
# Overbuff data from GitHub CSV online (NOTE: Data is from May 2020).

import requests
import pandas as pd

# URL: https://github.com/jamie-ralph/overbuff-webscrape/blob/master/all_data.csv
overbuff = pd.read_csv("https://raw.githubusercontent.com/jamie-ralph/overbuff-webscrape/master/all_data.csv", sep = ',')

# Dropping the first column as it is not needed.
overbuff = overbuff.drop('Unnamed: 0', axis = 1)

# Creating a new dataframe for PC only.
overbuff_pc = overbuff.loc[overbuff['Platform'] == 'PC']

# Renaming columns to keep the same column-naming schema.
overbuff_pc = overbuff_pc.rename(columns = {'Pick_rate': 'Pick %', 'Win_rate': 'Win %', 'Tie_Rate': 'Tie %', 'On_fire': 'On Fire %'})

# Creating a unique list of heroes in Overwatch 1 and sorts them in alphabetical order.
hero_list = []
for i in range(256):
    if overbuff_pc['Hero'][i] not in hero_list:
        hero_list.append(overbuff_pc['Hero'][i])
hero_list.sort()

display(overbuff_pc)  # NOTE: Displays data for PC. This contains each rank separately, including all ranks combined.

# Creating a new dataframe for PSN (Playstation) only.
overbuff_psn = overbuff.loc[overbuff['Platform'] == 'PSN']


# Creating a new dataframe for Xbox only.
overbuff_xbox = overbuff.loc[overbuff['Platform'] == 'XBox']


# Creating a new dataframe for PC by each rank.
overbuff_pc_all = overbuff_pc.loc[overbuff_pc['Rank'] == 'All']
overbuff_pc_bronze = overbuff_pc.loc[overbuff_pc['Rank'] == 'Bronze']
overbuff_pc_silver = overbuff_pc.loc[overbuff_pc['Rank'] == 'Silver']
overbuff_pc_gold = overbuff_pc.loc[overbuff_pc['Rank'] == 'Gold']
overbuff_pc_diamond = overbuff_pc.loc[overbuff_pc['Rank'] == 'Diamond']
overbuff_pc_master = overbuff_pc.loc[overbuff_pc['Rank'] == 'Master']
overbuff_pc_grandmaster = overbuff_pc.loc[overbuff_pc['Rank'] == 'Grandmaster']

display(overbuff_pc_all) # NOTE: Shows PC data for all ranks combined.

# The same dataframe of all ranks as above just sorted alphabetically by hero.
overbuff_pc_all_sorted = overbuff_pc_all.sort_values('Hero')


# Creating a new dataframe for PC by each role (NOTE: this combines every rank)
overbuff_pc_tank = overbuff_pc.loc[overbuff_pc['Role'] == 'TANK']
overbuff_pc_support = overbuff_pc.loc[overbuff_pc['Role'] == 'SUPPORT']
overbuff_pc_offense = overbuff_pc.loc[overbuff_pc['Role'] == 'OFFENSE']
overbuff_pc_defense = overbuff_pc.loc[overbuff_pc['Role'] == 'DEFENSE']

# Resetting indices for PC tank data (for cleaner display purposes).
overbuff_pc_tank = overbuff_pc_tank.reset_index()

# Dropping old index table.
overbuff_pc_tank = overbuff_pc_tank.drop('index', axis = 1)

display(overbuff_pc_tank) # NOTE: Shows PC data for the role 'TANK'.


# SECOND, Overbuff Grandmaster data from Overbuff, from GitHub CSV online (NOTE: Data is from Overbuff.com December 2021).


import requests
import pandas as pd

# URL: https://github.com/rshah5324/rshah5324.github.io/blob/main/overbuff2021grandmaster.csv
overbuff_2021 = pd.read_csv("https://raw.githubusercontent.com/rshah5324/rshah5324.github.io/main/overbuff2021grandmaster.csv", sep=',')

# Dropping rows 31-35 as these characters are exclusive to Overwatch 2 and also lack enough data still.
overbuff_2021 = overbuff_2021.drop(range(32,36))

# Sorting the dataframe alphabetically by hero.
overbuff_2021 = overbuff_2021.sort_values('Hero')

# Reindexing the dataframe to start at 1 and stop at 32.
overbuff_2021.index = pd.RangeIndex(start = 1, stop = 33, step = 1)

for i in range(1,33):
    
    # Changing 'Hero' column to fix hero names.
    overbuff_2021.loc[i,'Hero'] = hero_list[i-1]
    
    # Changing 'Pick Rate' column values to be floats.
    str = overbuff_2021.loc[i,'Pick Rate'][:-1]
    overbuff_2021.loc[i,'Pick Rate'] = float(str)
    
    # Changing 'Win Rate' column values to be floats.
    str = overbuff_2021.loc[i,'Win Rate'][:-1]
    overbuff_2021.loc[i,'Win Rate'] = float(str)

# Renamig 'Pick Rate' column to 'Pick %' and 'Win Rate' column to 'Win %' (keeping column naming consistent).
overbuff_2021 = overbuff_2021.rename(columns = {'Pick Rate': 'Pick %', 'Win Rate': 'Win %'})

# Dropping old index column from CSV file.
overbuff_2021 = overbuff_2021.drop('Unnamed: 0', axis=1)

# Adding roles to every hero.
overbuff_2021.insert(1,'Role', overbuff_pc_all_sorted['Role'].tolist())

# Printing cleaned up Overbuff dataframe.
display(overbuff_2021)


# THIRD, importing necessary packages that we will use and data from esportstales.com (NOTE: Data is from April 2022)
import pandas as pd

import plotly.io as pio
pio.renderers.default='notebook'

pd.set_option('display.max_rows', 100)
pd.set_option('display.max_columns', 50)
pd.set_option('display.width', 300)



# Getting our data from esportstales.com (NOTE: Data is from April 2022)

# esportstales URL
esports = pd.read_html('https://www.esportstales.com/overwatch/tier-list-and-most-played-heroes')

# The esports April 2022 Most Popular Hero by Rank
esports_rank = esports[0]

# Renaming '#' to 'Popularity Ranking' to make the column name more descriptive.
esports_rank = esports_rank.rename(columns = {'#': 'Popularity Ranking'})

# Displaying the first table (Most Popular Hero by Rank)
display(esports_rank)



# The following esports April 2022 'tiers' display the pick % and the win % (they are just the Grandmaster data, NOT Master, Diamond..etc)/
esports_tier1 = esports[1]
esports_tier2 = esports[2]
esports_tier3 = esports[3]
esports_tier4 = esports[4]

# Creating a new column, 'Tier', and displaying the respective tier for each hero.
esports_tier1['Tier'] = 1
esports_tier2['Tier'] = 2
esports_tier3['Tier'] = 3
esports_tier4['Tier'] = 4

# Combining the different tier tables into one large table with every hero in Overwatch 1.
frames = [esports_tier1, esports_tier2, esports_tier3, esports_tier4]
df1 = pd.concat(frames)

df1 = pd.DataFrame(df1)

