One shot stops, what's the best criteria?

Status
Not open for further replies.

john l

Member
Joined
Mar 1, 2003
Messages
206
Location
slc, ut
Guys,
If you were in charge of creating a database for one shot stops (oss), then what would be your criteria?

An example: Upper torso hits from the sternum to the neck excluding arms/shoulders. Incapacitation means that within 3 seconds the person is on the ground and not moving. Just an example.

I know that there are many of you out there who think that Marshall and Sanow have flawed ways of collecting their data, so how would you do it?
I still like the idea of using real-life shootings as a reference point against using ballistic gelatin. However, I think that ballistic gel serves a useful purpose. I think that shooting mythical goats to sell your bullet only pisses people off.

And, yes, I am aware that there are many, many variables in each shooting, but I am curious what you all think and why.

Question #2:
Using existing data from Marshal/Sanow and ballistic gel tests, are there any results that directly conflict each other, i.e. a bullet that has performed well on the streets but gel tests conclude that it is a poor performer?
thanks,
john l
 
IMHO, the Marshall/Sanow definition of a OSS is very good. Their flaw was throwing out shootings where more than one shot was made. This would seem to throw out a bunch of one shot failures that required additional shot(s), thus skewing the results for those loads and making them look better than they are.

Ballistic gel is great for comparison, and probably the very best practical way to compare terminal effects. However, unless performance in gel has been correlated to real life, it has very limited value. Fortunately, such comparisons have been done.

Having never shot a mythical goat, I am unaware of the consequences. Shooting real goats has rarely caused a problem, unless the owner objects.

I suspect that you will find the answer to #2 in one of the three Marshall/Sanow stopping power books.
 
45critter,
I am not cracking wise here, but if there was more than one shot to the individual, then I don't see a problem with throwing that incident out, because it wouldn't be a oss.
perhaps I am not understanding what you mean.
I haven't bought any books from Marshall/Sanow because of all the hype, so I asked about the comparison because I figured someone out there knew the answer off the top of their head.
The goat reference is the dubious " Strassbourg" test printed in G&A circa 1993 or 1994.
john l
 
Hey John:

What I meant was it should be counted as a failure to OSS, rather than throwing it out and not counting it either way - in other words the first shot was a one shot stop failure, and the remainder were not "one" shots.

Yes, I know about Strasbourg, but understood they were French Alpine goats, rather than mythical goats.
 
IMHO The Marshal & Sanow data are not flawed from the criteria used. The data is however frequently misinterpeted .

The data was collected to compare bullet effectiveness only . If you use it for that and that only then I believe that it has usefull data.

It is NOT valid at all to say I use round "X" which has a "One Shot Stop" rating of 92% therefore if I shoot someone with this round in the chest with one bullet there is a 92% chance that the person will be "stopped". Again that is not valid at all.

Many of the attacks I have read on M & S were by authors who either did not read or purposlely misquoted the books. That is of course their right to do so but I think actually reading the books where they (M&S) suggest that any round in the upper 1/3 of the ratings is fine as long as it functions 100% reliable in your gun.

I am not saying M&S work is perfect but I do think that many of the criticisms come from misunderstanding the work.

As always I could be wrong.

NukemJim
 
Quote:
-------------------------------------------------
Their flaw was throwing out shootings where more than one shot was made.
--------------------------------------------------

Correct -- that skews the picture.

The correct standard should be stops/encounters. That does away with the problem of determining if a hit was a "torso hit," and goes to the meat of the question -- "What do winners use?"
 
Far too much emphasis is placed on exact numbers ( my 96% is better than your 94%) .The fact is there are many variables in the real world but what can be said is that a 357sig,40S&W, or 45acp, with premium ammo do a fine job of stopping. Remember also that the rule is to shoot and continue to shoot until the BG is no longer a threat..... Only good hits count so the effort should be to place the rounds in the center of mass.....You cannot substitute more power for poor hits and and if a 9mm (less effective than the above rounds) isn't enough for you don't go to +p, go to a more potent round.
 
Using any handgun, there is absolutely no guarantee of a one shot stop using. Even a .45 ACP to the head doesn't.
 
