)
ax.set_ylabel(ylabel = 'Distribution', fontsize = 10)
ax.set_title(label = 'Disstribution of Wages in some Popular Clubs', fontsize = 20)
plt.xticks(rotation = 90)
plt.show()
In [33]:
# Distribution of Wages in some Popular clubs

some_clubs = ('CD Leganés', 'Southampton', 'RC Celta', 'Empoli', 'Fortuna Düsseldorf', 'Manchestar City',
             'Tottenham Hotspur', 'FC Barcelona', 'Valencia CF', 'Chelsea', 'Real Madrid')

data_club = data.loc[data['Club'].isin(some_clubs) & data['International Reputation']]

plt.rcParams['figure.figsize'] = (16, 8)
ax = sns.boxenplot(x = 'Club', y = 'International Reputation', data = data_club, palette = 'copper')
ax.set_xlabel(xlabel = 'Names of some popular Clubs', fontsize = 10)
ax.set_ylabel(ylabel = 'Distribution of Reputation', fontsize = 10)
ax.set_title(label = 'Distribution of International Reputation in some Popular Clubs', fontsize = 20)
plt.xticks(rotation = 90)
plt.grid()
plt.show()

Query Analysis¶

Best Players per each position with their age, club, and nationality based on their Overall Scores

In [34]:
# best players per each position with their age, club, and nationality based on their overall scores

data.iloc[data.groupby(data['Position'])['Overall'].idxmax()][['Position', 'Name', 'Age', 'Club',
                             'Nationality','Overall']].sort_values(by = 'Overall',
                                        ascending = False).style.background_gradient(cmap = 'pink')
Out[34]:
Position Name Age Club Nationality Overall
1 ST Cristiano Ronaldo 33 Juventus Portugal 94
0 RF L. Messi 31 FC Barcelona Argentina 94
2 LW Neymar Jr 26 Paris Saint-Germain Brazil 92
5 LF E. Hazard 27 Chelsea Belgium 91
7 RS L. Suárez 31 FC Barcelona Uruguay 91
4 RCM K. De Bruyne 27 Manchester City Belgium 91
3 GK De Gea 27 Manchester United Spain 91
8 RCB Sergio Ramos 32 Real Madrid Spain 91
12 CB D. Godín 32 Atlético Madrid Uruguay 90
11 LCM T. Kroos 28 Real Madrid Germany 90
17 CAM A. Griezmann 27 Atlético Madrid France 89
14 LDM N. Kanté 27 Chelsea France 89
24 LCB G. Chiellini 33 Juventus Italy 89
20 CDM Sergio Busquets 29 FC Barcelona Spain 89
21 LS E. Cavani 31 Paris Saint-Germain Uruguay 89
33 LM P. Aubameyang 29 Arsenal Gabon 88
35 LB Marcelo 30 Real Madrid Brazil 88
28 LAM J. Rodríguez 26 FC Bayern München Colombia 88
25 RM K. Mbappé 19 Paris Saint-Germain France 88
45 RDM P. Pogba 25 Manchester United France 87
69 RB Azpilicueta 28 Chelsea Spain 86
67 CM Thiago 27 FC Bayern München Spain 86
56 RW Bernardo Silva 23 Manchester City Portugal 86
129 RAM J. Cuadrado 30 Juventus Colombia 84
271 CF Luis Alberto 25 Lazio Spain 82
474 LWB N. Schulz 25 TSG 1899 Hoffenheim Germany 80
450 RWB M. Ginter 24 Borussia Mönchengladbach Germany 80

Best Players for each Skill

In [35]:
## Let's Analyze the Skills of Players

@interact
def skill(skills = ['Defending', 'General', 'Mental', 'Passing', 
                       'Mobility', 'Power', 'Rating','Shooting'], score = 75):
    return data[data[skills] > score][['Name', 'Nationality', 'Club', 'Overall', skills]].sort_values(by = skills,
                                    ascending = False).head(20).style.background_gradient(cmap = 'Blues')

interactive(children=(Dropdown(description='skills', options=('Defending', 'General', 'Mental', 'Passing', 'Mo…

Let's make a function to get the list of Top 15 Players from each Country¶

In [36]:
# lets make an interactive function for getting a report of the players country wise

# lets make a function to see the list of top 15 players from each country
@interact
def country(country = list(data['Nationality'].value_counts().index)):
    return data[data['Nationality'] == country][['Name','Position','Overall',
                    'Potential']].sort_values(by = 'Overall',
                            ascending = False).head(15).style.background_gradient(cmap = 'magma')

interactive(children=(Dropdown(description='country', options=('England', 'Germany', 'Spain', 'Argentina', 'Fr…

