Distribution Plots: Histograms & KDE

Distribution plots examine how continuous numbers are spread across a range—whether they cluster centrally, skew to one side, or show multimodal peaks.

1. Histograms (histplot)

A histogram divides continuous numbers into sequential buckets called bins and counts how many observations land in each bucket.

Basic Histogram

import seaborn as sns
import matplotlib.pyplot as plt
 
tips = sns.load_dataset('tips')
sns.histplot(data=tips, x='total_bill')
plt.show()

Controlling Bin Resolution

Choosing the number of bins is critical:

• Too few bins: Oversimplifies the shape into coarse blocks.
• Too many bins: Can introduce excessive noise and jagged spikes.

import seaborn as sns
import matplotlib.pyplot as plt
 
tips = sns.load_dataset('tips')
sns.histplot(data=tips, x='total_bill', bins=30)
plt.show()

Figure 3.1: More bins provide finer detail across total bill amounts.

Comparing Distributions Across Groups

You can overlay distributions across categories using hue:

import seaborn as sns
import matplotlib.pyplot as plt
 
tips = sns.load_dataset('tips')
sns.histplot(data=tips, x='total_bill', hue='time')
plt.show()

2. Kernel Density Estimation (kdeplot)

What KDE Actually Means

Kernel Density Estimation (KDE) replaces rigid, discrete bins with a smooth probability density curve. It places a smooth kernel (usually a Gaussian bell curve) over every observation and sums them up into a single flowing line.

import seaborn as sns
import matplotlib.pyplot as plt
 
tips = sns.load_dataset('tips')
sns.kdeplot(data=tips, x='total_bill')
plt.show()

Layering KDE on Top of a Histogram

The most popular practice is combining both: the histogram shows raw observation counts while the KDE curve reveals the underlying distribution shape. Set kde=True in histplot:

import seaborn as sns
import matplotlib.pyplot as plt
 
tips = sns.load_dataset('tips')
sns.histplot(data=tips, x='total_bill', kde=True)
plt.show()

Figure 3.2: The smooth KDE curve overlaid directly on the histogram bins.

Comparing Two KDE Curves

import seaborn as sns
import matplotlib.pyplot as plt
 
tips = sns.load_dataset('tips')
sns.kdeplot(data=tips, x='total_bill', hue='time')
plt.show()

Interactive Data Distribution Check

Inspect distribution summary statistics programmatically:

Try It Yourself

Problem 1: Custom Bin Histogram

Plot a histogram of tip from the tips dataset with 20 bins.

Problem 2: Comparative KDE Curves

Plot a KDE curve of tip, split by sex.