Throwing out instances where there was more than one shot also has the opposite problem of masking possibly more oss. If a typical split is say .15sec for a controlled pair (I'm not a gamer, just guessin';) ) then if a shooting was 2 quick shots, and gets thrown out, the first shot could have been a oss, but the next one followed too quickly to know. Since shooting until the BG is down makes the most sense it seems the data is culled way too much...including all gunfights and classifying rounds by which ones take the least amount of shots too stop would be better, or the "what the winners use" idea.

Using gunfights with one shot only results in a sample size that is way too small.
 
some field notes

here is an interesting set of notes I ran across a while back

at least it has contact info if you are so inclined

<snip started here>


RESULTS FROM THE STREET


Snip.....<sorry extraneous junk ads etc if I recall>


The criteria I established were a set of rules that I felt
comfortable with. They are as follows:

Only torso hits were included. I and co-researcher, Ed Sanow, found
that it was unrealistic to include shootings where the victim was hit
in a non vital area and then use that incident as proof of a
particular rounds ineffectiveness.
Multiple hits were also discarded. I just couldn't come up with a way
to equate, for example, how three hits from a .380 ACP could be
compared with two 2 hits from a .357 Magnum.
A stop was defined as the assailant collapsing within 10 feet. If the
perpetrator had been engaged in an assault, he or she would not be
able to strike any more blows or fire any more shots, whether they
struck their intended target or not.
In order to include a shooting in my study, I had to have access to
the officer-involved shooting files, autopsy or medical treatment
reports, police reports, homicide reports and files, press accounts,
and when possible, conversations with emergency room personnel.
Recovered bullets were either personally examined and photographed by
us, or we were provided with photographs of the bullets.
Interestingly enough, there was often little correlation between the
expansion of hollow points and the final result.
Finally, a minimum of ten shootings were required before a load could
be included in this study. Of course, I would like to have a lot
more, but I had to start somewhere, and fortunately in most cases, I
have significantly larger totals than that.
The collection of a data base of appropriate size was a difficult and
time-consuming process for several reasons. First of all, it was self-
financed, and police sergeants with seven children don't have much
discretionary capitol. Second, there was a reluctance to share data
because departments feared they would see themselves identified in
gun magazines as the source of such sensitive material. It took
almost a decade to develop a network of sources who trusted my
discretion. Third, this was just one of many things that occupied my
time including the completion of a masters degree and service as a
scoutmaster.

It finally came into focus with the publication of "Handgun Stopping
Power-The Definitive Study" published in the early 90's. This work
is, in the opinion of many experts, still the most definitive study
of the various stopping power theories, followed by the inclusion of
what is the best way to determine what are appropriate loads to use
for law enforcement/self-defense.

Let's take a look at the most current results from the street in the
handgun caliber's most often used by law enforcement personnel. Feel
free to draw your own conclusions from the data and I will share my
own opinions with you as well. Perhaps we can find a common ground.


.380 ACP
The .380 ACP has been used in law enforcement for years as either a
plain clothes/off-duty pistol or as a second gun. This caliber has
seen a lot of attention from the ammunition manufacturers. There are
some good performers in this caliber.
LOAD TOTAL STOPS %
Federal Hydra Shok 78 55 71
Cor Bon Jacketed Hollow Point (jhp) 23 16 70
Federal jhp 62 43 69
Winchester Silver Tip 85 52 61
Remington jhp 58 33 57
CCI jhp 63 36 57
Federal full metal jacketed 154 79 51


.38 SPECIAL-2" BARREL
For decades the short-barreled .38 Special revolver was the plain
clothes and off-duty weapon of American police. In spite of the
massive inroads made by semiautomatic pistols, these compact
revolvers are still extremely popular in the second gun/off-duty
role.
LOAD TOTAL STOPS %
Winchester 158gr lead hp 119 80 67
Federal 158gr lead hp 144 97 67
Remington 125gr jhp 104 70 67
Remington 158gr lhp 92 62 67
Federal 125gr jhp 111 72 65
CCI 125gr jhp 62 40 65
Winchester 125gr jhp 70 43 61
Federal 125gr Nyclad hp 36 22 61
Fed 158gr semi-wadcutter 204 101 50
Fed 158gr round nose 381 185 49
.38 SPECIAL-4" BARREL
The four-inch barreled, fixed sight .38 Special revolver was the
police duty hand gun for decades. While the .357 Magnum made some
inroads into its popularity, it was the adoption of the Beretta 9mm
pistol by the US military that sounded its death knell. It offers
good performance in a moderate recoil, proven design.
LOAD TOTAL STOPS %
Cor Bon 115gr jhp +P+* 29 24 83
Winchester 110gr jhp +P+ 36 30 83
Winchester 158gr lead hp 375 287 77
Federal 158gr lead hp 249 189 76
Remington 125gr jhp 144 104 72
Federal 125gr jhp 239 171 72
Remington 158gr lead hp 156 109 70
CCI 125gr jhp 79 56 70
Federal 158gr swc 312 163 52
Federal 158gr lead 504 259 51
* NO LONGER MADE


