Much as I agree with your implied point that it would be better if there were fewer guns out there, your third point is focusing on the wrong statistics. You seem concerned that 90% of children injured by firearms are injured by a “family” gun. But would it make a difference to you if these children were injured by “non-family” guns, instead? I doubt it.

You seem to be focusing on (1) “given that a child was injured with a gun, what’s the probability of X [gun was owned by their parents]?” But what’s important is HOW MANY CHILDREN are injured. What we should be interested in is, (2) “total probability of child’s injury by gun”, or possibly (3) “total probability of child’s injury by a parent’s gun”. The latter is calculated as (3a) “given that a child’s parents own a gun, probability of injury from that gun” times (3b) “probability of parent owning a gun”.

Your implied interest (1) is the inverse of (3a) – so I think you’re committing a classic prosecutor’s fallacy.

I don’t take Freakonomics as gospel, but I do believe their basic numbers: From http://scienceblogs.com/deltoid/2001/07/levittpoolsvsguns.php, in 1998, there were something like 60 children under age 10 killed by a gun in a non-homicide (which I interpret to mean, generally, accidentally). 60 kids. In a country with 200 million guns. Including all deaths by gun, the numbers are 175 in 1998.

The numbers under 10 aren’t the whole story by any means, but they are indicative. The problem is actually quite small in absolute size. Every single one of those deaths is horrible, without a doubt. But the actual public health concern here may get overplayed.

Lying is wrong.
Lying on the internet, to people who obviously have access to the internet, is simply stupid.
Everyone has a different truth, and I respect your right to have your own truth, but Facts are immutable things.
Learn the facts.

http://www.gunfacts.info/