Bottom line up front
Only the most recent public fills are visible, so this audit covers the data covered rather than full account history. The account is near break-even in that window: $167.73 realised PnL on $31.8M gross volume, with a highest balance in this window of $19.115M and a lowest balance of $19.111M, representing a deepest decline in this window of 0.02%. The headline masks a structural problem: five oversized revenge trades—SOL short for $1.99M notional, VVV short for $239K notional, xyz:MU short for $448K notional, HYPE short for $453K notional, and CRV short for $29K notional—consumed $4.8M in realised losses, while a handful of short-side scalps and one long xyz:MU trade kept the account afloat. Fees of $340.46 net drag further eroded what was already a thin edge. The account is currently short BTC, ETH, ATOM, DYDX, and SOL with no stops in place.
What the data shows
This is a 38-closed-episode window spanning 25 hours, from 23:46 UTC on 18 May to 00:05 UTC on 20 May 2026. The account opened with an estimated $19.113M and closed at $19.114M, a gain of $167.73 after $340.46 in net fee drag. Realised PnL before fees was negative: -$5,302 closed, offset by $5,469.73 in realised wins.
The money was made on the short side: six short-only trades (ZEC, TON, xyz:CL, DASH, MON, PENDLE) returned $2,819.51 combined. Long trades contributed $2,608.43, driven almost entirely by two xyz:MU long scalps that netted $2,090.53. The short-side edge was real but narrow: 56% win rate on shorts, 61.5% on longs.
The money was lost in revenge trades. After a small loss on DYM, the account opened a $1.99M short SOL position on 18 May at 23:46, which closed 12 hours later at a loss of $1,893.27. Immediately after, a $239K short VVV opened and lost $2,311.62 in 18 hours. Then a $448K short xyz:MU opened and lost $571. A $453K short HYPE lost $453.79. A $29K short CRV lost $136.08. These five trades alone account for $5,365.96 in realised losses—nearly the entire loss bucket. The pattern is unmistakable: each loss triggered a larger, more leveraged position in a different coin within minutes.
Long trades showed a 61.5% win rate versus shorts at 56%, but the sample is small (13 longs, 25 shorts). Fees consumed $340.46 net, equivalent to 203% of the final realised PnL. Without fees, the account would have been -$172.73, still negative.
Trade quality
Win rate: 57.89%. Profit factor: 1.03. Expectancy: $4.41 per closed trade. Win/loss ratio: 0.75 (average winner $261.98, average loser -$349.74).
These numbers describe an account that is barely profitable on a per-trade basis and only because of a small number of outsized wins. The 1.03 profit factor means that for every dollar won, $0.97 is lost. The expectancy of $4.41 per trade, applied across 38 closed episodes, yields $167.73—exactly the realised PnL. The win/loss ratio of 0.75 indicates that winners are smaller than losers; the account survives only because it wins more often than it loses. Fees erased 203% of net PnL, so the gross edge (before execution costs) was marginally positive, and the net edge is negligible.
Post-mortems
VVV short, 18–19 May, $239K notional, -$2,311.62. Opened 23:46 UTC on 18 May at an unspecified entry price, closed 17:49 UTC on 19 May at $15.03. Held 18 hours. This was flagged as both an oversized loser (95.5× the median loss) and a revenge trade, opened immediately after the SOL short loss. The position size was extreme relative to the loss magnitude: a $239K short that moved only $2,311 represents a 0.97% move against the position. This suggests either a very tight entry or a liquidation-adjacent close.
SOL short, 18–19 May, $1.99M notional, -$1,893.27. Opened 23:46 UTC on 18 May, closed 12:04 UTC on 19 May. Held 12 hours. Flagged as oversized loser (78.21× median) and revenge trade, opened after a $9.35 loss on DYM. The $1.99M notional short SOL closed at $84.97. This was the largest single loss in the window and triggered the cascade of revenge trades that followed.
xyz:MU short, 19 May, $448K notional, -$571. Opened 00:01 UTC on 19 May at $690.65, closed 18:03 UTC at $703.94. Held 18 hours. Flagged as oversized loser and revenge trade. The entry was 0.38% below the exit price (MAE), and the position had a 4% structural stop. This loss came after the VVV short closed, continuing the pattern of re-entry after losses.
What the risk simulator reveals
Under a 1% stop-loss rule applied historically to this window, the account would have realised $197,435.50 with a maximum decline of 1.2%. Under 2%, $394,871. Under 4%, $789,742.01. Win rate across all simulations held at 60%.
These figures are gross of fees and represent a counterfactual: if every trade had been capped at 1%, 2%, or 4% loss, the account would have been dramatically more profitable. The actual deepest decline in this window was 0.02%, so the account never approached a 1% loss on any single closed trade. The simulator is showing that the account's edge—when protected by mechanical stops—scales linearly with risk capital. The absence of stops in the actual trading (and the presence of revenge-sized positions) is the difference between $167.73 and $789K.
Open positions
The account is currently short BTC at $76,774.50 entry with 40× leverage, unrealised loss of $1,246.52, no stop in place. This is the largest open exposure by notional. Short ETH at $2,114.26 with 25× leverage, unrealised gain of $160.37, no stop. Short ATOM at $2.0489 with 5× leverage, unrealised gain of $17.84, no stop. Short DYDX at $0.14221 with 5× leverage, unrealised loss of $565.69, no stop. Short SOL at $84.3767 with 20× leverage, unrealised gain of $535.12, no stop. Combined unrealised PnL across all five positions is -$1,099.54. None of these positions have stops, and the BTC short is the only one with material leverage risk.
Honest summary