FAQ
Short answers, stated limits.
If a question is about method or privacy, follow the linked detail page.
What is the difference between an unscramble and an exact anagram?
Subset mode may use some or all letters. Exact mode requires every supplied character to be used once, including duplicates.
How do wildcards work?
Each ? or * covers one missing letter anywhere in a rack search. It does not represent a fixed crossword position.
Are all results valid in Scrabble?
No universal list exists. Verify unfamiliar words in the official dictionary required by your region, tournament, or app.
Why is a common word missing?
Our browser list is intentionally curated and incomplete. Email the word and context so we can review it.
How are scores calculated?
Displayed values use familiar English tile face values only. They do not include board multipliers, cross-words, bonuses, or blank-tile adjustments.
Does the solver upload my letters?
No. Matching runs locally in the browser. Hosting and advertising may still process ordinary technical data; see Privacy.
Can I use it during a live game?
Only when the agreed rules allow assistance. It is always suitable for solo practice and post-game review.
Who runs the site?
Letter Solver Pro is an independent project operated by Stephen Liu from Hong Kong SAR.
More questions
The questions that come up after the first result list
The FAQ above covers the mechanics. These are the questions about limits — what the result list does not tell you, and why some of it looks the way it does.
Where does the word list come from, and how big is it?
The list is bundled with the page and runs in your browser, not on a server. It is curated rather than exhaustive: it favours words that appear in general reading, and deliberately leaves out a long tail of obscure entries. A comprehensive tournament dictionary contains far more words than any page should ship to a phone. The trade-off is stated plainly — you will sometimes fail to see a valid but rare word, and you will almost never see an invalid one.
Why is there no score or probability next to each result?
Because any number would be invented. A solver can tell you a word exists in its list; it cannot tell you how likely a player is to find it, or how many points it scores in a game whose dictionary it does not have. Where tile values are shown, they follow standard English letter values and are labelled as such — they are a reference, not a promise about the game you are playing.
Why do the same letters produce different lists in subset and exact mode?
Exact mode requires every supplied letter to be used once, including duplicates, so a seven-letter rack returns only seven-letter answers. Subset mode allows shorter answers, which is why the list is longer and why the first results look unrelated to the full rack. Neither mode is more correct; they answer different questions.
Can I combine wildcards with a subset search?
Yes, and the combination is where most surprises come from. A wildcard covers exactly one missing position, so two wildcards in a five-letter query effectively mean "three known letters, any two others, in order". If the result list looks too broad, remove one wildcard before adding more letters.
Does anything I type get sent anywhere?
No. Matching happens in the page, and the rack you enter never leaves the browser — it is not sent to a server, not logged, and not stored after the tab closes. That is also why the tool works with the network tab open and no requests firing.
What if a word I know is missing?
Send it. Missing common words are treated as a defect in the list rather than a difference of opinion, and reported words are reviewed together rather than one at a time. The contact route is on the contact page; include the rack you used so the case can be reproduced.