An approval step protects a trading plan when an AI-generated signal exceeds the position-size rule. The pause gives the trader a chance to reject risk that looks attractive in the moment but does not fit the limits set before the market opened.
At 8:42 a.m., the signal is already queued. The setup looks clean: entry, stop, and target are visible. The proposed share count is the problem.
The trader’s rule caps risk at 1% of the account on any single trade. With the planned stop distance, the queued size would risk more than that cap. The AI may have identified a plausible setup. It has not earned permission to override the rule.
The cursor moves to Reject.
That decision can feel unsatisfying. A fast market rewards action often enough to make a pause look expensive. But the rule exists for the trades that feel hardest to skip. A position-size limit has no value if it disappears whenever a signal seems unusually convincing.
A signal is evidence, not an instruction
In September 1983, Lieutenant Colonel Stanislav Petrov was on duty at the Serpukhov-15 early-warning facility near Moscow when the Soviet system reported a U.S. missile launch. It later reported more launches. Petrov did not treat the alert as sufficient confirmation of an attack. He judged that the report required further verification, and the alarm proved false.
The episode is documented in BBC News reporting on Petrov’s role in the Soviet early-warning system. Its relevance to trading is narrow but useful: a system can surface urgent-looking information, while the person responsible for the decision still has to evaluate whether that information clears the required threshold.
A trading signal has a much smaller consequence, but it carries the same structural risk. A model may detect a pattern. It cannot decide that your maximum risk per trade suddenly no longer matters.
The approval step turns that distinction into a practical action. Before an order executes, you can check the position size against the account risk limit, inspect the stop, and reject the trade if the numbers do not fit.
Position size is where discipline becomes measurable
“Keep risk small” is too vague to protect an account. A usable rule names a maximum loss and calculates size from the distance between entry and stop.
For example, if a trader’s per-trade limit is $35 and the planned loss from entry to stop is $0.70 per share, the maximum size is 50 shares. A queue suggesting 80 shares changes the possible loss to $56 before fees or slippage. The setup may still be valid. The size is not.
That is why an approval gate belongs inside the risk system. It creates a deliberate check between an analytical output and a real order.
A rejection does not mean the AI failed. It means the trader’s rule held. That distinction matters when reviewing performance later. If a trader records every rejected signal as a missed opportunity, they may start weakening controls to avoid the discomfort of missing a winner. A better journal records why the order was rejected: position size exceeded the cap, stop distance was too wide, daily loss limit was near, or the trade repeated an existing exposure.
The point is to make rule deviations visible before they become losses. What Happens When 40 Shares Risk $100 Against a $35 Limit? walks through that calculation in more detail.
The pause reveals what the rule is for
The easiest trade to approve is usually the one that already fits the plan. The harder case is a signal that looks compelling while breaking one number.
That is the moment to ask a short set of questions:
- What is the maximum dollar loss if the stop is hit?
- Does this size fit the rule set before the session began?
- Has today’s loss changed the amount I am willing to risk?
- Am I approving the setup, or reacting to the fear of missing it?
Those questions take seconds. Their value appears over a sequence of trades, especially during a drawdown. A single oversized loss can require a much larger gain to recover. Limits reduce the chance that one decision changes the account’s trajectory.
Petrov did not have the benefit of knowing the alert was false when he chose to seek confirmation. The trader also cannot know whether a rejected signal would have won. That uncertainty is exactly why a predefined size rule matters. It replaces a guess made under pressure with a limit chosen when the trader had more distance from the moment.
Review rejections as carefully as approvals
A rejected queue item can teach more than an executed trade because it shows where the system and the plan disagreed. Review those rejections at the end of the week.
If the same mismatch appears repeatedly, investigate it. The position-sizing inputs may need adjustment. The stop assumptions may be inconsistent. Or the rule may be clear while execution discipline is still catching up. None of those findings require a larger order tomorrow.
Keep the rejection reason beside the proposed size, allowed size, stop distance, and account-risk limit. Over time, that record shows whether approval decisions protected the plan or whether rules were bent during volatile periods.
The lesson from Petrov’s 1983 decision is not to distrust every system alert. It is to define what must be verified before acting. In trading, a position-size rule is one of those requirements. Leave the order unapproved when it fails.
Educational content, not financial advice.
Comments
No comments yet.