TTAG Simulated School Shooting Experiment — Results and Analysis

On Sunday, December 30, 2012, The Truth About Guns joined forces with King 33 Training to perform a series of live fire simulation of active shooter scenarios. We wanted to assess the potential impact of armed personnel on school property during an event like the tragedy in Newtown, Connecticut. This article will serve as the “Key Findings and Executive Summary” of the results from that experiment, providing a single article that discusses the results. It contains a link to the full report at the end . . .

Executive Summary

At the King 33 facility in Connecticut, 11 volunteers and five staff members enacted a series of simulated shooting scenarios with the intent of determining whether an armed teacher or armed guard at a school such as Sandy Hook Elementary would have been able to successfully confront and interdict an active shooter.

When designing the scenarios for this experiment, care was taken to identify moments during the progression of a “typical” active shooter case where armed intervention may have been effective in interdicting the shooter. Three such moments were identified:

1. The moment the shooter enters the school building

2. The moment the shooter enters a classroom

3. The moment an armed response arrives on scene.

One of these scenarios (when the shooter entered the classroom) was enacted both with and without any advanced warning that the shooter was coming. In the first case, the control allowed sufficient time for the teacher to enact a standard “lockdown” procedure such as the one implemented at Sandy Hook Elementary (approximately 15 seconds).

For scenarios where no advanced notice was given, unarmed participants were instructed to leave and re-enter the area being defended at random to simulate normal traffic. This protocol was implemented keep the defender from being able to react to an event (e.g., the door opening) instead of the first sight of a gun or the sound of a gunshot, as would be the case during a real shooting.

Results: Scenario 1

Scenario 1 consisted of a single armed guard seated behind a desk, facing the entryway. This position is consistent with many current configurations for schools, enabling the security guard to watch the front door and challenge visitors to the school during the day.

Unarmed participants were then walked out an opening in the wall and past the guard until the shooter emerged from the opening and engaged the guard. The guard was not aware of exactly when the shooter would emerge and therefore had no specific advanced warning.

[This type of scenario attempts to replicate situations where the shooter is carrying the firearm concealed into the school, or the security guard does not have sufficient warning through cameras or other means to alert them that an armed intruder is about to enter the school.]

According to the results from our testing, in this scenario the armed security guard appears to be ineffective at interdicting the active shooter. While the “shooter” can choose their moment to attack and prepare themselves to fire the moment they are seen by the security guard, the guard has insufficient time to respond.

In future experiments we would like to investigate the effectiveness of an armed guard when there is advanced warning, such as would be provided if the shooter was visible through a door.

Results: Scenario 2

Scenario 2 involves a shooter making entry without warning into a classroom setting where the teacher is carrying a concealed weapon. As previously mentioned, students were encouraged to exit and enter the room at random to keep the teacher from knowing when the shooter would enter the room. The teacher had no specific advanced warning of the shooter’s presence, and was not given any instructions or training on how to react.

The results of this scenario were mixed. They depended on the level of training of the shooter and teacher.

For example, in round 2 participant #5 played the role of the “teacher.” Participant #5 had prior military experience and a current concealed carry licensee. Participant #1, an active member of the military, played the role of the “shooter.” While the shooter inflicted casualties in the room, the teacher was able to successfully engage the shooter and keep them from entering the room.

By contrast, round 5 pitted a national-level competition shooter as the “shooter” against an older retired law enforcement officer as the “teacher.” The shooter was able to neutralize the teacher before he could even begin to draw his weapon.

While not all “teachers” were able to successfully stop the shooter, the data suggests that armed teachers would be beneficial in situations where there is no advanced warning.

If the teacher is ineffective then there is no difference from the baseline “risk” associated with a shooting scenario such as this; the natural tendency seemed to be for the students to retreat and not get in the line of fire between the teacher and shooter. Thus, the probability of students being shot accidentally by a firefight in the classroom is very small.

If the teacher is effective in stopping the shooter (as they were in some situations) there is a definite decrease in the number of potential casualties in the room, as the threat has been completely removed.

The experiment did not investigate the possible deterrence effect of an armed teacher or on-scene first responder. It’s important to note the probable psychological effect that the knowledge of an immediate armed response would have to prospective active shooters.

Future experiments should investigate whether the direction that the door opens has an impact on the effectiveness of the teacher.

