Start with a rule, then inspect the chart
Freedman-Diaconis is a strong default for skewed data and outliers. Scott is useful for near-normal data. Sturges is simple and often works for smaller datasets.
Histogram guide
How to Choose Histogram Bins
Bin size changes how a histogram reads. Too few bins hide structure, while too many bins can make noise look like a pattern.
The main histogram maker lets you compare common binning rules and switch to custom bins when a report or assignment requires a specific interval.
Use custom bins when needed
Custom bin count and custom bin width are best when you need consistent intervals across multiple charts or must match a published method.
FAQ
Should every histogram use the same bin rule?
No. The best binning rule depends on sample size, the presence of outliers, the skewness of the data, and the intended audience. A rule that works well for a near-normal distribution of 200 values may produce too few or too many bins for a heavily skewed dataset of 30 values.
Why does bin size change the shape?
Bin width controls how values are grouped together. Wider bins smooth over detail and can hide multiple peaks or gaps, while narrower bins can make natural variation look like meaningful structure. The same dataset can appear unimodal, bimodal, or irregular depending on the bin width chosen.
Can I set exact bin width?
Yes. Select the custom width option in the bin method selector and enter the exact interval width. This is useful when you need consistent bin boundaries across multiple charts or must match a bin width specified in a published method.
Make a histogram from your data
Paste numbers, upload a CSV or TXT file, choose bins, customize labels and styles, then export SVG, PDF, PNG, CSV, or JSON from the main tool.
Go to the histogram maker