Capacity, Hit Rate, Multiple Assailants and some thoughts...

Bart said:
No; but we can look at what the mean hit rates of people who have been in gun fights as a rough guide to what to expect from similarly trained shooters. If we want to be especially cautious we can even halve those numbers.

And the numbers really need not be precise; because unlike the Price is Right, we don't have to be the closest to the actual number. We just need to not go over it.

Bart,

I wish it were that simple.

The numbers had better be precise because what goes down on the asphalt is staying on the asphalt.

It's your life to gamble with.

I can't delude myself into thinking that this model tells me anything that is not already apparent without the model and the risk of depending upon it for all that is dear to me is not an option. It could offer an "incorrect" prediction just as easily as it offers a "correct" prediction and there is no way to say for sure until it happens.

I can get to the same place by just flipping a coin- "heads" I make it, "tails" I don't.
 
Yes.

That carrying a 5 shot revolver with no reload is a bad idea no matter the number of assailants I meet requires no model to ascertain.

Why make the process more complicated than it need be?

How do you know this?

there is no way to say for sure until it happens.

Have you personally lost gunfights with this combinations enough times to know that it is a bad idea? ;)

Or are you, like most people, making an accidental statistical analysis without actually putting pen to paper or punching a calculator?:eek:

Based on your "common sense" approach, there is no way to know that 5 shots won't be "enough" until the event actually happens.

So what makes this analysis suspect but your personal analysis "common sense"?:confused:
 
CrowHunter said:
How do you know this?

I used common sense. Didn't need a calculator either!

Have you personally lost gunfights with this combinations enough times to know that it is a bad idea?

That's very funny. At least you haven't succumbed to an attack of "the vapors" like someone else who recently bailed out of this thread. :D

You've got a good sense of humor and that's gotta count for something. And I didn't need a model to figure that out either. :)

CrowHunter said:
Or are you, like most people, making an accidental statistical analysis without actually putting pen to paper or punching a calculator?

Based on your "common sense" approach, there is no way to know that 5 shots won't be "enough" until the event actually happens.

So what makes this analysis suspect but your personal analysis "common sense"?

Nope, I simply carry as many rounds as I can comfortably carry on my person using the platform of my choice. In this case, I go with a Glock 17 and an "extra" magazine (equipped with a +2 floor plate) loaded with the same JHPs that are in already in the gun. That gives me 37 rounds of 9mm JHPs "on tap". I also practice and train a lot.

No model required. ;)
 
Posted by 481: Sure, it's fun to play with numbers and calculators, but the only answers that matter are the ones that actually occur.
Well, yeah. The result is what is important.

The question at hand is , how many rounds should one prudently carry? And no, that is not an "assumption." It is an independent variable, to be decided upon in advance. Once the decision is made and acted upon, the number is a fact.

Most of us have been making that decision based on rather subjective thought processes. Few of us have ever really analyzed the "what ifs" that, one way or the other, will ultimately lead to the "answers that will actually occur", which in this case are whether we will survive a violent encounter.

After endless back and forth discussion of whether five rounds are "enough" or whether one might "feel comfortable" with six, John decided to apply a simple analytical approach to test the reasonableness of some of the opinions offered on the subject.

Fact is, one will never know how many rounds are sufficient until the event occurs, and as I have said before on many occasions, even then the number cannot be relied upon for a subsequent encounter. But that does not mean that we cannot do some assessment, based on certain assumptions, of what a reasonable number of rounds might be.

Many people have been happy with five. John's calculations show that five rounds is probably far from the best number to head out of the house with, though you say that that was obvious to you from the beginning.

And John's projection only shows a probability. If one doesn't know how to assess that kind of result, one should not try to use the numbers.

Four rounds might suffice; one may need nine; but how does one decide?

How does one decide? On anything? With projections base on assumptions. The key lies in testing and assessing those assumptions. That takes up back one more time to this:

...the only answers that matter are the ones that actually occur.

That's true for everything, but one has to make decisions, and to do so one has to make some assumptions. If one is looking for a bank loan or for investments to launch a business venture, one is going to have to make and support some assumptions, one is going to have to show some calculations, and one is going to have to decide upon a plan. And even if one had enough money to start on his or her own, one will have to make a lot of decisions. and those will require projections based on assumptions.

