Marcus’s oversized position. One trade created most of the week’s drawdown.

A weekly drawdown can come from one oversized position, even when the rest of the week’s losses stayed within a manageable range. A Friday review should separate ordinary losing trades from the single position whose size changed the week’s risk profile.

The Friday review that exposed the real loss

At 4:47 p.m. on Friday, Marcus sat at his kitchen table in Manchester with six losing trades open in his journal and a cold mug of coffee beside his laptop. The account was down for the week. His first reaction was familiar: maybe the strategy had stopped working.

Then he sorted the trades by risk at entry.

Six losses were small and broadly consistent with his plan. Each had a defined stop, a position size chosen before entry, and a loss that stayed within the range he had accepted. The seventh trade was different. Marcus had increased the position after a strong signal and moved the stop farther away when price went against him.

That one position created most of the week’s drawdown.

The uncomfortable question was still open: if he treated the trade as evidence that the whole strategy had failed, would he abandon a process that had mostly followed its rules? Or would he admit that one decision had changed the outcome?

The review gave him a narrower answer. The problem was not seven identical failures. It was one oversized position combined with a stop adjustment.

Six manageable losses can hide one oversized decision

Losses feel similar when they appear in a weekly total. A red number at the bottom of the account does not show whether the damage came from repeated small mistakes or one trade that carried too much exposure.

That distinction matters because the response should differ.

If six trades each lost a planned amount, the review may point toward normal variance, weak entries, or a strategy that needs more testing. If one trade represented several times the intended risk, the first correction belongs to position sizing and rule enforcement.

A commonly used trading rule limits risk on one trade to 2% of account equity. The threshold is an adjustable convention, not a universal law. Your own limit may be lower, depending on your account, strategy, liquidity, and tolerance for drawdown. The useful part is the discipline of setting the limit before the trade and measuring the actual risk against it.

Marcus’s review showed that the position had exceeded his normal risk before he moved the stop. The stop move did not create the original sizing error, but it extended the trade’s ability to damage the account.

This is why a weekly review should record more than entry, exit, and profit or loss. Add planned risk, actual risk, position size, stop distance, and any rule changed after entry. Those fields turn a vague feeling of “a bad week” into a sequence you can inspect.

The approval gate belongs before the order

For traders using AI assistance, the same review can include the point where a signal became an order.

An AI system can generate a trade idea, calculate a proposed size, and place the order automatically. That removes a pause at exactly the point where a human may need to inspect the risk. An approval-gated assistant keeps the trade queued until a person accepts or rejects it. The control is simple: analysis can inform the decision, while execution waits for approval.

Before approving a queued order, check four numbers:

• The account equity used for the calculation.
• The distance between entry and stop.
• The dollar amount at risk if the stop is hit.
• The total exposure if the trade overlaps with existing positions.

A signal can look convincing and still deserve rejection because the size is wrong. A strong setup does not make an oversized position safer.

For a deeper look at this boundary, Who Has Permission to Turn an AI Signal Into a Live Order? examines why approval authority matters. The same principle applies during a Friday review: identify who, or what, had permission to change the risk.

Turn the weekly loss into a rule you can test

Marcus did not rewrite his entire strategy after the review. He marked the oversized trade separately, recorded when the stop moved, and added a pre-approval check for maximum position risk.

The following week, that check took less than a minute. It did not predict the next trade. It did something more practical: it made the risk visible before commitment.

You can use the same review process without copying Marcus’s limits. For every losing trade, compare planned risk with actual risk. Group the results into three categories: within plan, outside plan because of execution, and outside plan because of a rule change after entry.

The third category deserves attention first. It shows where a loss became larger than the system intended.

A trading journal should make that moment hard to miss. Record the original stop, the revised stop, the position size, and the reason for any change. If the reason is simply “conviction increased,” treat that as information about decision-making, not proof that the trade deserved more capital.

By Friday evening, Marcus’s account was still down. The number had not changed. His understanding of the week had.

Six losses were data about the strategy. One oversized position was data about control. Those are different problems, and they require different next steps.

Educational content, not financial advice.