Three plausible signals can create one concentrated bet when the assets rely on the same market move. Review correlation, shared catalysts, and total loss at the stop before approving any of them.
At 10:07 p.m., Jonah is at his kitchen table in Chicago, one cold mug beside his laptop, looking at three queued orders. One is for a large crypto asset. One is for a crypto-linked equity. One is for a technology stock that has been moving with the same risk-on mood.
Each chart has a clean case. The first shows a breakout above a recent range. The second has regained a moving average. The third has pulled back to an area Jonah had marked earlier. Taken one at a time, each order looks measured.
Then he adds the numbers.
If all three stops are hit, the combined loss would exceed the risk limit he set for a single market view. If crypto weakens overnight, the crypto asset and the linked equity could fall together. If broader appetite for risk reverses, the technology stock may join them. Three order tickets have become one thesis expressed three times.
That is the risk hiding behind a stack of individually reasonable signals.
Separate charts can share the same risk
Diversification comes from owning exposures that can behave differently when conditions change. It does not come from counting tickers.
A trader can hold three positions in different symbols and still depend on one idea: rates will fall, crypto sentiment will stay strong, large-cap technology will keep rising, or volatility will remain low. When the idea breaks, the positions may move together precisely when diversification was supposed to help.
Charts can make this easy to miss. Each chart has its own entry, stop, trend line, and signal. The screen creates separation. The portfolio may not have any.
Before approval, ask what needs to remain true for each trade to work. If the answer repeats across the orders, treat them as related exposure. A breakout in a crypto asset and a long position in a crypto-linked stock may have different charts, but both can be vulnerable to the same headline or market shift.
This does not mean correlated positions are automatically wrong. It means their combined downside belongs in the decision.
Add risk before you add positions
Position sizing starts with the distance between entry and stop, then turns that distance into a defined dollar risk. The next step is often skipped: add the risks for positions that share a thesis.
Jonah had planned to risk a fixed amount on one trade. His three queued orders each fit that amount when viewed alone. Together, their maximum planned loss at the stops was several times larger than the limit he meant to place behind one market idea.
He had two choices. He could reject two orders and keep the strongest setup. Or he could reduce the size of all three so their combined loss fit his risk budget. The charts did not decide that for him. His written limit did.
A simple review table can expose concentration:
- List each pending order and its maximum loss at the planned stop.
- Write the market condition or catalyst each trade depends on.
- Group orders that depend on the same condition.
- Compare the total risk in each group with the amount you allow for one thesis.
The numbers will rarely be perfect. Correlation changes, especially during sharp moves. The purpose is to make a hidden dependency visible before a real order is placed.
An approval gate creates a pause for portfolio context
An AI can identify setups and queue trade signals. That does not make the AI responsible for deciding how much related risk belongs in your account.
The approval step is where a trader can move from chart-level reasoning to portfolio-level reasoning. Look beyond whether an entry appears valid. Look at what else is open, what is already queued, where the stops sit, and what shared move could pressure the positions at once.
That pause matters most when the signals arrive close together. A sequence of alerts can feel like confirmation. It can also be repetition.
Responsible AI in financial markets has drawn attention to the need for human judgment and careful controls. In trading, an approval gate provides a practical control: no queued signal becomes an order until a person reviews it in context.
For a deeper look at why historical results cannot answer every live execution question, see [Backtest confidence versus approval-gated execution]( /blog/backtest-confidence-versus-approval-gated-execution-what-historical-testing-can-estimate-what-it-cannot-know-and-why-a-human-decision-remains-necessary-before-a-real-order-433ccdae/).
The order Jonah rejected changed the next morning
Jonah rejected the crypto-linked equity order. He reduced the size of the remaining two trades until the combined planned loss matched his limit for a single thesis.
The next morning, the market opened lower. Both remaining positions moved against him. His stops defined the loss he had accepted. The rejected order never added a third loss to the same move.
The lesson was not that every related trade should be rejected. It was that a portfolio cannot be reviewed one chart at a time when the same market condition connects the charts.
Before approving the next signal, name the thesis in one sentence. Then calculate what the entire thesis can cost if it fails.
Educational content, not financial advice.
Comments
No comments yet.