Well, considering we have a data set that consists of zero events/yr for decades except for 1 in 2001 and 1 in 2003? (shoe bomber, don't remember the year), I'd

Well, considering we have a data set that consists of zero events/yr for decades except for 1 in 2001 and 1 in 2003? (shoe bomber, don't remember the year), I'd say we are firmly in the region of small number statistics and I'm very doubtful any meaningful inference can be made about what the rate itself if, let alone whether it's changed after 2001...