Precision
Of the cases a filter selects, the share that really are what it's looking for. When the target is rare, precision can be low even for an accurate filter.
In plain English
A filter can catch most of what it's looking for (high sensitivity) and reject most of what it isn't, and still have low precision if the target is rare, because its small error rate is applied to a big pile of non-targets. Precision rises when the target is common, when the filter is stricter, or when independent filters are stacked, at the price of missing more real targets.
Uncle Gary explains
If 10% of ideas are good and a panel is right 80% of the time, only 31% of the ideas it approves are good. A second independent round lifts that to 64%.
Why it matters for your money
A stock-picker who is 'right most of the time' about rare big winners still ends up mostly holding non-winners. Ask how rare the target is before trusting a hit rate.
Where we tested it
Related terms
All glossary terms. Educational content, not financial advice.