Bottom line up front
This account is -0.01% all-time, down $352.38 on a $6.85M balance, but the headline masks severe structural damage. The account suffered a 39.8% drawdown from peak to trough, and the damage is concentrated in a single instrument: XYZ100 short trades account for -$398.46 of realised PnL across 435 episodes at a 31.72% win rate. NVDA is the only edge—51.8 realised PnL on 39 episodes—but it is dwarfed by systematic losses elsewhere. The core problem is not volatility; it is behaviour. Five of the top losses and both post-mortem trades are flagged for averaging down, FOMO re-entry, revenge trading, and oversized positioning. The account is technically near break-even, but operationally it is a loss-making machine with a single profitable instrument being slowly buried by compulsive re-engagement with losing positions.
What the data shows
The account opened on 14 November 2025 and has been active for 139 days across 518 total episodes (514 closed, 4 open). The balance peaked at $5.74M on 25 March 2026 and troughed at $3.22M on 21 January 2026—a $2.53M swing in a single account. Realised PnL is -$362.18 after fees; net fee drag is -$11.33 on $1.11M gross volume, so execution costs are not the primary culprit.
The account's entire edge comes from NVDA: 51.8 realised PnL on a 41% win rate across 39 episodes. The largest single win is a short from 190.36 to 181.34 (14 November to 19 November), closed for $51.67 profit. This trade had no behavioural flags and ran for 101.54 hours—a patient, structural trade. Everything else is noise or loss.
XYZ100 is the graveyard. 435 episodes, 31.72% win rate, -$398.46 realised PnL. The account has taken five oversized losses in XYZ100 that are each 3–8× the median loss size: -$6.10, -$3.98, -$3.26, -$2.24, -$2.24. These are not variance; they are position-sizing failures. TSLA is also negative: -$5.71 realised PnL on 39 episodes at 33% win rate, with long trades underperforming (-$5.86 long vs. +$0.15 short).
Long-side trading is weaker than short-side: -$203.06 long PnL at 30.86% win rate versus -$149.32 short PnL at 34.11% win rate. The account is marginally better at shorting, but neither side is profitable.
Trade quality
Win rate is 32.49% across 514 closed trades. Profit factor is 0.25—for every dollar won, the account loses four dollars. Expectancy is -$0.69 per trade. Win/loss ratio is 0.52, meaning the average loss (-$1.35) is 1.9× the average win (+$0.70). These numbers describe an account that is losing money on nearly every dimension of execution quality.
The account paid $10.00 in gross fees and incurred -$11.33 net fee drag, implying a small rebate. Maker percentage is 94.3%, so the account is primarily a liquidity provider. The fee structure is not the problem.
Post-mortems
XYZ100 short, 18 November 2025, 24635.22 to 24710.0, -$10.38 (0.1 hours)
Opened at 24635.22, closed at 24710.0 for a 75-point loss. Notional was $3,463.82 at peak. This trade is flagged for averaging down, FOMO re-entry, oversized positioning, and revenge trading. The structural stop (ATR 14 1h) was 0.86% away—a tight stop relative to the 0.31% move against the position. The trade was closed within 6 minutes. This is a revenge trade opened after a loss in NVDA, sized aggressively, and exited immediately when it moved against the account. The oversizing relative to the account's median loss is the critical failure.
XYZ100 short, 17 November 2025, 25157.67 to 25240.0, -$9.69 (0.37 hours)
Opened at 25157.67, closed at 25240.0 for an 82-point loss. Notional was $2,998.95. Flagged for averaging down, FOMO re-entry, oversized positioning, and revenge trading. The structural stop was 0.3% away. This trade was held for 22 minutes and closed at a loss. It is a revenge trade opened after a loss in XYZ100, sized near the account's median position notional, and stopped out quickly. The pattern is identical to the 18 November trade: loss triggers re-engagement with the same instrument at oversized notional, followed by rapid exit at a loss.
What the risk simulator reveals
Under a 1% stop-loss rule applied historically, the account would have realised -$3.36M PnL with a -49.66% drawdown and 16 early stops. Under 2%, the loss balloons to -$6.71M with -98.07% drawdown. Under 4%, the loss reaches -$13.42M with -191.35% drawdown. These are not edge-case scenarios; they are the natural consequence of the account's current position-sizing and re-entry behaviour applied to historical fills. The simulator reveals that the account's actual -39.8% drawdown is only survivable because stops were not consistently applied. Tighter risk rules would have bankrupted the account multiple times over.
Open positions
The account has 4 open episodes but no dominant position. Current open positions are empty or immaterial to the analysis.
Honest summary
- Structural edge exists but is buried: NVDA shorts generated $51.67 in a single trade with clean entry, exit, and hold duration. The account can identify and execute a profitable setup. This edge is real but represents less than 15% of total episode count and is being systematically eroded by losses elsewhere.
- Averaging down and revenge trading are systematic: Five of the five largest losses are flagged for averaging down, oversized positioning, or revenge trading. The account does not lose money randomly; it loses money by re-engaging with losing positions at larger sizes after small losses. Both post-mortem trades exemplify this pattern within 22 minutes of each other.
- Position sizing is inverted: The account sizes largest into XYZ100, the instrument with the worst win rate (31.72%) and largest cumulative loss (-$398.46). It sizes smallest into NVDA, the only profitable instrument. This is the inverse of rational capital allocation.
- The drawdown was severe and persistent: The $2.53M swing from peak to trough over 63 days indicates the account was not managing risk dynamically. A 39.8% drawdown on a $6.85M account is a structural failure, not a market event.
- Sample size is adequate but behaviour is the constraint: 514 closed trades is sufficient to identify patterns. The patterns are clear: the account can trade profitably in isolation (NVDA) but cannot manage the emotional and positional discipline required to avoid revenge trading and oversizing after losses.