.357 MAGNUM
The .357 Magnum was the first choice for decades by those officers
who were dissatisfied with the performance levels of the
available .38 Special loads. It was also the first choice of many
highway patrol units who were concerned about penetration of motor
vehicles. It has produced the best stopping power results of any
handgun caliber.
LOAD TOTAL STOPS %
Federal 125gr jhp 556 536 96
Remington 125gr jhp 245 231 94
CCI 125gr jhp 169 154 91
Federal 110gr jhp 239 214 90
Remington 110gr jhp 67 59 88
Winchester 125gr jhp 111 97 87
Winchester 145gr Silver Tip 94 81 86
Remington 125gr jhp-mv 29 24 83
Remington 158gr jhp 45 37 82
Federal 158 gr Nyclad hp 49 40 82
Winchester 158gr swc 117 85 73


9mm
The 9mm is a favorite weapon in law enforcement at the present time.
This weapon is easy to fire, accurate, and reliable. Care should be
given to selecting a model which can be handled safely.
LOAD TOTAL STOPS %
Cor Bon 115gr jhp 35 32 91
Federal 115gr jhp +P+ 149 136 91
Winchester 115gr jhp +P+ 139 126 91
Remington 115gr jhp +P+ 74 67 91
Federal 124gr HS +P+ 85 75 88
Federal 124gr Nyclad hp 265 222 84
Winchester 115gr ST 365 299 82
Federal 115gr jhp 263 216 82
Federal 124gr HS 129 106 82
Remington 115gr jhp 221 180 81
CCI 115gr jhp 149 121 80
Cor Bon 147gr jhp 10 8 80
Federal 147gr HS 461 371 79
Federal 147gr jhp 34 27 79
Winchester 147gr jhp 291 216 74
Winchester 115gr fmj 312 201 64


.40 S&W
The .40 S&W has become the premiere law enforcement caliber. It
serves as a bridge caliber, satisfying those who feel more
comfortable with a bigger caliber than the 9mm and more bullets than
the .45 ACP. It has started to produce excellent results and appears
it will be the law enforcement round of the future replacing
the .45ACP.
LOAD TOTAL STOPS %
Federal 155gr Hydra Shok 56 54 96
Cor Bon 155gr jhp 24 23 96
Remington 165gr Golden Sabre 68 64 94
CCI 155gr Gold Dot 39 37 94
Federal 155gr jhp 34 32 94
Cor Bon 150gr jhp 38 34 93
Winchester 155gr Silver Tip 29 26 90
Remington 155gr jhp 25 22 88
Federal 180gr Hydra Shok 58 51 88
PMC 155gr Star Fire 33 29 88
CCI 180gr Gold Dot 45 39 87
Cor Bon 180gr jhp 22 19 86
Remington 180gr GS 48 41 85
PMC 180gr Star Fire 29 24 83
Black Hills 180gr jhp 34 28 82
Federal 180gr jhp 88 72 82
Winchester 180gr jhp 107 88 82
Winchester 180gr Black Talon 69 56 81
Winch. 180gr full metal jacketed 42 30 71


.45 ACP
The .45 Auto, of course, has an extremely large and emotional
following. Unfortunately, much of this loyalty is tied to the alleged
superiority of .45 ACP hard ball. I used to a believer in it too,
until I started to collect actual shooting data and found that while
there are some excellent loads in this caliber, the 230 grain full
metal jacketed offering is not one of them, regardless of bullet
design.
LOAD TOTAL STOPS %
Federal 230gr Hydra Shok 107 102 95
Remington 185gr Golden Sabre 66 62 94
Cor Bon 185gr jhp 12 11 92
Remington 185gr jhp +P 59 54 92
CCI 230gr Gold Dot 32 29 91
Federal 185gr jhp 114 100 88
CCI 200gr jhp 139 123 88
Winchester 185gr Silver Tip 101 84 83
Winchester 230gr Black Talon 67 54 81
Remington 230gr fmj 145 90 62
Winchester 230gr fmj 201 124 62
Federal 230gr fmj 198 123 62


