Lesson 18

Histogram Styles

Switch between the bar, step, and stepfilled histogram styles, thin the bars with rwidth, and set the fill and edge colours.

Intermediate25 min

Histogram Styles hero Module 25 Lesson 18

The default histogram is a row of filled bars, and that is the right choice most of the time. Two other styles earn their place when the chart gets crowded: step draws the outline only, and stepfilled fills that outline without drawing a border between neighbouring bins.

The four histogram styles with a description of each

The three you will use

histtype chooses the style. Side by side the difference is obvious: bar separates each bin, step reduces the chart to a line you can see through, and stepfilled sits in between.

python

import matplotlib.pyplot as plt

minutes = [
    22, 24, 25, 27, 28, 28, 29, 30, 31, 31, 32, 33, 33, 34, 34, 35,
    35, 36, 36, 37, 38, 38, 39, 40, 41, 42, 44, 46, 48, 52,
]

plt.figure(figsize=(11, 4))
for index, kind in enumerate(["bar", "step", "stepfilled"], start=1):
    plt.subplot(1, 3, index)
    plt.hist(minutes, bins=6, histtype=kind, color="#2DD4BF", edgecolor="#0F172A")
    plt.title(kind)

plt.tight_layout()
plt.show()

print("same counts, three outlines")

Output:

text

same counts, three outlines

Thinner bars with rwidth

rwidth is the fraction of each bin the bar actually fills, so rwidth=0.8 leaves a fifth of every bin as white space. It only affects the bar style, and it is worth knowing that the gaps it creates make a histogram look like a bar chart, which is a little misleading.

python

import matplotlib.pyplot as plt

minutes = [
    22, 24, 25, 27, 28, 28, 29, 30, 31, 31, 32, 33, 33, 34, 34, 35,
    35, 36, 36, 37, 38, 38, 39, 40, 41, 42, 44, 46, 48, 52,
]

plt.hist(minutes, bins=6, histtype="bar", rwidth=0.85, color="#7C3AED")
plt.title("rwidth leaves a gap in each bin")
plt.xlabel("minutes")
plt.ylabel("deliveries")
plt.show()

print("rwidth is a fraction of the bin, not a number of units")

Output:

text

rwidth is a fraction of the bin, not a number of units

Fill and edge

color fills the bars and edgecolor outlines them. An outline is close to essential on the bar style: without it, neighbouring bars of the same colour merge into one block and you cannot see where a bin ends. For step there is no fill, so the colour is the line.

python

import matplotlib.pyplot as plt

minutes = [
    22, 24, 25, 27, 28, 28, 29, 30, 31, 31, 32, 33, 33, 34, 34, 35,
    35, 36, 36, 37, 38, 38, 39, 40, 41, 42, 44, 46, 48, 52,
]

plt.subplot(1, 2, 1)
plt.hist(minutes, bins=8, color="#2DD4BF")
plt.title("no edge")

plt.subplot(1, 2, 2)
plt.hist(minutes, bins=8, color="#2DD4BF", edgecolor="#0F172A", linewidth=1.2)
plt.title("with an edge")

plt.tight_layout()
plt.show()

print("the right chart shows where each bin ends")

Output:

text

the right chart shows where each bin ends

Step for comparing distributions

step comes into its own with two or three distributions on one chart. Filled bars hide each other however much transparency you add; outlines simply cross, and every shape stays readable.

python

import matplotlib.pyplot as plt

edges = [20, 26, 32, 38, 44, 50, 56]
morning = [22, 25, 27, 28, 30, 31, 33, 34, 35, 36, 38, 41, 44, 48]
evening = [26, 29, 31, 33, 34, 36, 37, 39, 40, 42, 43, 45, 49, 53]

plt.hist(morning, bins=edges, histtype="step", linewidth=2.5,
         color="#7C3AED", label="morning")
plt.hist(evening, bins=edges, histtype="step", linewidth=2.5,
         color="#F59E0B", label="evening")
plt.title("Two outlines, both readable")
plt.xlabel("minutes")
plt.ylabel("deliveries")
plt.legend()
plt.show()

print("no transparency needed")

Output:

text

no transparency needed

A histogram over six shared bins used for a comparison

Key takeaways

histtype picks bar, step, stepfilled, or barstacked
rwidth thins the bars as a fraction of the bin, on the bar style only
Always give bar histograms an edgecolor so the bins stay visible
step is the best style for laying two distributions over each other
Next: Pie Charts
Previous: Choosing Histogram Bins

Loading playground…

Check your understanding

Histogram Styles Quiz

Which argument chooses the histogram style?

1 / 10

← Choosing Histogram Bins

Pie Charts →