# Renaming the columns 'Name' to 'Hero Name', 'Pick%' to 'Pick %', and 'Win%' to 'Win %'.
df1 = df1.rename(columns = {'Hero': 'Delete','Name': 'Hero', 'Pick%': 'Pick %', 'Win%': 'Win %'})

# Dropping the 'Hero' column (Removing redundancy).
df1 = df1.drop('Delete', axis = 1)

# Sorting the table in alphabetical order by hero name.
df1 = df1.sort_values('Hero')

# Resetting the indices on the far left to make them start from 1.
df1.index = pd.RangeIndex(start = 1, stop = 33, step = 1)

# Change Cassidy to McCree
df1.Hero = df1.Hero.apply(lambda x: 'McCree' if 'Cassidy' in x else x)

# Printing out the dataframe (NOTE****, this is DATA from GRANDMASTER ONLY***).
df1 = df1.sort_values('Hero')


# Resetting the indices on the far left to make them start from 1.
df1.index = pd.RangeIndex(start = 1, stop = 33, step = 1)

# Adding roles to dataframe.
df1.insert(1, 'Role', overbuff_pc_all_sorted['Role'].tolist())

# Sorting by win percentage.
esports_win_percent = df1.sort_values('Win %', ascending = False)

# Sorting by pick percentage.
esports_pick_percent = df1.sort_values('Pick %', ascending = False)


# Sorting by hero name alphabetically.
esports_name_order = df1.sort_values('Hero')

# Stripping '%' from Win % and making it a float.
esports_name_order['Win %'] = esports_name_order['Win %'].str.rstrip('%').astype('float') 

# Stripping '%' from Pick % and making it a float.
esports_name_order['Pick %'] = esports_name_order['Pick %'].str.rstrip('%').astype('float')


# Displaying order by name.
display(esports_name_order)


# FOURTH, getting our data from overbuff.com for Overwatch 2 November 2022 Data.

import requests
import pandas as pd

# URL: https://raw.githubusercontent.com/rshah5324/rshah5324.github.io/main/overbuff2022.csv
overbuff_2022 = pd.read_csv("https://raw.githubusercontent.com/rshah5324/rshah5324.github.io/main/overbuff2022.csv", sep=',')

# Dropping rows 31-35 as these characters are exclusive to Overwatch 2 and also lack enough data still.
overbuff_2022 = overbuff_2022.drop(range(32,36))

# Sorting the dataframe alphabetically by hero.
overbuff_2022 = overbuff_2022.sort_values('Hero')

# Reindexing the dataframe to start at 1 and stop at 32.
overbuff_2022.index = pd.RangeIndex(start = 1, stop = 33, step = 1)

for i in range(1,33):
    
    # Changing 'Hero' column to fix hero names.
    overbuff_2022.loc[i,'Hero'] = hero_list[i-1]
    
    # Changing 'Pick Rate' column values to be floats.
    str = overbuff_2022.loc[i,'Pick Rate'][:-1]
    overbuff_2022.loc[i,'Pick Rate'] = float(str)
    
    # Changing 'Win Rate' column values to be floats.
    str = overbuff_2022.loc[i,'Win Rate'][:-1]
    overbuff_2022.loc[i,'Win Rate'] = float(str)

# Renaming 'Pick Rate' column to 'Pick %' and 'Win Rate' column to 'Win %' (keeping column naming consistent).
overbuff_2022 = overbuff_2022.rename(columns = {'Pick Rate': 'Pick %', 'Win Rate': 'Win %'})

# Dropping old index column from CSV file.
overbuff_2022 = overbuff_2022.drop('Unnamed: 0', axis = 1)

# Adding roles to every hero.
overbuff_2022.insert(1, 'Role', overbuff_pc_all_sorted['Role'].tolist())

# Printing cleaned up Overbuff dataframe.
display(overbuff_2022)


# Here, we will find the average pick rate of every role and print out how often a rank chooses it.

import matplotlib.pyplot as plt
import pandas as pd
import numpy

plt.rcParams["figure.figsize"] = (10,6)

# Getting a list of each unique rank (Bronze, Silver, Gold, etc.).
ranks = overbuff_pc['Rank'].unique()

# List to hold each average pick percentage.
average_percent = []

# Color for each bar that will be graphed.
c = ['red', 'brown', 'grey', 'yellow', 'silver', 'blue', 'orange', 'black']

for rank in ranks:

  # Creating a new temporary dataframe for each.
  curr_table = overbuff_pc.loc[overbuff_pc['Rank'] == rank]

  # Getting the tanks.
  curr_table = curr_table.loc[curr_table['Role'] == 'TANK']

  # Addding up the pick rate of each tank.
  total_percent = curr_table['Pick %'].sum()

  # Finding the average pick rate of tanks for each rank.
  average_pick_rate = total_percent / len(curr_table)

  average_percent.append(average_pick_rate)


fig = plt.figure()
ax = fig.add_axes([0,0,1,1])
ax.bar(ranks, average_percent, color = c)
plt.title('Average Pick Rate of Tank Role Based on Rank', fontsize = 20)
plt.ylabel('Average % Pick Rate', fontsize = 20)
plt.show()


import matplotlib.pyplot as plt
import pandas as pd
import numpy

plt.rcParams["figure.figsize"] = (10,6)

# Getting a list of each unique role.
ranks = overbuff_pc['Rank'].unique()

# List to hold each percent.
average_percent = []

# Color for each bar that will be graphed.
c = ['red', 'brown', 'grey', 'yellow', 'silver', 'blue', 'orange', 'black']

for rank in ranks:

  # Creating a new temporare dataframe for each.
  curr_table = overbuff_pc.loc[overbuff_pc['Rank'] == rank]

  # Getting the supports.
  curr_table = curr_table.loc[curr_table['Role'] == 'SUPPORT']

  total_percent = curr_table['Pick %'].sum()

  # Finding the average pick rate of support for each rank.
  average_pick_rate = total_percent / len(curr_table)


  average_percent.append(average_pick_rate)


fig = plt.figure()
ax = fig.add_axes([0,0,1,1])
ax.bar(ranks, average_percent, color = c)
plt.title('Average Pick Rate of Support Role Based on Rank', fontsize = 20)
plt.ylabel('Average % Pick Rate', fontsize = 20)
plt.show()


import matplotlib.pyplot as plt
import pandas as pd
import numpy

plt.rcParams["figure.figsize"] = (10,6)

# Getting a list of each unique role.
ranks = overbuff_pc['Rank'].unique()

# List to hold each percent.
average_percent = []

# Color for each bar that will be graphed.
c = ['red', 'brown', 'grey', 'yellow', 'silver', 'blue', 'orange', 'black']

for rank in ranks:

  # Creating a new temporare dataframe for each.
  curr_table = overbuff_pc.loc[overbuff_pc['Rank'] == rank]

  # Getting the offense.
  curr_table = curr_table.loc[curr_table['Role'] == 'OFFENSE']

  total_percent = curr_table['Pick %'].sum()

