Skip to reversion research
Mean Reversion StrategyReference / Excursion / Evidence

Conviction grades are not probabilities

A seller's A-D grade is a label until its definition and calibration are supplied. No verified formula linking Vector Ridge grades to z-scores or position size was provided. Do not convert a grade into risk or an implied win probability.

Original editorial revision: 9 October 2026. Expanded edition: 14 October 2026.

Educational diagnostics, not a signal service. A standardized price displacement is not a calibrated probability; invented examples and user-supplied observations are not verified market performance.

The research question

A seller's A-D grade is a label until its definition and calibration are supplied. No verified formula linking Vector Ridge grades to z-scores or position size was provided. Do not convert a grade into risk or an implied win probability.

An inspectable record

For a reversion claim, request the frozen reference definition, observation availability, complete eligible event population, treatment of crossing and overshooting, exit clock and cost model. An alert tool can express a condition without proving that a seller has tested it. Separate a model summary from a subscriber account record.

Next check

Write the price series, adjustment policy, prior lookback, standard-deviation convention, reference update clock, event threshold and recovery definition. Save a version before inspecting later outcomes. Count every eligible observation, including zero-variance exclusions, missing bars and events that never return.

Source documents

Documentation establishes the stated mechanism, not tested trading performance. Checked 9 October 2026; a later version can differ.

Detailed research

Do not infer odds from a letter

A grade needs a defined source population, effective rule and calibration evidence before it can imply a numerical probability. A displacement score and a direction label are different quantities. The new path tool accepts declared prices rather than assigning grades, and its fictional hits cannot calibrate a seller's labels. Keep an absent probability record absent. Changing a grade threshold after viewing outcomes has selection history; it is not automatically an unchanged independently evaluated rule.

Reference ownership remains necessary

A genuine grade-conditioned reversion study must preserve destination, scale, endpoint and native action. Otherwise apparent improvement may come from a changed question rather than a more informative grade. Report complete opportunities, exclusions and unresolved outcomes for every grade category. A small authored case can demonstrate this requirement without estimating a real success rate. If original model records are unavailable, publish the specific evidence request and do not invent a grade-to-odds table.

Category boundaries need effective versions

An unchanged letter can describe a changed category after its reference or scale is revised. Preserve the exact effective rule and which rows it classified; do not transfer a prior category result to a successor simply because its public label matches.