Firearms Tactical Institute
Reprinted with express written consent from the International Wound Ballistics Association (IWBA) and the author. Publication of this document on our web site does not constitute an endorsement of Firearms Tactical Institute by IWBA or the author.
Web Site Index and Navigation Center
Reprinted from Wound Ballistics Review; 3(1), 1997: 26-31.
Review By Martin L. Fackler, MD.
Marshall E.P., Sanow E.J.: Street Stoppers --The Latest Handgun Stopping Power Street Results, CO, Paladin Press, 1996. (374 Pages, $39.95)
The authors of this book are gunwriters who have close ties to bullet companies that specialize in lightweight handgun bullets shot at higher than usual velocities. They have published numerous articles, and a previous book, Handgun Stopping Power -- the Definitive Study, which are essentially unabashed advertisements for this type of bullet.
Rather than a "study," their work consists of unsupported pronouncements. A "study" implies serious intellectual work, using accepted scholarly guidelines. A "study" provides a source of verification so that serious readers can peruse the raw data and decide for themselves if the authors have interpreted it correctly. The more marginal the reputation for truthfulness of the authors, the more critical such independent verification becomes. The authors, however, have repeatedly refused to allow any independent review of their "data" -- claiming some strange need to protect their sources.
Fortunately, the great majority of law enforcement groups have ignored the Marshall and Sanow "Definitive Study" and opted for the heavier, slower bullets, which have proved far more reliable than the faster, lighter bullets that they replaced.1
Marshall - Sanow Can't Beat the Long Odds -- Wound Wizards' Tally Too Good To Be True (Soldier of Fortune Jan 94, pp. 64-65) presents the facts in a way that allows the interested layman to comprehend why the impossible regularity of the Marshall - Sanow data has caused professional statisticians, unanimously, to declare it bogus -- or "too good to be true." More recently, Duncan MacPherson has written, "it is almost impossible for a layman with no knowledge of statistics to avoid the 'too good to be true' trap in manufacturing or doctoring data," and "Any claim that WTI (Wound Trauma Incapacitation) can be assessed within a few percent based on combat shooting data is based on ignorance, or fraud, or both."2 "Statistician Dan Watters from the University of South Carolina" is mentioned on page 330 of "Street Stoppers." It appears that Marshall and Sanow are implying that there exists a professional statistician who doesn't burst out laughing when presented with their "data." I was especially interested in this Watters because all of the professional statisticians, who I know have seen the Marshall - Sanow "data," were in clear and unequivocal agreement that the "one-shot stop statistics" were so flagrantly bogus that nobody competent in statistics could believe them to be genuine. I tried to reach Watters: the University of South Carolina directory does not have any such person listed -- and the Department of Statistics there denies that any such person works for them, or has in the past. I remain eager to contact any professional statistician who thinks he or she can support the Marshall - Sanow "data." Further evidence regarding these authors' credibility (or lack thereof) is contained in the details reviews of "Handgun Stopping Power." 3,4
In mid-1992, Sanow made an abortive attempt to delude law enforcement by publishing several articles in the popular gun press in which he claimed that the heavier, slower bullets were failing with great regularity. He included details of about a dozen purported incidents and mentioned the departments involved. In response to Sanow's onslaught on the 9mm WW 147 grain JHP bullet, SGT Mike Dunlap, Rangemaster at Amarillo, TX, PD contacted every department for which Sanow claimed poor results with this bullet in his "anti-subsonic" articles. Mike submitted his results to Law and Order: they showed that Sanow had misrepresented what these departments found. In the November 1992 issue, Law and Order published three letters contradicting Sanow's "data" (p. 90). SGT William Porter, head of the Michigan State Police Marksmanship Unit wrote, "I hope that those who read this article will not be influenced by what Sanow wrote about what happened in the Michigan State Police shooting, because it didn't happen that way." In a note introducing these letters, Bruce Cameron, Editorial Director of Law and Order wrote, concerning Sanow's article, "...we do apologize for printing information that has proven to be in error."
