Help & Docs

Copying means that when an account you follow opens or closes a trade, the same trade is placed in your own account, at your size. Your money never leaves your account.

LinceBot is our automated strategy: it trades for you in your own Hyperliquid accounts at the risk level you choose (×1 to ×4).

The stars (1 to 5) say how easy it is to copy an account in time. They do not say whether it is good: an account can be excellent and impossible to follow, or easy to follow and mediocre.

Drawdown is the biggest fall from a high point before recovering. It tells you how much you would have had to endure: a 40% drawdown means that at some point the account was 40% below its best.

The rest of this page goes deeper; the Glossary tab explains every term.

Linceo shows you who really wins on Hyperliquid, using the platform's own official data. Everything traded on Hyperliquid is public, so anyone can check any account's record. We do it for you: we look at where its profit came from and whether you could follow it in time. If you want, its trades are copied into your own account. Your money never leaves your account.

Every wallet goes through five steps:

1. Audited risk metrics: return against risk (Sharpe) from the official daily data, the biggest drop measured on profits (not on the balance, which moves with deposits and withdrawals), Calmar and month-by-month consistency. How each one is computed is in the Metrics tab.
2. Where the profit comes from (skeleton check): tells legitimate withdrawals apart from real losses and flags wallets whose PnL was ever negative.
3. Style and structure — the trading style from the real median hold time, and structure patterns seen in the fills: market-maker grids, martingale-style averaging into a losing position, many positions open at once.
4. Copyability ★ 1-5 — could you mirror this wallet's entries and exits in time, with the real latency between a signal and your order? It scores timing and structure: how long positions are held, how many entries are open at once and how many close within 30 seconds. It is not a quality verdict: Sharpe, ROI, drawdown, ticket size and recency are the leaderboard filters, applied separately.
5. Realistic backtest — the portfolio backtest counts and flags the entries that, at your capital, would fall below Hyperliquid's $10 minimum order; the curve itself is not changed.

• Free: auditing any account, the full leaderboard, every metric and deep tab, alerts, the watchlist, side-by-side compare, the portfolio backtest and CSV/JSON export. No subscription and no card.
• 0.01% of what you trade, only if you copy or run LinceBot on Hyperliquid. Nothing else to us. Example: $1,000 traded = $0.10. Hyperliquid charges it on your fills and passes it to Linceo (technically, a "builder fee"). You approve that exact rate in your wallet first. Hyperliquid's own trading fees and funding are separate. On Polymarket there is nothing to pay, because no order is ever placed for you.

Go to the Leaderboard. Use the filter strip to set the quality bar — minimum profit, minimum capital, Sharpe, recency — then a minimum copyability (★) and a sort (Sharpe, PnL, drawdown or median account value). Losing wallets, wallets with negligible capital and wallets rated below 3★ (not copyable) are hidden by default; lower the ★ filter to see them.

Click any row to open the wallet's full page. The Add to portfolio button puts it in your cart.

A wallet's page shows:

• a header with the size and profit tags, the copyability stars and the watchlist toggle;
• the key figures — the PnL is Hyperliquid's own figure;
• the deep tabs: Capital, Trades, Coins, Temporal, Episodes, Structure and Skeleton.

Some tabs are rebuilt from the wallet's full activity. The first visit downloads it; later visits only fetch what is new and are fast.

From the Leaderboard or a wallet's page, add wallets to your cart. Open Portfolio to:

• adjust the weight of each one;
• set the starting capital (from $100);
• turn Realistic mode on or off (on by default) — it applies Hyperliquid's $10 minimum order;
• see the combined figures (PnL, return %, final value).

The cart is saved in your browser and works across tabs.

Auditing is free. Copying or running LinceBot on Hyperliquid costs 0.01% of what you trade ($1,000 = $0.10) and nothing else to us, on top of Hyperliquid's own trading fees and funding. Donations are optional, grant no service and are non-refundable.

It answers one question: given the latency between a signal and our order, can we mirror this wallet's entry and exit before the edge is gone? It scores timing and structure, not quality.

Quality — Sharpe, ROI, drawdown, win rate, ticket size, recency — lives in the leaderboard filters, applied separately. A mediocre but slow trader is very copyable; a brilliant scalper is not. Filter on quality, then copy what the stars say you can.

