Letter Frequency in English, Explained

Open any long English text and count the letters, and the same ranking emerges every time: E first by a wide margin, then T, A, O, I, N. The full sequence — immortalized in printing lore as 'ETAOIN SHRDLU', the first two columns of a Linotype keyboard — is one of the most stable statistical facts about the language. E alone accounts for roughly one letter in eight; Q, J, X, and Z together account for less than one in a hundred. That imbalance is not trivia. It is the hidden engine behind every word game you have ever played.

The rough numbers

• E ≈ 12–13% — the undisputed champion, which is why a text written without it (it has been done) reads so strangely.
• T, A, O ≈ 7–9% each — the workhorses of articles, pronouns, and common verbs.
• I, N, S, H, R ≈ 6–7% each — completing the top nine that cover well over half of all text.
• D, L, U, C, M ≈ 2–4% each — the comfortable middle class.
• J, Q, X, Z — each under 0.2%, the rare-letter cellar that Scrabble prices at 8–10 points.

Two subtleties make the picture richer than a single ranking. First, position matters: S is the most common first letter of English words even though it ranks only around seventh overall, because it starts so many everyday words and plural-adjacent forms. E, meanwhile, is far more likely at the end of a word than at the beginning. Second, the ranking shifts with the text: dictionary headwords, telegrams, and casual conversation each have slightly different profiles. Game strategy mostly cares about the dictionary profile — the set of words that could be the answer — which is the one this site's word lists reflect.

How frequency wins Wordle

A Wordle opener built from high-frequency letters — the logic behind classics like 'slate', 'crane', and 'raise' — tests more of the answer space per guess than any clever-looking rare-letter word. Positional frequency refines it: starting S and ending E match how five-letter words actually distribute their letters, which converts more yellows into greens. When your second guess deliberately avoids every letter from the first, the two together can cover ten of the twelve most frequent letters — usually enough to see the word's skeleton by guess three.

How frequency wins Hangman — twice

Hangman is letter frequency played from both sides of the board. The guesser's optimal opening is exactly the frequency table: E, T, A, O, I, N, then S and R — sweeping the letters most likely to appear in any word. The word-picker's optimal strategy is the mirror image: choose short words built from the cellar of the table, ideally with repeated rare letters, so the standard sweep comes up empty. That is the entire reason 'jazz' is harder than 'encyclopedia' despite being a third of the length: it contains one common letter, and the guesser has a limited budget of misses.

Scrabble: the table with prices attached

Scrabble's tile values are, quite literally, a letter-frequency table converted to points — its designer counted letters on newspaper front pages in the 1930s to set them. One-point tiles (E, A, I, O, N, R, T, L, S, U) map to the frequency elite; the 8- and 10-point tiles (J, X, Q, Z) map to the cellar. Strategically, this means rack balance is frequency management: a rack of one-point tiles plays smoothly but scores slowly, while a rare-letter tile is a bounty that needs the right opening. Knowing which letters combine well — and which endings like -ED, -ER, -ING absorb your surplus — is frequency knowledge doing quiet work.

Explore it yourself

The fastest way to internalize the table is to browse words by letter and feel the difference in list sizes: compare the crowd of words starting with S to the handful starting with X. Generate batches filtered by starting letter, guess how big each list will be before you look, and you will be carrying a working frequency table in your head within a week — which is precisely the advantage the games above reward.