How letter frequency counter works
Letter frequency analysis counts how many times each letter of the alphabet appears in a body of text. It is the classic technique for breaking monoalphabetic substitution ciphers (Caesar, Atbash, simple keyword ciphers): in English the most frequent letters are usually E, T, A, O, I, N, so the most frequent ciphertext letter is a strong candidate for E.
This counter scans your text and tallies only alphabetic letters (A-Z), ignoring digits, punctuation, spaces, and symbols. By default it folds case so 'A' and 'a' count together; toggle case sensitivity to count them separately. Results are sorted by count descending, with each letter's percentage of all letters shown.
Compare the resulting distribution against the known English letter-frequency profile to identify likely plaintext mappings when attacking a cipher, or use it for text analysis, stylometry, and language detection. All processing happens in your browser — no text is uploaded.