In mid-1993, the results of an authorless "study" done purportedly by shooting more than 600 goats in Strasbourg, France, were circulated, anonymously, throughout the handgun community. A copy of these "Strasbourg Tests" was sent to the Firearms Training Unit of the FBI just before a scheduled meeting of the Wound Ballistics Committee. The committee members, all respected pathologist or trauma surgeons, were unanimous in their opinion that these "tests" were, in fact, a hoax -- and had been fabricated, most likely by somebody without a medical background. A detailed analysis of these tests was published in the Wound Ballistics Review.5
Conceptual Problems in "Street Stoppers"
This book is filled with contradictions, inconsistencies, incongruities, and outright errors. These include:
On page two we find "...there is no substitute for bullet placement...." Yet, the "one-shot stop" concept contradicts the necessity for good bullet placement: a shot that disrupts nothing but skin and the fat of the abdominal wall is counted equal to one that goes through the heart or aorta. This "one-shot stop" concept is of great advantage, however, to these gunwriter/bullet salesmen authors: it says, "buy the bullets we recommend and you can forget about all those hours of practice -- just hit your adversary, with one shot, in any part of the torso, and 96% of them (more than 19 out of 20) will immediately cease their aggression."
On page 201 they wrote "The issue of lethality is strictly one of shot placement; the issue of stopping power is more one of energy transfer." So again, buy our bullets and shot placement doesn't matter.
On page 161, "we have had a number of calls from agencies who...selected loads that looked great in gelatin but proved to be dismal failures in the most realistic laboratory of all -- the street." On page 162, "Bullets recovered from people rarely resemble those recovered from gelatin," and on page 251, "Generally, bullets which expand perfectly in bare gelatin and water are too strong for good expansion in humans." Despite this and other gelatin bashing, we find a photograph of a bullet being fired through gelatin on the cover of "Street Stoppers," and 85 photographs of bullets shot in gelatin throughout the book. Indeed, Marshall and Sanow claim to use shots in gelatin as the basis for their "predicted one-shot stop" percentages. Unfortunately, however, we found no calibration BBs in any of the gelatin blocks shown. Without this calibration none of these shots can be considered valid: gelatin consistency varies greatly with temperature and how it was made. The mandatory calibrating of gelatin is as fundamental to scientific method as verifying the accuracy of a sensitive balance with a known weight.
On page 193 we are told that those who assumed that the FBI Miami shootout would have turned out differently if Agent Dove's bullet had perforated Platt's heart (rather than stopping before reaching it) were wrong. Marshall and Sanow evidently just discovered that some physical activity can occur after a person is struck in the heart by a bullet. Unfortunately, they overlooked the fact that this activity usually ceases within a dozen seconds -- Platt took much longer than that as he ran around while killing two FBI agents and wounding five.
On page 198, "The 147-grain subsonic is a low momentum round." In fact, the 147 grain subsonic possesses more momentum (0.65 lb-sec) than the 115 grain Silvertip bullet (0.62 lb-sec) which it replaced.
Comments Related To Specific Chapters
CHAPTER 4 -- "Strasbourg Goat Tests." Here Marshall and Sanow reproduced the aforementioned anonymous "Strasbourg Tests." In analyzing these purported test results, Marshall and Sanow found an "extremely high rank correlation" with their very own "actual street results." Interestingly, if we compare the shot trajectories in the purported "Strasbourg Tests" with that of the most common shots in humans, we find:
A bullet fired into a goat from side to side, above the heart and behind the shoulder, will pass through or very near the major pulmonary vessels at a penetration depth of three to five inches, and must pass through the mediastinum, either near or through other very large blood vessels.
Conversely, with a shot passing front to back in the human torso, most bullets do not pass near or through the aorta or vena cava until more than six inches of penetration depth in a small slender person and at greater penetration depth in a larger person, or if penetrating at a significant angle.
Due to human anatomy, most shots from the front do not come near major blood vessels. Most go through perforating just lungs near their periphery or just loops of bowel.
Given these facts, the near perfect correlation of Marshall's random torso "one-shot stops" with the purported goat shot results is strong evidence that the anonymous "Strasbourg Test" results have been fabricated or doctored; or the "one-shot stop" results have, or both have.
Some might argue that the "Strasbourg Test" results could be from a real experiment; but one planned with incredible incompetence.5 A few things, however, do not ring true: for example, they mention great difficulty in finding enough goats for the study. Yet, strangely, each of the more than 600 goats found purportedly weighed within four pounds of 160 pounds. Anybody familiar with large animal experimentation realizes that here Marshall and Sanow apparently fell into another "too good to be true" trap.
CHAPTER 5 -- "Navy/Crane 9mm Ammo Tests." This chapter consist of ten pages of "excerpts" from a series of six year old tests done by the US military. Then Marshall and Sanow spent three pages pointing out the Navy researchers' "misunderstanding of the police shooting scenario," pontificating, and correcting "the errors in the cycle testing and in the subjective opinions." The reason for all this quibbling was that the Navy researchers had a 147 grain subsonic 9mm JHP bullet as their top choice. That couldn't be right, it disagreed with the gunwriter/bullet salesmen. So Marshall and Sanow reworked the "numbers" and guess what: a 115 grain +P+ round came out on top. Yes, "reworking the numbers" so that bullets they are touting come out on top seems to be a specialty of these gunwriter/bullet salesmen.
