The opening paragraph
Only the most recent public fills are visible in this data covered.
You're short 479.25 shares of MU at $1,159.80, currently worth $469,444.54, sitting on an unrealised gain of $86,390.31—a 155.42% ROE on 10x leverage. That's still open exposure. The unrealised PnL is not locked in. The liquidation price at $4,950.73 is a theoretical downside boundary, but the outcome is unknown until the position closes. Meanwhile, the rest of the book is a carnival of underwater positions: APT long is down $95,608 on a -798% ROE, kPEPE long is bleeding $6,210, and SOL long is -$825.76 on -156% ROE. The MU short is carrying the entire account on its back while the open portfolio hemorrhages $270,398 in unrealised losses across 25 positions.
The greatest hits
- MU short dominates the risk profile: The $469k position accounts for 44.6% of total open notional value. If it closes at current mark, you lock in $86k. If it doesn't, you're exposed to a 4,950-point liquidation boundary on a 10x margin cross account. The position is still open.
- APT long is a $95,608 crater: Entered at $3.1159, now marking $0.6292. The funding bleed is $18,070 data-covered, with $6,744 paid since open. This is not a position; it's a slow-motion liquidation watch. No stop in place.
- kPEPE long entered at $0.004, now $0.0028: $6,210 down on a 10x leverage micro-cap. Funding paid $292.83. The size is 5.4m tokens. This reads like a revenge entry after a loss, and the evidence pack flags exactly that pattern across five separate re-entries.
- The closed trades that got you here: VVV long closed for $59,541 (7.99 hours), ORCL for $6,535, MSTR for $3,003. Three clean wins. Then the day turned into a revenge-trade demolition: CL long lost $2,659 (flagged as averaging-down, oversized, revenge), SNDK long lost $2,126 (same flags), MU long lost $3,934 on a 3-minute hold. You closed 69 episodes in the data covered and finished +$61,441 net, but you're carrying $270k in unrealised losses to get there.
- Five re-entries, four of them red: The evidence pack flags FOMO re-entries on CL, XYZ100, MU, CRCL, and XYZ100 again. CRCL cost you $1,603 on the re-entry alone. MU re-entry cost $178. Only XYZ100's second re-entry printed $53. The pattern is: close a loss, wait 4-45 minutes, re-enter the same coin at a worse price, hope for mean reversion, get punched again.
The pattern
You have a profitable closed record—53.62% win rate, 3.63 profit factor, $890 expectancy per trade—but you're funding it by holding oversized, under-stopped positions that move against you. The closed wins (VVV, ORCL, TSLA, NVDA, SKHX) are real edges. The closed losses (CL, SNDK, MU, CRCL) are all flagged as revenge trades or averaging-down episodes after small losses. You then re-enter the same coins within minutes at worse prices. The MU short is the only open position that's working; everything else is a margin call waiting for a wick.
The reluctant compliments
- The closed PnL is legitimate: $59,786 realised after $545 in fees on $23.2m in closed trade volume. That's 0.26% net edge on closed trades. The win-loss ratio of 3.14 and profit factor of 3.63 are real.
- Maker rebates are helping: 82.29% maker fills means you're getting paid the spread on most entries. That's discipline in execution, even if the position sizing and re-entry logic aren't.
- VVV and ORCL closed clean: No averaging, no revenge flags, just entry, hold, exit. That's the template that works.
The verdict
You're profitable on closed trades but insolvent on open ones. The MU short is a $86k unrealised win that's masking a $270k unrealised loss portfolio. Close the APT, kPEPE, and SOL positions. They're not waiting for recovery; they're waiting for liquidation.