Results: Scenario 3

Scenario 3 was identical to the configuration of scenario 2 with the exception that the classroom had approximately 15 seconds warning that there was an active shooter on scene. When the teacher heard gunshots, they were instructed to perform a basic “lockdown” procedure as implemented in some schools: gathering the children in one corner of the room before attempting to interdict the shooter as they entered the room.

In every instance where scenario 3 was played out, the “teacher” successfully prevented the shooter from entering the room, often without a single casualty among the “students” or even a single round fired on the part of the “shooter”. In this scenario, an armed teacher was extremely effective in stopping an active shooter.

Future experiments should explore the difference in shooter’s behavior when they are uncertain that there is an armed teacher.

Results: Scenario 4

Scenario 4 began with a shooter entering a classroom full of students with no armed teacher present and opening fire. On hearing the gunshots, an armed responder (either an armed guard or another teacher, for the purposes of this experiment it was not important) who was standing 25 yards from the classroom door was instructed to rush to the scene of the shooting and engage the shooter.

Like scenario 2, the successful interdiction of the shooter depended on the level of training of the participants.

When responders were participants with high levels of training, such as a military background or law enforcement training, they were successful in immediately engaging and stopping the shooter.

However, responders with little or no training the results were notably less capable. Not only did some responders fail to effectively engage the shooter, in one instance they actually shot a “student” fleeing the room instead of the shooter.

For this scenario, an armed response was only completely effective when the responder had adequate training. Those with insufficient training were able to engage the shooter as well, but results were less impressive.

Conclusion

This experiment was a preliminary test, providing a proving ground for the methodology and scenarios selected for testing before being implemented in a large scale test at a later date.

Based on the limited data collected from this experiment it appears that an armed teacher would save lives in an active shooter scenario. The caveat: the teacher’s effectiveness depends on their level of training. Maximum effectiveness of an armed teacher of any skill level is achieved with advanced warning of the approaching shooter and implementation of a classroom “lockdown.”

Click here for the full report.

avatar

About Nick Leghorn

Nick Leghorn is a gun nerd living and working in San Antonio, Texas. In his free time, he's a competition shooter (USPSA, 3-gun and NRA High Power), aspiring pilot, and enjoys mixing statistics and science with firearms. Now on sale: Getting Started with Firearms by yours truly!

57 Responses to TTAG Simulated School Shooting Experiment — Results and Analysis

  1. avatarCurzen says:

    Did you have a hundred kids run around there as well and tried it with your average teacher instead of what appears to be (semi)professionals to get somewhat usable results?

    • avatarC says:

      “semi-pro” baddies vs “semi-pro” goodies are going to be roughly analogous to newbie B’s vs newbie G’s

      • avatarEd says:

        Except semi-pros will artificially suppress the collateral damage results, a very dangerous oversight considering the purpose of the study.

        If a semi-pro responder killed a kid, how well can we trust amateurs to do the same?

        • avatarSteve says:

          You’re right, let’s just sit there and do nothing while the bad guy kills everyone at their leisure, because that works so well.

  2. avatarStacy says:

    Kids are pretty good at not running around with their hair on fire when they’ve practiced a particular response. That’s the benefit of lockdown drills, fire drills, etc.

    Even though the results are pretty well what you’d intuit given the scenarios, it’s nice to have them in hand, especially along with the videos. I was afraid TTAG would get itself in hot media water for doing this in CT, but I’ll be the first to say I’m glad they did it (and that the media crucifixion didn’t materialize)

    • avatarIng says:

      At least the crucifixion hasn’t happened *yet*. If the national networks pick up on it at all, it’s bound to be negative.

      I’m betting the likelihood of national media picking this up is inversely proportional to the quality of the evidence. If it has usable information and unbiased analysis, they’ll ignore it, just like they ignore everything else that doesn’t fit their preconceived storylines.

      • avatarCasey T says:

        The media did pick up on it, as I saw an article on Fox News about it. The grabbers didn’t take hold though. I thought this was going to get crucified and I am elated that I was wrong.

      • avatarMariana says:

        excellent article I’m cntemoming to let you be aware of of the amazing discovery my child developed browsing the blog. She came to find a good number of details, including how it is like to have a great coaching character to get a number of people clearly completely gr

    • avatarWilliam says:

      I don’t have any data on drill effectiveness. Does anyone here? It’s my notion that the real thing causes panic, while they KNOW a drill when they see one, and behave as expected.

  3. avatarSaul Feldstein says:

    If you have kids, home school.

    Its borderline child abuse to send them to these govt brainwash camps anyway.