Yes, there will be considerable uncertainty in the projections, and yes, the ultimate results will vary, but no, you won't even get to first base if you simply tell your banker or investors that they will just have to see what occurs when all is said and done.

Or if you prefer--the question of how many troops to commit to take an objective will have to be decided upon in advance, based on assumptions and projections and calculations. One cannot just throw up one's hands because there is uncertainty.

The same thing applies here. One cannot simply say, "I'll just grab the gun the guy at the shop sold me. After all, I won't know whether five shots are sufficient until they are not".

You have said, "That carrying a 5 shot revolver with no reload is a bad idea no matter the number of assailants I meet requires no model to ascertain." OK. Many people have offered the opinion that, because many gunfights are said to be over in three rounds, five should be plenty. It is for those people for whom John's calculations will likely prove illustrative. You obviously did not need them.

Personally, I believe that John's contribution here has been among the best I have seen. By the way, I often carried a five shot revolver. Not any more.

John's model is simplistic. One could make it more elegant, and one could put a lot more effort into assessing the hit rate assumptions based on extensive FoF simulations, but toward what end? The uncertainty in the number of hits required would remain very significant. John did not set out to create an accurate estimation tool. He set out to do what he said he set out to do, and I think he succeeded admirably.
 
Posted by 481: Nope, I simply carry as many rounds as I can comfortably carry on my person using the platform of my choice. In this case, I go with a Glock 17 and an "extra" magazine (equipped with a +2 floor plate) loaded with the same JHPs that are in already in the gun. That gives me 37 rounds of 9mm JHPs "on tap".
Wow!
 
All statistic predictions are guesswork. The difference is that there is a framework that is utilized that approximates a result based on the inputs used. If done correctly, those results will approximate reality within the confidence interval used.

You mean a S.W.A.G.? :D
 
I can see 481's problem with the process.

There is no real data, just 5 assumptions lumped together mathematically to prove a point.

I do not disagree with the point, just the methodology.

I have written a number of operational plans, in that context an assumption is more like; if we call for a fire truck we can assume that one will respond. It is an assumption of fact, not fantasy.

Assumptions without fact will get you in trouble every time. the assumptions may be Correct for John, but not for everyone. Not everyone possesses the same skill level or the same ability to maintain a level head.

Every shooting is different, was it at bad breath distance or 10 yards? That makes a huge difference in the amounts of hits, current LE data shows this. In the famous Miami FBI shootout, one agent missed 15 shots a bad breath distance, it happens.

Carry enough gun, with enough ammo and master the gun you carry. Get training and realistic practice. Like Bill Jordan said, "There is no second place winner".
 
Thanks, Nanuk. That makes two of us. :cool:

Nanuk said:
Carry enough gun, with enough ammo and master the gun you carry. Get training and realistic practice.

Now there's a model!

As I've said before, if someone lacks the common sense to make these realizations, no amount of math is going to help them and they probably shouldn't be carrying a gun in the first place. No amount of math can fix stupid.

I am sure that JohnKSa's intentions are noble. The point (carry enough ammo) is a good one, but the methodology (e.g. SWAG) leaves much to be desired.
 
I do not disagree with the point, just the methodology.

I have written a number of operational plans, in that context an assumption is more like; if we call for a fire truck we can assume that one will respond. It is an assumption of fact, not fantasy.

Assumptions without fact will get you in trouble every time. the assumptions may be Correct for John, but not for everyone. Not everyone possesses the same skill level or the same ability to maintain a level head.

It's all about assumptions and context, and everyone's assumptions and context are different. However, an assumption is not a fact. An assumption is a probability for people who don't like numbers... just a guess without numbers to back it up.

Depending on where we live and what's going on during any given time, we may or may not be able to safely assume the fire truck will come in time to help us. If we carry 5 and assume it will be enough, then we have to live with it if it's not. If we choose not to carry at all we live with that. And if we carry 50 and get taken out by a guy with a knife or pointy stick, because we were too lazy to train hand-to-hand, then we live with that.

The only reason to cc a gun at all in the first place is "what if", statistics, models, or gut-feel and assumptions. I'm less likely to need a gun than many, but I'll carry what I damn well please whether it's 1 round or 50. John's data gives us all something to think about whether we chose to think or not, and his presentation of it is excellent. If anyone has any improved what-ifs or assumptions, either numeric or otherwise, then I'd welcome them.

