
Why Simulated Results Cannot Promise Returns
A simulated track record is produced inside a model with friendly fills, no queue, no impact, and no capital at risk. Extending it into a claim about live performance requires assuming away everything that makes live trading hard — an assumption no journal should smuggle into its summary.
Regulators treat simulated performance presented as predictive as misleading for exactly this reason. An educational archive holds itself to the same standard even though it sells nothing.
The Limits of Extrapolation
Extrapolation fails in both directions. A strong simulated record does not imply live success, and a weak one does not imply live failure — the model's fill generosity and absent emotional stakes cut both ways. The record measures familiarity with a process, nothing more.
This is why the Reviewbook never publishes aggregate 'returns.' Totals would invite exactly the comparison the archive exists to prevent: simulated outcomes read as performance.
Language That Stays Honest
Certain phrasings quietly convert practice into performance claims. 'The strategy returned 12% in simulation' reads as evidence. 'The simulated model displayed a 12% change under stated assumptions' keeps the claim inside the model where it belongs.
- Say 'simulated outcome' or 'displayed result' — never 'earned.'
- Attach the model's assumptions to any number you cite.
- Label every record HYPOTHETICAL where it appears.
- Never summarize many entries into a performance figure.
What You Can Claim Instead
A well-kept journal can legitimately claim process evidence: that plans were written before outcomes, that deviations were recorded, that reviews separated decision quality from result. These are claims about discipline — verifiable, honest, and genuinely valuable to a learner.
That is the archive's actual product: not a track record of hypothetical profits, but a track record of honest documentation. The second kind of record is the one that transfers to real learning.
Claim Language Audit
| Claim in Journal | Category | Review Treatment |
|---|---|---|
| 'The setup worked' | Outcome language | Rewrite: 'the setup produced X under model Y' |
| 'I was right about the move' | Prediction claim | Rewrite: 'pre-written expectation matched scripted move' |
| 'This proves the strategy' | Performance claim | Flag: unprovable in sandbox — delete or reframe |
| 'Next time I'll catch it' | Future claim | Convert to testable: define 'it' as a rule |
| 'The model showed a fill' | Observation | Acceptable — keep with SIMULATED tag |
Journal Checklist
- Is every outcome statement tagged as model-produced?
- Are performance claims ('this works', 'proven') flagged or removed?
- Is language audited for real-market implications?
- Does the journal distinguish 'practiced' from 'proven'?
- Are SIMULATED / HYPOTHETICAL labels on every data field?
Review Note
If a sentence about a simulated result would look different to a reader who didn't know it was simulated, rewrite the sentence.
Frequently Asked Questions
A sandbox outcome demonstrates one model path — 'worked' implies real-world efficacy the simulation cannot contain. The honest sentence is 'the model produced X under assumptions Y'.
Only as documentation of practice — labeled SIMULATED, with the model's assumptions attached. Presented as track record, they become a performance claim, which is exactly what this guide exists to prevent.
Write the observation plainly, then add the model caveat in parentheses. It reads naturally once it's a habit — and it keeps every sentence checkable.
Continue Exploring
Paper Fills Are Not Live Fills
Why a simulated execution differs from a real market fill, and how to record the gap honestly in an educational trade journal.
Recording the Order Idea Before the Outcome
How to fix your assumptions in writing before running a simulated order, so the review compares your real plan — not your memory of it.
Market Versus Limit Orders in Simulation
What market and limit order types actually mean inside an educational simulator, and how to choose between them for a learning record.