Let's make a function to get the list of Top 15 Players from each Club¶

In [37]:
# lets make an interactive function to get the list of top 15 players from each of the club

# lets define a function 
@interact
def club(club = list(data['Club'].value_counts().index[1:])):
    return data[data['Club'] == club][['Name','Jersey Number','Position','Overall','Nationality','Age','Wage',
                     'Value','Contract Valid Until']].sort_values(by = 'Overall',
                                ascending  = False).head(15).style.background_gradient(cmap = 'inferno')

interactive(children=(Dropdown(description='club', options=('AS Monaco', 'Real Madrid', 'Fortuna Düsseldorf', …

youngest Players from the FIFA 2019

In [38]:
# finding 5 youngest Players from the dataset

youngest = data[data['Age'] == 16][['Name', 'Age', 'Club', 'Nationality', 'Overall']]
youngest.sort_values(by = 'Overall', ascending = False).head().style.background_gradient(cmap = 'magma')
Out[38]:
Name Age Club Nationality Overall
11457 W. Geubbels 16 AS Monaco France 64
11732 A. Taoui 16 Toulouse Football Club France 64
12496 Pelayo Morilla 16 Real Sporting de Gijón Spain 63
12828 Guerrero 16 CF Rayo Majadahonda Spain 63
13293 H. Massengo 16 AS Monaco France 62

15 Eldest Players from FIFA 2019

In [39]:
# finding 15 eldest players from the dataset

data.sort_values('Age', ascending = False)[['Name', 'Age', 'Club',
                              'Nationality', 'Overall']].head(15).style.background_gradient(cmap = 'Wistia')
Out[39]:
Name Age Club Nationality Overall
4741 O. Pérez 45 Pachuca Mexico 71
18183 K. Pilkington 44 Cambridge United England 48
17726 T. Warner 44 Accrington Stanley Trinidad & Tobago 53
10545 S. Narazaki 42 Nagoya Grampus Japan 65
7225 C. Muñoz 41 CD Universidad de Concepción Argentina 68
1120 J. Villar 41 No Club Paraguay 77
12192 H. Sulaimani 41 Ohod Club Saudi Arabia 63
15426 M. Tyler 41 Peterborough United England 59
4228 B. Nivet 41 ESTAC Troyes France 71
10356 F. Kippe 40 Lillestrøm SK Norway 65
16264 P. van der Vlag 40 FC Emmen Netherlands 58
9484 B. Castillo 40 Atlético Huila Colombia 66
4187 C. Lucchetti 40 Atlético Tucumán Argentina 71
2821 S. Bertoli 40 Patronato Argentina 73
3550 S. Nakamura 40 Júbilo Iwata Japan 72

The longest membership in the club

In [40]:
# The longest membership in the club

now = datetime.datetime.now()
data['Join_year'] = data.Joined.dropna().map(lambda x: x.split(',')[1].split(' ')[1])
data['Years_of_member'] = (data.Join_year.dropna().map(lambda x: now.year - int(x))).astype('int')
membership = data[['Name', 'Club', 'Years_of_member']].sort_values(by = 'Years_of_member', ascending = False).head(10)
membership.set_index('Name', inplace=True)
membership.style.background_gradient(cmap = 'Reds')
Out[40]:
Club Years_of_member
Name
O. Pérez Pachuca 29
M. Al Shalhoub Al Hilal 22
H. Sogahata Kashima Antlers 22
M. Ogasawara Kashima Antlers 22
S. Narazaki Nagoya Grampus 21
M. Wölfli BSC Young Boys 20
K. Kitamoto Vissel Kobe 20
C. Källqvist BK Häcken 19
Y. Endo Gamba Osaka 19
S. Pellissier Chievo Verona 18
In [41]:
import ipywidgets as widgets
from ipywidgets import interact
@interact
def check(column = 'Years_of_member', 
          club = ['FC Barcelona', 'Real Madrid', 'Chelsea'], x = 4):
    return data[(data[column] > x) & (data['Club'] == club)][['Name', 'Club',
                                             'Years_of_member']].sort_values(by = 'Years_of_member',
                                                                ascending = False).style.background_gradient(cmap = 'magma')

interactive(children=(Text(value='Years_of_member', description='column'), Dropdown(description='club', option…

