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: $184.04 realised PnL after fees on $6.87M in closed trade volume, with a highest balance in this window of $9.16M and a lowest balance in this window of $6.93M. The deepest decline in this window was -20.21%. The headline masks a structural problem: long trades work (59.89% win rate, +$212.35), but short trades consistently fail (54.87% win rate, -$28.31), and the account has been bleeding capital into oversized, revenge-driven shorts since mid-May. Five open positions now sit on the book, none dominant, but all unhedged and without stops.
What the data shows
This account opened in mid-May 2026 and has run 434 total episodes (377 closed, 57 open) across a 52-day visible window. The arc is clear: early May saw profitable long-biased trading in SILVER, SNDK, CL, and DRAM, with win rates above 65% on those instruments. By late May, the account pivoted hard into short-side revenge trades and oversized positions, triggering a cascade of losses that eroded the early edge. The deepest decline in this window occurred on 9 June, when the account fell from $9.16M to $6.93M—a $2.23M swing driven entirely by short-side blowups in INTC, SILVER, MU, and CL.
Realised PnL sits at $663.25 gross, but fees consumed $166.15 net, leaving $184.04 after execution costs. The fee-to-PnL ratio of 30.29% is material; the account is paying roughly one-third of its gross profit in fees. Maker participation is 51.26%, suggesting mixed execution quality. Long trades generated $212.35 of the $663.25 gross profit; short trades lost $28.31. The asymmetry is stark: longs averaged $4.00 per win with a 59.89% win rate; shorts averaged $4.22 per loss with a 54.87% win rate. This is not noise—the account has a directional edge that it has systematically traded against.
The by-instrument breakdown confirms the pattern. SILVER (19 episodes, 68.42% win rate, +$103.88), SNDK (16 episodes, 56.25% win rate, +$68.04), CL (23 episodes, 69.57% win rate, +$64.67), DRAM (13 episodes, 69.23% win rate, +$32.98), and XYZ100 (17 episodes, 64.71% win rate, +$28.43) are all profitable. MU, INTC, and NVDA are all loss-making, with win rates below 60% and negative realised PnL. The account has no edge on those names and should have stopped trading them after the first 5–10 episodes.
Trade quality
Win rate of 57.29% with a profit factor of 1.27 and expectancy of $0.49 per trade. This is marginally profitable before fees, barely profitable after. The win/loss ratio of 0.95 means winners are slightly smaller than losers in absolute terms, which is a structural weakness: the account is winning more often but making less per win. Average win is $4.00; average loss is $4.22. That 0.55% edge per trade is fragile and entirely consumed by execution costs on any trade with slippage.
The max win streak is 8; the max loss streak is 5. Neither is exceptional. The account has not demonstrated the kind of consistency that would justify the leverage and position sizing it is using.
Post-mortems
INTC short, 19 May, entry 111.63, exit 112.07, -$95.18 loss. Opened and closed on the same day in 9 minutes. Flagged for averaging down and oversized loser (5.82x median loss). Position notional reached $22,060.83 on a $7.88M account—0.28% of capital on a single short that moved 0.38% against entry. The structural stop (ATR 14 1h) was 2.78% away; the account blew past it. This was a revenge trade following a loss in BRENTOIL. The account added into a failing short instead of exiting.
SILVER short, 19 May, entry 76.30, exit 76.35, -$37.17 loss. Opened and closed in 4.8 minutes. Flagged for FOMO re-entry and oversized loser (5.55x median loss). Position notional reached $25,536.89. This was a re-entry after a previous SILVER close at 76.30 on the same day. The account closed a trade, watched the price move 0.05%, and re-entered the same direction into a position that was immediately underwater. The structural stop was 1.35% away; the account ignored it.
Both trades share a common signature: they were opened after losses, sized aggressively relative to account equity, held for minutes, and exited at a loss without respecting the structural stop. Neither trade had a stop in place. Both were short-side. Both occurred on 19 May, the day the account's balance fell from $7.4M to $6.93M.
What the risk simulator reveals
Under a 1% fixed stop rule, the account would have generated $148,175.86 in realised PnL with a max decline of -4.53%. Under 2%, $296,351.71 with -8.93%. Under 4%, $592,703.43 with -17.33%. These are gross of fees. The simulator stopped 15 episodes early across all three scenarios, indicating that hard stops would have prevented some of the worst trades from running to full loss. The win rate remained stable at 54.59% across all three rules, suggesting that stops do not materially reduce the frequency of winning trades, only their magnitude when they fail.
The contrast with actual performance is instructive: the account ran 377 closed trades with no mechanical stops, generated $184.04 net PnL, and experienced a -20.21% decline. A 1% rule would have produced 800x better PnL and a quarter of the decline. This is not a prediction; it is a historical counterfactual. The account's refusal to use stops has cost it hundreds of thousands of dollars in the data covered.
Open positions
Five positions are currently open, totalling $29,068.05 notional with -$1.56 unrealised PnL:
- SPCX short, 72.31 size, entry 146.1039, mark 146.09, 3x cross, liquidation 3267.74. Held since 10 July (0 days visible). Unrealised PnL +$1.01 (0.03% ROE). No stop in place. The liquidation price is so far away it is immaterial.
- GOLD long, 0.8323 size, entry 4120.42, mark 4121.0, 3x cross. Unrealised PnL +$0.47 (0.04% ROE). No stop. No liquidation price.
- HOOD long, 0.476 size, entry 111.255, mark 111.26, 3x isolated, liquidation 78.077. Held since 10 July (0 days visible). Unrealised PnL flat