
Scenario Overview
This log documents a replayed crash scenario: synthetic technology instruments gap lower across multiple sessions while the replay engine introduces modeled execution latency, forced liquidations, and sentiment reversal scripts.
Crash replays are the most seductive material a simulator can offer — they let the participant feel like they 'lived through it' at zero cost. This log exists to expose that seduction: every 'I would have sold here' written during a replay is written by someone who knows it is a replay.
The exercise therefore shifts the graded unit from reaction to documentation: the journal is scored on whether it captured its own decision-making in real time, not on how the synthetic P&L ended.
Crash Replay Configuration
| Replay Element | Configuration | Review Relevance |
|---|---|---|
| Gap script | 3 sessions of −4%, −7%, −2% synthetic index moves | Defines the stress regime the journal describes |
| Latency model | 2-tick simulated execution delay on all order types | Makes 'instant fill' assumptions visibly false |
| Margin script | Simulated margin call on leveraged legs, session 2 | Tests whether exits were pre-planned or forced |
| Sentiment script | Randomized counter-rallies inside the downtrend | Separates disciplined exits from lucky ones |
| Participant plan | Pre-written: exit rules, size limits, no-add rule | The only real variable in the scenario |
Simulated Observations
- Latency changed everything the participant expected from their own orders: hypothetical market orders displayed fills two ticks away from the quote seen at decision time — a small mechanical detail that overturns most replay confidence.
- The scripted counter-rally on session 3 produced the log's most honest moment: the participant's written-in-advance exit at a 'bounce' fired, the tape then fell further, and the journal could record 'correct by rule' without self-congratulation.
- Simulated margin calls forced exits on legs the participant had intended to hold — the journal marks these 'externally resolved' to distinguish forced action from decision.
- Post-hoc review found the participant's most confident annotations clustered at local bottoms — even knowing it was a replay, the urge to declare the turn was documented and preserved.
What the Log Actually Graded
Three variables survived scoring: (1) was every action traceable to a pre-written rule, (2) were externally-forced exits labeled as such rather than claimed as decisions, and (3) did the journal's emotional annotations get preserved rather than edited for dignity.
Nothing about the crash's path was scored — because the path was a script. This inversion is the point of the exercise: in replayed disasters, the market is fiction and the discipline is real.
The log's final entry is deliberately mundane: a table of planned vs. executed actions. The drama stays in the screenshots; the learning stays in the table.
Methodology and Limits
A replay removes the two things that make real crashes hard: irreversibility and unknowing. The participant cannot experience either — they can, however, be observed writing decisions under simulated pressure, which is the data this study collected.
Modeled latency and margin scripts are approximations chosen for teaching clarity; real crash microstructure (halts, widening spreads, broken quotes) is categorically harsher. The log carries a standing 'best-case simulator' caveat on every fill.
Review Checklist
- Does every executed action cite the pre-written rule that authorized it?
- Are forced exits (margin script) separated from chosen exits?
- Are confident-sounding annotations preserved with their timestamps, including the wrong ones?
- Does the review decline to score the participant on the crash's outcome?
Review Takeaways
Filed in the journal, this scenario becomes the participant's reference for 'what I do when the tape is screaming': which rules held, which were improvised, and how honestly the improvisation was recorded.
The study closes with an uncomfortable observation: participants replay the same crash better on attempt two — not because they learned markets, but because they learned the script. The journal exists to tell those two improvements apart.
Frequently Asked Questions
No — replay removes the two things that make real crashes hard: irreversibility and unknowing. What it can train is documentation under pressure, which is what this log scored.
Exits forced by the simulation's margin script rather than chosen by the participant. The log labels them separately because claiming a forced exit as a decision is the most common self-deception in replay journaling.
Because an edited journal is a diary, not a record. The confident 'this is the bottom' note at the local low is precisely the entry future-you needs to read unedited.
Usually because they learned the script, not the market. The journal's job is to separate script-memorization from genuine process improvement — one is memorization, the other is method.
Review Note
In replayed crashes the only real variable is the participant's discipline — grade that, not the tape.
Continue Exploring
Federal Reserve Rate Decision Scenario
Synthetic modeling of rate cut announcements, short-term treasury curve reaction, and resulting simulated foreign exchange fluctuations.
Large and Small Caps Divergence
Studying Russell 2000 vs. Nasdaq 100 tracking differentials and liquidity friction within simulated small-firm equity baskets.
Janus Henderson Markets Overview
A comprehensive appraisal of sovereign debt benchmarks, corporate credit spreads, and cross-asset correlation models.