CHAPTER 6 -- "Police Marksman/Fairburn Tests." Here we find the Marshall - Sanow spin on an abortive collection of subjective opinions submitted to Police Marksman magazine purportedly by law enforcement officers. I recall Dick Fairburn calling me before his study started and mentioning that Marshall was very upset at the prospect of such a study. Cases came in slowly, after three years of collection Dick had less than 200 shootings with handguns. He wisely stopped the study, but unwisely Police Marksman published what had been collected. In 1992, I discussed with Dick the problems with his study, which included:
How could he tell that some of the reports sent to him were not being made up? or slanted by the reporter? or misrepresented through ignorance? With his study as described there was no way he could avoid being victimized by those with a need to have his study support the bullets that have already declared to be the most effective.
Any data collection, short of gathering every shooting for a particular time period from a particular police department, is invalid. Unless all cases are included, the unscrupulous investigator can "prove" anything he wants by just selectively including the cases that tend to support his preconceived theory, and omitting the ones that don't.
Here again, Marshall and Sanow appear to have "reworked the numbers" in claiming that Fairburn's study supports their "data." "The 147-grain subsonic JHP is the least effective expanding 9mm caliber according to Fairburn...." Eleven shootings were reported for this bullet: interestingly it had a "50.0%" success ratio. Given the success or nonsuccess type of compilation (yes or no -- no maybes), getting 50.0% outcome from a sample of 11 is interesting mathematics indeed. However, even if we assume that most of the cases sent to Dick Fairburn were honest and unbiased, there are far too few cases to support any valid conclusions for any particular bullet. The only possible indication of any value might come from a look at the comparative percentages for the entire group of cases. One would expect Fairburn's "success ratio" to be higher than Marshall's "one-shot stop" percentages: the "one-shot stop" being a sort of super success. Yet we find just the opposite -- and dramatically so. The overall "success ratio" for all of Fairburn's 187 handgun shooting incidents was only 49% while Marshall's overall "one-shot stop" percentages for the same four handgun calibers was 78%. Yes, I think this, the only statistic with enough cases to assume any kind of validity tells us a great deal about Marshall's "one-shot stop" data.
CHAPTER 7 -- "Royal Canadian Mounted Police Ammo Tests." In this chapter, Marshall and Sanow lifted, verbatim, ten pages of text and half of the data summary from the Technical Report "Comparative Performance of 9mm Parabellum, .38 Special and .40 Smith & Wesson Ammunition in Ballistic Gelatin," by Dean Dahlstrom and Kramer Powley, published by the Canadian Police Research Centre in September 1994. Unfortunately, half the shots in this study were done at a distance of three meters and half at a distance of fifty meters and Marshall and Sanow failed to indicate which half of the data summary they included: the shots at 3 or those at 50 meters -- or possibly a combination of the two. They also failed to indicate which shots were made from a four inch barrel and which from a nine inch barrel.
In Canada, as in the USA, reports from government laboratories are not copyrighted, so the material can be reproduced without violating any law. It is scholarly practice, however, to ask permission of the authors and acknowledge this permission if granted. Permission was not requested from Dahlstrom or Powley. These Canadian researchers are not happy about having their names associated with a book by Marshall and Sanow; they are even more disturbed that their meticulously referenced text was altered by deleting not only the list of references but the superscripts; making it appear as if Dahlstrom and Powley did not use any references. This not only makes the Canadian authors look scientifically illiterate but robs the authors they cited (and quote) of proper recognition for their work. By deleting these references Marshall, Sanow and Paladin Press have published information and quotations from six articles that appeared in the Wound Ballistics Review (which is copyrighted) without acknowledgment or permission.
CHAPTER 8 -- "Secret Service Ammo Tests." Here a "1972 Secret Service Ammo Evaluation" is presented. This seems pointless in view of the changes in handgun ammunition in the past 24 years. This is not the only problem, however, with this data. I visited the Secret Service lab in the early eighties and observed some of the their testing. They were using National Institute of Justice (NIJ) gelatin "standards" 20% gelatin shot at 10 degrees Centigrade. Like everybody but the NIJ, the Secret Service didn't have a refrigerator they could set at 10 C. They kept the gelatin at 4 degrees C and let it sit at room temperature before shooting for a time period they had determined by sticking a thermometer in the middle of the block and determining how long it took to warm to 10 C. It doesn't take a great deal of scientific acumen to recognize that these blocks warm from the outside in, and that when the temperature of the middle of the block is 10 C, most of the block will be at a higher temperature. As with many US Government agencies, the technician doing the bullet testing had no training in science or experimental method, and had no supervision by anybody who did. Any scientist who has worked with ordnance gelatin knows that its characteristics change greatly with temperature, and recognizes that any tests done using the "setting out to warm" method are clearly invalid.
