In real-world projects, I was analyzing survey results from employees. company or reviewing sales performance across different states, it wasn’t enough to just show the average values. I needed to communicate how reliable those values were. That’s where error bars in Matplotlib bar charts come in.
Adding error bars to a bar chart in Python makes the visualization more accurate and trustworthy. It tells the audience not only what the value is but also how much it might vary.
In this tutorial, I’ll show you step by step how to plot bar charts with error bars in Matplotlib. I’ll share multiple methods, full Python code, and some best practices that I’ve learned from years of hands-on experience.
Methods to Use Error Bars in Python Matplotlib Bar Charts
Error bars are a simple yet powerful way to represent variability in your dataset.
For example, if you’re comparing average household income across different U.S. states, you might calculate a mean value for each state. But income data often varies widely. Error bars help you show that variation, so your audience understands the context behind the numbers.
In Python, Matplotlib makes it easy to add error bars to bar charts with just a few lines of code.
Method 1 – Basic Bar Chart with Error Bars in Python
The first method I usually teach beginners is the easy way of using the plt.bar() function with the yerr parameter.
This is the simplest approach, and it works well when you already have mean values and standard deviations (or some measure of error) calculated.
Here’s the full Python code:
import matplotlib.pyplot as plt
import numpy as np
# Example data: Average monthly expenses for U.S. households (in dollars)
categories = ['Housing', 'Food', 'Transportation', 'Healthcare', 'Entertainment']
values = [1800, 600, 400, 300, 200]
# Standard deviations (error values)
errors = [150, 80, 60, 50, 40]
# Create bar chart with error bars
plt.figure(figsize=(8, 6))
plt.bar(categories, values, yerr=errors, capsize=8, color='skyblue', edgecolor='black')
# Add labels and title
plt.title('Average Monthly Expenses for U.S. Households with Error Bars', fontsize=14)
plt.xlabel('Category')
plt.ylabel('Cost in USD')
# Show plot
plt.show()I executed the above example code and added the screenshot below.
I like this method because it’s quick and clean. You just pass the error values using yerr, and Matplotlib automatically adds vertical error bars.
Method 2 – Horizontal Bar Chart with Error Bars in Python
Sometimes, horizontal bar charts are easier to read, especially when you have long category names.
In Python’s Matplotlib, you can use the barh() function and add error bars with the xerr parameter.
Here’s how I do it:
import matplotlib.pyplot as plt
import numpy as np
# Example data: Average weekly working hours across U.S. industries
industries = ['Tech', 'Healthcare', 'Education', 'Retail', 'Construction']
hours = [42, 38, 36, 34, 40]
# Standard deviations (error values)
errors = [3, 2, 1.5, 2.5, 3]
# Create horizontal bar chart with error bars
plt.figure(figsize=(8, 6))
plt.barh(industries, hours, xerr=errors, capsize=6, color='lightgreen', edgecolor='black')
# Add labels and title
plt.title('Average Weekly Working Hours by Industry in the U.S. with Error Bars', fontsize=14)
plt.xlabel('Hours Worked')
plt.ylabel('Industry')
# Show plot
plt.show()I executed the above example code and added the screenshot below.
I’ve used this approach many times when presenting results to business teams in the U.S., especially when category labels are long and vertical bars look cluttered.
Method 3 – Use NumPy for Random Error Bars in Python
Sometimes, you don’t have predefined error values but want to simulate or demonstrate how error bars work.
In Python, I often use NumPy to generate random errors and then add them to the bar chart. This is useful for teaching or for quick exploratory analysis.
Here’s an example:
import matplotlib.pyplot as plt
import numpy as np
# Categories: Sales regions in the U.S.
regions = ['East', 'West', 'North', 'South']
sales = [250, 300, 200, 280]
# Generate random errors using NumPy
np.random.seed(42) # for reproducibility
errors = np.random.randint(10, 30, size=len(sales))
# Create bar chart with random error bars
plt.figure(figsize=(8, 6))
plt.bar(regions, sales, yerr=errors, capsize=5, color='orange', edgecolor='black')
# Add labels and title
plt.title('Quarterly Sales by Region in the U.S. with Random Error Bars', fontsize=14)
plt.xlabel('Region')
plt.ylabel('Sales (in thousands USD)')
# Show plot
plt.show()I executed the above example code and added the screenshot below.
This method is handy when you just want to visualize variability without worrying about exact error calculations.
Method 4 – Customize Error Bars in Python Matplotlib
One of the things I love about Matplotlib is how customizable it is. You can change the color, line style, width, and more for error bars.
Here’s an example where I make the error bars red and dashed:
import matplotlib.pyplot as plt
import numpy as np
# Example data: Average internet speed in U.S. cities (Mbps)
cities = ['New York', 'Los Angeles', 'Chicago', 'Houston', 'Phoenix']
speed = [120, 110, 95, 100, 105]
# Error values
errors = [10, 8, 12, 7, 9]
# Create customized bar chart with error bars
plt.figure(figsize=(8, 6))
plt.bar(cities, speed, yerr=errors, capsize=6, color='lightcoral', edgecolor='black',
error_kw=dict(ecolor='red', elinewidth=2, linestyle='--'))
# Add labels and title
plt.title('Average Internet Speed in U.S. Cities with Customized Error Bars', fontsize=14)
plt.xlabel('City')
plt.ylabel('Speed (Mbps)')
# Show plot
plt.show()This method gives you full control over how the error bars look, which is especially useful when preparing charts for presentations or reports.
Best Practices for Using Error Bars in Python
Over the years, I’ve learned a few best practices when working with error bars in Python Matplotlib:
Always explain what the error bars represent (standard deviation, standard error, confidence interval).
- Keep error bars simple and easy to read. Don’t overload the chart with too much decoration.
- Use consistent colors and styles to avoid confusing your audience.
- If presenting to a U.S. audience, use familiar units like dollars, hours, or Mbps.
In this tutorial, I showed you how to create a Matplotlib bar chart with error bars in Python using different methods.
We started with the basic yerr parameter, moved on to horizontal bar charts, explored random error generation with NumPy, and finally customized error bars for better presentation.
Error bars are a small addition, but they make a big difference in how your data is perceived. By using them, you provide clarity, transparency, and professionalism in your Python visualizations.
You may also read:
- Plot a Bar Chart from a Dictionary in Python Matplotlib
- Bar Chart from a DataFrame in Python Matplotlib
- Plot a Bar Chart with Dates in Matplotlib
- Create a Bar Chart with Values in Matplotlib
I am Bijay Kumar, a Microsoft MVP in SharePoint. Apart from SharePoint, I started working on Python, Machine learning, and artificial intelligence for the last 5 years. During this time I got expertise in various Python libraries also like Tkinter, Pandas, NumPy, Turtle, Django, Matplotlib, Tensorflow, Scipy, Scikit-Learn, etc… for various clients in the United States, Canada, the United Kingdom, Australia, New Zealand, etc. Check out my profile.
