Project: Compare Baseball Player Statistics using Visualizations

# Run this cell to begin
# Import the necessary packages
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
%matplotlib inline

# Load Aaron Judge's Statcast data
judge = pd.read_csv('judge.csv')

# Load Giancarlo Stanton's Statcast data
stanton = pd.read_csv('stanton.csv')

# Display all columns (pandas will collapse some columns if we don't set this option)
pd.set_option('display.max_columns', None)

# Custom Functions
def assign_x_coord(row):
    """
    Assigns an x-coordinate to Statcast's strike zone numbers. Zones 11, 12, 13,
    and 14 are ignored for plotting simplicity.
    """
    # Left third of strike zone
    if row.zone in [1, 4, 7]:
        return 1
    # Middle third of strike zone
    if row.zone in [2, 5, 8]:
        return 2
    # Right third of strike zone
    if row.zone in [3, 6, 9]:
        return 3
    
def assign_y_coord(row):
    """
    Assigns a y-coordinate to Statcast's strike zone numbers. Zones 11, 12, 13,
    and 14 are ignored for plotting simplicity.
    """
    # Upper third of strike zone
    if row.zone in [1, 2, 3]:
        return 3
    # Middle third of strike zone
    if row.zone in [4, 5, 6]:
        return 2
    # Lower third of strike zone
    if row.zone in [7, 8, 9]:
        return 1
    
# Display the last five rows of the Aaron Judge file
judge.tail()

# Filter the data for 2017
judge_2017 = judge[judge['game_year'] == 2017]
stanton_2017 = stanton[stanton['game_year'] == 2017]

# Count the events and sort in descending order
judge_events_2017 = judge_2017['events'].value_counts()
stanton_events_2017 = stanton_2017['events'].value_counts()

# Create a bar chart for Judge's events in 2017 using Seaborn with flipped X and Y
plt.figure(figsize=(10, 6))
sns.barplot(x=judge_events_2017.values, y=judge_events_2017.index, palette='Blues_d')
plt.title("Aaron Judge's Events in 2017")
plt.xlabel('Count')
plt.ylabel('')
plt.xticks(rotation=45)
plt.show()

# Create a bar chart for Stanton's events in 2017 using Seaborn with flipped X and Y
plt.figure(figsize=(10, 6))
sns.barplot(x=stanton_events_2017.values, y=stanton_events_2017.index, palette='Reds_d')
plt.title("Giancarlo Stanton's Events in 2017")
plt.xlabel('Count')
plt.ylabel('')
plt.xticks(rotation=45)
plt.show()

import seaborn as sns

# Filter data for home runs
judge_hr = judge[judge['events'] == 'home_run']
stanton_hr = stanton[stanton['events'] == 'home_run']

# Create the plot
# fig1, ax1 = plt.subplots(1, 2, figsize=(16, 6))
fig1, ax1 = plt.subplots(ncols=2, sharex=True, sharey=True)

# KDE Plot for Judge
sns.kdeplot(x=judge_hr['launch_angle'], y=judge_hr['launch_speed'], ax=ax1[0], fill=True, cmap="Blues", shade=True, shade_lowest=False)
ax1[0].set_title("Aaron Judge\n Home Runs 2015 - 2017")
ax1[0].set_xlabel("Launch Angle (degrees)")
ax1[0].set_ylabel("Launch Speed (mph)")

# KDE Plot for Stanton
sns.kdeplot(x=stanton_hr['launch_angle'], y=stanton_hr['launch_speed'], ax=ax1[1], fill=True, cmap="Reds", shade=True, shade_lowest=False)
ax1[1].set_title("Giancarlo Stanton\n Home Runs 2015 - 2017")
ax1[1].set_xlabel("Launch Angle (degrees)")
ax1[1].set_ylabel("Launch Speed (mph)")

plt.tight_layout()
plt.show()

# Calculate average launch angle and speed for each player
judge_avg_angle = judge_hr['launch_angle'].mean()
judge_avg_speed = judge_hr['launch_speed'].mean()
stanton_avg_angle = stanton_hr['launch_angle'].mean()
stanton_avg_speed = stanton_hr['launch_speed'].mean()

# Determine which player hit home runs slightly lower and harder
if stanton_avg_angle < judge_avg_angle and stanton_avg_speed > judge_avg_speed:
    player_hr = "Stanton"
else:
    player_hr = "Judge"

print(f"Average Launch Angle - Judge: {judge_avg_angle:.2f}, Stanton: {stanton_avg_angle:.2f}")
print(f"Average Launch Speed - Judge: {judge_avg_speed:.2f}, Stanton: {stanton_avg_speed:.2f}")
print(f"{player_hr} hit home runs slightly lower and harder.")

# Combine the data for both players into a single DataFrame
judge_stanton_hr = pd.concat([judge_hr, stanton_hr])

# Create box plots for pitch velocity comparison
fig2, ax2 = plt.subplots(figsize=(10, 6))

# Create a color palette to differentiate players
palette = {'Aaron Judge': 'blue', 'Giancarlo Stanton': 'red'}

sns.boxplot(x='player_name', y='release_speed', data=judge_stanton_hr, ax=ax2, palette=palette,
            medianprops=dict(color="white"))
ax2.set_title('Pitch Velocity for Home Runs (2015 - 2017)')
ax2.set_xlabel('Player')
ax2.set_ylabel('Release Speed (mph)')

plt.show()

# Calculate median release speed for each player
judge_median = judge_hr['release_speed'].median()
stanton_median = stanton_hr['release_speed'].median()

# Determine which player hit home runs off faster pitches
if stanton_median > judge_median:
    player_fast = "Stanton"
else:
    player_fast = "Judge"

print(f"Median Release Speed - Judge: {judge_median:.2f}, Stanton: {stanton_median:.2f}")
print(f"{player_fast} hit home runs off faster pitches on average.")