Is E4 Truly More Exciting?

# import modules 
import pandas as pd
import numpy as np
import re
import matplotlib.pyplot as plt
import seaborn as sns

# establish dataframe

df = pd.read_csv('club_games_data.csv')

display(df.head())

# Code in this cell was borrowed from Kaggle user ADITYAJHA1504

feature_names = ['Event', 'Site', 'Start_Date', 'End_Date', 'Start_Time',
                 'End_Time', 'Eco', 'EcoName', 'Round', 'Result']
feature_positions = [0, 1, 2, -6, -7, -5, -15, -14, 3, 6]

#Takes in the name you want to give the feature, and the position of the feature in 
#the pgn.split('\n') and creates the feature with feature name in the dataframe

for feature_name, position in zip(feature_names, feature_positions):
    df[feature_name] = df['pgn'].apply(
        lambda x: x.split('\n')[position].split('"')[1])
    
    
#Extract the moves
def extract_move(pgn):
    if(pgn.find('{[') == -1):
        original_list = pgn.split("\n")[-2].split()
        toberemoved_list = pgn.split("\n")[-2].split()[::3]
        new_list = [x for x in original_list if x not in toberemoved_list]
        return new_list
    else:
        return pgn.split("\n")[-2].split()[1::4]
df['Moves'] = df['pgn'].apply(extract_move)

# Isolate first move and remove square brackets and apostrophes
df['Moves'] = df['Moves'].astype(str).str.replace('[','').str.replace(']','').str.replace(",",'').str.replace("'",'').str.strip()

df['first_move'] = df['Moves'].str.split().str[0]

# Create Boolean columns for e4 and d4 games
df['e4'] = (df['first_move'] == 'e4').astype(int)
df['d4'] = (df['first_move'] == 'd4').astype(int)

display(df)

# Define a function to categorize the ratings
def categorize_rating(rating):
    if rating < 1200:
        return "Beginner"
    elif rating < 1600:
        return "Intermediate"
    elif rating < 2000:
        return "Advanced"
    else:
        return "Expert"

# Create new columns for the categorized ratings
df['white_level'] = df['white_rating'].apply(categorize_rating)
df['black_level'] = df['black_rating'].apply(categorize_rating)

# Create a function to identify checkmates
def is_checkmate(result):
    if result == 'checkmated':
        return 1
    else:
        return 0

# Create a new column to indicate if the game ended in a checkmate
df['checkmated'] = df['white_result'].apply(is_checkmate) + df['black_result'].apply(is_checkmate)

# Create a function to identify resignations
def is_resignation(result):
    if result == 'resigned':
        return 1
    else:
        return 0

# Create a new column to indicate if the game ended in a resignation
df['resigned'] = df['white_result'].apply(is_resignation) + df['black_result'].apply(is_resignation)

# Create a function to identify timeouts
def is_timeout(result):
    if result == 'timeout':
        return 1
    else:
        return 0

# Create a new column to indicate if the game ended in a timeout
df['timeout'] = df['white_result'].apply(is_timeout) + df['black_result'].apply(is_timeout)

# Create a function to identify the result of the game
def game_result(result):
    if result == '1-0':
        return 'white_win'
    elif result == '0-1':
        return 'black_win'
    elif result == '1/2-1/2':
        return 'draw'

# Create new columns for the game result
df['game_result'] = df['Result'].apply(game_result)
df['white_win'] = (df['game_result'] == 'white_win').astype(int)
df['black_win'] = (df['game_result'] == 'black_win').astype(int)
df['draw'] = (df['game_result'] == 'draw').astype(int)

# Create a function to identify whether the game was a win or a tie
def overall_result(result):
    if result == '1-0':
        return 'win'
    elif result == '0-1':
        return 'win'
    elif result == '1/2-1/2':
        return 'draw'
    
df['overall_result'] = df['Result'].apply(overall_result)

# Create a new column to calculate the length of the Moves column
df['moves_length'] = df['Moves'].str.split().apply(len)