Which Histogram Type and Bin Settings Should You Choose on Histogram Maker?

Updated

On histogrammaker.net, the histogram type controls what the bar height means, and the bin size controls how your data is grouped. Choose Count to see how many values fall in each range, Probability Density to compare distributions of different sizes, and Cumulative to see what share of values fall at or below a point. Leave Bin Size empty to let the tool pick an automatic interval, or set a number when you need a specific grouping. Bar Gap should normally stay at 0, since histograms represent continuous data with no gaps between bars.

What the three histogram types mean

The tool offers three histogram types, and each answers a different question about the same dataset.

Type Bar height represents Best for
Count Number of data points in each bin Seeing the raw shape of one dataset
Probability Density Frequency scaled so the total area equals 1 Comparing datasets of different sizes
Cumulative Running total of values up to each bin Finding medians, percentiles, or thresholds

Count is the default mental model most people have: a bar of height 12 means 12 values landed in that range. Use it when you only care about your own data and want the simplest read.

Probability Density rescales the bars so the whole histogram integrates to 1. This matters when you put two histograms side by side — a 500-row dataset and a 5,000-row dataset would look wildly different in Count mode, but in Probability Density mode their shapes become comparable.

Cumulative adds each bin to the ones before it, so the last bar always reaches the total. This is the type to pick when the question is "what fraction scored below 70?" rather than "how many scored around 70?"

Automatic vs. manual bin size

Bin Size is the width of each range your data is grouped into. The page lets you keep it empty for automatic sizing or enter your own value.

  • Automatic (leave empty): the tool calculates an interval for you. This is the right starting point for almost every first look at a dataset, because it removes the guesswork and gives a reasonable overview.
  • Manual: enter a number to force a specific width. Use this when you need bins that match a meaningful unit — for example, test scores grouped in 10-point bands, or ages grouped in 5-year bands.

The trade-off is the classic one: too few bins hides detail and flattens the shape; too many bins turns the histogram into a noisy picket fence where every bar holds one or two points. If the automatic result looks too coarse or too spiky, override it manually and regenerate.

What Bar Gap does

Bar Gap controls the spacing between bars, with options from 0 up to 0.9. The page describes histograms as continuous, with no gaps between bars, because each bar represents an interval of data rather than a separate category.

That means 0 is the correct setting for a true histogram. Raising the gap makes it look like a bar chart, which visually implies the categories are discrete and unrelated. Only increase it if you deliberately want that separated look for a presentation, and be aware you are changing what the chart communicates.

Choosing based on your goal

  • "What does my data look like?" → Count, automatic bin size, Bar Gap 0.
  • "How do these two groups compare?" → Probability Density for both, same bin size on each.
  • "What percentage is above/below a cutoff?" → Cumulative.
  • "I need bins that match real units (10-point bands, 5-year groups)" → Count or Probability Density with a manual Bin Size.

How to generate and verify

  1. Enter your values separated by spaces or commas, or import a CSV/Excel file.
  2. Set the chart title and X/Y axis titles so the output is self-explanatory.
  3. Pick the histogram type, then leave Bin Size empty or enter a value.
  4. Keep Bar Gap at 0 for a standard histogram.
  5. Adjust bar, background, title, and axis text colors if needed.
  6. Click Generate Histogram, then download as PNG, JPEG, or SVG.

To verify the settings are right, check that the bars cover the full range of your data, that no single bar dominates purely because the bins are too wide, and — for Cumulative — that the final bar equals your total count. If any of those fail, change the bin size and regenerate.