  # Finding the average pick rate of offense for each rank.
  average_pick_rate = total_percent / len(curr_table)


  average_percent.append(average_pick_rate)


fig = plt.figure()
ax = fig.add_axes([0,0,1,1])
ax.bar(ranks, average_percent, color = c)
plt.title('Average Pick Rate of Offense Role Based on Rank', fontsize = 20)
plt.ylabel('Average % Pick Rate', fontsize = 20)
plt.show()


import matplotlib.pyplot as plt
import pandas as pd
import numpy

plt.rcParams["figure.figsize"] = (10,6)

# Getting a list of each unique role.
ranks = overbuff_pc['Rank'].unique()

# List to hold each percent.
average_percent = []

# Color for each bar that will be graphed.
c = ['red', 'brown', 'grey', 'yellow', 'silver', 'blue', 'orange', 'black']

for rank in ranks:

  # Creating a new temporary dataframe for each.
  curr_table = overbuff_pc.loc[overbuff_pc['Rank'] == rank]

  # Getting the defense.
  curr_table = curr_table.loc[curr_table['Role'] == 'DEFENSE']

  total_percent = curr_table['Pick %'].sum()

  # Finding the average pick rate of defense for each rank.
  average_pick_rate = total_percent / len(curr_table)


  average_percent.append(average_pick_rate)


fig = plt.figure()
ax = fig.add_axes([0,0,1,1])
ax.bar(ranks, average_percent, color = c)
plt.title('Average Pick Rate of Defense Role Based on Rank', fontsize = 20)
plt.ylabel('Average % Pick Rate', fontsize = 20)
plt.show()


# Based on the above data, we will find out how often, on average, a certain rank will win using a certain role.

import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

plt.rcParams["figure.figsize"] = (15,10)

# Getting a list of each unique role.
ranks = overbuff_pc['Rank'].unique()

# Tank characters.
characters = overbuff_pc_tank['Hero'].unique()

# List to hold each percent.
average_percent = []


# Iterating through each unique role and their tank win rate.
for hero in characters:

  # Creating a new temporary dataframe for each.
  curr_table = overbuff_pc.loc[overbuff_pc['Hero'] == hero]

  # Getting the tanks.
  curr_table = curr_table.loc[curr_table['Role'] == 'TANK']

  plt.plot(curr_table['Rank'], curr_table['Win %'], color=np.random.rand(3,), label = hero, marker ='o')

plt.title('Winning Rate of Tanks Based on Rank', fontsize = 15)
plt.xlabel('Rank', fontsize = 15)
plt.ylabel('Win Rate %', fontsize = 20)
plt.grid(True)
plt.legend()  # Displaying the hero on the corner.
plt.show()


import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

plt.rcParams["figure.figsize"] = (15,10)

# Getting a list of each unique role.
ranks = overbuff_pc['Rank'].unique()

# Support characters.
characters = overbuff_pc_support['Hero'].unique()

# List to hold each percent.
average_percent = []


# Iterating through each unique role and their support win rate.
for hero in characters:

  # Creating a new temporary dataframe for each.
  curr_table = overbuff_pc.loc[overbuff_pc['Hero'] == hero]

  # Getting the supports.
  curr_table = curr_table.loc[curr_table['Role'] == 'SUPPORT']

  plt.plot(curr_table['Rank'], curr_table['Win %'], color=np.random.rand(3,), label = hero, marker ='o')

plt.title('Winning Rate of Support Based on Rank', fontsize = 15)
plt.xlabel('Rank', fontsize = 15)
plt.ylabel('Win Rate %', fontsize = 20)
plt.grid(True)
plt.legend()  # Displaying the hero on the corner.
plt.show()


import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

plt.rcParams["figure.figsize"] = (15,10)

# Getting a list of each unique role.
ranks = overbuff_pc['Rank'].unique()

# Offense characters.
characters = overbuff_pc_offense['Hero'].unique()

# List to hold each percent.
average_percent = []


# Iterating through each unique role and their offense win rate.
for hero in characters:

  # Creating a new temporary dataframe for each.
  curr_table = overbuff_pc.loc[overbuff_pc['Hero'] == hero]

  # Getting the offense.
  curr_table = curr_table.loc[curr_table['Role'] == 'OFFENSE']

  plt.plot(curr_table['Rank'], curr_table['Win %'], color=np.random.rand(3,), label = hero, marker ='o')

plt.title('Winning Rate of Offense Based on Rank', fontsize = 15)
plt.xlabel('Rank', fontsize = 15)
plt.ylabel('Win Rate %', fontsize = 20)
plt.grid(True)
plt.legend()  # Displaying the hero on the corner.
plt.show()


import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

plt.rcParams["figure.figsize"] = (15,10)

# Getting a list of each unique role.
ranks = overbuff_pc['Rank'].unique()

# Defense characters.
characters = overbuff_pc_defense['Hero'].unique()

# List to hold each percent.
average_percent = []


# Iterate through each unique role and their defense win rate.
for hero in characters:

  # Create a new temporary dataframe for each.
  curr_table = overbuff_pc.loc[overbuff_pc['Hero'] == hero]

  # Getting the defense.
  curr_table = curr_table.loc[curr_table['Role'] == 'DEFENSE']

  plt.plot(curr_table['Rank'], curr_table['Win %'], color=np.random.rand(3,), label = hero, marker ='o')

plt.title('Winning Rate of Defense Based on Rank', fontsize = 15)
plt.xlabel('Rank', fontsize = 15)
plt.ylabel('Win Rate %', fontsize = 20)
plt.grid(True)
plt.legend()  # Display the hero on the corner.
plt.show()


import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

plt.rcParams["figure.figsize"] = (15,10)


# Tank characters.
characters = overbuff_pc_grandmaster['Hero'].unique()


# Getting the Grandmaster dataframe for tanks only.
overbuff_pc_grandmaster_tank = overbuff_pc_grandmaster.loc[overbuff_pc_grandmaster['Role'] == 'TANK']


# Tank characters.
characters = overbuff_pc_grandmaster_tank['Hero'].tolist()

win_per = overbuff_pc_grandmaster_tank['Win %'].tolist()

pick_per = overbuff_pc_grandmaster_tank['Pick %'].tolist()

# Plotting the points.
plt.scatter(win_per, pick_per, label='Tank Hero', color='purple')

# Write the hero names above the plot.
for i in range(len(win_per)):
  plt.text(win_per[i], pick_per[i] + 0.1, characters[i], ha="center", va="bottom")


# Obtaining the slope, m, and intercept, b, of the plot.
m, b = np.polyfit(win_per, pick_per, 1)

# Plotting the regression line.
plt.plot(win_per, m*(np.array(win_per))+b, color='orange')

plt.title('Tank Win Rate to Pick Rate in Grandmaster', fontsize = 15)
plt.xlabel('Win Rate %', fontsize = 15)
plt.ylabel('Pick Rate %', fontsize = 15)
plt.grid(True)
plt.legend()  # Displaying the hero on the corner.
plt.show()


import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

plt.rcParams["figure.figsize"] = (15,10)