Levels

• ★★★★★ Copy at leisure — discrete entries, held long enough to mirror both the entry and the exit
• ★★★★☆ Easy to copy — a little brisk, or a few entries open at once
• ★★★☆☆ Workable — fairly fast, or several entries open at once; copyable with effort
• ★★☆☆☆ Marginal — fast and with many entries open at once; easy to miss fills
• ★☆☆☆☆ Uncopyable — scalping, HFT or a heavy market-maker grid: no time to react

What forces 1★

Any one of these, measured on the wallet's fills: more than 30% of positions closed within 30 seconds; 30 or more entries open at once while holding for under an hour (a market-maker grid); a median hold under one minute.

On the leaderboard, where there are no fills, timing is estimated from how many times a day the wallet turns over its own capital: more than 50× is 1★, more than 20× is 2★, more than 8× is 3★, more than 2× is 4★ and the rest 5★. The wallet's page uses the real hold time and the entries open at once.

A copy is sized in proportion: the wallet's order × your capital × weight ÷ the wallet's peak account value. With a small bankroll copying a big wallet, that proportional copy can fall below Hyperliquid's $10 minimum — an order the venue would reject.

Realistic mode detects it and tells you how many orders would fall below the $10 minimum; the curve doesn't change. In real copying those orders are raised to the minimum, so each weighs much more in your bankroll than in the leader's, which amplifies both gains and losses.

The minimum: Hyperliquid documents a minimum order value of $10. The real floor per coin ranges from $10.00 to $14.07 with its size decimals and mark price; we use a flat $10, which is right about 95% of the time.

Returns come from the trader's PnL over a constant capital base, one return per day (the official series is resampled to the last value of each day; a day without a point keeps the previous one): mean(ΔPnL_day ÷ base) ÷ std × √365, × a small-sample correction that shrinks it towards 0 under ~60 days or ~20 data points.

The official Hyperliquid series is irregular: hourly points recently, about weekly further back. Taking one return per point would weigh an hour and a week the same, so we resample to one value per day first. And we do not take returns from the account value: it moves with deposits and withdrawals, not only with trading — computed that way, a wallet that had lost $11.2M once ranked #5.

We compute the maximum drawdown on the cumulative PnL series, not on the account value. Why: a wallet that withdraws $500k of profit shows a 99% drop in account value, but the trader lost nothing — they took money home. The MDD measures only what the trader controls: how the PnL builds up. As a %, it is divided by the account's peak over its whole life: if the wallet deposited more after the drop, the % falls short of what that drop was at the time.

Most wallets ride an almost straight PnL curve, so a near-zero MDD is normal. Below about 1% we show Calmar as “—” instead of an absurd ratio.

Six ratios that summarise the economics of an account's wins against its losses. Each shows “—” when it is mathematically undefined (for example, no losses yet) instead of a fake 0 or ∞:

• Sortino — like Sharpe, but only the downward swings count as risk. High Sharpe with low Sortino means the volatility is mostly losses.
• Profit factor — dollars won ÷ dollars lost. The unit is the month on Hyperliquid and the decided market on Polymarket, never the single trade. Above 1 makes money; 2 wins twice what it loses. Shown up to “50+”.
• Payoff ratio — average win ÷ average loss, whatever the frequency of each. The unit is again the month on Hyperliquid and the decided market on Polymarket.
• Expectancy — the average dollars THIS account nets per closed trade, at ITS size, not yours. The unit is a closing order on perps on Hyperliquid (an order filled in several pieces counts once) and a decided market on Polymarket. Your copy is sized to your own bankroll and per-order cap, so read it as the shape of their edge, never as your expected profit.
• Recovery factor — net profit ÷ worst drawdown: how many worst-case holes the total gain is worth.
• Kelly — the textbook Kelly figure for this account (half-Kelly, capped at 50%), from its win rate and payoff; on Hyperliquid it is computed on months and assumes a losing month loses the whole amount at risk; on Hyperliquid it is shown only with at least 3 months and 3★ (on Polymarket, always). It is a reference for reading the account, not a recommendation of how much to put in.

Heads up: an extremely high Sharpe or Sortino (8 or more) is almost always a high-frequency market maker — a very predictable PnL, but uncopyable. Always read these next to the copyability stars.

