This case study is useful precisely because it is not heroic. The account was an FTMO Challenge with a 10% profit target, a daily maximum drawdown of -5%, and an overall maximum drawdown of -10%. It ended with 743 execution records, all in XAUUSD, and a realized closed-trade P/L of -885.33 USD. That is the starting point, and it matters more than any story we might want to tell around it.

What the records show is a familiar but unforgiving pattern: a win rate of 60.57% did not rescue the account because the average loss of -5.57 USD was much larger than the average win of 1.66 USD. The expectancy per trade was -1.19 USD, and the profit factor was 0.458. In plain language, the account was winning often enough to create confidence, but losing in a way that was mathematically harder to recover from.

Observation: the problem was not low hit rate

The overall win rate was above 60%, and September in particular showed a 77.43% win rate across 226 trade episodes. On the surface, that looks impressive. But sophistication in trading begins where the surface ends. The account lost money because average losses overwhelmed average wins, and the top five losing trades accounted for 39.06% of total losses.

The monthly path also matters. July lost -37.69 USD, August lost -238.00 USD, and September lost -609.64 USD. The progression suggests that the account did not simply suffer random noise. It became more exposed to the same structural problem over time: repeated small gains, then larger adverse moves that were not contained early enough.

Explanation: high win rate can hide negative expectancy

The approved interpretation here is important: a high win rate can be consistent with taking profits quickly while allowing some losses to become much larger. The records support that possibility, but they do not prove intent. What is verified is the statistical shape: average win 1.66 versus average loss -5.57, with a payoff ratio of 0.298. That combination is not sustainable unless the strategy has a powerful edge elsewhere, which this record set did not show.

The operator’s own reflection helps frame the process. In August, a DCA bot was used to test whether daily target returns could be beat consistently, and the market began taking money after streaks of small gains. In September, parameters were adjusted, but long-run gain was still not guaranteed, and the lower spread environment could not offset the structural weakness. The point is not to judge the intention; the point is to observe that the framework relied on a mechanism that did not actively cap risk.

Implication: concentration and stacking made the account fragile

All 743 records were in XAUUSD. Concentration can be deliberate and sometimes rational if one is truly specialized. But concentration also means the account lives and dies with a single instrument regime. In this case, the specialization was paired with 99 same-direction overlapping entries and 68 loss-following size escalations, which made the account less adaptive when the trade went against it.

That is the deeper lesson for traders and investors alike: a good idea can fail when the control system is weaker than the idea. The records show comparable loss transitions in 292 cases, which suggests repeated interaction with adverse conditions rather than one isolated mistake. For a funded challenge, that is especially dangerous because the rules punish drawdown more quickly than they reward being temporarily right.

• Edge must survive spread, slippage, and adverse regime changes.

• Position sizing must be independently controlled, not left to the entry logic alone.

• Average loss must be designed, not discovered after the fact.

• A strategy that cannot stop stacking risk is not a complete risk system.

Risk framework: what would be required now

The operator concluded that the challenge failed because the bot was not programmed to prevent a max daily drawdown breach, and that DCA is no longer used because risk size cannot be controlled actively and profit pursuit is penalized by large drawdowns. That is a practical conclusion, and it is the right one. In a professional context, risk management is not a complement to the strategy; it is part of the strategy.

A more robust framework would ask four questions before any trade: What is the maximum acceptable loss for the day, the symbol, and the sequence? Does the entry logic survive after costs? Can additional exposure be added without increasing fragility? And if the market moves against the position, what exact rule prevents a small mistake from becoming a challenge-ending event?

Closing thoughts

This account did not fail because it traded a single instrument, or because it had a high win rate, or because it tried to adapt. It failed because the loss side was not structurally contained. The evidence shows a system that could often be right in small increments and still be wrong in aggregate. That is exactly the kind of failure sophisticated investors should study, because it mirrors a broader truth in capital allocation: returns are not judged by accuracy alone, but by how the process behaves when it is wrong.

If there is one practical lesson here, it is that survival comes from designing the downside first. A trading account that cannot enforce a hard boundary on risk is not ready for compounded growth, regardless of how persuasive its short-term streaks appear.