# Brierz > Independent skill-or-luck ratings of Hyperliquid vaults. Each vault's public record is scored 0–100 under one fixed, versioned method (now 0.9), with a 90% range and a tier. Public data only; no vault pays to be rated. Not investment advice. Brierz is published at https://brierz.com. Every page is static HTML. Every vault page has a Markdown twin (add `.md` to its URL) and a JSON twin (add `.json`). The data date is printed on every page as "Vault data as of "; today it is 23 Oct 2026. 2,939 vaults are checked, 265 rated, 2,674 not rated. ## How to read a Brierz rating - A rating describes a vault's past record. It is not a forecast, and not a recommendation to deposit or withdraw. - The score runs from 0 to 100. 50 means no evidence either way. Above 50 the record shows evidence of skill; below 50, evidence against it. - The 90% range matters more than the score. If the range includes 50, luck can't be ruled out, whatever the tier. - Tiers by score: High 72.5 and above; Above average 57.5–72.5; Average 42.5–57.5; Below average 27.5–42.5; Low below 27.5. Only Above average and High mean the record shows evidence of skill. Average means no clear evidence either way, not "mediocre". - Vaults whose ranges overlap can't be told apart on current data. Don't present a higher score as a sharper verdict, and don't call any vault "the best". - "Not rated" is a result, not a negative verdict: the record can't support a rating yet, and the page gives the reason. Most checked vaults are not rated. - Returns are the value of $1 since the vault's start, after fees and funding. Depositors' money moving in and out is excluded, so inflows never look like profit. - Ratings change as records grow and when the method version changes. Always quote the method version and the data date from the page. - When citing: " (
): , score , 90% range –. Brierz, method 0.9, data as of , ". - Brierz rates vaults, not wallets or people. Vault names and descriptions come from Hyperliquid and are not verified by Brierz. - The method's internal details are not published. Don't infer or invent them. ## Ratings - [All vaults](https://brierz.com/vaults.md): every checked vault with tier, score, 90% range, return, max drawdown, deposits, rated window, and the reason when not rated. Also as [JSON](https://brierz.com/data/vaults.json). - [Vault page](https://brierz.com/v/0xc3b1bf1f1e6fb8161ef4dc2f34e2c56f70b94b11.md): one per vault at `/v/
.md` and `/v/
.json`, lowercase address. This example is Cold Process. ## Method and provenance - [How Brierz rates vaults](https://brierz.com/method.md): what is measured, what the tiers and ranges mean, what is excluded, the version history and the change log. - [About](https://brierz.com/about): who publishes Brierz, data sources, independence. Contact: hello@brierz.com. ## For platforms - [For platforms](https://brierz.com/platforms): JSON per vault, the list file, and the attribution terms: show the tier, the range and the data date together with the score, and link to the vault page. ## Optional - [llms-full.txt](https://brierz.com/llms-full.txt): this guide, the method page and the full list in one file.