fifa-data-visualization.py

#!/usr/bin/env python
# coding: utf-8

# # FIFA Players Performance Data Analysis and Visualization 

# ## Importing the Libraries

# In[48]:


# basic operations
import numpy as np

# for dataframe manipulations
import streamlit as st
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('dark_background')


# **Reading the Data**

# In[49]:


# reading the data and also checking the computation time
get_ipython().run_line_magic('time', "data = pd.read_csv('data.csv')")

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


# In[50]:


# lets check the column names present in the data
data.columns


# In[51]:


#describring data for numerical columns
pd.set_option('max_columns',100)
data.iloc[:,2:].describe().style.background_gradient(cmap = 'viridis')


# ## Cleaning Data

# In[52]:


# 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[53]:


# 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[54]:


# 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[55]:


pd.set_option('max_rows', 100)
data.isnull().sum()


# In[56]:


# impute with 0 for rest of the columns
data.fillna(0, inplace = True)

# check whether the data still has any missing values
data.isnull().sum().sum()


# ## Feature Engineering

# In[57]:


# 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[58]:


# 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)

#Checking the column names in the data after adding new features
data.columns


# ## Data Visualization

# In[59]:

st.write("""
#Data Visualization


""")

# 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 = 'yellow')
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 = 'yellow')
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 = 'yellow')
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 = 'yellow')
plt.grid()

plt.suptitle('Score Distributions for Different Abilities')
plt.show()


# In[60]:


# 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[97]:


# 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()


# **Checking the Players with International Reputation as 5**

# In[62]:


data[data['International Reputation'] == 5][['Name','Nationality',
                            'Overall']].sort_values(by = 'Overall',
                                        ascending = False).style.background_gradient(cmap = 'magma')


# In[63]:


# 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[64]:


# 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[65]:


# 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[66]:


# 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[67]:


# 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[68]:


# 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[69]:


data[(data['Skill Moves'] == 5.0) & (data['Age'] < 20)][['Name','Age']]


# In[70]:


# 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[71]:


# 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[72]:


# 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[73]:


# 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[74]:


# 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[75]:


# 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[76]:


# 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[77]:


# 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[78]:


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[79]:


# 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[80]:


# 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[81]:


# 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[82]:


# 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')


# **Best Players for each Skill**

# In[83]:


## 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')


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

# In[84]:


# Making an interactive function for getting a report of the players country wise

# Making 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')


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

# In[85]:


# 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')


# **youngest Players from the FIFA 2019**

# In[86]:


# 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')


# **15 Eldest Players from FIFA 2019**

# In[87]:


# 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')


# **The longest membership in the club**

# In[88]:


# 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')


# In[89]:


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')


# **Defining the features of players**

# In[90]:


# 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)))


# **Top 10 left footed footballers**

# In[91]:


# 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')


# **Top 10 Right footed footballers**

# In[92]:


# 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')


# In[93]:


# 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()