Statistics can be accurate and not applicable at the same time.
Take the "1 in 200" stat as an example. The problem with that number is that it is all-inclusive. It is ENTIRELY accurate that only 1 in 200 of ALL THE PEOPLE WHO CARRY in the ENTIRE COUNTRY will ever need their guns. What that number fails to do however is take into account regional variations. In my area, there is probably a 1 in 10,000 chance of ever needing my gun. In Portland, OR, especially right now, there might be a 1 in 10 chance.
The other problem is that we very often get no information except the "average", usually the "mean". While that information may be helpful, in and of itself there is no way of knowing what it tells us. Look at this example:
I have a set of 500 numbers with an average of 500. What sort of distribution do those numbers occupy? You have NO WAY of knowing without additional information. I could have 499 "1's" and 1 "249,501" and the average would be 500. I could have 500 "500's" and the average would still be 500. Much helpful information, like Standard Deviation and Median, are left out of the information available to the public because, frankly, most people wouldn't have the foggiest idea what it meant anyway.
Why am I talking about this on a gun forum? I forget now, but it's interesting.
Still happily answering to the call-sign Peetza.
The problem, as you so eloquently put it, is choice.
He is no fool who gives what he can not keep to gain what he can not lose.
-Jim Eliott, paraphrasing Philip Henry.