I’m looking at an FBI report with figures similar (but not identical) to yours, but it also has
Firearms, type not stated 1,611
With rifles, shotguns, and “other guns” all broken out separately, it seems fair to conclude that most of the “type not stated” were handguns, no?
… And the numbers are similar enough, year-to-year, that your “a different year” hypothesis is rejected.
So your “5000” should actually have been “about 8000”; correct? And here’s another government document (for 2013 instead of 2012) that states
Firearm homicides: Number of deaths: 11,208
The FBI reports 13,000 homicides with about 9000 by firearm. CDC reports 16,000 homicides with 11,000 by firearms. It might be interesting to understand the difference. But none of these numbers are anywhere near 5000.
But whether 5000, 8000, or 11000, what’s the difference, right? One side will characterize any of these numbers as “very rare”; the other side as “much too common.”
But I get very annoyed when incorrect statistics are prattled about on message boards, hoping that no one will check. If I remember a number as “6500”, am too lazy to Google, and the smaller the number the better it suits my case, I’ll write “less than 7000” or “about 6500.” I won’t pull a Sam Stone, write “6000” and accuse the one who corrects me of being a “nit-picker.” Writing “5000” is right out.
Since we’re in the Pit, I’ll just come out and say it: If you don’t know a number, just stifle your impudence and shut up!