This is wrong. Even your quote shows this. A false positive rate is how often it triggers when the true result is false. > A false positive error, or in short

This is wrong. Even your quote shows this. A false positive rate is how often it triggers when the true result is false. > A false positive error, or in short a false positive, commonly called a "false alarm", is a result that indicates a given condition exists, when it does not. So 50% false positive rate means 50% of people who should not be flagged are flagged. https://en.wikipedia.org/wiki/False_positive_rate https://en.wikipedia.org/wiki/False_positive_rate