Compares the account-value timeline with the PnL timeline to tell apart:

• ▼ Withdrawals — the account value drops and the PnL stays flat: a legitimate cash-out
• ✗ Real losses — both drop: a trading loss
• ? Mixed events — both change, in different proportions

It also detects whether the cumulative PnL was ever negative, counts consecutive losing months and checks recent activity. It gives one of three severities:

clean attention skeleton (past losses dominate)

Every wallet carries three labels: how big it is, how much it has won or lost, and how it trades. None of them is a recommendation — they summarise what the account has actually done, computed from official exchange data. A big wallet can be a terrible trader, and a small one can be excellent.

Based on the wallet's TYPICAL capital (the median account value over its life), not on a lucky peak. It tells you whether its results can be reproduced at your size: a strategy that works with $50M may be impossible with $500.

Lifetime profit or loss, read from the platform's official API — never reconstructed by us. The red tags matter as much as the green ones: seeing what a losing account looks like is half of the lesson.

Measured from the REAL median hold time (entry → exit), taken from the wallet's own fills. That single number decides the style, because it is also what decides whether YOU could copy it in time.

When we have no fills for a wallet we fall back to its turnover — how many times a day it rotates its own capital (daily volume ÷ typical capital). It is a proxy, so the confidence drops. If there is not even enough capital to make turnover meaningful, the style is shown as Unknown: we never invent one.

Confidence: HIGH needs at least 30 days of life, 50 recorded changes and 3 months with activity. LOW means the speed signal is not reliable (no fills and no usable capital) or there is too little history. Treat a low-confidence style as a hint, not a fact.

Next to the style you may see one or more traits. They are what a single label hides, and several of them are warnings:

• laddered — 15 or more entries open at once (a ladder or a grid): hard to mirror, and the reason a slow trader can still score low on copyability
• high win-rate grinder — wins often but small: it lives on its hit rate, so one bad streak hurts more than it looks
• asymmetric — lets winners run — a few big wins pay for many small losses (average win at least 2.5× the average loss)
• balanced win/loss — average win between 1.5× and 2.5× the average loss
• consistent month-to-month — green in 70% of months or more, with the profit spread across months instead of piled into a few
• streaky / few green months — green in fewer than 40% of months: the result depends on when you got in
• one-bet dependent — a single MONTH made 60% or more of the profit: luck until proven otherwise
• deep drawdowns — has lost 60% or more from a peak at some point
• drained / near-zero balance — the account has been emptied or is close to zero
• inactive >30d — has not traded in more than 30 days

A tag is a summary of the past, computed from public data. It is not a prediction, not advice and not a promise that the account will keep doing what it did. We publish the exact thresholds precisely so you can disagree with our reading and form your own — that is the whole point of this site.

No. Your money never leaves your Hyperliquid account and only you can withdraw it. To audit, we only read public data. To sign in you sign a message that proves the address is yours. It moves nothing. If you turn on copying or LinceBot, your wallet signs a permission to trade on Hyperliquid (an "agent wallet"). Orders are then signed by that permission's key: it can open and close trades, but it can never withdraw, transfer or touch staking. We keep the agent's key, encrypted on our server; your wallet's key never leaves your wallet. You can revoke that permission on Hyperliquid at any time, and we cannot stop you. What that permission means is in Risk disclosure.

Yes. Curves and profits are Hyperliquid's own official figures, taken from its public API and shown as they are. No insider data: anyone can query the same and check any number. How we compute each metric is in How we audit. (For the technical: the /info endpoint.)

Auditing is free: the leaderboard, every metric, the wallet pages and their deep tabs, the portfolio backtest, alerts, exports and the watchlist. No subscription and no card. If you copy or run LinceBot, you pay 0.01% of what you trade ($1,000 traded = $0.10), and nothing else to us. You approve it in your wallet before anything trades. Hyperliquid's own trading fees and funding are separate.

LinceBot is Linceo's automated strategy: it trades for you in your own Hyperliquid accounts. It uses three Hyperliquid accounts of yours, each signed separately, and your money never leaves them. You choose the risk level, from ×1 to ×4, and the level multiplies gains and losses alike. Its rules are not published. Its results are: you see them in its backtest, updated every day.

Almost nothing. Each part of LinceBot sizes its trades on one third of the total of your three accounts, and over time some accounts grow more than others. Once a week, on Sunday (17:00 UTC), we check them: if one has drifted more than 10 points from its third, we tell you on Telegram (and you see it in LinceBot) how much to move and between which accounts. You do it yourself on Hyperliquid, within a day or two; LinceBot cannot move your money. In the backtest, at ×4 it was needed about 9 times a year, and very few at the lower levels. If you don't do it, an account can run out of margin.