As I've said before, if someone lacks the common sense to make these realizations, no amount of math is going to help them and they probably shouldn't be carrying a gun in the first place. No amount of math can fix stupid.

Agreed. No amount of ammo will either. :D
 
Depending on where we live and what's going on during any given time, we may or may not be able to safely assume the fire truck will come in time to help us. If we carry 5 and assume it will be enough, then we have to live with it if it's not. If we choose not to carry at all we live with that. And if we carry 50 and get taken out by a guy with a knife or pointy stick, because we were too lazy to train hand-to-hand, then we live with that.

The only reason to cc a gun at all in the first place is "what if", statistics, models, or gut-feel and assumptions. I'm less likely to need a gun than many, but I'll carry what I damn well please whether it's 1 round or 50. John's data gives us all something to think about whether we chose to think or not, and his presentation of it is excellent. If anyone has any improved what-ifs or assumptions, either numeric or otherwise, then I'd welcome them.

Thats the point. Self defense is a system, the lethal option is one part of that system.

You cannot assign random hit probabilities to random people in random encounters, with other random people. People defy logic everyday. The numbers John generated give as much credence to a 380 ACP as it does to a 357 Magnum, but, that is another discussion.

What the theory would indicate is that an untrained person has the same hit probability of a trained pro in the same situation. I cannot even wrap my little brain around that one. Stacking statistics and guesses is mental masturbation at its worst
 
Posted by Nanuk: What the theory would indicate is that an untrained person has the same hit probability of a trained pro in the same situation. I cannot even wrap my little brain around that one.
No!

One chooses the hit probability. John chose one that can be supported, and one can choose another one, perhaps based on extensive simulation, but there is considerable variation around any average. John's purpose was not to make an accurate projection, but to see just how differing assumptions might affect the probabilities in the low round count range. For him and for many of us, the result was interesting. Don't like the assumptions? Use your own.

Don't like the model itself? Develop your own, if you like. You could model distributions around the average hit probability for each shot, and around the mean number of shots needed to stop an attacker, if you wanted to. You could include other algorithms.

I really do not think it would be worth the doing. The assumption regarding the number of hits required to stop an attacker involves so many uncertainties that the model would be little better than what we have now. And to display the output in an understandable manner manner would likely require 3D animation.

No manner what one might do, there is considerable uncertainty, in terms of spread and distribution, in each of the input variables, except of course for the number of rounds carried which is a constant. And the result is only a probability.

None of this speaks against the value of John's work. Far from it.

A number of people had settled upon low round counts, based on reported average occurrences. John decided to test the distribution, just for his own benefit, and then he decided to share the results....

...which indicated that starting with five rounds is demonstrably a high-risk proposition, no matter how much you vary the assumptions. I found it very helpful.

I can certainly see how someone who carries thirty seven rounds would not gain anything from the exercise.
 
Oldmarksman said:
One chooses the hit probability.


OM,

You cannot choose your hit probability. In a real, honest-to-God gun fight, it will be whatever it is. You may miss every shot. You may make every shot. You might not have a chance to even attempt a shot. All this hypothesizing, this fantasizing, it is all just mathematical busy work.

Lately, our thugs have taken to simply walking up to a robbery victim with their gun already in hand tucked behind their leg/buttock and shoving it into the victim's gut and firing a shot. Where's your "hit probability" now? If you encounter one of these vermin, it ain't gonna be like you are standing "toe-to-toe" and "face-to-face" trading shot-for-shot.

Attempts at such modeling are exercises in futilty and they have the property of leading the unintiated and unknowing into the perception that they will have more control, know more about what is about to transpire, and expect something other than what they should from such a dynamic and infinitely unpredictable event.

Oldmarksman said:
Don't like the model itself? Develop your own, if you like.

No point in it. It is a waste of time for all of the reasons given above. I'd rather spend my time training and practicing drills, honing my skills. "Button time" on the calculator does not translate to "trigger time" behind the gun. You- no one- needs a model to tell them that 5 rounds < 15 rounds or that no reload < 1 reload.

Carry the most gun and as much ammo as you can and train often. Prepare for the worst, pray for the best. That's the "model" for sucess- no calculator needed. :)
 
Last edited:
Posted by 481: You cannot choose your hit probability. In a real, honest-to-God gun fight, it will be whatever it is. You may miss every shot. You may make every shot. You might not have a chance to even attempt a shot.
True fact.

