Core terms
Conditional probability
The probability of an outcome given that a condition is true, written P(A | B). In trading, the rate of an outcome among only the situations that met an extra condition.
A conditional rate narrows the sample. In the FennEdge Gapper HOD study, 34.7% of the whole sample closed at or above the open (233 of 671), but 69.5% of the names whose high printed at 10:00 ET or later did (n = 200; 2026-07-01 to 2026-10-02; FennEdge archive; as of 2026-10-02).
A condition is only useful for a decision if it is knowable at the decision time. "The high came after 10:00" is known only after the session, so that split describes two kinds of day rather than a way to sort stocks in advance. Narrower conditions also leave fewer observations, so conditional rates usually carry wider uncertainty than the base rate.
What traders should verify
For each condition in a plan, ask two questions: was it visible before entry, and how many logged observations met it? Record conditions with every practice trade so your own conditional rates can be computed later.
Do not conflate
A large difference between a conditional rate and the base rate can come from hindsight (the condition uses later information), from a small subsample, or from chance. It does not by itself show cause.