A false positive is a match produced by a rule that does not correspond to a meaningful outcome. In trading, it is a flagged pattern that matches the geometric criteria but is followed by no significant move, trend change, or other signal the trader cares about.
How it works
Looser scanner tolerance finds more shapes and more false positives; tighter tolerance finds fewer, cleaner matches but can miss valid setups.
The false-positive rate is measured by running a rule on a defined sample of history and counting how many matches were followed by nothing.
Why it matters
Counting matches tells you about the rule, not the market. A scanner flagging dozens of setups a day is finding shapes, not opportunities.
Filtering by false-positive rate is evidence-based tuning; tuning until a few charts you liked look good is curve-fitting.
A simple market example
A scanner set to find a specific breakout shape flags twenty charts in a week. Reviewing them, eighteen show no move after the flag. Those eighteen are false positives: the geometry matched, and the market did nothing.
Common mistakes
Judging a scanner by how many setups it flags rather than by its false-positive rate.
Assuming a high match count means the market is full of opportunities.
Frequently asked questions
Is it possible to have zero false positives?
Only by tightening the rule so far that it stops finding real setups too. Some noise is the price of catching anything.
How do I reduce false positives?
Measure the false-positive rate on a defined sample and adjust one parameter against that evidence, then re-test on different history.
Does a false positive mean the rule is broken?
Not necessarily. It means the rule and the market are producing noise at that tolerance. Some of that is unavoidable.
Educational content only. Definitions describe common market usage and may vary by jurisdiction, instrument, or institution.