  4. avatarTRP says:

    Right on Saul. Mine have both attended private schools, and a relatively conservative public high school, so they are less “brain washed” than most. We have also had a full time police officer at the local high school for many years. Not that the anti-2A loones would care, but there have been ZERO major incidents.

  5. avatarimrambi says:

    Have you thought about partnering with a local PD and rerunning drills? One of the PDs near me just did some training to respond to a school shooter.

  6. avatarJosh says:

    This study is a great start. I commend TTAG for attempting this, and the study appears to have been performed as well can be expected scientifically by (please don’t be offended) amateur researchers.

    The sample sizes of the scenarios and control groups were much too small to minimize variables and it wouldn’t be prudent to determine generalized conclusions or statistical relevance. Yet.

    But it lays the groundwork for additional iterations to be run and then peer reviews when enough data has been collected. The observation that training appears to play an extremely important role in success/failure was probably not something many readers here expected, and I think it would be a good idea to run more scenarios to nail that hypothesis down better. It gives us all something to think about, that the “arm the teachers!” call needs more thought and study. Hopefully the NRA studies will be thoroughly conducted as well to draw meaningful solutions.

    So a foundation has been laid. I hope TTAG doesn’t just say “Mission Accomplished” but instead gathers support to continue the study to really extract something meaningful. My best advice would be: Start out with a hypothesis and then design your experiments to try to prove your hypothesis wrong. If you fail to prove yourself wrong, then you may be onto something. Please don’t fall into the new age political scientific method of trying to prove yourself right and succeeding. Those results will almost undoubtedly be politically useful but scientifically useless.

    Keep up the good work!

    • avatarg says:

      +1

      As a teacher, the results are too surprising to me, though as Josh mentioned, more tests + more people would probably improve the test validity. This is good start, though.

      Along with the door variable, what about the use of desks / tables as barricades to provide better concealment for the students / teacher? I also noticed in one of the videos, one of the students held up a chair to protect himself. Was that part of testing protocol?

      Great work, Nick, and TTAG!

    • avatarSanchanim says:

      I think this is just the tip of the iceberg Josh.
      Nick myself and many others know that there will need to be bigger case studies and further peer reviews.
      To that end it would be great if we could organize real teachers that actually went through the CCW style classes that they would get being teachers, and see how they fair. If anything it might help improve training methods and ideas. It would also, while a bit scary, be great training for the teachers themselves!!
      This is only the beginning but an important first step…

  7. avatarMatt in SD says:

    Excellent work with an honest analysis. I hope more exercises like this is done. I’m sure you guys already have ideas that would make running these scenarios better and more accurate for data collection. It’s just sad to me to think that it has taken nearly 15 years since Columbine and other tragedies for anyone actually explore this.

  8. avatarDon says:

    What can’t be measured is the deterrent effect on crazy people if it is known that there may be a armed security guard and staff in schools. Since they seem to be drawn to known gun free zones, it is possible that just the well publicized change in policy to have armed staff would move the target to another gun free zone like shopping malls.
    Remember (the joke), you don’t have to outrun the bear, just your friend that is also trying to outrun the bear.

    • avatarMatt in SD says:

      I hope this gets forwarded, with video, to every police department, school superintendent, principle, and in the country. A huge step in the right direction for keeping our kids safe. Not only does this lay the foundation for effectively dealing with an active shooter situation in a school but it also goes leaps and bounds towards deterring someone from trying.

  9. avatarOHgunner says:

    Oh if only some benevolent organization had only offered to train the educators for free, we could have a proven safeguard for our children…. Oh wait, the EVIL NRA already has! But I thought they ate children by the school bus load and cause all crime? How can this be? ;)

  10. avatarMilsurp Collector says:

    I regret saying this whole simulation was a horrible idea and I’m damn proud of you all for going through with it. You did a thorough job, and as a bonus, didn’t get royally crucified by the media. Cheers.

  11. avatarjwm says:

    It appears to me that all these scenarios were set up based on 1 armed responder per school. The security guard being killed by gunfire should bring a “Dog Pile” response by armed staff and volunteers against the shooter before he gets to far into the school.

    Also, you had no scenario where there were community volunteers on patrol outside the school building. The patrol may scare off or engage the shooter before he gets near the kids.

    As I’ve stated before. I’m retired. X military with weapons training and experience. Not more than a month ago I was finger printed and FBI background checked, not for the first time as I’ve worked a lot of years in jobs that require a clean background as a minimum.

