Histogram Maker
Paste your numbers and get an instant histogram, a frequency table, and summary stats. Switch the binning method live, then download a PNG or SVG. Free, private, works on your phone.
Chart labels and custom range
Summary statistics
- Count
- 40
- Mean
- 80.83
- Median
- 80.5
- Mode
- none (too many ties)
- Std dev (sample)
- 8.2
- Range
- 31
- Min
- 65
- Max
- 96
- Q1 / Q3
- 74.75 / 87.25
Frequency table
| Bin | Count | Relative | Cumulative |
|---|---|---|---|
| 65 to 69.43 | 4 | 10.0% | 10.0% |
| 69.43 to 73.86 | 4 | 10.0% | 20.0% |
| 73.86 to 78.29 | 8 | 20.0% | 40.0% |
| 78.29 to 82.71 | 7 | 17.5% | 57.5% |
| 82.71 to 87.14 | 7 | 17.5% | 75.0% |
| 87.14 to 91.57 | 6 | 15.0% | 90.0% |
| 91.57 to 96 | 4 | 10.0% | 100.0% |
Everything runs in your browser. Your data is never uploaded, and the tool remembers your last dataset on this device only.
What this histogram maker does
Paste a column or a comma-separated list of numbers and this tool draws the histogram right away. It groups your values into bins, counts how many fall in each, and shows the bars, a frequency table, and the summary statistics all at once. Nothing is uploaded, and it works on your phone.
The one thing most histogram tools get wrong
A histogram is only as honest as its bins. Too few bins and you flatten real detail; too many and you turn noise into fake peaks. Most makers pick one bin count and hide the choice from you. This one puts it front and center: switch between Sturges, Freedman-Diaconis, Scott, Rice, square-root, or a manual bin count and watch the shape update instantly, so you can see how the binning changes the story before you trust it.
How to read a histogram
Each bar covers a range of values, called a bin, and its height is how many data points land in that range. Because the bars represent ranges, they sit right next to each other with no gaps. That is the main difference from a bar chart, where each bar is a separate category.
- A single central hump that tapers both ways is bell-shaped, close to a normal distribution.
- A long tail to the right is right (positive) skew; a long tail to the left is left (negative) skew.
- Two humps usually means two groups are mixed together.
- Flat bars mean a roughly uniform spread.
Choosing the number of bins
The default here is Sturges' rule, which sets the bin count to ceil(log2 n) + 1. It works well for small, roughly normal datasets. For skewed data or data with outliers, Freedman-Diaconis is more robust because it uses the interquartile range instead of the full spread. When in doubt, try two or three methods and keep the one that shows the structure most clearly without inventing spikes.
Worked example
Say you have the ages of 20 people. With Sturges' rule that is 6 classes. The mean age is 42.8 and the median is 43, which are close, so the data is fairly symmetric. Paste your own list above, pick a method, and download the chart as a PNG or SVG when it looks right.
Frequently asked questions
- How do I make a histogram from a list of numbers?
- Paste or type your numbers into the box above. The tool splits them on commas, spaces, or new lines, groups them into bins, and draws the histogram instantly. You can then switch the binning method or set the bin count by hand.
- Is this histogram maker free?
- Yes, completely free with no sign up and no watermark. It runs entirely in your browser, so your data never leaves your device.
- Can I download the histogram?
- Yes. Use Download PNG for a sharp image to drop into a report or slide, or Download SVG for a scalable vector you can edit. You can also copy the frequency table or print the page.
- How many bins should a histogram have?
- A common starting point is Sturges rule, ceil(log2 n) + 1. For skewed data or data with outliers, Freedman-Diaconis usually works better. Try a couple of methods and keep the one that shows the real structure without adding fake spikes.
- Does it work with negative numbers and decimals?
- Yes. Negative values and decimals are valid data and are binned normally. Only text and blank entries are skipped, so pasted headers or labels will not break the chart.