CHAPTER 15 -- "Updated Street Results." The figures presented here allow us to compare the present "one-shot stop" statistics with earlier versions of this same "data base" (from Petersen's Handguns November 1988). Here, again, Marshall and Sanow's "too good to be true" problem rears its ugly head. If, as claimed, they simply increased their "data base" by adding more cases, about as many "one-shot stop" percentages would be expected to decrease as would be expected to increase. The "Updated Street Results," however, show that the "one-shot stop" percentages for 48 of the 60 comparable bullets increased, while only eight decreased (four remained the same). All of those that decreased were found in the lowest two of their respective caliber groups. Marshall and Sanow did not comment on these extremely unlikely ("too good to be true") results. Possibly an alternative explanation is that shooters are becoming much more accurate. No, I forgot, according to Marshall and Sanow, only lethality is dependent on shot placement. "Stopping power" is what they claim to be measuring and, again, according to their dogma, "the issue of stopping power is more one of energy transfer." Could it be that the same bullets, shot at the same velocity, are "transferring more energy" these days? Maybe the ozone hole or global warming has something to do with it.
CHAPTER 19 -- "Black Talon and Winchester Supreme SXT." It is obvious, that in order to explain their "street results" Marshall and Sanow must convince their readers that the reliable and permanent mechanisms by which bullets disrupt tissue, that have been known throughout history, are no longer valid. They have apparently been disturbed by the success of the Winchester "Talon" bullets (which their "street results" show to be "one of the least effective hollowpoints" -- while the Hydra-Shoks are always right up there near the top). Consider (from page 229): "It seems clear that the talons themselves were not adding to the load's overall wounding and stopping power, in spite of the medical sense it made that they should add to the effectiveness." The cutting mechanism of the talons, of course, does add to the effectiveness of this bullet, putting the talon in a class by itself, well above hollow point bullets that have no added cutting mechanism. The denial of this indisputable fact is an almost evangelistic appeal which says to the reader "throw out all your previous knowledge, all learning, all science: we will save you from ignorance, believe in us, we are the sole source of truth regarding bullet effects.
CHAPTER 22 -- This chapter is a 24 page advertisement for Hydra-Shok, Starfire, and Quik-Shok ammunition by the inventor of these "gimmick" rounds. Just like the rest of the book, this advertisement is couched in terms to make it appear to be a valid, objective article by a nonbiased observer. It is clear that the Hydra-Shok, with erect post, gets some extra penetration depth compared to other bullets of equal expanded diameter, velocity and weight. To those who understand that in physics there is no "free lunch" it is also clear that this extra penetration depth can only be gained at the expense of a lessened effective diameter of permanent tissue crush. To recognize this, however, apparently takes more understanding of how bullets disrupt tissue than possessed by most bullet testers: so the Hydra-Shok has sneaked by without having this trade-off detected.
CHAPTER 23 -- "New Ammo: Rhino/Razor, Quik-Shok, and Omega Star." Marshall and Sanow devoted eight pages to the "Rhino/Razor" bullet. Those were the "magic" bullets that hit the national headlines between Christmas and the end of 1994. A Glaser/MagSafe like projectile with the shot held in place with a polymer formerly used to paint airplanes, this bullet was advertised to cause instant death from a hit anywhere in the body -- a real "one-shot stopper." As one might predict, Marshall and Sanow loved it -- wrote eight pages about it: but evidently just before "Street Stoppers" was to go to press the "Rhino/Razor" ammo was found to be in violation of SAAMI pressure standards -- its inventor recalled it and withdrew from the ammunition market. Well, can't win them all -- but at least it filled eight pages!
