Self Defense Statistics (Or Lack Of Them) For The Armed Citizen

[Fred Fuller]

http://thinkinggunfighter.blogspot.com/2012/03/self-defense-findings.html
Self Defense Findings

Many thanks to Claude Werner of Firearms Safety Training LLC for the analysis reported here, based on cases originally covered in the NRA's Armed Citizen column over a five-year period, from 1997 - 2001.

A common refrain from those who are seriously interested in 'the numbers' regarding the self defense experiences of armed citizens is that there just is so little reliable data upon which to base analysis. So we take what we can get. A certain number of people criticize Tom Givens for making reference to the experiences of his students who have been involved in self defense shootings ... but at the very least, Tom's students represent a statistically valid sample from which to draw useful conclusions. Until there are larger populations/more data available from someplace/anyplace, we have to do the best we can with what we have, while recognizing the shortcomings in the available data we necessarily must deal with in the process. The unknown and unavailable "real number" is what my wife the criminologist refers to as "the dark figure of crime."

For example, many instances of self defense by armed citizens are not statistically accessible, simply because they are never formally reported in such a way as to make them available to anyone seeking to analyze the statistics. Some numerical conclusions have been extrapolated by various researchers using available reports, but we still don't have a way to capture every single use of defensive firearms by private citizens. Probably we never will have such a thing, but we can hope for better days.

In idle moments I've sometimes wished that every single state that offers concealed carry permits, in the process of renewing said permits, would conduct a simple anonymous survey covering defensive use of firearms by the permit holder in the previous permit period. It still wouldn't be perfect, but it could offer us a clearer picture of just how often armed citizens were forced to access their weapons in self defense, even with no shots fired.

ETA: In the event you don't read the comments at the original link above, be sure and go to http://blog.hsoi.com/2012/03/08/looking-deeper-into-the-findings/ for more discussion.

 

[Loosedhorse]

The shooting distance in the vast majority of cases was slightly in excess of arm's length.
At this distances, even .22s and .25s are highly immediately lethal.

If an author is going to make a highly questionable and unqualified statement like this one about immediate lethality, maybe don't make it in the sixth sentence of the report. Kinda hurts the credibility of the whole exercise.

I think the dataset problems here are obvious: including only those cases that made it to the "Armed Citizen" column? And we hear complaints that the Marshall & Sanow numbers suffered from selection bias!

 

[Averageman]

What you have is not a statistic of anything , but headlines.
"The Armed Citizen" is a series of compiled dramatic cases used to promote the Second Amendment. This is not a crime statistic, nor is it relevant to anything other than showing that people can and do defend themselves in dramatic headline grabbing circumstances.

 

[Fred Fuller]

OK, the Armed Citizen is a less than perfect sample. I think anyone who followed all the links in the OP and read the discussion there is aware of that to some degree.

So please provide more reliable data from which to draw better conclusions.

Someone?

Anyone?

 

[MachIVshooter]

The biggest problem here is that, even if the cases are accurately and truthfully reported, they still only exemplify successful self-defense scenarios. We can learn a lot more from the unsuccessful ones. Like the figure of 80% retrieving guns from another location and having plenty of time to do so; It's wonderful that those folks had time, but the way this article reads implies that there will be time to get your gun. Dangerous.

 

[Librarian]

One can stretch the sample quite a bit - Armed citizen, http://www.nraila.org/gun-laws/armed-citizen.aspx, has 160 pages of about 40 per page, or around 6400 incidents back to 1958.

I've read more than a few. Newspaper reports generally are missing a lot of information that many people would like. For example

A 16-year-old Dallas, Tex., schoolgirl, awakened by an intruder in her bedroom, fired a revolver and sent him flying through a closed window. The girl's mother then aimed and fired and later identified a suspect captured by the police.

Fired how many rounds? Did both mother and daughter use the same revolver? At what distance(s)? Hit? Either have any training at all?

Or this one from 2011:

A Boiling Springs, S.C. homeowner was driving home when he noticed a car was following him. After the homeowner turned into his driveway, two people got out the suspicious car and attempted to rob him. The homeowner retrieved a gun from the glove box of his vehicle and fired at the criminals, killing one and causing the other to flee

Distance? Number of shots fired? We know one was hit - he was killed.

If someone wants to turn that info into a master's or doctoral thesis, s/he could look for the police reports that correspond to the incidents. Many of those will have more information.