.45 COLT
The .45 Colt has seen only limited use in law enforcement primarily
due to the introduction of large frame revolvers chambered in this
caliber. It had seen use in those departments who allowed their
officers to carry privately-owned, non magnum revolvers.
LOAD TOTAL STOPS %
Federal 225gr lead hollow point 73 57 78
Winchester 225gr Silver Tip 62 45 73
Winchester 255gr round nose lead 72 50 69
Remington 255gr rnl 24 15 63


The recent violent confrontation in Los Angeles brought cries for
more powerful handguns. While I can understand the reason for such
demands, the facts of this case make it clear that handguns chambered
for .45 ACP would have not performed any better against the heavy
armor the two bank holdup men wore. If we examine carefully the best
performers in the most popular semi-automatic pistol calibers, we can
see that they vary by less than 5%. Anyone who thinks that the felon
shot with a load that is producing one shot stops 96% of the time
will collapse significantly faster than the felon shot with a load
that produces one shot stops 91% of the time is in for a severe
disappointment. Gravity can only pull a body to the ground so fast
and no handgun round produces knock down power.


We need to remember that the three most important components of
handgun stopping power are: bullet placement, bullet placement, and
bullet placement. I carry a double action 9mm Beretta because I shoot
it very well. The fact that I'm giving up five and four percentage
points respectively by not carrying a .45 ACP or .40 S&W, has not
caused me to suffer sleep deprivation.
We should remember that if we know we are going into harm's way, we
should be equipped with an appropriate shoulder weapon. Handguns are
last ditch, emergency weapons that are as much a badge of office as
they are for our self-defense. I cannot see the value of giving up
seven rounds of ammunition in my magazine to gain four percentage
points of stopping power. The .45 ACP is an excellent caliber and I
own several hand guns chambered for this round, but after 20+ years
in law enforcement I've found that the 9mm meets my needs.


It should also be pointed out that the majority of ammunition R & D
is and will be focused primarily on the .40 S&W, at least until the
next super caliber comes along. There are a number of ammunition
manufacturers that are producing new loads and bullet designs almost
every day. All of them, of course, claim their load is the ultimate
stopper and while some of them are working rather well, the one that
appears to hold the most potential is the Quik Shok load from Triton
Cartridge. Designed by Tom Burzinski, who gave us Hydra Shok and Star
Fire bullets, it is a pre-stressed bullet that breaks into three
pieces that go their separate ways causing tissue damage over a much
broader area. Pre-production versions of this load worked extremely
well in the controversial Strasbourg Tests. Of course, we'll have to
wait to see how it works in actual use, but it looks extremely
promising.


CONCLUSION


I am hopeful the information I have provided in this short article
will assist you in selecting a handgun in which you have absolute
confidence. In my opinion, it doesn't make much difference whether
you select a 9mm, .40 caliber, a .357 caliber, or a .45 caliber as
long a you use the proper ammunition. Any weapon of these calibers is
adequate for law enforcement use. My advice is to select a weapon
that you can fire most accurately given the different size, weights,
and recoil of each weapon. I think the data I have provided makes
that point very clear. I also urge you to buy a weapon that is safe
to handle in our rough and tumble world and can be stored at home
without any worry about the potential of an unintentional discharge
by a child. Finally, for God's sake, learn how to handle the weapon
safely and shoot accurately. In the final analysis, safety and being
able to shoot the weapon accurately are the two most important
aspects in the selection of a law enforcement handgun.