CHAPTER 24 -- "Effects of Multiple Bullet Impacts." "Street Stoppers," advertised as the sequel to "Handgun Stopping Power," contains the same errors and misconceptions. Again, they have proven Duncan MacPherson right in his assertion that it is nearly impossible for those unfamiliar with statistics to avoid the "too good to be true trap" in manufacturing or doctoring data.2 This chapter purports to show that two shots in the torso have essentially no more effect than one. For the 42 loads listed, two shots produced a 0% higher "one-shot stop" percentage in 8 loads, 1% higher in 18 loads, 2% higher in 10 loads, 3% higher in 3 loads, and 4% higher in 3 loads. This Marshall - Sanow "data" show that two hits produce only about one percent higher "stopping power" than one hit over a wide range of calibers and loads. The absurdity of this "finding" is exceeded only by the preposterous regularity of the "data." Of course, a single shot cannot cause less tissue disruption, in reality, than that same shot plus another hit; but any set of real data has data scatter (especially shots into the human body). If the superiority of two shots is only 1% as shown by the "data," the probability that none of the 42 cases would show better results for the single shot (due to data scatter) is about one in 80 million. To repeat, the chance that these "data" are real and not fabricated or doctored in some way is about one in 80 million.
CHAPTER 25 -- "Shotgun and Rifle Results." Here, some purported "one-shot stop" statistics are given for long guns. They show a 100% success rate for the top two .223s -- but only 98% for all three 12 gauge slugs. The .308 MatchKing averaged between 98 and 99% and was declared "intensely lethal" by our gunwriter/bullet salesmen. That is certainly not the story found in testing or actual practice by police snipers.6,7 This round was anything but "intensely lethal" when it passed through Randy Weaver's shoulder (four inches of muscle tissue) at Ruby Ridge without deforming at all -- or even slowing Weaver down.
APPENDICES -- There are a lot of pages in "Stopping Power" filled with things that appear strangely out of place and have nothing to do with the rest of the book's contents. Appendix "B" consists of an autopsy report describing two shots in the chest and one in the hand. Appendix "C" consists of another autopsy report describing a gunshot wound of the head. These two rather ordinary autopsy reports just sit there, with no commentary or explanation of any reason for their inclusion; just filling pages.
In view of the contradictions, credibility problems, "too good to be true" statistics, and clear indications that the authors of "Street Stoppers" fail to understand the most basic of scientific principles and discourse, nothing in this book can be relied upon. The authors appear to have painted themselves into the corner of the gun culture belonging to the intellectually challenged "true believers." No intelligent reader will tolerate a nonfiction book devoid of references: the astute reader checks references, knowing that even the valid scientific literature authors misinterpret their sources from time to time.
"Street Stoppers" is a compilation of fantasy: written in the arrogant, dead certain tone of the con man. Everything echoes "trust me." The reader is constantly preached to, with evangelistic fervor: and without equivocation implored to believe in nonsense with no basis in established fact. This book is the antitheses of honest, intelligent, scientific discourse in which the evidence is laid out, dispassionately, always with clear references so skeptical readers can check out the sources and raw data for themselves. In scientific discourse readers are not asked to believe, but to consider -- and to think for themselves.
ACKNOWLEDGMENT -- I thank Duncan MacPherson for his help, not only for the probability calculations, but for the entire manuscript and making many useful suggestions.
Fackler ML: "FBI 1993 Wound Ballistics Seminar: Efficacy of Heavier Bullets Affirmed." Wound Ballistics Rev 1994; 1(4): 8-9.
MacPherson D: Bullet Penetration -- Modeling the Dynamics and the Incapacitation Resulting from Wound Trauma. El Segundo, CA, Ballistic Publications, pp. 22-23.
Roberts GK, Wolberg EJ: "Book Review, Handgun Stopping Power: The Definitive Study." Assn of Firearm and Toolmark Examiners Journal 1992; 24(4): 383-387.
Roberts GK, Wolberg EJ: "Definitive Study on Handgun Stopping Power?" Soldier of Fortune, December 1992, pp. 40-41, 69-71.
Fackler ML: "The 'Strasbourg Tests:' Another Gunwriter/Bullet Salesman Fraud?" Wound Ballistics Rev 1994; 1(4): 10-11.
Haag LC: "Federal Premium .308 Win. 168 gr. JHP-BT - a SWAT/HRT Round with Some Idiosyncrasies." Wound Ballistics Rev 1995; 2(2): 8-10.
Fackler ML: "Matchking Bullet -- Past, Present and Future." Wound Ballistics Rev 1995; 2(2): 11-12.
This review is published with the permission of Fighting Firearms who invited the review and originally published it.
Delivering you informative multimedia essays about the "battlefield problem-solving" tactical aspects of armed self-defense.
Web Site Index and Navigation Center
© 2000 Firearms Tactical Institute. All Rights Reserved.
FirearmsTactical™, Salus In Periculo, and logo are trademarks of Firearms Tactical Institute.