    Commentor WC made the statement that she did not want a bunch of George Zimmerman’s running around the schools armed. I am not George Zimmerman and every community in America has resources just like me that would be more than happy to protect our schools at little or no cost to the taxpayers.

    But what are the chances that the anti’s would get behind a program that would actually secure our kids and schools? I predict 0.

    • avatarGregolas says:

      +1 jwm! I’m prepared to train others if needed.
      Thanks for the SUPER job, TTAG crew!

    • avatarSanchanim says:

      JWM and I are in the same boat.
      To that end, you are right they didn’t go through every possible scenario, but this was a first step. With limited people time and money it is the best they could do..
      My hope is this will get the ball rolling…

      JWM and myself know that there needs to be multiple rings of defense. Maybe it is just a guard at the front door at first, followed by faculty who could be anywhere in the halls library etc. The LAST line of defense is the teacher standing between the madman and their students.
      In most case with a multilayer-ed approach there is a good chance that any situation would essentially be defending with warning which as you saw by their tests was very successful!

      • avatarLarry says:

        Let’s discuss the costs, here. How about a couple thousand for a comfortable lounge area in the library, a couple hundred for identifiable name tags or vests, and untold zillions to remove the damn “gun free school zone” laws, and every school in the country could have armed response hanging out in their library. Not saying that would be the best approach, but was it even CONSIDERED by these anti fruitcakes?

  12. avatarJason says:

    I’m curious in any iteration was the presence of an armed person ever a detriment to safety? You mentioned a bystander was shot by someone playing the first responder role. I’d be curious to know if the casualty rate was higher in that iteration then the iterations with no armed personnel.

    Also did any of the experiments look at possibility of escape. One would think that even if the shooter isn’t stopped, the process of engaging an armed person would allow more bystanders to be able to escape. Not sure if the experiments allowed for variables such as “students” exiting windows or other means of escape. Even if an armed guard didn’t stop the shooter at the entrance, the act of stopping to engage the guard, or even the fact the first shots happen before the shooter even can target children, could give valuable seconds of warning in a real world scenario.

    It’s good to see a lot of the theories actually be tested.

  13. avatarO.E says:

    Good training, however none of the fleeing students tackled the armed man with the police outfit I say this because we are all aware of what has and continues to happen in these times. Some spree shooters have and may choose to continue wearing the uniform of 2nd responders!

    There are Wolves in Sheep’s clothing.

  14. avatarMark says:

    Interesting that testing showed the very results one would expect.

  15. avatarDerryM says:

    Nice work, gents! These results indicate that some scenarios which could be implemented in the real world would work better than others. It would be good if another Test could be performed focusing the scenarios that worked best, more “Students” in the classroom, and (Dare I suggest it?) younger, smaller volunteer students. Also, the classroom door swing aspect is probably more critical to determining what would work best.

    Obviously, the assailant benefits from the good old “element of surprise”…to paraphrase Monty Python, “…No one expects an armed assailant..”.

    You guys are to be commended for undertaking this Project, especially in a time of high emotionalism and irresponsible rhetoric flying about post Newtown Tragedy.

  16. avatarGyufygy says:

    Sometimes, I like being proven wrong. Yay for no media incidents! Also, yay for simulations!

  17. avatarBob says:

    Good data. Would be curious to see what could be done to minimize the skill side of the skill=positve results equation. Does data sugggest it was a lack of training with the handgun (accurate shot placement) or lack of training engaging targets in an intense situation?

  18. avatarLance says:

    Seems interesting wish you had a scenario where you had more than one armed guard.

  19. avatarWilliam says:

    The question I don’t see anyone asking is, “whose opinion are you expecting to sway with this study”? ‘Cause I can GUARANTEE it will be ignored by any and all of officialdom.

    It’s nice to have the data and know the truth, but is there anyone else who wants to be confused by facts?

  20. avatarLoyd says:

    A buddy of mine sent me a video a few days ago of ABC trying to do similar to this, but it was full of fail all around, trying to prove the point that CCW is dangerous. I just sent him this link to show how its really done.

    • avatarGyufygy says:

      If it’s the same video I’ve seen, all it proves is that being ambushed and flanked sucks. Great insight there, ABC.

    • avatarlolinski says:

      You mean the video where the students were wearing long wool sweaters(which make it harder to reach for a gun) and restrictive paintball helmets?

  21. avatarmarkshere2 says:

    Still no discussion on the common thread through most of the non terrorist shootings:
    prescription and illegal drugs: ritalin, prozac paxil and so forth .