Not very enlightening, but true.

One chooses (and varies) an assumption of hit probability to learn something about its impact on the likely outcome. Just as one chooses an assumed sales volume to assess projected profitability. Yes, the sales volume will be what it will be--period. But one needs to know how the different variables will drive the results.

Or, if you prefer, one estimates casualty rates when selecting from among alternative military plans. But the causalities will be what they will be.

All this hypothesizing, this fantasizing, it is all just mathematical busy work.
Not if it makes a point, which it does.

Attempts at such modeling are exercises in futilty and they have the property of leading the unintiated and unknowing into the perception that they will have more control, know more about what is about to transpire, and expect something other than what they should from such a dynamic and infinitely unpredictable event.
Only for those who do not understand what they are looking at.

No point in it. It is a waste of time for all of the reasons given above.
So I said.

You- no one- needs a model to tell them that 5 rounds < 15 rounds or that no reload < 1 reload.
Alrighty then.

But what the model showed us was that, no matter which reasonable assumptions one might choose, the prognosis with five rounds is markedly dimmer than many had believed, and that the situation improves markedly when a few rounds are added.

John did not set out to try to make reliable, accurate projections. He simply wanted to know something about the validity of the claim that, since the average number of round expended in a defensive action is reportedly close to five, five should be good enough.

He succeeded very well.
 
Oldmarksman said:
John did not set out to try to make reliable, accurate projections.

That is quite apparent.

Oldmarksman said:
He simply wanted to know something about the validity of the claim that, since the average number of round expended in a defensive action is reportedly close to five, five should be good enough.

The model doesn't tell you anthing that you don't know already- it simply gives you a "probability" that 5 rounds might be enough.

It took all that (the monstrous exercise in statistical gymnastics) to come up with a "definite maybe"?

Hell, I coulda told you that without all of the mathematical pretentiousness and dancing 'round the bush.

Watch...

Five rounds might be good enough or it might not.

See? Easy enough.

Oldmarksman said:
But what the model showed us was that, no matter which reasonable assumptions one might choose, the prognosis with five rounds is markedly dimmer than many had believed, and that the situation improves markedly when a few rounds are added.

No kidding. Having more rounds is better than having fewer. Now there's an earth-shattering revelation!

You need a model to tell you that? :rolleyes:


Oldmarksman said:
He succeeded very well.

He sure did. He succeeded at complicating a simple guess with lots of unnecessary math.
 
Last edited:
Posted by 481: The model doesn't tell you anthing that you don't know already- it simply gives you a "probability" that 5 rounds might be enough.
Really? That's all it tells you? Look again.

It shows what that probability is, within an order of magnitute and given the base assumptions, and it shows that if one varies those base assumptions within reasonable ranges, things do not really get any better. And it shows that that probability is frighteningly low.

No, I did not know that already. Did you?

No kidding. Having more rounds is better than having fewer. Now there's an earth-shattering revelation!

You need a model to tell you that?
No, no more than I need one to know that "5 rounds < 15 rounds or that no reload < 1 reload."

John showed more than that. He showed that no matter which reasonable assumptions one might choose, the prognosis with five rounds is markedly dimmer than many had believed, and that the situation improves markedly when a few rounds are added.

That's not the same thing as saying that more are better than fewer.

On more time: the prognosis with five rounds is markedly dimmer than many had believed, and the situation improves markedly when a few rounds are added.

Wouldn't matter to you, considering how many rounds you carry, but we have had a number of people on this board saying, without having really thought it through analytically, that they "feel comfortable" with five rounds.

I am one of two whom I know of who have changed my habits as a result.
 
Oldmarksman said:
John showed more than that. He showed that no matter which reasonable assumptions one might choose, the prognosis with five rounds is markedly dimmer than many had believed, and that the situation improves markedly when a few rounds are added.

And you needed a model to tell you that?

Oh, wait...

Oldmarksman said:
I am one of two whom I know of who have changed my habits as a result.

...I see that you did.

Nevermind. :rolleyes:
 
Last edited:
Let's tone it down a bit gentlemen.

If the rollyeyes guy is becoming a routine part of your discussion technique.... find a new technique.
 
Back
Top