Letter Frequency Counter

How to use it

1. Paste or type your text in the box, open a plain-text file with Open file, or click Example to load the opening of A Tale of Two Cities.
2. Choose the options: Letters only, Case-sensitive, Count accented letters as plain, and the Order of the rows.
3. Read the table: each row shows the letter, its count, its share of everything counted, the typical share in English and a bar scaled to the most frequent letter.
4. Use Copy or Download .txt to take the result as plain text, one row per line: letter, count and percent separated by tabs, ready to paste into a spreadsheet.

The line under the table shows the total, for example “129 characters counted, 20 different”.

The options

Here is what the options do to one input, Hello, World! 2026:

How the counting works

• What a character is. The text is split into the characters you see, not into code units, so an emoji with a skin tone such as 👍🏽 counts as one character (with “Letters only” off) rather than two or four.
• Whitespace (spaces, tabs, line breaks) is never counted, whatever the options.
• Letters from any script count. With “Letters only” on, Greek, Cyrillic, Arabic, Chinese and other letters are counted too: 東京 Москва gives 8 letters, each its own row. The “In English” column shows a dash for anything outside a–z.
• Accents. Without the fold option, é is a different letter from e: Café crème, naïve gives 11 different letters, with é, è and ï in their own rows. With the option on, the same text gives 9, and e rises to 4. Letters that are not built from a base letter plus a mark stay as they are, so ø, æ and ß are not changed into o, ae or ss.
• Percentages are worked out from everything counted under the current options, so changing “Letters only” changes every share.
• Long text. The table shows the first 120 rows, which only matters with many scripts or symbols mixed in; the copied text result always has every row.
• Two ways of typing an accent. An accented letter typed as two code points (a plain e followed by a combining accent, common in text copied from some PDFs and file names) is normalized first, so it lands in the same row as the usual one-character é: the two forms of é together give one row, é 2.
• Turkish İ. Without “Case-sensitive”, the capital dotted İ counts as i, so İstanbul gives 8 letters with one i. With “Case-sensitive” on, İ keeps its own row.

Worked example: the sample text

The Example button loads the first lines of Dickens’ A Tale of Two Cities. With the default options it counts 129 letters, 20 different:

E and T lead, as in typical English, but W, S and F are far above the norm because the passage repeats “it was”, “times” and “of”. This is the main thing to remember about letter frequency: one sentence or paragraph can sit far from the averages, and the figures only settle near the English table for long texts.

Frequency analysis of a cipher

In a substitution cipher each letter is always replaced by the same other letter, so the letter counts stay the same; only the labels change. That is why counting letters is the first step in breaking one. Take this sentence, shifted by 3 with the Caesar cipher tool:

Phhw ph qhdu wkh rog wuhh dw vhyhq; eulqj wkh ohwwhu dqg wkh nhb.

The counter finds 50 letters, and h is far ahead with 14 (28%), followed by w with 8 (16%). Guessing that h stands for e, the most common letter in English, gives a shift of 3 (e → h), and w → t fits the same shift. Shifting back by 3 gives the message: “Meet me near the old tree at seven; bring the letter and the key.” Short messages do not always follow the English pattern, so test the second and third most frequent letters too if the first guess gives nonsense.

When people use it

• Puzzles and word games. Paste a list of candidate words for Wordle, Scrabble or a crossword to see which letters appear most often in it. The tool counts letters anywhere in the word, not by position.
• Teaching and classwork. Students compare the letters in their own essay or a novel chapter with the English table, or count letters in another language to see how its pattern differs.
• Typography and design. Before choosing or subsetting a font, check which letters and accented characters a text really uses (untick “Letters only” to see punctuation too).
• Checking text data. A quick scan with “Letters only” off shows stray characters such as curly quotes, dashes or symbols you did not expect; the remove special characters tool can then clean them out.

Letter frequency vs word frequency

Letter frequency answers “which characters appear most”; word frequency answers “which words repeat most”, which is the one to use for spotting overused words in writing. For the total number of characters, with and without spaces, use the character counter.

Questions