Digit Frequency Analyzer

Count how often digits 0–9 appear in your data. Optional Benford's Law and entropy views for rough distribution checks.

Digit Frequency Analyzer

Analyze the frequency distribution of digits (0-9) in text, numbers, or data files

Frequency Analysis

Analyze digit frequency patterns in your data

What digit frequency shows

Natural datasets rarely use every digit equally. First digits of many real-world magnitudes follow Benford's Law (1 appears more than 9). Flat distributions show up in random IDs, hashes, or fabricated uniform data. Neither pattern proves fraud by itself; they are screening signals.

Entropy summarizes how spread out the digit mix is. High entropy leans toward even use of all digits; low entropy means a few digits dominate. Interpret it against what you expect for that dataset.

Benford's Law caveats

  • Works best on magnitudes spanning several orders (populations, invoices, river lengths)
  • Fails on constrained ranges (percentages 0–100), assigned IDs, and tiny samples
  • Need enough observations; dozens of numbers are weak, hundreds are more useful

Analysis modes

  • Digits only: count 0–9 (Benford-style first-digit work)
  • Alphanumeric: letters plus digits for codes
  • All characters: full character mix including punctuation

Frequently Asked Questions

What does entropy tell me?

It measures how unpredictable the digit mix is. Compare it to a baseline for your data type rather than treating any single number as a verdict.

When does Benford's Law not apply?

Assigned numbers, fixed ranges, normal-ish samples with a narrow span, and very small datasets. Prefer naturally occurring, scale-free magnitudes.

What is a uniformity score?

A simple summary of how even digit counts are. Near-even often means random or synthetic IDs; skewed often means natural magnitudes. Context decides which is "good."

Can I analyze non-English text?

Yes for extracting 0–9 digits. Letter-mode analysis depends on the script you paste.

How large should the dataset be?

Frequency tables work on small inputs. Benford-style claims need more numbers; aim for hundreds or more when you can.

Is my data uploaded?

Analysis runs in your browser. Nothing is sent to a server for this tool.

Can I export results?

Use copy or download options in the tool UI when available (text/CSV).

Related tools

Related tools

Related tools