Simulation and fairness

STRATHEX can simulate a field after STRATHMARK returns predicted cutting times, marks and performance standard deviations. Simulated races show how the field might behave under those assumptions.

Read the report

Reports can show win rates, podium chances, finish distributions and margins. Large differences can help a judge decide which fields need closer review.

The simulation uses performance spread for race-to-race noise. A forecast interval is uncertainty in the prediction and is not a substitute for that spread.

A simulation cannot prove equal real-world chances or rulebook compliance. Poor predictions produce misleading simulations. Review warnings, degraded cases, history quality and the complete field alongside the numerical report.

Run limits

The championship simulator is capped at 250,000 races and a two-million competitor-cell budget. Normal local CLI modes and STRATHMARK's REST /simulate have separate limits. The REST endpoint is capped at 250,000 runs and its own competitor-cell budget; it is not an interchangeable local mode.

Official decisions

Judges determine legal outcomes and authorized adjustments. Historical narrative LLM tools do not supply numeric authority. A close finish does not by itself prove a valid handicap or cheating. See Handicaps explained and Rules and officials.