# Support characters.
characters = overbuff_pc_grandmaster['Hero'].unique()


# Getting the Grandmaster dataframe for support only.
overbuff_pc_grandmaster_support = overbuff_pc_grandmaster.loc[overbuff_pc_grandmaster['Role'] == 'SUPPORT']


# Support characters.
characters = overbuff_pc_grandmaster_support['Hero'].tolist()

win_per = overbuff_pc_grandmaster_support['Win %'].tolist()

pick_per = overbuff_pc_grandmaster_support['Pick %'].tolist()

# Plotting the points.
plt.scatter(win_per, pick_per, label = 'Support Hero', color ='green')

# Writing the hero names above the plot.
for i in range(len(win_per)):
  plt.text(win_per[i], pick_per[i] + 0.1, characters[i], ha="center", va="bottom")

# Obtaining the slope, m, and intercept, b, of the plot.
m, b = np.polyfit(win_per, pick_per, 1)

# Plotting the regression line.
plt.plot(win_per, m*(np.array(win_per))+b, color='orange')

plt.title('Support Win Rate to Pick Rate in Grandmaster', fontsize = 15)
plt.xlabel('Win Rate %', fontsize = 15)
plt.ylabel('Pick Rate %', fontsize = 15)
plt.grid(True)
plt.legend()  # Displaying the hero on the corner.
plt.show()


import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

plt.rcParams["figure.figsize"] = (15,10)


# Offense characters.
characters = overbuff_pc_grandmaster['Hero'].unique()


# Getting the Grandmaster dataframe for offense only.
overbuff_pc_grandmaster_offense = overbuff_pc_grandmaster.loc[overbuff_pc_grandmaster['Role'] == 'OFFENSE']


# Offense characters.
characters = overbuff_pc_grandmaster_offense['Hero'].tolist()

win_per = overbuff_pc_grandmaster_offense['Win %'].tolist()

pick_per = overbuff_pc_grandmaster_offense['Pick %'].tolist()

# Plotting the points.
plt.scatter(win_per, pick_per, label='Offense Hero', color='red')

# Writing the hero names above the plot.
for i in range(len(win_per)):
  plt.text(win_per[i], pick_per[i] + 0.05, characters[i], ha="center", va="bottom")

# Obtaining the slope, m, and intercept, b, of the plot.
m, b = np.polyfit(win_per, pick_per, 1)

# Plotting the regression line.
plt.plot(win_per, m*(np.array(win_per))+b, color='orange')

plt.title('Offense Win Rate to Pick Rate in Grandmaster', fontsize = 15)
plt.xlabel('Win Rate %', fontsize = 15)
plt.ylabel('Pick Rate %', fontsize = 15)
plt.grid(True)
plt.legend()  # Displaying the hero on the corner.
plt.show()


import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

plt.rcParams["figure.figsize"] = (15,10)


# Defense characters.
characters = overbuff_pc_grandmaster['Hero'].unique()


# Getting the Grandmaster dataframe for defense only.
overbuff_pc_grandmaster_defense = overbuff_pc_grandmaster.loc[overbuff_pc_grandmaster['Role'] == 'DEFENSE']


# Defense characters.
characters = overbuff_pc_grandmaster_defense['Hero'].tolist()

win_per = overbuff_pc_grandmaster_defense['Win %'].tolist()

pick_per = overbuff_pc_grandmaster_defense['Pick %'].tolist()

# Plotting the points.
plt.scatter(win_per, pick_per, label = 'Defense Hero', color='blue')

# Writing the hero names above the plot.
for i in range(len(win_per)):
  plt.text(win_per[i], pick_per[i] + 0.05, characters[i], ha="center", va="bottom")



# Obtaining the slope, m, and intercept, b, of the plot.
m, b = np.polyfit(win_per, pick_per, 1)

# Plotting the regression line.
plt.plot(win_per, m*(np.array(win_per))+b, color='orange')

plt.title('Defense Win Rate to Pick Rate in Grandmaster', fontsize = 15)
plt.xlabel('Win Rate %', fontsize = 15)
plt.ylabel('Pick Rate %', fontsize = 15)
plt.grid(True)
plt.legend()  # Displaying the hero on the corner.
plt.show()


import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

# Data for Grandmasters in May 2020.
overbuff_pc_tank_grandmaster = overbuff_pc_tank.loc[overbuff_pc_tank['Rank'] == 'Grandmaster']

overbuff_pc_tank_grandmaster = overbuff_pc_tank_grandmaster.sort_values('Hero')

# Data for Grandmasters in December 2021.
overbuff_2021_tank_grandmaster = overbuff_2021.loc[overbuff_2021['Role'] == 'TANK']

# Data for Grandmasters in April 2022.
esports_name_order_tank_grandmaster = esports_name_order.loc[esports_name_order['Role'] == 'TANK']

# Data for Grandmasters in November 2022.
overbuff_2022_tank_grandmaster = overbuff_2022.loc[overbuff_2022['Role'] == 'TANK']

fig, (ax1, ax2, ax3, ax4) = plt.subplots(nrows=1, ncols=4)

# Plotting violin plot on axes 1.
ax1.violinplot(overbuff_pc_tank_grandmaster['Win %'], showmedians = True)
ax1.set_title('Tank Win Rate in May 2020')
ax1.set_ylabel("Win %")


# Plotting violin plot on axes 2.
ax2.violinplot(overbuff_2021_tank_grandmaster['Win %'].tolist(), showmedians = True)
ax2.set_title('Tank Win Rate in December 2021')

# Plotting violin plot on axes 3.
ax3.violinplot(esports_name_order_tank_grandmaster['Win %'].tolist(), showmedians = True)
ax3.set_title('Tank Win Rate in April 2022')

# Plotting violin plot on axes 4.
ax4.violinplot(overbuff_2022_tank_grandmaster['Win %'].tolist(), showmedians = True)
ax4.set_title('Tank Win Rate in November 2022 (OW2)')

plt.show()


import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

# Data for Grandmasters in May 2020.
overbuff_pc_support_grandmaster = overbuff_pc_support.loc[overbuff_pc_support['Rank'] == 'Grandmaster']

overbuff_pc_support_grandmaster = overbuff_pc_support_grandmaster.sort_values('Hero')

# Data for Grandmasters in December 2021.
overbuff_2021_support_grandmaster = overbuff_2021.loc[overbuff_2021['Role'] == 'SUPPORT']

# Data for Grandmasters in April 2022.
esports_name_order_support_grandmaster = esports_name_order.loc[esports_name_order['Role'] == 'SUPPORT']

# Data for Grandmasters in November 2022.
overbuff_2022_support_grandmaster = overbuff_2022.loc[overbuff_2022['Role'] == 'SUPPORT']

fig, (ax1, ax2, ax3, ax4) = plt.subplots(nrows=1, ncols=4)

# Plotting violin plot on axes 1.
ax1.violinplot(overbuff_pc_support_grandmaster['Win %'], showmedians=True)
ax1.set_title('Support Win Rate in May 2020')
ax1.set_ylabel("Win %")


# Plotting violin plot on axes 2.
ax2.violinplot(overbuff_2021_support_grandmaster['Win %'].tolist(), showmedians=True)
ax2.set_title('Support Win Rate in December 2021')