 

[Loosedhorse]

So please provide more reliable data from which to draw better conclusions.

Reliable data on what?

What is this "data" purporting to show?

And are we to use such data politically ("See how good it is that law-abiding citizens can legally go armed!") or--as expected in this forum--for preparedness ("Guess I don't need more than two shots--derringer it is!").



Look at this stat, stated in the report but certainly not generated by its own numbers:

Eighty per cent of gunshot wounds are self-inflicted.

On what basis is this statement made? An undisclosed survey of accidental, non-fatal gunshot victims treated in ERs? A (gu)estimated from some other source? Does the figure include suicides (as I think, if it's got any change of being accurate, it must)? And if so, how in the world can the author then conclude:

...safe gunhandling qualities are much more important characteristics than its ability to be shot accurately and reloaded quickly. Revolvers are much less likely than autoloaders to AD in the hands of novices.

This reads like satire. While the statements may be arguably true, the "data" in no way supported them.

 

[Fred Fuller]

Once again...

Offer something better.

 

[Odd Job]

So please provide more reliable data from which to draw better conclusions.
Someone?
Anyone?

It is a tough ask, there is no doubt about it.

I've seen several thousand gunshot patients. I don't have an exact number of how many I have been able to speak to, but here are some common themes (as explained to me by the gunshot patient):

1) I was robbed.
2) I was sitting minding my own business and a stray bullet hit me.
3) People broke into my house/car and when I went to investigate I was fired on.

You get some detailed honest stories also, backed up by the evidence at hand and other witnesses.

Once in a while a guy will admit he screwed up and had an ND. As I sit here now I can only think of two cases where a person has volunteered information to the effect that this was an ND. One was a gunshot foot and the other was a gunshot thigh whilst handling a Lorcin pistol without a holster in a car. That was a Ranger SXT that missed that man's jewels by a few millimetres.

Often the clothing or radiological features will conflict directly with the patient's story. I know for a fact that at least two yutes I X-rayed in 2002 had self-inflicted GSWs caused by IWB discharges, one of which definitely involved Mexican carry. In other cases the fracture patterns show trajectories that are incompatible with the patient's version of events.

Now consider this: not all of these cases prove that the patient is a liar. In a lot of cases they are simply confused, cannot recall clearly or are otherwise influenced by the trauma of the event or treatments being applied.
In other situations the patient does not realise that the projectile did not behave as he has expected, for example a deflection through auto glass, or observed effects on intervening materials which may bias his interpretation of what has happened.

That's the ones who were shot, some of who will have fired back (or maybe they fired first, who knows?)

The variables surrounding these shootings are complex and varied. If all you are after is documenting damage, that is a much easier thing.

But circumstances, that's a different animal...

 

[Loosedhorse]

Something better? Ice cream is better!

I will assume (lacking a response) that you are looking for some statistics that will depict the "average" lawful, non-LE SD shooting, in various particulars? I still do not know to what purpose we would put that data--just curious?

If we can reasonably expect that these statistics are systematically biased, and therefore misleading (that is, they do not in fact describe anything close to the "average" scenario, but are presented as such), then offering NOTHING is offering something better.

And ice cream is clearly better than nothing.

 

[Fred Fuller]

So, you're going to send everyone ice cream? How nice.

But that's not going to offer any constructive help with the question at hand - is there any more reliable source of unbiased statistics on the use of firearms in self defense by armed citizens that is publicly available? Which, when compiled and analyzed, can offer a reasonably accurate picture of what actually happens to armed citizens who are called upon to defend themselves?

Or do we just get an unlimited supply of snark and sharpshooting?

I'm all in favor of critical thinking and analysis, but it is less than helpful when nothing more constructive is offered in addition to criticism.

 

[Loosedhorse]

Or do we just get an unlimited supply of snark and sharpshooting?

One man's snark is another's help. You (and the authors) are presenting this data as somehow and to some purpose "useful" (?) or "interesting (?). If I feel that is inaccurate, I think saying so may be something besides snark.

I feel as if you have presented us a map of territory we are already a little familiar with. You say the map has more detail than our intuition, and I point out that, if the detail is false, it is actually worse than having no map at all.

That's sharp-shooting?

Which, when compiled and analyzed, can offer a reasonably accurate picture of what actually happens to armed citizens who are called upon to defend themselves?

How about this?

