The 80 Cents That Turned $25 of Planned Risk Into $65

The due diligence you actually perform is the pause between receiving a signal and risking capital. You compare the signal with current price, liquidity, position size, portfolio exposure, invalidation conditions, and your own readiness to accept the loss.

In 1983, Stanislav Petrov was on duty at the Serpukhov-15 early-warning command center near Moscow when the Soviet satellite system reported an incoming American missile. It then indicated additional launches. The system had produced a clear warning, and Petrov had minutes to decide how to classify it.

A signal can be correct and still deserve rejection

Petrov did not treat the alert as proof. He considered the surrounding facts.

The reported attack involved only a handful of missiles, which did not match the large first strike he expected. Ground radar had not confirmed the launches. He reported a false alarm.

The warning was wrong. The satellite system had mistaken sunlight reflected from high-altitude clouds for missile launches. The BBC later documented Petrov’s decision and the uncertainty surrounding it.

A trading signal has the same structural limitation, with far smaller stakes. It converts selected inputs into an output. The calculation may be consistent with its rules while missing information outside those rules.

A bot might detect a breakout because price crossed a threshold and volume increased. Before approving the trade, you may notice that the spread has widened, the move already ran farther than expected, or two existing positions would respond to the same market shock. Those observations can change the decision without disproving the original signal.

That review is due diligence. You are testing whether the signal still belongs in the market and portfolio that exist now.

The checks traders often perform without naming them

Retail traders frequently make several judgments in a few seconds, then call the result a feeling. Naming each check makes the process easier to repeat and audit.

First, check whether the price is still close enough to the proposed entry. A setup calculated at $50.00 may carry different risk at $50.80. If the stop remains fixed, the extra 80 cents increases the amount at risk per share. If position size stays unchanged, planned risk quietly becomes actual risk.

This is the problem examined in The $50.80 Entry That Turned $25 of Planned Risk Into $65. The key lesson concerns changed inputs: approval should use the executable price, rather than the price that existed when the signal was created.

Next, check what would prove the setup wrong. “Price went down” gives you little guidance. A defined level, failed breakout, or invalidated market condition creates a decision boundary before money is at risk.

Then inspect the portfolio. A new position can appear modest by itself while increasing exposure to one asset class, sector, currency, or market narrative. Four nominally different crypto positions may behave like one concentrated bet during a broad selloff.

Finally, check execution conditions and timing. A thin order book, an upcoming market event, or a signal generated hours earlier may change the trade enough to require recalculation. The signal’s age matters because markets continue moving after analysis ends.

A bot’s due diligence stops at its inputs

Automated systems can apply explicit rules with speed and consistency. They can compare price with a moving average, calculate a position size from a stop distance, or reject a setup that exceeds a programmed risk limit.

Their blind spot begins where the specification ends.

If the system does not measure correlated exposure, it cannot flag correlated exposure. If it assumes the quoted entry remains available, it cannot recognize that slippage changed the risk. If the strategy defines success through returns while ignoring max drawdown, the backtest may reward a path the trader cannot afford to survive.

Human review also has weaknesses. Fatigue, impatience, recent losses, and fear of missing a move can turn approval into a reflex. The answer is a short, recorded review that makes intuition inspectable.

Before approving a queued trade, write down:

• The current executable price and recalculated amount at risk.
• The specific observation that invalidates the setup.
• The position’s effect on total and correlated exposure.
• Any relevant condition the signal does not measure.
• The reason for approval or rejection in one sentence.

This turns “something felt off” into evidence you can examine later. Rejected signals become part of the track record, especially when you record what happened after each rejection. When Do Rejected Trade Signals Reveal a Weakness in Your Decision Process? explains how those decisions can expose inconsistent thresholds and missing rules.

Make the approval gate earn its place

An approval gate only works when approval requires fresh judgment. Clicking approve because the system produced a confident score leaves the machine’s assumptions unchallenged.

Petrov’s contribution in 1983 came from comparing the warning with facts the warning did not settle. Your review follows the same mechanism: treat the output as evidence, look for missing context, and decide whether the proposed action still fits your risk boundaries.

For the next 20 queued signals, record the five checks above before acting. Then compare approvals, rejections, planned risk, actual risk, and later outcomes. The useful result is a visible decision process that can be corrected.

Educational content, not financial advice.