# Plotting violin plot on axes 3.
ax3.violinplot(esports_name_order_support_grandmaster['Win %'].tolist(), showmedians=True)
ax3.set_title('Support Win Rate in April 2022')

# Plotting violin plot on axes 4.
ax4.violinplot(overbuff_2022_support_grandmaster['Win %'].tolist(), showmedians=True)
ax4.set_title('Support Win Rate in November 2022 (OW2)')

plt.show()


import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

# Data for Grandmasters in May 2020.
overbuff_pc_offense_grandmaster = overbuff_pc_offense.loc[overbuff_pc_offense['Rank'] == 'Grandmaster']

overbuff_pc_offense_grandmaster = overbuff_pc_offense_grandmaster.sort_values('Hero')

# Data for Grandmasters in December 2021.
overbuff_2021_offense_grandmaster = overbuff_2021.loc[overbuff_2021['Role'] == 'OFFENSE']

# Data for Grandmasters in April 2022.
esports_name_order_offense_grandmaster = esports_name_order.loc[esports_name_order['Role'] == 'OFFENSE']

# Data for Grandmasters in November 2022.
overbuff_2022_offense_grandmaster = overbuff_2022.loc[overbuff_2022['Role'] == 'OFFENSE']

fig, (ax1, ax2, ax3, ax4) = plt.subplots(nrows=1, ncols=4)

# Plotting violin plot on axes 1.
ax1.violinplot(overbuff_pc_offense_grandmaster['Win %'], showmedians=True)
ax1.set_title('Offense Win Rate in May 2020')
ax1.set_ylabel("Win %")


# Plotting violin plot on axes 2.
ax2.violinplot(overbuff_2021_offense_grandmaster['Win %'].tolist(), showmedians=True)
ax2.set_title('Offense Win Rate in December 2021')

# Plotting violin plot on axes 3.
ax3.violinplot(esports_name_order_offense_grandmaster['Win %'].tolist(), showmedians=True)
ax3.set_title('Offense Win Rate in April 2022')

# Plotting violin plot on axes 4.
ax4.violinplot(overbuff_2022_offense_grandmaster['Win %'].tolist(), showmedians=True)
ax4.set_title('Offense Win Rate in November 2022 (OW2)')

plt.show()


import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

# Data for Grandmasters in May 2020.
overbuff_pc_defense_grandmaster = overbuff_pc_defense.loc[overbuff_pc_defense['Rank'] == 'Grandmaster']

overbuff_pc_defense_grandmaster = overbuff_pc_defense_grandmaster.sort_values('Hero')

# Data for Grandmasters in December 2021.
overbuff_2021_defense_grandmaster = overbuff_2021.loc[overbuff_2021['Role'] == 'DEFENSE']

# Data for Grandmasters in April 2022.
esports_name_order_defense_grandmaster = esports_name_order.loc[esports_name_order['Role'] == 'DEFENSE']

# Data for Grandmasters in November 2022.
overbuff_2022_defense_grandmaster = overbuff_2022.loc[overbuff_2022['Role'] == 'DEFENSE']

fig, (ax1, ax2, ax3, ax4) = plt.subplots(nrows=1, ncols=4)

# Plotting violin plot on axes 1.
ax1.violinplot(overbuff_pc_defense_grandmaster['Win %'], showmedians=True)
ax1.set_title('Defense Win Rate in May 2020')
ax1.set_ylabel("Win %")


# Plotting violin plot on axes 2.
ax2.violinplot(overbuff_2021_defense_grandmaster['Win %'].tolist(), showmedians=True)
ax2.set_title('Defense Win Rate in December 2021')

# Plotting violin plot on axes 3.
ax3.violinplot(esports_name_order_defense_grandmaster['Win %'].tolist(), showmedians=True)
ax3.set_title('Defense Win Rate in April 2022')

# Plotting violin plot on axes 4.
ax4.violinplot(overbuff_2022_defense_grandmaster['Win %'].tolist(), showmedians=True)
ax4.set_title('Defense Win Rate in November 2022 (OW2)')

plt.show()


# Resetting the indices of the grandmaster dataframes of all four time periods for defense heroes.
overbuff_pc_defense_grandmaster = overbuff_pc_defense_grandmaster.reset_index()
overbuff_2021_defense_grandmaster = overbuff_2021_defense_grandmaster.reset_index()
esports_name_order_defense_grandmaster = esports_name_order_defense_grandmaster.reset_index()
overbuff_2022_defense_grandmaster = overbuff_2022_defense_grandmaster.reset_index()

# Getting the unique defense heros.
characters = overbuff_pc_defense_grandmaster['Hero'].unique()

# Indices for the x-axis.
time = ['May 2020', 'December 2021', 'April 2022', 'November 2022']

# Temporary list to hold win percentages for each defense hero.
temp = []
counter = 0

# Iterating through each defense hero.
for hero in characters:
    
    # Appending each win percentage each year for each defense hero.
    temp.append(overbuff_pc_defense_grandmaster['Win %'][counter])
    temp.append(overbuff_2021_defense_grandmaster['Win %'][counter])
    temp.append(esports_name_order_defense_grandmaster['Win %'][counter])
    temp.append(overbuff_2022_defense_grandmaster['Win %'][counter])
    
    # Plotting the win percentage each hero for each defense hero.
    plt.plot(time, temp, color = np.random.rand(3,), label = hero, marker ='o')
    counter += 1
    temp = []

plt.title('Win Percentage of Defense Heroes Over Time', fontsize = 15)
plt.xlabel('Years', fontsize = 15)
plt.ylabel('Win Percentage', fontsize = 20)
plt.grid(True)
plt.legend()  # Displaying the hero on the corner.
plt.show()


# Resetting the indices of the grandmaster dataframes of all four time periods for tank heroes.
overbuff_pc_tank_grandmaster = overbuff_pc_tank_grandmaster.reset_index()
overbuff_2021_tank_grandmaster = overbuff_2021_tank_grandmaster.reset_index()
esports_name_order_tank_grandmaster = esports_name_order_tank_grandmaster.reset_index()
overbuff_2022_tank_grandmaster = overbuff_2022_tank_grandmaster.reset_index()

# Getting the unique tank heros.
characters = overbuff_pc_tank_grandmaster['Hero'].unique()

# Temporary list to hold win percentages for each tank hero.
temp = []
counter = 0

# Iterating through each tank hero.
for hero in characters:
    # Appending each win percentage each year for each tank hero.
    temp.append(overbuff_pc_tank_grandmaster['Win %'][counter])
    temp.append(overbuff_2021_tank_grandmaster['Win %'][counter])
    temp.append(esports_name_order_tank_grandmaster['Win %'][counter])
    temp.append(overbuff_2022_tank_grandmaster['Win %'][counter])
    
    # Plotting the win percentage each hero for each tank hero.
    plt.plot(time, temp, color=np.random.rand(3,), label = hero, marker ='o')
    counter += 1
    temp = []

plt.title('Win Percentage of Tank Heroes Over Time', fontsize = 15)
plt.xlabel('Years', fontsize = 15)
plt.ylabel('Win Percentage', fontsize = 20)
plt.grid(True)
plt.legend()  # Displaying the hero on the corner.
plt.show()