It is a bit old, so comes from a time when crime rates were higher and CCWs rarer. It suggests that non-police shoot their guns about 2 out of 7 times that they "use" them for SD. It says that guns were used most often to prevent assault.

This study is "better" in that it has a better methodology than the one you offer. It has its own problems (as Kleck has pointed out). And it gives fewer details of the defensive scenarios. Depending on one's impression of the accuracy and importance of such details, their absence might be a good or a bad thing.

As to ice cream, I do suggest that enjoying a (small, healthy) portion will be better than "enjoying" this study.

 

[Fred Fuller]

As I said, it isn't the saying so that I object to.

It's the tone.

And I'm still looking for material that covers what some refer to as Gun Culture 2.0, which essentially began in 1987. As indicated, the study you linked is set just at the beginning of that period and leaves most of the time (1990-present) unreported. And since then, as indicated, crime rates have declined and the number of concealed carriers nationwide has increased significantly.

And I'd like a small slice of pie with my ice cream, please.

 

[Loosedhorse]

It's the tone.

Then I apologize. Humor that doesn't come across as humor isn't humor.

Again, you posted this thread in ST&T. The implication is that this is not a "well that's kinda interesting" thread, but that it has implications for our strategic thinking and preparations. I believe it doesn't. When you asked for "better," I was serious: if such a study gives us false info, not reading it is better than using it for planning.

Perhaps you are a fan of the Marshall and Sanow "One-shot stop" numbers. It is a similar situation. Some folks think they are misleading and dismiss them as worse than nothing; others say, well, they're the best numbers out there, so let's look at them.

If you had posted this in General with a "this was fun to read" label, there would have been no implication that this study has relevance to strategy or tactics, and I would have made no comment.

And I'd like a small slice of pie with my ice cream, please.

You are free to follow my suggestion and get your ice cream, just as I was free to "get" the study (by following the link, using my own ISP resources). I guess, to be fair, I should provide you a link.

Enjoy!

 

[Fred Fuller]

Then I apologize. Humor that doesn't come across as humor isn't humor.

As do I. I didn't realize it was intended as humor.

Let's see. I seem to recall titling the thread "Self Defense Statistics (Or Lack Of Them) For The Armed Citizen." What part of (Or Lack Of Them) is so opaque? I thought there was sufficient discussion of the realities of the material posted in the link at http://blog.hsoi.com/2012/03/08/looking-deeper-into-the-findings/, which was also pointed out in the OP.

But perhaps not.

And I have about as much realistic expectation of coming up with better data upon which to base more useful conclusions anytime soon, as I do a delivery of pie and ice cream.

But I'm still looking forward to someone out there deciding that this is a potentially rewarding subject for serious research. As my wife the recently retired professor said in discussing the dearth of current material available on the subject - "Why don't I have grad students when I need them?"

 

[Loosedhorse]

Ah, graduate students: the slave labor of the academic plantation!

Well, let's not overlook what we do know. As Rory Miller reminds us, violence happens between people, so I will be attacked either in a place that I frequent (like home or work) and the attacker chooses to "visit", or a place that the attacker frequents (like deserted places if he's a "professional" attacker, or a bar if he's a dabbler) and I've been unwise enough to visit.

I will likely be attacked when I am tired and unattentive. Because of the training and practice I have engaged in, I think I know something about how the attack will proceed from there. But I can't know exactly.

I guess it is true that until we can look at statistics for the "average" attack these days (perhaps broken down for region, sex, age, race, etc. for greater correspondence to each individual), we can't tell what might surprise us, and cause us to approach our preparations in a different way. But I am biased, and think that it would change little in how I practice, or in the equipment that I choose.

And figure that any such study would be publicly available, so at least the "professional" attackers would have the opportunity to change their MOs, and lessen the study's validity.

Oh, can't help you with pie, but for ice cream, don't abandon hope: these guys deliver!

 

[JRH6856]

So please provide more reliable data from which to draw better conclusions.

Someone?

Anyone?

Ok, here you go. Although this is an analysis of firearm performance, it is bit more in depth and possibly more useful for some of the areas touched upon.

As to the AC data, it is the Armed Citizen. It relates incidents of gun related self defense. Having that specific purpose, it is not a general compilation of crime statistics. But trying to draw conclusions from it because it is "all we have" is like one of the 6 blind men insisting an elephant is like a rope while holding its tail.

