Conditional scenarios
A model generates paths from defined data and rules. Target-before-stop frequency depends on entry, horizon, volatility and costs. A later-confirmed pending entry has less time remaining than immediate execution.
Quantity and precision
More paths stabilize calculations conditional on the model. They do not fix stale data, an unrepresented regime or incorrect execution assumptions. A numerical sampling interval is not a complete confidence interval about the market.
Subsequent evidence
Compare forecasts recorded before the outcome with later observations. Separate unfilled, expired and ambiguous outcomes and examine calibration and actual costs. Waiting can be useful when the model has not demonstrated a net edge.
Illustrative example
Equal path counts can produce different estimates when one entry needs an hourly close and the other uses first touch. Retain each calculation's horizon, prices and rules. Repeating an incorrect assumption 10,000 times does not validate it.
Review checklist
- Show data and reference time.
- Define entry confirmation and horizon.
- Include costs and unfilled outcomes.
- Separate later validation from simulation.