About SeeMyRank
What this is
SeeMyRank answers one question: where does a chess player actually stand? Type a Chess.com or Lichess username and you get a percentile for each time control — not a rating you already knew, but the share of rated players you are above.
It also does the two things people immediately want next. It says what a rating on one site is worth on the other, using anchors measured here rather than repeated from forum posts. And it compares two players head to head, with the record between them where one exists.
There are no accounts and nothing to sign up for. You do not need to own the username you look up, which is the point — most people arrive wanting to check a friend, a rival, or somebody famous.
How the ranks are worked out
Percentiles are computed against published rating distributions, never estimated from a formula. Lichess publishes the full distribution for each of its rating pools, and those figures are counted directly: when this site says 1600 is the 61st percentile in Lichess blitz, it is because 61% of the 657,474 players with an established blitz rating are below it.
Chess.com publishes no equivalent distribution. The figures for it are DERIVED — mapped across from the Lichess distribution through the conversion anchors — and every page that shows one says so and rounds it to make the second-hand status visible. That is a real limitation, stated rather than hidden behind a precise-looking number.
A provisional rating gets no percentile at all. The rating systems mark a new account as unsettled for a reason, and placing an unsettled number inside a distribution of settled ones produces a figure that looks precise and means nothing.
Every table on this site is computed when the page is rendered, from the same data files the lookup tool uses. An article cannot quietly disagree with the product, and where the prose states a figure, a test derives that figure from the data file and fails if it stops appearing.
Where the data comes from
Ratings and rating histories come from the public APIs that Chess.com and Lichess publish for this purpose. Requests identify themselves honestly, are rate-limited on our side, and back off when either service asks them to. Nothing here is scraped, and no endpoint that requires an account is used.
FIDE figures are counted from FIDE's own published monthly lists, including fourteen years of its archive for the rating-inflation study. Real names for titled players come from Wikidata, which is CC0, and are shown for titled players only.
Every page built on data carries the date that data was pulled. Where a measurement was made here rather than taken from somewhere, the page says how many players it covered and what it could not reach.
Who this is, and who it is not
SeeMyRank is an independent site. It is not affiliated with, endorsed by, or connected to Chess.com, Lichess or FIDE, and it does not speak for any of them. Those are their trademarks and their data; this site reads what they publish and says where it got it.
It is not a coaching service, a training tool, or a rating you can put on anything official. It is a measurement of where a public rating sits in a public population, and it is careful about the difference between what it measured and what it inferred.
The site is free and paid for by advertising. Advertisements never appear on a page that has nothing on it and never sit above an answer, because an accidental tap on a result is not a click anybody wanted.
Corrections, and getting your name off it
If a number here is wrong, saying so is genuinely useful. Several figures on this site were corrected because they were re-checked against the sources, including a cross-platform conversion that had its sign backwards at the top of the range and was quietly producing nonsense for the strongest players.
Player pages are built on demand from public profiles. They are never listed in the sitemap and always carry a noindex instruction, so they do not accumulate in search results. If you would rather your username did not resolve here, say so and it will be blocked — the report stops being served and the name disappears from the leaderboard, the climbers and the autocomplete as well. No reason is needed, and the site keeps no record of who asked: the block is stored as a one-way digest, so the list cannot be read as a roster of people who objected.
Translations are the other thing worth reporting. Nine of the twelve languages started as machine passes and are marked as such in the site's own files; if one reads badly in yours, that is a bug and not a style choice.
Contact
Questions, corrections, removals, or a translation that reads badly in your language: admin@yatedlabs.com