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 exceptional: +0.2% net PnL ($11,469.70) on a $5.78M balance after fees, with a 57.5% win rate and a 2.04 profit factor. The headline obscures severe fragility: a deepest decline in this window of -49.98%, driven by catastrophic oversized losses in BTC and ETH that dwarf the edge built elsewhere. Short-side trades work (58.9% win rate, +$14,781 realised); long-side trades bleed (56% win rate, -$3,311 realised). The account survives on a handful of high-conviction shorts—notably LIT short for +$8,171—while revenge trades, FOMO re-entries, and averaging down into micro-cap volatility (FARTCOIN, 119 episodes, -$11.36 realised) consume capital at scale.
What the data shows
The account opened on 2 November 2025 and has executed 339 closed episodes across eight instruments in the data covered. The arc is sharp: early aggression into oversized positions on 2–3 November triggered a cascade of losses that carved the account from an estimated starting balance of $5.78M down to $577K (the lowest balance in this window), then recovered to $5.79M by mid-May 2026. This is not a smooth equity curve; it is a near-extinction event followed by a slow rebuild.
Money is made almost entirely on the short side. BTC shorts contribute +$3,803 realised PnL across 20 episodes (70% win rate); ETH shorts contribute +$794 across 40 episodes (52.5% win rate); SOL shorts contribute +$1,358 across 29 episodes (58.6% win rate). LIT shorts, despite only two episodes, delivered +$8,171 realised PnL at 100% win rate. The long side is a net drain: BTC longs cost -$2,585, ETH longs cost -$392. XRP is a black hole: 100 episodes, -$115 realised PnL, 55% win rate, no edge.
Fees are a secondary but material tax. Gross fees paid total $121.65, but net fee drag is -$865.15, indicating a net rebate of $743.50 (98.2% maker ratio). The account is highly efficient on execution but the fee drag still represents 7.5% of realised PnL. Realised PnL before fees is $10,643.49; after fees, $11,469.70. The account is profitable, but only narrowly.
The long/short split reveals the core problem: shorts generate +$14,781 realised PnL at 58.9% win rate; longs generate -$3,311 at 56% win rate. The account has built a short-side edge but has not learned to avoid long-side traps. This asymmetry is not noise—it is structural.
Trade quality
Win rate of 57.5% with a profit factor of 2.04 means the account wins more often than it loses and wins bigger than it loses. Average win is $117.40; average loss is -$77.97. Win/loss ratio of 1.51 confirms this: for every dollar lost on a losing trade, the account makes $1.51 on a winning trade. Expectancy of $34.34 per closed episode is modest but positive. Over 334 closed episodes, this compounds to the observed realised PnL.
The profit factor of 2.04 is respectable but not exceptional. It means gross profit is 2.04 times gross loss. For a 57.5% win-rate account, this is consistent with the observed average win/loss spread. The account is not extracting outsized edge per trade; it is surviving on volume and consistency.
Post-mortems
XRP long, 2 November 2025, 0.62 hours: Opened at $2.51, closed at $2.50, loss of -$815.91 on a $163K notional position. This trade was flagged as a revenge trade (following a small FARTCOIN loss), an averaging-down episode, and an oversized loser (124x the median loss). The structural stop distance was 1.13%, meaning the ATR-14 1H stop would have been at $2.48. The account held through the stop, took the loss, and immediately re-entered other positions. This is the first domino in the 2–3 November collapse.
BTC long, 3 November 2025, 1.27 hours: Opened at $110,318, closed at $109,756.05, loss of -$2,808.91 on a $657K notional position. Flagged as a revenge trade (following XRP losses) and an oversized loser (46.7x the median loss). The structural stop was 0.71% away at $109,483. The account held through, took the full loss, and continued trading. This single trade consumed 24.5% of the realised PnL for the entire data covered.
Both post-mortems share a pattern: oversized notional exposure (>$150K), tight structural stops (0.7–1.1%), and immediate re-entry after losses. The account is not managing position sizing relative to volatility; it is using leverage to amplify micro-cap and major-pair volatility into account-threatening losses.
What the risk simulator reveals
Under a 1% hard stop rule, the account would have realised -$10,897.31 PnL with a max decline of -1.02%, stopping out 8 episodes early. Under a 2% rule, -$21,794.63 and -2.03% max decline. Under a 4% rule, -$43,589.26 and -4.05% max decline. All three scenarios show the account turning negative. The simulator is gross of fees, so actual net results would be slightly worse.
This is the critical insight: the account's profitability in the data covered depends entirely on the absence of hard stops. The two largest losses (BTC -$2,809, ETH -$1,703) would have been capped at 1–2% of notional under a disciplined stop regime, but instead they ran to full loss. The account is profitable because it survived; a 1% stop rule would have bankrupted it.
Open positions
One open position: HYPE long, 3x leverage, entry price $40.876, unrealised PnL +$1,514.20. No stop in place. The position is still open and the outcome is unknown until it is closed.
Honest summary
- Short-side edge is real: 58.9% win rate on shorts, +$14,781 realised PnL. The account has identified a genuine edge in short-side execution, particularly in BTC, ETH, and SOL. This is the only reason the account is profitable.
- Oversized losses on longs are catastrophic: The five largest losses total -$6,151 and are all long positions or short positions that were averaged down. The account lacks a coherent long-side strategy and instead uses long entries as revenge trades after short-side losses. Position sizing on longs is 2–4x larger than on shorts, amplifying the damage.
- Revenge trading and FOMO re-entry are endemic: Five FARTCOIN re-entries, five revenge trades flagged in the data covered, and averaging down across BTC, ETH, HYPE, and FARTCOIN. These are not isolated lapses; they are habitual responses to