Before CCW expanded in the last 10 years, the 3 most common places one might have a gun were the home, a business, or in a vehicle. And gun related incidents tend to occur where guns are present. I suspect further analysis would show that the incidents in businesses were business owners protecting themselves from robbers, not customers defending themselves or foiling the robbery.

As to the shootings taking place at close range and the question as to why someone would go to another room to retrieve a gun and then return to the area of danger, perhaps it is that it seemed like the best way to handle the situation. Barracading oneself while shouting warnings does nothing to stop a burglar from burglarizing the rest of your house while you are hiding in a closet. Before the expansion of the Castle Doctrine, a duty to retreat required one to do so even in ones own home. And I believe Texas is still the only state that allows deadly force to protect property. Perhaps the homeowner placed himself in danger because that was the only way to create the elements necessary for an affirmative self defense argument which was necessary to justify what was actually a defense of his property. Just a thought.

 

[GEM]

Some commentary on Claude's piece by John Hsoi:

http://blog.hsoi.com/2012/03/08/looking-deeper-into-the-findings/

I know all the players and like them. I think John makes some good points.

Glenn

 

[MachIVshooter]

Does the figure include suicides (as I think, if it's got any change of being accurate, it must)?

It has to. In a given year, suicides do account for roughly half of the firearm related deaths in the USA. But even at that, it would mean that the extreme majority of a large number of non-lethal gun incidents would have to have been self-inflicted to come up with the 80% figure (if all the non-lethal gun injuries were self inflicted, there would still have to be 60k of them).

I'm not buying it

 

[HeadJunter]

Since I wrote the original study, I'll just make a few comments.

The reason I did it was to compile some kind of composite picture of what Private Citizen firearms related self defense incidents. Since every self defense shooting is an anomaly, I didn't intend for it to be taken as a portrayal of what any particular Private Citizen's incident will look like.

Since I like to vet things I believe and write about, after I did the study I ran several RSS feeds for two years on topics such as "armed robbery" "self defense shooting" "liquor store robbery" etc. During that time, I received several stories each week. The results were not remarkably different from what my study indicated.

What I did want to accomplish is get past the thinking that unless one carries a service size weapon, three spare magazines, two flashlights, a pair of fighting knives, a backup gun that is a smaller version of the service weapon, a "blowout kit", and a partridge in pear tree that one is not going to survive a criminal encounter. That doesn't mirror my own experience and I can demonstrate it doesn't mirror the experiences of a lot of other people.

I just had a conversation with a smart friend who made the comment: "I find it interesting how many people are adamant about proving me wrong without proving me wrong." That made me laugh.

 

[JRH6856]

"I find it interesting how many people are adamant about proving me wrong without proving me wrong." That made me laugh.

Oh I think you are probably right, I just don't think the study proves you are right.

 

[Loosedhorse]

What I did want to accomplish is get past the thinking that unless one carries a service size weapon, three spare magazines, two flashlights, a pair of fighting knives, a backup gun that is a smaller version of the service weapon, a "blowout kit", and a partridge in pear tree that one is not going to survive a criminal encounter.

Are you offering a guarantee that we won't need that stuff?

An instructor once told me, "I'm a pretty smart guy and I've thought about this stuff a lot, but there's two things I can't tell you: what your fight is going to look like and what it's going to take to win." Not sure he had a study to back that up, but it sounded right to me.

Oh: I no longer carry the pear tree. The partridge, George, comes along sometimes, anyway.



As I said, I thought that bit about .22s and .25s being "highly immediately lethal" got things started on the wrong foot.

 

[Fred Fuller]

Lots of us don't like to look this particular grinning skull in the empty eye socket, but it's still sometimes the case - you can do everything right and still get killed.

That is NOT to say "give up" or "don't try" or "don't train" or "don't prepare." All of us should be working to train, test, improve, and practice despite the unpleasantness of some of the possibilities we face.

But the simple fact is, we have no guarantees. No one can offer us any. None of us can reasonably demand any, either. We are citizens, after all, not Imperial Stormtroopers. We'd get shuttled downtown most rickytick if we went out in public in full battle rattle every day. Some of us prefer to carry bigger guns than others. Some of us prefer to carry more guns than others. Some of us prefer to carry more ammunition than others. That's just the way it is. But does any of that actually make us safer, or does it just make us feel safer?