You can lose what you risk. With leverage, in a fast market, it can be more than you expected. Past results do not guarantee future ones, and future drops can be bigger than the backtest's. With LinceBot, the level multiplies gains and losses: check each level's biggest drop before you choose. The full detail is in Risk disclosure.

Whenever you want. In LinceBot, Pause stops opening new positions, and what is already open keeps being managed until it exits. Stop makes the bot stop sending orders to your three accounts. Neither closes your open positions or removes the permission on Hyperliquid: you do that yourself at app.hyperliquid.xyz → More → API, without going through us. To stop copying, use “Remove permission and stop” in Portfolio › Overview.

Your money stays in your Hyperliquid account, and so do your open positions and their orders (stops and pending entries), with nobody managing them. In that case, cancel the open orders, manage or close the positions in each account and remove the permission to trade.

To audit, nothing: no sign-up and no card. To copy or run LinceBot, a wallet with a balance on Hyperliquid. LinceBot also uses three Hyperliquid accounts of yours, each signed separately, and nothing trades until all three are signed.

They say how easy it is to copy an account in time, from 1 to 5. They do not say whether it is good. They score timing and structure: could we mirror the entry and the exit with our latency? The signals are the median hold time (entry → exit), how many entries are open at once and the share closed within 30 seconds; on the leaderboard, how many times a day the capital turns over. Sharpe, ROI, drawdown, win rate, ticket size and recency are quality filters you set on the leaderboard: they are not part of the stars. See Metrics → Copyability score.

Sharpe (a quality filter) rates how good the wallet is in absolute terms: return adjusted for risk. It can be stellar and the wallet still uncopyable, because it runs sub-second HFT or keeps dozens of market-making positions open at once, so there is no time for us to mirror the trade.

Copyability ★ answers a different question: can we replicate the entries in time? They measure different things — quality against timing. What you are looking for is both: a high-Sharpe account with real capital that is also 4-5★ copyable.

With little capital, copying a big wallet in proportion gives trades below Hyperliquid's $10 minimum. Realistic mode counts how many and warns you; the curve doesn't change (it's the proportional projection of the official data). In real copying those trades are raised to the minimum, so they weigh more in your capital and the result would differ. See Metrics → Realistic backtest mode.

• HL / Hyperliquid — a high-performance perpetuals exchange on its own L1 blockchain
• Perpetual (perp) — a futures contract with no expiry that follows an asset's price; you can go long or short, with leverage
• Leverage — holding a position larger than the collateral behind it: at 10×, a 1% move is 10% of your money
• Liquidation — the exchange closes your position when losses eat the collateral that backs it
• AV — account value: collateral plus unrealised PnL at a given moment
• Permission to trade (agent wallet) — Hyperliquid's own secondary key, with limited permissions, which your wallet authorises with a signature. It can trade but never withdraw, and you can revoke it on Hyperliquid at any time
• Linceo's fee (builder fee) — the fee Hyperliquid lets apps charge for routing your orders. Linceo's is 0.01% of the volume of each order it places for you, and you approve it in your wallet: $1,000 traded = $0.10
• Venue — the exchange where the trading happens: here, Hyperliquid or Polymarket
• Fill — an order that actually executed, in full or in part
• Notional — the size of a position in dollars: quantity × price
• Ticket — the size in dollars of a single order
• Scalp — opening and closing within seconds or minutes to catch tiny moves
• Grid / market maker — keeps many buy and sell orders resting at once and earns the spread between them, with dozens of positions open at the same time
• Cohort — the size band or profit band an account falls in (see Tags & styles)
• ROI — return on investment: profit ÷ capital put in
• PnL — profit and loss, cumulative or realised
• MDD — maximum drawdown, the worst run from a high: the largest drop from a peak to a later low
• Sharpe — return adjusted for risk: mean ÷ standard deviation × √annualisation
• Calmar — annual return ÷ maximum drawdown
• Gini — how concentrated the monthly PnL is: 0 = spread evenly, 1 = all in one month
• mPos% — the share of months with a positive PnL
• Copyability ★ — a verdict on timing and structure: can we mirror entry → exit before the edge is gone? It is NOT a quality score
• SIWE — Sign-In With Ethereum (EIP-4361): logging in with your wallet, without passwords
• Non-custodial — your money never leaves your account: the platform does not hold it and cannot withdraw it, and you keep control through your wallet's keys