When do statistics count?

[BrianSoddingBoru4]

A slightly more simplistic (but also more illustrative way) to look at it is:

80 000 000 civilians kill 45 000 people per year with guns.

700 000 cops kill 1000 people per year with guns.

If there were as many cops as there are armed civilians, - and nothing else changed - deaths by cop would be in the neighbourhood of 114 000 per year, with no change in the number of gun deaths by civilians. The chief reason that cops kill - in raw numbers - fewer people than armed civilians do is that there are fewer cops.

Boru

Cops deal with dangerous criminals on a regular, if not daily basis, and so, there is a selection effect going on here.  Imagine buying a pair of handcuffs.  You as a private citizen could carry them around with you all day; you'll never use them as part of some (legal) citizen's arrest.

Cops deal with dangerous suspects on a sporadic basis (while you didn't explicitly say it, I just want to note that I'm fed up with the notion that copping is a dangerous profession).

But I'm not sure the reasons for shooting people is germane to the OP's point, which seems to be that you're less likely to be shot by a cop than by a civilian, full stop. That's simply not true.

Boru