View Single Post
Old February 8, 2014, 08:03 PM   #39
JohnKSa
Staff
 
Join Date: February 12, 2001
Location: DFW Area
Posts: 24,928
Quote:
It doesn't seem like an outlier to me, John. TOTAL shots fired, yes, but not average or the distance or the basic scenario.
I can't imagine how a barricaded suspect or officer with multiple officers on the scene could seem like a typical LE shooting as opposed to an outlier. I think this speaks strongly to my comment about trying to judge what is typical by personal recollections.
Quote:
Yes, the NRA numbers come from "confirmed " SD and only successful ones but isn't that what we want to know?
Well, it's part of what we want to know. But we really need to see what goes on in ALL the shootings, not just the ones that turned out right. It's not possible to find out what is typical if we only focus on the ones that were all easy enough for the typical, untrained, unskilled, gun owner to survive.
Quote:
i don't want to know about claimed SD that was criminal.
First of all, it doesn't appear that they even include the ones that might appear questionable, so the fact that they only include the clearcut ones doesn't mean that they're only eliminating criminal shootings. It means they're eliminating anything that might even appear to be questionable.

Second, what we want to know about is what we need to think about in terms of preparation. The fact that a defender might have made a serious legal mistake in the process of saving his life certainly doesn't mean that there's no value in understanding and preparing for the kind of situation that put him in need of self-defense in the first place.
Quote:
Unsuccessful events are not included but I don't know why that would skew the numbers...
Generally speaking, statistics theory suggests that you're more likely to skew results by throwing away data altogether than by including outliers. The outliers tend to occur on both sides of the average and often have a tendency to cancel each other. But when you throw away an entire class of shootings (questionable legalities, LE involvement, unsuccessful self-defense), you have a very good chance of ending up with a data set that is completely unrepresentative of reality.

I think there's a little bit of a double standard going on. In spite of the fact that we all realize that the ACA stats have been "pre-filtered", you suggest we should be willing to accept the filtered results as typical of what we should expect and prepare for but at the same time, based on personal recollection of what is typical for LE shootings, you say we should discard all the LE shooting information as irrelevant.

It doesn't make sense to accept the ACA data as valid for determining what is "typical" in spite of the fact that we know for a fact it is heavily filtered before we ever see it and then use something as unreliable as personal recollections/impressions of what is typical for LE shootings to completely invalidate the use of LE shooting data for anything not related to LE.
__________________
Do you know about the TEXAS State Rifle Association?
JohnKSa is offline  
 
Page generated in 0.03151 seconds with 8 queries