	Hero	Role	Pick %	Win %	Tie %	On Fire %	Rank	Date	Platform
0	Reinhardt	TANK	12.08	52.10	2.61	7.15	All	2020-05-24 11:54:16.967151	PC
1	Ana	SUPPORT	10.80	49.92	2.36	9.81	All	2020-05-24 11:54:16.967151	PC
2	Moira	SUPPORT	8.28	49.95	2.45	10.52	All	2020-05-24 11:54:16.967151	PC
3	Zarya	TANK	7.51	52.90	2.47	12.39	All	2020-05-24 11:54:16.967151	PC
4	Mercy	SUPPORT	6.56	51.32	2.42	2.46	All	2020-05-24 11:54:16.967151	PC
...	...	...	...	...	...	...	...	...	...
251	Junkrat	DEFENSE	0.72	57.72	0.73	7.28	Grandmaster	2020-05-24 12:00:03.487597	PC
252	Symmetra	SUPPORT	0.51	50.53	1.04	6.50	Grandmaster	2020-05-24 12:00:03.487597	PC
253	Sombra	OFFENSE	0.44	62.05	0.60	13.35	Grandmaster	2020-05-24 12:00:03.487597	PC
254	Bastion	DEFENSE	0.22	49.41	0.00	5.88	Grandmaster	2020-05-24 12:00:03.487597	PC
255	Reaper	OFFENSE	0.21	53.75	1.23	11.46	Grandmaster	2020-05-24 12:00:03.487597	PC

	Hero	Role	Pick %	Win %	Tie %	On Fire %	Rank	Date	Platform
0	Reinhardt	TANK	12.08	52.10	2.61	7.15	All	2020-05-24 11:54:16.967151	PC
1	Ana	SUPPORT	10.80	49.92	2.36	9.81	All	2020-05-24 11:54:16.967151	PC
2	Moira	SUPPORT	8.28	49.95	2.45	10.52	All	2020-05-24 11:54:16.967151	PC
3	Zarya	TANK	7.51	52.90	2.47	12.39	All	2020-05-24 11:54:16.967151	PC
4	Mercy	SUPPORT	6.56	51.32	2.42	2.46	All	2020-05-24 11:54:16.967151	PC
5	McCree	OFFENSE	4.72	47.43	1.79	8.60	All	2020-05-24 11:54:16.967151	PC
6	Lúcio	SUPPORT	4.36	51.76	2.06	11.86	All	2020-05-24 11:54:16.967151	PC
7	Ashe	OFFENSE	3.83	53.39	2.20	9.85	All	2020-05-24 11:54:16.967151	PC
8	Roadhog	TANK	3.40	49.70	1.75	9.34	All	2020-05-24 11:54:16.967151	PC
9	Sigma	TANK	3.07	50.85	2.22	12.90	All	2020-05-24 11:54:16.967151	PC
10	Genji	OFFENSE	2.93	50.54	2.17	11.95	All	2020-05-24 11:54:16.967151	PC
11	Soldier: 76	OFFENSE	2.38	49.48	1.86	9.78	All	2020-05-24 11:54:16.967151	PC
12	Junkrat	DEFENSE	2.15	52.12	2.65	11.61	All	2020-05-24 11:54:16.967151	PC
13	Wrecking Ball	TANK	2.15	50.94	1.86	3.02	All	2020-05-24 11:54:16.967151	PC
14	D.Va	TANK	2.05	48.38	2.11	6.53	All	2020-05-24 11:54:16.967151	PC
15	Echo	OFFENSE	2.01	50.35	1.94	7.51	All	2020-05-24 11:54:16.967151	PC
16	Winston	TANK	1.99	47.72	1.75	6.19	All	2020-05-24 11:54:16.967151	PC
17	Reaper	OFFENSE	1.94	50.56	1.86	13.61	All	2020-05-24 11:54:16.967151	PC
18	Brigitte	SUPPORT	1.93	55.70	1.79	7.04	All	2020-05-24 11:54:16.967151	PC
19	Baptiste	SUPPORT	1.83	46.50	2.59	11.44	All	2020-05-24 11:54:16.967151	PC
20	Doomfist	OFFENSE	1.63	52.35	1.88	7.41	All	2020-05-24 11:54:16.967151	PC
21	Orisa	TANK	1.62	49.77	1.95	6.77	All	2020-05-24 11:54:16.967151	PC
22	Hanzo	DEFENSE	1.60	47.44	2.09	10.41	All	2020-05-24 11:54:16.967151	PC
23	Tracer	OFFENSE	1.52	49.48	1.50	7.16	All	2020-05-24 11:54:16.967151	PC
24	Pharah	OFFENSE	1.51	54.15	1.85	14.75	All	2020-05-24 11:54:16.967151	PC
25	Zenyatta	SUPPORT	1.49	53.46	1.90	8.28	All	2020-05-24 11:54:16.967151	PC
26	Widowmaker	DEFENSE	1.28	50.44	1.49	6.62	All	2020-05-24 11:54:16.967151	PC
27	Torbjörn	DEFENSE	1.16	57.40	2.79	7.69	All	2020-05-24 11:54:16.967151	PC
28	Mei	DEFENSE	0.77	48.07	2.48	6.75	All	2020-05-24 11:54:16.967151	PC
29	Symmetra	SUPPORT	0.66	58.08	2.90	9.13	All	2020-05-24 11:54:16.967151	PC
30	Sombra	OFFENSE	0.47	47.34	1.84	12.21	All	2020-05-24 11:54:16.967151	PC
31	Bastion	DEFENSE	0.31	55.00	2.38	16.33	All	2020-05-24 11:54:16.967151	PC