So far in my almost six decades on this planet, I have not had to shoot people with single stack .45ACPs. I've not had to shoot people with J-frame .38 Specials. I've not had to shoot people with double stack Austrian wondernines. And I've not had to shoot people with Beretta .22LRs as well. So far. I've carried all of the above at one time or another. And there have been times when it's been comforting to be able to put a hand on any one of them, and times when being able to put a hand on one or another has kept me from having to present said piece and/or fire it.

None of that guarantees (there's that word again) that I'll be able to get through tomorrow without having to pull a trigger on someone.

What statistics we have seem to say that most of the time we won't have to draw. If we do have to draw, most of the time we won't have to fire. And if we do have to fire, most of the time it will be a few rounds at close range and it will all be over very quickly. Most of the time. But certainly not all of the time. I don't think anyone with any of their compilations or analysis of statistics so far has tried to say anything any different than that.

Human nature is human nature. We want things to be predictable. We expect things to happen pretty much the same way they have happened before in our experience. Reality being what it is, most of us are doomed to be disappointed in our desires and expectations sooner or later.

So. Here we are back at the beginning of the circle again. Please ask yourself - do you have a gun on your person right now?

Do you have a reload on your person right now?

Do you have a gun on your person all day every day, wherever and whenever it's legal to do so?

If you do, it's for sure that you're in a definite minority. Some say only a fraction of one percent of Americans actually carry their defensive pistol habitually every day. I don't know if that's true, I don't have any way to prove or disprove it and I'm not looking to start another nutroll over lies, damn lies and statistics (there's a book by that title, by the way*).

But is even carrying a pistol every day all day enough? Some argue vociferously that it's not necessary to carry at all, while some argue just as vehemently that more is needed.

Me, I don't profess to know. I do think that the only way to have a gun when you need one is to have a gun all the time, and that it doesn't do much good to have a gun that you are not mentally and physically prepared to use if necessary. You need to carry a gun that is as reliable as possible and one in a caliber that will do what you want it to do. You don't have to satisfy me with what you do or do not carry, and I'm not looking for anyone to dictate to me what I should or shouldn't carry.

And I'm willing to let it go at that.

*http://www.amazon.com/Lies-Damn-Statistics-Manipulation-Opinion/dp/0440351219

 

[JRH6856]

Security is a state of mind, safety is an actual condition. What it takes to make you feel secure may or may not make you safer. It may even make you less safe, but if it makes you feel secure, that is probably what you will do.

 

[Double Naught Spy]

OK, the Armed Citizen is a less than perfect sample. I think anyone who followed all the links in the OP and read the discussion there is aware of that to some degree.

So please provide more reliable data from which to draw better conclusions.

Someone?

Anyone?

Once again...

Offer something better.

Hmm, to offer more reliable data, one would first have to establish the reliability of the data in question. Being the only data set offered at this point does not make it reliable.

A certain number of people criticize Tom Givens for making reference to the experiences of his students who have been involved in self defense shootings ... but at the very least, Tom's students represent a statistically valid sample from which to draw useful conclusions. Until there are larger populations/more data available from someplace/anyplace, we have to do the best we can with what we have, while recognizing the shortcomings in the available data we necessarily must deal with in the process.

The problem is, we tend to skate over the shortcomings and treat the data as reliable. I am not familiar with the arguments against Givens using his students' events as data, other than from statsitistic aspects where his students are in no way representative of the norm. As such, extrapolating the information to the general population would have some serious limits. Talking about what works or doesn't work based on examples of folks trained in given methods would be less so.

David Armstrong brings up some valid points in the comments on the summarized data. Probably the one that bothers me most is that the Armed Citizen is all about success stories...people with guns doing good. Conclusions draw from the Armed Citizen don't include all the times people performed in the exact same ways and were killed as a result. In order for the data to be reliable in terms of being usefully representative, you need the failures, but also, a lot of those folks are dead.

For all we know, things done in those successful engagments may actually be things often unsuccessful in the failed engagments. That is a real problem and represents a huge bias in the data.

Is there a more reliable data set? Probably not, but given that we know that the Armed Citizen data are heavily biased to self defense success stories at the expense of not including failure stories, the notion of "more reliable data" is something of a joke. The Armed Citizen stories may be individually accurate, but as a data set should not be considered reliable in any way. In fact, the outright bias indicates the data would be unreliable for trying to understand the general self defense situation.