Every proposed order should pass a review of position size, maximum acceptable loss, and exit conditions before any capital moves. An AI-generated signal can be useful research, but the approval decision belongs to the trader.
At 9:31 a.m., Eli is standing at his kitchen counter in Queens, coffee cooling beside a scratched notebook. The opening minute has already pushed a stock higher, and an AI-generated buy order is waiting in his queue. It shows an entry, a stop, and a target. His finger is one tap from approval.
Then he sees the planned loss. The stop sits far enough below the entry that the proposed share count would put more of his account at risk than his rule allows. If he approves it because the chart is moving, the bad ending is simple: one ordinary reversal could consume the risk he meant to spread across several trades.
The order stays queued.
That pause is the point of an approval gate. A proposed order has no special authority because software produced it, because the market just opened, or because a move looks convincing on a one-minute chart.
Review the loss before you review the opportunity
The first question is not, “How far could this run?” It is, “What is the maximum amount I am willing to lose if this idea is wrong?”
A trade plan needs an entry, an invalidation point, and a position size that fits the distance between them. If an entry is $50 and the exit condition is $48, each share carries $2 of planned risk. A trader willing to risk $40 on that setup could size around 20 shares before considering fees, slippage, or other positions already open. The numbers are illustrations, not a recommendation.
This is why position size comes after the exit condition. Choosing the number of shares first invites the trader to stretch the stop until the trade “fits.” The price then becomes a defense of the position rather than evidence that the original idea still holds.
Eli writes the proposed entry and stop in his notebook. The math says the setup requires a smaller position than the queued order suggests. He reduces the size, then notices the potential reward no longer justifies the distance to the stop. The trade may still rise. That does not make the plan acceptable.
Schwab’s trade-planning guidance centers the same sequence: position size, maximum acceptable loss, exit conditions, and the relationship between entry and exit. The order review is where that sequence becomes real.
Define the exit condition in plain language
A stop price by itself can hide an unfinished plan. Review what would make the original thesis invalid, then check whether the exit level reflects that condition.
“Exit if it drops” is vague. “Exit if price closes below the prior session’s low” gives the trader something observable to test. “Take profit when it feels extended” creates room for impulse. A defined target or trailing rule creates a decision rule before the position becomes emotionally loaded.
The approval screen should also surface what the plan does not say. Is there a cancellation condition if the price runs beyond the intended entry before approval? Does the trade open before an event that changes the risk? Is the target based on a chart level, or is it simply large enough to make the trade look attractive?
A missing cancellation condition can turn a reasonable plan into a chase. The cancellation condition this trade plan didn’t have, and what it cost examines that gap in more detail.
Approval is a decision point, not a rubber stamp
Automation can make analysis arrive faster. It can also make a weak plan feel more official than it is.
An approval gate creates a small piece of friction at the moment when speed feels most persuasive. That friction gives the trader room to compare the proposed order with account-level rules: total open risk, a daily loss limit, concentration in one asset, and the number of trades already taken.
The review matters most after a loss, when the urge to recover can disguise itself as conviction. It also matters after a win, when larger size can feel earned even though the next trade has its own entry, stop, and uncertainty. Yesterday’s result does not change today’s maximum acceptable loss.
Eli checks his existing positions and sees that another open trade would react to the same broad market move. Approving both would create more combined risk than either order shows alone. He rejects the queued order and keeps the smaller, separate position under review for later.
The rejection does not prove the original signal was bad. It protects a rule that matters even when a signal later would have worked.
Build a record of decisions, including the rejected ones
A trading journal is more useful when it records the proposal, the review, and the reason for the final decision. Over time, that record can show whether losses came from poor entries, oversized positions, moved stops, overlapping exposure, or trades taken without a defined exit.
Rejected orders belong in that record. They reveal the conditions that repeatedly pressure discipline: the first minutes after the open, a sharp overnight move, a third trade after two losses, or a chart that seems too active to miss.
At 10:07 a.m., Eli closes the notebook and leaves the order unapproved. The price has moved again, but his maximum loss, exit condition, and total exposure never met his review standard. His account remains available for the next setup, which is the outcome the rule was built to protect.
Educational content, not financial advice.
Comments
No comments yet.