The National Executive Institute Associates Leadership Bulletin
editor is Edward J. Tully. He served with the FBI as a Special Agent
from 1962 to 1993. He is presently the Executive Director of the
National Executive Institute Associates and the Major City Chiefs.
You can reach him via e-mail at tullye@a... or by writing to 308
Altoona Drive, Fredericksburg, Virginia 22401

<end quoted>

hope this helps. I'm not much on stats I tend to say hit early and often with what ever you have handy

says the dude with .32 61gr jhp out of a pocket pistol ;-)
 
John, I think you may have opened up a can of worms my friend. :)

My problem with any sort of analysis of OSS is that there is not really a good way to measure the physical or psycological make up of the person getting shot. There are people out there who have died quickly from otherwise non-fatal hits with weak calibers, and there are people from the other end of the spectrum who soak up tons of heavy caliber rounds who refuse to stop for a long time.

I just don't see a really scientific way to get good data from actual shootings.
 
Quote:
----------------------------------------------
I just couldn't come up with a way to equate, for example, how three hits from a .380 ACP could be compared with two 2 hits from a .357 Magnum.
-------------------------------------------------

Quote:
----------------------------------------------
I just don't see a really scientific way to get good data from actual shootings.
-------------------------------------------------

The way to accomplish both is to use stops/encounters as the standard.

That way we answer the question, "What do most winners use?"

Once we have that answer, we can probe deeper. Perhaps we find that the .380 IS more effective than the .38 Special under our standard (Just to pick a possible example at random), and find the reason is that .380 shooters get more hits per encounter.

We are now standing at the point where we can look at issues like how controlability and ergonometrics contribute to winning gunfights.
 
good point

Vern,

that is in essence what is often missing from the debate,

what winners use give light to other factors that need to be address ie spray and pray etc.

r
 
Incapacitation means that within 3 seconds the person is on the ground and not moving.

A lot can happen in three seconds. There's no way in @#$%^&! I'm going to wait three seconds before the first pair of warning shots to the center of mass and the second pair of warning shots to the head if I ever need to defend my life.
 
One problem with including multiple shots as 'failures' is that it would unfairly penalize lighter rounds. Why? Because people are going to get off more shots in the same amount of time with the smaller loads. You'd have to start factoring things like the average number of shots in the 3 seconds it takes any gun to effect a 'stop'. A .40 might average 1.8 shots vs a 9mm's 2.3 (numbers pulled out of ether). As I've only shot a .45 once, I can't really say how much it'd slow me down. I know I shoot my .40 (CZ40) slower than my 9mm(CZ75BD),

I also train for Doubletapping & rapid aimed fire. If a 'one stop shot' is defined as the BG going down in less than 10 ft/3 seconds, I'll have already shot a second or third time into them.

Now a seperate "Stop" study that checks the 'average effectivness in an encounter' would be a good idea. Just don't try to factor it into a One Shot Stop study. A .50BMG handgun might come in first place for a OSS study, but it's not practical in a number of ways.
 
Last edited:
In a hunting situation my definition would be to instantly drop something in it's tracks which did not require a follow-up shot. Why modify that for a self defense situation?
 
Quote:
---------------------------------------------------
In a hunting situation my definition would be to instantly drop something in it's tracks which did not require a follow-up shot. Why modify that for a self defense situation?
----------------------------------------------------

Because it puts us back where we started -- throwing out any encounter where multiple shots were fired.

Remember, on dangrous game, many professional hunters urge their clients to fire again, until the quarry is definitely down, not to sit around and "admire the shot."

Also consider this -- most people don't hunt large animals with handguns designed for concealed carry. When you combine the relatively low power of these weapons with the danger from the opponent, "Shoot 'til he's down" is the obvious tactic.

Why adopt a criterion that penalizes the obvious tactic?
 
Tell me what premise you want to "prove" and I will provide the statistics to support it.

If one deals with statistics, one should read "How to Lie with Statistics" by Darrell Huff. Its a classic in the mathematics literature.

Marshall & Sanow take a very SMALL sample of SELECTED shootings to generate their statistics. By using a SMALL sample instead of ALL shootings they can arrive at results which exhibit a low confidence level when it comes to the real world.

Marshall can be forgiven if he gets the stats wrong. He was a Detroit cop. On the other hand, Sanow has a degree in industrial engineering (some would call it imaginary engineering) from Purdue. IE's deal heavily in statistics and Sanow should know better.

On top of that, there was the Sanow - Dayton, OH PD messup. A web search will provide more info on that SNAFU.
 
There are no magic bullets....

