Letter Frequency Analyser
Count the letters in any English text and the same shape appears every time: E far ahead, then T, A and O, with J, Q, X and Z barely present. That regularity is what makes simple substitution ciphers breakable. The index of coincidence goes one better — it measures how uneven the distribution is, and because shuffling the alphabet does not change it, it identifies English even when every letter has been swapped.
What this generator does
Counts every letter in your text, shows each one's share beside the standard English figure, and computes two summary measures: the index of coincidence, and a chi-squared comparison against English frequencies.
How to use this tool
- Paste text into the box — a few hundred letters gives a much steadier picture than a sentence.
- Compare each letter's share against the English figure beside it.
- Read the index of coincidence: near 0.067 is English, near 0.038 is random.
- Try enciphering your text by shifting every letter, and watch that figure stay put.
Understanding the controls
- Your text
- At least 40 characters, though a few hundred gives a far steadier reading. It is counted on your own device and never sent anywhere.
Common use cases
- Frequency analysis when breaking a simple substitution cipher
- Telling enciphered English apart from random characters
- Teaching why substitution ciphers are weak
- Checking whether generated text has natural letter statistics
- Puzzle and escape-room construction
How this generator works
The index of coincidence is the probability that two letters picked at random from the text are the same. Uneven distributions raise it, so English sits near 0.067 while a flat distribution sits near 0.038. Because it depends only on how often letters repeat and not on which letters they are, replacing every letter with another leaves it completely unchanged — which is exactly why it is the first measurement a codebreaker takes.
Randomness and fairness
Nothing here is random. The counts and both measures follow deterministically from the text you enter.
For how randomness is produced across the whole site, see how Generate Random works.
Limitations and good to know
- English reference frequencies only; other languages have different profiles and will look wrong against these.
- Short samples are unreliable — under a hundred letters the figures move a great deal on very little.
- Only the 26 unaccented letters are counted; digits, punctuation, spaces and accented characters are ignored.
- The index of coincidence identifies the shape of a distribution, not the message — it cannot decipher anything on its own.
Privacy and your data
Your text is counted in your browser and never leaves the device. Analytics records only that the tool ran and roughly how much text was involved, never the text itself.
Related generators
- Cryptogram Puzzle GeneratorTurn a message into a substitution cryptogram, with letter-frequency aids for solving and a reveal for checking.
- Readability Score CalculatorFlesch Reading Ease, Flesch-Kincaid, Gunning Fog and SMOG for your own text, with each figure recomputed and its limits stated plainly.
- Soundex Phonetic Code GeneratorSoundex codes for your own list of names, grouped by the code they share, and the pairs the algorithm can never match however alike they sound.
- Text Statistics GeneratorCounts words, sentences and distinct vocabulary in a passage, with every figure re-derived from the text and the ratios kept honest.