Bottom line up front
Only the most recent public fills are visible, so this audit covers the data covered rather than full account history. This account is -33.66% in the data covered, having fallen from a highest balance in this window of $22.3M to a lowest balance in this window of $854k—a deepest decline in this window of -96.17%. The headline loss masks a structural asymmetry: short-side commodity trades (crude oil, Brent) generated $795k in realised profit, while long-side crypto bets consumed $4.5M. The single dominant pattern is oversized, unhedged long exposure to illiquid or volatile altcoins opened in mid-December, held for weeks, and closed into losses of $1.5M (HYPE), $339k (LTC), and $255k (ENA). This is not a skill problem; it is a position-sizing and conviction problem.
What the data shows
The data covered spans 184 days of trading across 58 closed episodes and 31 open positions. The account started with an estimated $10.9M, peaked at $22.3M on 19 November 2025, then deteriorated sharply. The lowest balance in this window of $854k arrived on 18 February 2026. Current balance stands at $7.2M, implying recovery from the lowest balance in this window but still deeply underwater relative to the highest balance in this window.
Money was made in two places: short crude oil (CL) for $475k closed 10 April 2026, and short Brent oil for $320k closed the same day. Both trades were held roughly 195 hours, entered with ATR-based structural stops at 2.68% and 2.01% respectively, and executed with notional positions of $2.6M and $2.1M. These are the only two trades marked "primary_edge" in the by-instrument summary. Win rate on shorts was 75%; on longs, 4.35%. Short PnL was +$807k; long PnL was -$4.5M.
The losses came from a concentrated burst of long entries opened 16 December 2025. HYPE was the catastrophe: a $2.5M notional long held 1,117 hours, closed 31 January at $24.94, realising -$1.47M and flagged as an oversized loser (10.6x the median loss). LTC followed: $1.06M notional, same entry date, same exit date, -$339k. ENA was -$255k. AVAX was -$238k. PUMP was -$226k. STRK across two episodes was -$263k. These six coins account for -$2.8M of the -$4.5M long loss. All were opened within a 48-hour window in mid-December and held into a period of sharp crypto weakness in January.
Fees paid were $8,390 gross, a modest drag relative to gross volume of $57.3M. The net fee drag was $8,390. Realised PnL after fees was -$3.49M. The account did not benefit from rebates; maker percentage was 49.69%, suggesting balanced order flow.
Trade quality
Win rate was 14.81% across 27 closed episodes. Profit factor was 0.2—for every dollar won, the account lost five. Average win was $221k; average loss was -$198k. Win-loss ratio was 1.12, meaning wins were slightly larger in absolute terms, but far too infrequent to offset the frequency of losses. Expectancy was -$135k per closed episode. Max loss streak was 18 consecutive closed trades in the red.
These numbers describe an account with no edge in the data covered. The two commodity shorts were profitable; everything else was not. The long-side win rate of 4.35% is not noise—it is a signal that the long-side thesis was broken.
Post-mortems
HYPE long, 16 December 2025 to 31 January 2026. Opened at an unspecified entry price, closed at $24.94 after 1,117 hours. Max notional was $2.46M. Realised loss: -$1.47M. This trade is flagged as an oversized loser, 10.6 times the median loss in the data covered. No structural stop was in place. The position was held through a full month of weakness and closed into a collapse. This is the single largest leak in the account.
LTC long, 16 December 2025 to 31 January 2026. Opened at an unspecified entry price, closed at $73.24 after 1,116 hours. Max notional was $1.06M. Realised loss: -$339k. Held alongside HYPE for the identical duration, closed on the same day. No structural stop. The pattern is identical: conviction in a thesis, oversized entry, no exit discipline.
What the risk simulation reveals
The risk simulator applied historical stop-loss rules to the data covered. Under a 1% stop-loss rule, the account would have realised -$138k with a max decline of -2.66%. Under 2%, it would have realised -$276k with a max decline of -5.32%. Under 4%, it would have realised -$553k with a max decline of -10.64%. All three scenarios show a 50% win rate, and one episode was stopped early in each run. The simulator is gross of fees.
The contrast is stark: actual loss was -$3.66M with a deepest decline in this window of -96.17%. A mechanical 1% stop would have reduced the loss by 96.2%. This is not a statement about what should have been done; it is a statement about what the data shows: the account's losses were concentrated in a small number of unmanaged positions that were allowed to run to catastrophic levels.
Open positions
Five positions are currently open, all short, totalling -$1.78M in unrealised PnL. None have stops in place.
ZEC short is the dominant open loss: $420.343 entry, 10x leverage, -$1.60M unrealised. This position is underwater and has no stop.
xyz:XYZ100 short: $28,322.56 entry, 20x leverage, -$61k unrealised, no stop.
BTC short: $77,034.70 entry, 20x leverage, +$9.2k unrealised, no stop.
xyz:INTC short: $123.48 entry, 10x leverage, +$55k unrealised, no stop.
xyz:AMD short: $377.00 entry, 10x leverage, -$182k unrealised, no stop.
The absence of stops across all five positions is consistent with the closed-trade pattern: conviction-driven entries without mechanical exit discipline.
Honest summary
- Visible strength: The two commodity short trades (crude oil, Brent) were well-structured, held appropriate notional size, and closed profitably with ATR-based stops in place. This demonstrates the account can execute disciplined, profitable trades when the thesis is sound and position management is applied.
- Visible weakness: Long-side crypto exposure in mid-December was sized catastrophically (up to $2.5M notional per position), held without stops, and closed into losses that consumed 80% of the account's highest balance in this window balance. The 4.35% win rate on longs and 18-trade loss streak indicate the long thesis was not validated by price action and was not abandoned when evidence mounted.
- Data scope caveat: Only the most recent 10,000 fills are visible. Earlier trading history is not available, so this audit cannot assess whether the mid-December losses represent a