Human-reviewed decision replay sample
A static sample of how RiskDrive AI reviews closed trade records to separate outcome from process: clean wins, lucky wins, good losses, bad losses, and the critical moments that changed risk. This page is not investment advice and does not provide buy/sell recommendations.
Use anonymized closed trades. No broker login required. Human review happens before any reply is shared.
Intake page: /intake.html
RiskDrive does not claim one true cause for profit or loss. It separates realized outcome from process evidence, then shows sample size, confidence, and limitations next to each claim.
First-review relationship: 3 closed trades create the initial manual review, this 25-row sample is a public format example, and the replay lab is an interactive local preview with synthetic data or record-only imports.
Input quality note: 19 of 25 rows included stop status, 17 included a pre-trade rule note, and 25 included realized P/L. Missing fields lower confidence.
Confirm the report is based on 25 closed historical rows and a medium confidence label.
Each pattern should point back to submitted rows such as D-014, D-018, or D-021.
Missing stop status, weak timestamps, or absent pre-trade notes should reduce claim strength.
The report stays historical and educational; it does not turn a pattern into trade direction.
RiskDrive separates the final outcome from the quality of the process. This sample focuses on closed loss rows, but the review model is designed to label both winning and losing outcomes when the records support it.
Result was positive and the process stayed inside the trader's documented rules.
Result was positive, but the process drifted: size, stop, timing, or notes indicate avoidable risk.
Result was negative, but the record shows controlled risk and documented process.
Result was negative and the process broke: missing rules, expanded risk, or repeated friction cues.
Reviews closed historical records, identifies repeated risk patterns, checks user-defined rules, and gives educational notes for human review.
It does not provide buy/sell recommendations, instrument recommendations, account management, order execution, order placement, order modification, order cancellation, or profit guarantees.
RiskDrive does not provide buy/sell recommendations, manage funds, or place/modify/cancel orders.
These findings are based only on the 25 historical sample rows. Counts can overlap because one trade may trigger more than one replay cue.
Position size was above the trader's own stated rule on 16 rows. The review marks this as process drift, not as a prediction about the next trade.
Confidence: High Evidence: size field + rule note Reviewer check: rule wordingNine historical rows show missing, moved, or unclear stop status. This is reported as a record-quality and rule-following issue, not a future trade instruction.
Confidence: Medium Evidence: stop status + notes Limitation: 6 rows incompleteSeven rows happened within 24 hours of a prior loss or journaled frustration event. RiskDrive marks this as a replay cue for review, not a diagnosis.
Confidence: Medium Evidence: timestamp + trader note Limitation: notes may be incompleteCritical moments are historical review points. They do not say what a trader should buy or sell; they show where the submitted record suggests risk expanded, reduced, or became unclear.
| Moment | Historical action | Review label | Decision cost note |
|---|---|---|---|
| Entry | Opened above documented size rule | Risk Drift | Initial exposure exceeded the user's stated cap. |
| Volatility expansion | Held full size while the stop note was missing | Watch Zone | Evidence is too thin to score strongly; missing notes lower confidence. |
| Stop move | Stop status moved or became unclear | Process Break | The review flags expanded downside exposure relative to the submitted rule. |
| Final exit | Closed after the repeated cue appeared | Late Save | Exit ended additional drift, but after the largest process break. |
The estimated historical risk added by a decision compared with the trader's own rule or a more conservative alternative.
A review score for rule clarity, risk control, and evidence quality. It is not a prediction of future returns.
A point in the closed trade where the submitted record shows risk changed or the evidence became too thin.
Ask whether the trader has a written size rule before discussing any pattern.
Confirm that every claim is tied to a submitted row, screenshot, or user-written note.
Keep the reply educational and historical. Do not translate a pattern into a trade action.
The table shows representative rows; full replay counts come from the 25-row sample CSV.
| Trade | Historical row | Observed replay cue | Replay review prompt | Confidence |
|---|---|---|---|---|
| D-014 | FX-A, closed loss, size above user rule | Size drift + missing pre-trade checklist | Verify whether the size rule existed before this trade. | High |
| D-018 | Index-B, closed loss, stop status unclear | Stop discipline gap | Ask for the original plan screenshot before making the note stronger. | Medium |
| D-021 | FX-C, closed loss within same-day cluster | Loss clustering after frustration note | Check whether the journal note came before or after the order. | Medium |
Send three recent closed trades through the intake page. The review stays historical and human-checked before any response is shared.