	Hero	Role	Pick %	Win %	Tie %	On Fire %	Rank	Date	Platform
0	Reinhardt	TANK	12.08	52.10	2.61	7.15	All	2020-05-24 11:54:16.967151	PC
1	Zarya	TANK	7.51	52.90	2.47	12.39	All	2020-05-24 11:54:16.967151	PC
2	Roadhog	TANK	3.40	49.70	1.75	9.34	All	2020-05-24 11:54:16.967151	PC
3	Sigma	TANK	3.07	50.85	2.22	12.90	All	2020-05-24 11:54:16.967151	PC
4	Wrecking Ball	TANK	2.15	50.94	1.86	3.02	All	2020-05-24 11:54:16.967151	PC
5	D.Va	TANK	2.05	48.38	2.11	6.53	All	2020-05-24 11:54:16.967151	PC
6	Winston	TANK	1.99	47.72	1.75	6.19	All	2020-05-24 11:54:16.967151	PC
7	Orisa	TANK	1.62	49.77	1.95	6.77	All	2020-05-24 11:54:16.967151	PC
8	Reinhardt	TANK	9.03	48.86	2.71	7.48	Bronze	2020-05-24 11:55:05.315769	PC
9	Zarya	TANK	4.92	48.83	2.59	8.94	Bronze	2020-05-24 11:55:05.315769	PC
10	D.Va	TANK	4.04	44.32	2.59	6.29	Bronze	2020-05-24 11:55:05.315769	PC
11	Sigma	TANK	3.59	47.81	2.42	10.79	Bronze	2020-05-24 11:55:05.315769	PC
12	Orisa	TANK	3.45	48.09	2.51	5.80	Bronze	2020-05-24 11:55:05.315769	PC
13	Roadhog	TANK	3.03	45.85	1.57	7.37	Bronze	2020-05-24 11:55:05.315769	PC
14	Wrecking Ball	TANK	1.99	47.83	2.21	2.62	Bronze	2020-05-24 11:55:05.315769	PC
15	Winston	TANK	1.94	46.27	1.47	5.53	Bronze	2020-05-24 11:55:05.315769	PC
16	Reinhardt	TANK	10.68	51.15	2.45	6.97	Silver	2020-05-24 11:55:53.912738	PC
17	Zarya	TANK	6.32	50.62	2.27	9.63	Silver	2020-05-24 11:55:53.912738	PC
18	Sigma	TANK	3.29	49.51	2.30	11.47	Silver	2020-05-24 11:55:53.912738	PC
19	Roadhog	TANK	3.11	46.85	1.78	7.78	Silver	2020-05-24 11:55:53.912738	PC
20	D.Va	TANK	2.88	47.58	2.23	6.44	Silver	2020-05-24 11:55:53.912738	PC
21	Orisa	TANK	2.25	49.21	1.97	6.06	Silver	2020-05-24 11:55:53.912738	PC
22	Winston	TANK	2.04	46.44	1.93	5.65	Silver	2020-05-24 11:55:53.912738	PC
23	Wrecking Ball	TANK	1.87	49.78	2.07	2.73	Silver	2020-05-24 11:55:53.912738	PC
24	Reinhardt	TANK	11.71	51.44	2.65	7.08	Gold	2020-05-24 11:56:43.966592	PC
25	Zarya	TANK	7.36	51.99	2.41	11.44	Gold	2020-05-24 11:56:43.966592	PC
26	Roadhog	TANK	3.33	48.78	1.87	8.85	Gold	2020-05-24 11:56:43.966592	PC
27	Sigma	TANK	3.05	51.00	2.13	12.78	Gold	2020-05-24 11:56:43.966592	PC
28	D.Va	TANK	2.15	48.59	2.09	6.89	Gold	2020-05-24 11:56:43.966592	PC
29	Winston	TANK	2.00	47.03	1.72	6.16	Gold	2020-05-24 11:56:43.966592	PC
30	Wrecking Ball	TANK	1.92	49.80	1.99	2.80	Gold	2020-05-24 11:56:43.966592	PC
31	Orisa	TANK	1.50	50.30	2.02	7.04	Gold	2020-05-24 11:56:43.966592	PC
32	Reinhardt	TANK	12.96	52.26	2.67	7.31	Platinum	2020-05-24 11:57:34.194215	PC
33	Zarya	TANK	8.32	53.60	2.53	13.17	Platinum	2020-05-24 11:57:34.194215	PC
34	Roadhog	TANK	3.70	50.96	1.75	10.12	Platinum	2020-05-24 11:57:34.194215	PC
35	Sigma	TANK	2.69	51.17	2.27	13.64	Platinum	2020-05-24 11:57:34.194215	PC
36	Wrecking Ball	TANK	2.20	51.69	1.76	3.18	Platinum	2020-05-24 11:57:34.194215	PC
37	Winston	TANK	1.88	47.48	1.61	6.50	Platinum	2020-05-24 11:57:34.194215	PC
38	D.Va	TANK	1.57	49.53	1.85	6.69	Platinum	2020-05-24 11:57:34.194215	PC
39	Orisa	TANK	1.14	50.83	1.56	7.22	Platinum	2020-05-24 11:57:34.194215	PC
40	Reinhardt	TANK	13.57	53.54	2.66	7.13	Diamond	2020-05-24 11:58:23.985963	PC
41	Zarya	TANK	8.74	55.04	2.63	14.53	Diamond	2020-05-24 11:58:23.985963	PC
42	Roadhog	TANK	3.66	52.05	1.55	10.44	Diamond	2020-05-24 11:58:23.985963	PC
43	Sigma	TANK	2.97	53.00	2.16	14.27	Diamond	2020-05-24 11:58:23.985963	PC
44	Wrecking Ball	TANK	2.31	53.23	1.73	3.20	Diamond	2020-05-24 11:58:23.985963	PC
45	Winston	TANK	1.99	49.35	1.92	6.54	Diamond	2020-05-24 11:58:23.985963	PC
46	D.Va	TANK	1.35	48.66	2.12	6.26	Diamond	2020-05-24 11:58:23.985963	PC
47	Orisa	TANK	1.14	49.17	1.36	7.33	Diamond	2020-05-24 11:58:23.985963	PC
48	Reinhardt	TANK	13.46	54.65	2.34	6.76	Master	2020-05-24 11:59:13.886732	PC
49	Zarya	TANK	7.38	54.39	2.48	14.34	Master	2020-05-24 11:59:13.886732	PC
50	Sigma	TANK	3.38	49.71	1.76	12.66	Master	2020-05-24 11:59:13.886732	PC
51	Roadhog	TANK	3.05	51.43	1.87	9.84	Master	2020-05-24 11:59:13.886732	PC
52	Wrecking Ball	TANK	2.97	50.90	1.46	3.30	Master	2020-05-24 11:59:13.886732	PC
53	Winston	TANK	2.32	51.52	1.99	6.13	Master	2020-05-24 11:59:13.886732	PC
54	Orisa	TANK	1.86	48.13	2.55	7.01	Master	2020-05-24 11:59:13.886732	PC
55	D.Va	TANK	1.41	51.03	2.02	4.87	Master	2020-05-24 11:59:13.886732	PC
56	Reinhardt	TANK	10.08	55.29	2.46	7.06	Grandmaster	2020-05-24 12:00:03.487597	PC
57	Sigma	TANK	5.92	51.74	2.64	13.20	Grandmaster	2020-05-24 12:00:03.487597	PC
58	Zarya	TANK	5.43	57.78	2.38	15.29	Grandmaster	2020-05-24 12:00:03.487597	PC
59	Wrecking Ball	TANK	4.15	52.30	1.65	3.50	Grandmaster	2020-05-24 12:00:03.487597	PC
60	Orisa	TANK	3.85	50.63	2.40	6.79	Grandmaster	2020-05-24 12:00:03.487597	PC
61	Winston	TANK	2.58	53.08	1.85	5.78	Grandmaster	2020-05-24 12:00:03.487597	PC
62	D.Va	TANK	2.45	52.97	1.94	5.12	Grandmaster	2020-05-24 12:00:03.487597	PC
63	Roadhog	TANK	2.23	51.62	1.18	8.81	Grandmaster	2020-05-24 12:00:03.487597	PC