Defining the features of players

In [42]:
# defining the features of players

player_features = ('Acceleration', 'Aggression', 'Agility', 
                   'Balance', 'BallControl', 'Composure', 
                   'Crossing', 'Dribbling', 'FKAccuracy', 
                   'Finishing', 'GKDiving', 'GKHandling', 
                   'GKKicking', 'GKPositioning', 'GKReflexes', 
                   'HeadingAccuracy', 'Interceptions', 'Jumping', 
                   'LongPassing', 'LongShots', 'Marking', 'Penalties')

# Top four features for every position in football

for i, val in data.groupby(data['Position'])[player_features].mean().iterrows():
    print('Position {}: {}, {}, {}'.format(i, *tuple(val.nlargest(4).index)))

c:\users\roshan\appdata\local\programs\python\python36\lib\site-packages\ipykernel_launcher.py:13: FutureWarning: Indexing with multiple keys (implicitly converted to a tuple of keys) will be deprecated, use a list instead.
  del sys.path[0]

Position CAM: Balance, Agility, Acceleration
Position CB: Jumping, Aggression, HeadingAccuracy
Position CDM: Aggression, Jumping, Balance
Position CF: Agility, Balance, Acceleration
Position CM: Balance, Agility, Acceleration
Position GK: GKReflexes, GKDiving, GKPositioning
Position LAM: Agility, Balance, Acceleration
Position LB: Acceleration, Balance, Agility
Position LCB: Jumping, Aggression, HeadingAccuracy
Position LCM: Balance, Agility, BallControl
Position LDM: Aggression, BallControl, LongPassing
Position LF: Balance, Agility, Acceleration
Position LM: Acceleration, Agility, Balance
Position LS: Acceleration, Agility, Finishing
Position LW: Acceleration, Agility, Balance
Position LWB: Acceleration, Agility, Balance
Position RAM: Agility, Balance, Acceleration
Position RB: Acceleration, Balance, Jumping
Position RCB: Jumping, Aggression, HeadingAccuracy
Position RCM: Agility, Balance, BallControl
Position RDM: Aggression, Jumping, BallControl
Position RF: Agility, Acceleration, Balance
Position RM: Acceleration, Agility, Balance
Position RS: Acceleration, Agility, Jumping
Position RW: Acceleration, Agility, Balance
Position RWB: Acceleration, Agility, Balance
Position ST: Acceleration, Jumping, Finishing

Top 10 left footed footballers

In [43]:
# Top 10 left footed footballers

data[data['Preferred Foot'] == 'Left'][['Name', 'Age', 'Club',
         'Nationality', 'Overall']].sort_values(by = 'Overall',
            ascending = False).head(10).style.background_gradient(cmap = 'bone')
Out[43]:
Name Age Club Nationality Overall
0 L. Messi 31 FC Barcelona Argentina 94
13 David Silva 32 Manchester City Spain 90
15 P. Dybala 24 Juventus Argentina 89
17 A. Griezmann 27 Atlético Madrid France 89
19 T. Courtois 26 Real Madrid Belgium 89
24 G. Chiellini 33 Juventus Italy 89
35 Marcelo 30 Real Madrid Brazil 88
37 H. Lloris 31 Tottenham Hotspur France 88
36 G. Bale 28 Real Madrid Wales 88
28 J. Rodríguez 26 FC Bayern München Colombia 88

Top 10 Right footed footballers

In [44]:
# Top 10 Right footed footballers

data[data['Preferred Foot'] == 'Right'][['Name', 'Age', 'Club',
                    'Nationality', 'Overall']].sort_values(by = 'Overall',
                            ascending = False).head(10).style.background_gradient(cmap = 'copper')
Out[44]:
Name Age Club Nationality Overall
1 Cristiano Ronaldo 33 Juventus Portugal 94
2 Neymar Jr 26 Paris Saint-Germain Brazil 92
3 De Gea 27 Manchester United Spain 91
4 K. De Bruyne 27 Manchester City Belgium 91
5 E. Hazard 27 Chelsea Belgium 91
6 L. Modrić 32 Real Madrid Croatia 91
7 L. Suárez 31 FC Barcelona Uruguay 91
8 Sergio Ramos 32 Real Madrid Spain 91
9 J. Oblak 25 Atlético Madrid Slovenia 90
10 R. Lewandowski 29 FC Bayern München Poland 90
In [45]:
# comparing the performance of left-footed and right-footed footballers
# ballcontrol vs dribbing

sns.lmplot(x = 'BallControl', y = 'Dribbling', data = data, col = 'Preferred Foot')
plt.show()