Acceptable False Positive Rates: Sky Marshalls on the No Fly list?

Bruce Schneier linked to a Washington Times article about Sky Marshalls whose names match names on the No Fly list. The first thing that came to mind for me was "Well, what is the acceptable false positive rate?".

The system currently primarily causes issues for individuals, rather than groups. When you take a Sky Marshall off of a plane, one would tend to believe that you increase the risk to the entire plane - they're there as another layer of security. Most security analysts would first get a good chuckle, then start worrying if they found out that their security system was actually stripping away a different layer.

What happens when we get pilots on the No Fly list?