	Hero	Role	Pick %	Win %	KDA
1	Ana	SUPPORT	9.05	48.28	1.99
2	Ashe	OFFENSE	2.12	50.36	2.91
3	Baptiste	SUPPORT	2.41	47.28	2.15
4	Bastion	DEFENSE	1.04	48.32	2.81
5	Brigitte	SUPPORT	1.81	52.96	2.05
6	D.Va	TANK	3.16	45.1	2.74
7	Doomfist	OFFENSE	3.06	48.61	4.87
8	Echo	OFFENSE	1.43	49.64	2.13
9	Genji	OFFENSE	0.79	50.96	3.11
10	Hanzo	DEFENSE	3.3	48.64	2.74
11	Junkrat	DEFENSE	2.88	46.69	2.82
12	Lúcio	SUPPORT	2.67	49.54	2.59
13	McCree	OFFENSE	3.72	52.1	2.41
14	Mei	DEFENSE	1.1	48.63	2.96
15	Mercy	SUPPORT	6.72	48.21	0.54
16	Moira	SUPPORT	4.93	50.57	3.76
17	Orisa	TANK	2.28	46.7	3.06
18	Pharah	OFFENSE	1.78	51.49	3.09
19	Reaper	OFFENSE	1.73	49.64	2.99
20	Reinhardt	TANK	3.75	51.48	2.39
21	Roadhog	TANK	3.0	46.93	3.57
22	Sigma	TANK	1.91	51.97	3.97
23	Soldier: 76	OFFENSE	3.19	47.7	3.16
24	Sombra	OFFENSE	1.27	44.44	3.20
25	Symmetra	SUPPORT	1.06	54.77	3.21
26	Torbjörn	DEFENSE	1.13	53.39	3.59
27	Tracer	OFFENSE	1.55	45.53	2.72
28	Widowmaker	DEFENSE	2.21	44.39	2.39
29	Winston	TANK	1.86	50.33	3.26
30	Wrecking Ball	TANK	1.28	50.3	3.34
31	Zarya	TANK	3.01	48.71	4.11
32	Zenyatta	SUPPORT	3.89	51.77	2.66

	Popularity Ranking	Bronze	Silver	Gold	Platinum	Diamond	Master	Grandmaster
0	1	Mercy	Ana	Ana	Ana	Ana	Ana	Sigma
1	2	Ana	Reinhardt	Reinhardt	Reinhardt	Reinhardt	Mercy	Roadhog
2	3	Reinhardt	Mercy	Mercy	Mercy	Roadhog	Roadhog	Wrecking Ball
3	4	Moira	Moira	Soldier: 76	Roadhog	Mercy	Sigma	Zenyatta
4	5	Soldier: 76	Soldier: 76	Zarya	Zarya	Cassidy	Reinhardt	Ana
5	6	D.Va	Zarya	Cassidy	Cassidy	Zarya	Baptiste	Baptiste
6	7	Junkrat	Cassidy	Roadhog	Soldier: 76	Sigma	Zenyatta	D.Va
7	8	Cassidy	Roadhog	Moira	Hanzo	Soldier: 76	Cassidy	Tracer
8	9	Lúcio	Baptiste	Baptiste	Sigma	D.Va	D.Va	Brigitte
9	10	Baptiste	D.Va	Hanzo	Baptiste	Zenyatta	Wrecking Ball	Mercy
10	11	Zarya	Junkrat	Sigma	D.Va	Baptiste	Tracer	Lúcio
11	12	Hanzo	Hanzo	D.Va	Zenyatta	Hanzo	Lúcio	Reinhardt
12	13	Sigma	Lúcio	Lúcio	Genji	Winston	Brigitte	Orisa
13	14	Roadhog	Sigma	Zenyatta	Lúcio	Lúcio	Hanzo	Cassidy
14	15	Reaper	Zenyatta	Genji	Winston	Genji	Soldier: 76	Zarya

Predicting What Heroes Got Nerfed or Buffed in Overwatch 2¶

Rushil Shah, Ethan Lin, Fabrizzio Perez¶

Table of Contents:¶

Introduction¶

Data Collection¶

Data Exploration and Analysis¶

Graphing Picks and Win Ratios by Rank¶

Average Pick Rate of Tank Based on Rank¶

Average Pick Rate of Support Based on Rank¶

Average Pick Rate of Offense Based on Rank¶

Average Pick Rate of Defense Based on Rank¶

Winning Rate of Tanks Based on Rank¶

Winning Rate of Support Based on Rank¶

Winning Rate of Offense Based on Rank¶

Winning Rate of Defense Based on Rank¶

Determining Which Characters are Overused or Underused¶

Comparing Tank Pick Rate to Win Rate in Grandmaster¶

Comparing Support Pick Rate to Win Rate in Grandmaster¶

Comparing Offense Pick Rate to Win Rate in Grandmaster¶

Comparing Defense Pick Rate to Win Rate in Grandmaster¶

Win Rate Anaylysis Throughout Periods¶

Predicting who got buffed and who got nerfed in Overwatch 2¶

Conclusion¶

	Hero	Role	Pick %	Win %	KDA
1	Ana	SUPPORT	8.23	43.1	1.82
2	Ashe	OFFENSE	1.71	46.72	3.14
3	Baptiste	SUPPORT	1.47	44.43	2.00
4	Bastion	DEFENSE	0.82	32.48	3.19
5	Brigitte	SUPPORT	1.57	51.73	2.43
6	D.Va	TANK	4.31	37.17	2.62
7	Doomfist	OFFENSE	4.06	41.58	4.90
8	Echo	OFFENSE	1.69	46.34	2.13
9	Genji	OFFENSE	0.48	48.57	3.31
10	Hanzo	DEFENSE	4.35	38.3	2.83
11	Junkrat	DEFENSE	3.64	36.55	2.81
12	Lúcio	SUPPORT	2.77	36.6	2.68
13	McCree	OFFENSE	3.84	42.79	2.38
14	Mei	DEFENSE	1.51	35.79	2.99
15	Mercy	SUPPORT	6.65	41.83	0.47
16	Moira	SUPPORT	3.93	46.72	3.86
17	Orisa	TANK	2.0	44.24	3.23
18	Pharah	OFFENSE	2.34	37.39	3.03
19	Reaper	OFFENSE	2.05	37.76	3.11
20	Reinhardt	TANK	6.26	45.34	2.26
21	Roadhog	TANK	3.87	39.96	3.48
22	Sigma	TANK	1.4	48.14	3.91
23	Soldier: 76	OFFENSE	3.15	37.72	3.22
24	Sombra	OFFENSE	1.45	35.47	3.12
25	Symmetra	SUPPORT	1.3	44.43	3.29
26	Torbjörn	DEFENSE	1.19	41.37	4.18
27	Tracer	OFFENSE	2.62	36.47	2.78
28	Widowmaker	DEFENSE	3.54	35.7	2.49
29	Winston	TANK	2.61	43.98	3.17
30	Wrecking Ball	TANK	1.58	47.57	3.07
31	Zarya	TANK	4.54	45.0	4.03
32	Zenyatta	SUPPORT	3.68	45.02	2.63