I think that when you figure in all the variables that it's impossible to quantify one shot stops. No small arm known to man, no bullet that fits in a small arm is capable of making one shot stops all the time. The numbers that M&S have come up with are worthless, not only because their science is wrong, but because no two shootings are alike enough to really be comparible.

For defensive purposes, get the most reliable handgun you can find, in any caliber .38 special or larger, choose ammunition that meets the FBI criteria in 10% ordnance gelatin and practice...a lot. Then train for the right mindset. Then train force on force, then practice some more.

Men win or lose gunfights. The equipment used is usually a minor consideration.

Jeff
 
No study's perfect

However, you can attempt to quantify. You've stated some problems like "They don't include all the data". Why did they not use multiple shot data? The additional shots invalidate the data for the purposes of this study. Why? You don't know if the first shot would have been an OSS if they're putting a second bullet in within a second.

As for 'what the winners use' you then get into questions of training and popularity. You'll get lots of data for 9mm and .45, followed by .40 S&W & .38.

What's that boil down to? You're back to carrying the caliber/gun you're most able to shoot quickly & accurately. Because accuracy is the defining characteristic to stopping power. Once you get up into 9mm territory, the percentage differences doesn't seem to mean much, especially if you can get follow-up shots out more quickly.

Here's a test:

Go out to the range with all the guns & ammo you're considering for SD. Set up an IPSC type target out 7-10 yards away. Bring a shot timer. Preferably bring a holster for each.

a:shoot a full magazine/cylinder for each gun using the timer.
b:Add up the number of A&B zone hits. Divide by the time it took you
c:Multiply the number by the OSS percentage.

Highest number wins. Starting from a holster would tend to be compensation for the tendency for smaller calibers to have larger magazine capacity in similarly sized guns.

Anybody have a better idea? I know it sounds alot like IPSC scoring. Heck you could even run an IPSC course and use IPSC scoring. If you don't differentiate minor/major power, then multiply by the OSS percentage, it might be even better. Of course, this will result in firing off quite a bit of ammo. But that's never bad, is it? :neener:
 
Quote:
-----------------------------------------------------------
Marshall & Sanow take a very SMALL sample of SELECTED shootings to generate their statistics. By using a SMALL sample instead of ALL shootings they can arrive at results which exhibit a low confidence level when it comes to the real world.
------------------------------------------------------------

That's why the first studies should use ALL the data -- hence the stops/encounters formula.

Quote:
-------------------------------------------------------------
a:shoot a full magazine/cylinder for each gun using the timer.
b:Add up the number of A&B zone hits. Divide by the time it took you
c:Multiply the number by the OSS percentage.

Highest number wins.
---------------------------------------------------------------

The flaw here is that it assumes the OSS percentages are based on valid data -- which is the issue in question.

What you would get IF the OSS data were correct is a backward-engineered stops/encounters outcome. So why not go there directly?
 
We need to remember that the three most important components of handgun stopping power are: bullet placement, bullet placement, and bullet placement.


I tend to think that Marshall pretty much summed up his position right there...
 
Vern, I think that the method I proposed uses the data in a mannor that utilizes it's value. It shows that using good quality, higher speed loads tends to improve stopping power more than even moving up to a larger caliber. The data shows basically how effective a single shot is at stopping somebody. It should not be used alone. Besides, the difference only seems to vary about 10-20%. Accuracy & speed would have a far greater effect in my testing. It ends up being more of a tiebreaker, or a compensation if you shoot the heavier calibers a bit slower.

What I proposed is a indivdual compensation for the data. You can argue about 'encounter' based evaluation, but what if you're better at a .45 than the average skill level of the user's of the weapons used in the sample? Or worse? Or you can't really handle a .45 effectivly, but are good with the 9mm? This is what I was trying to compensate for. Your self-defense weapon should be carefully selected by you. This study is also useful for determining what rounds/load you might want to carry. Subject to accuracy testing, of course. Once you start getting into the +P+ area, the extra kick might make it equivalent to the higher caliber.
 
Firethorn, the manner you proposed is fine -- IF the OSS data is valid.

But if you had stops/encounters data, you'd be in a position to select the weapon-cartridge combinations that winners use at the outset, and start practicing with them.

As it is, you're still stuck with data that is statistically invalid and which has been questionned by many investigators -- and which has never been released for peer review.
 
Status
Not open for further replies.
Back
Top