What Hyperliquid's Top Traders Reveal About Perp Risk

A leaderboard ranked by raw profit hides the more useful question. Two traders can both sit near the top of Hyperliquid's 7-day PnL table and be running completely different businesses: one is moving enormous size for a thin percentage, the other is compounding a small book hard. The number that separates them is ROI — return on the capital actually at work. Read PnL and ROI together and the leaderboard stops being a scoreboard and starts being a map of where perpetual-futures risk is being taken.

Two numbers, not one

PnL (profit and loss) is the dollar result of a position over a window — here, realized performance across the trailing seven days. ROI (return on investment) is that profit divided by the equity backing it. Equity, on a perps venue, is the margin in the account: the collateral that absorbs losses and gets liquidated if a trade goes wrong.

The relationship is simple but easy to skip past:

PnL ≈ ROI × equity

So a large PnL can come from a large ROI on a small book, a small ROI on a large book, or anything between. The leaderboard below shows both extremes living side by side.

The current top of the board

Hyperliquid publishes account performance openly, which makes this measurable rather than anecdotal. Here are the top six accounts by trailing 7-day realized PnL as of 22 June 2026:

Trader	7d PnL	ROI	Equity
0X393D…2109	$22,374,738	2.3%	$967,244,104
0XE611…98A7	$11,185,243	1.3%	$893,336,364
0XFC27…9D9D	$5,119,410	19.6%	$31,176,202
0X9DCF…302D	$4,609,337	25.4%	$22,779,769
0X4EC8…9A80	$3,617,938	0.8%	$431,318,522
0X3200…C407	$3,140,517	18.8%	$17,373,633

Rank by dollars and the order looks like a simple hierarchy. Rank by ROI and it inverts almost completely.

A worked example

Take the account at the top, 0X393D…2109. Its $22.4M of profit is the largest on the board by a wide margin — roughly double the second name. But against $967M of equity, that is a 2.3% weekly return. The PnL is a function of scale, not of an unusually sharp week.

Now 0X9DCF…302D. Its $4.6M of profit is barely a fifth of the leader's in dollar terms. Yet it sits on $22.8M of equity, so the return is 25.4% — more than ten times the leader's percentage. To match that ROI on its own book, the leader would have needed roughly $246M of profit instead of $22M.

Same table, opposite businesses. One is a capital-allocation operation where the edge per dollar is thin and the dollars are vast. The other is a high-conviction book where the edge per dollar is large and the dollars are modest. Neither is "better" in the abstract — they are answers to different constraints.

What the spread tells you about risk

Group the six accounts and a pattern appears. Three of them — 0X393D…2109, 0XE611…98A7, and 0X4EC8…9A80 — carry equity from $431M to $967M and post ROIs of 0.8% to 2.3%. The other three — 0XFC27…9D9D, 0X9DCF…302D, and 0X3200…C407 — run $17M to $31M of equity and post ROIs of 18.8% to 25.4%.

That clustering is the signal. The biggest dollar profits this week are being produced by low-percentage returns on very large balances — the behaviour of size-constrained capital that cannot take concentrated bets without moving the market against itself. The high-percentage returns come from books small enough to express conviction and exit cleanly.

For anyone reading the tape, the takeaway is about where fragility sits. A 2% return on a near-billion-dollar book is steady but unspectacular by design; a 25% week is the kind of return that comes with a matching drawdown distribution. Big ROI and small equity tend to travel together because the same flexibility that lets a small book win fast lets it lose fast.

How to read the next leaderboard you see

Before you anchor on whoever tops the dollar column, do three things. Find the equity, divide PnL by it to recover ROI, and ask which of the two businesses — scale or conviction — the number describes. The trader who made the most money and the trader who traded the best are rarely the same row, and the gap between them is exactly the risk you would be taking on if you treated the board as a signal.

See the funding, equity and ROI behind any perp on one screen at app.formion.ai.