    Here’s a list of similar shootings – see if you can spot a common thread.

    · Columbine mass-killer Eric Harris was taking Luvox – like Prozac, Paxil, Zoloft, Effexor and many others, a modern and widely prescribed type of antidepressant drug called selective serotonin reuptake inhibitors, or SSRIs. Harris and fellow student Dylan Klebold went on a hellish school shooting rampage in 1999 during which they killed 12 students and a teacher and wounded 24 others before turning their guns on themselves.Luvox manufacturer Solvay Pharmaceuticals concedes that during short-term controlled clinical trials, 4 percent of children and youth taking Luvox – that’s 1 in 25 – developed mania, a dangerous and violence-prone mental derangement characterized by extreme excitement and delusion.

    · Patrick Purdy went on a schoolyard shooting rampage in Stockton, Calif., in 1989, which became the catalyst for the original legislative frenzy to ban “semiautomatic assault weapons” in California and the nation. The 25-year-old Purdy, who murdered five children and wounded 30, had been on Amitriptyline, an antidepressant, as well as the antipsychotic drug Thorazine.

    · Kip Kinkel, 15, murdered his parents in 1998 and the next day went to his school, Thurston High in Springfield, Ore., and opened fire on his classmates, killing two and wounding 22 others. He had been prescribed both Prozac and Ritalin.

    · In 1988, 31-year-old Laurie Dann went on a shooting rampage in a second-grade classroom in Winnetka, Ill., killing one child and wounding six. She had been taking the antidepressant Anafranil as well as Lithium, long used to treat mania.

    · In Paducah, Ky., in late 1997, 14-year-old Michael Carneal, son of a prominent attorney, traveled to Heath High School and started shooting students in a prayer meeting taking place in the school’s lobby, killing three and leaving another paralyzed. Carneal reportedly was on Ritalin.

    · In 2005, 16-year-old Native American Jeff Weise, living on Minnesota’s Red Lake Indian Reservation, shot and killed nine people and wounded five others before killing himself. Weise had been taking Prozac.

    · In another famous case, 47-year-old Joseph T. Wesbecker, just a month after he began taking Prozac in 1989, shot 20 workers at Standard Gravure Corp. in Louisville, Ky., killing nine. Prozac-maker Eli Lilly later settled a lawsuit brought by survivors.

    · Kurt Danysh, 18, shot his own father to death in 1996, a little more than two weeks after starting on Prozac. Danysh’s description of own his mental-emotional state at the time of the murder is chilling: “I didn’t realize I did it until after it was done,” Danysh said. “This might sound weird, but it felt like I had no control of what I was doing, like I was left there just holding a gun.”

    · John Hinckley, age 25, took four Valium two hours before shooting and almost killing President Ronald Reagan in 1981. In the assassination attempt, Hinckley also wounded press secretary James Brady, Secret Service agent Timothy McCarthy and policeman Thomas Delahanty.

    · Andrea Yates, in one of the most heartrending crimes in modern history, drowned all five of her children – aged 7 years down to 6 months – in a bathtub. Insisting inner voices commanded her to kill her children, she had become increasingly psychotic over the course of several years. At her 2006 murder re-trial (after a 2002 guilty verdict was overturned on appeal), Yates’ longtime friend Debbie Holmes testified: “She asked me if I thought Satan could read her mind and if I believed in demon possession.” And Dr. George Ringholz, after evaluating Yates for two days, recounted an experience she had after the birth of her first child: “What she described was feeling a presence … Satan … telling her to take a knife and stab her son Noah,” Ringholz said, adding that Yates’ delusion at the time of the bathtub murders was not only that she had to kill her children to save them, but that Satan had entered her and that she had to be executed in order to kill Satan. Yates had been taking the antidepressant Effexor

    Read more at http://www.wnd.com/2013/01/the-giant-gaping-hole-in-sandy-hook-reporting/#OCcQVI97kGo3doqc.99

    • avatarQuiet Professional says:

      The real common thread is not the medication, but the underlying condition being treated.

  22. avatarJon says:

    I was just thinking that a study like this needs to be done. I would also suggest comparing different types of weapons used by the shooter. Does arming the shooter with a gun legal under the assault weapons ban have an effect on the outcome of a shooting situation? Do variables like high capacity magazines and assault weapon features really decrease the number of casualties? I would hypothesize that casualties would be decreased much more by the shooter meeting armed resistance than by changing the features of the gun he uses. Keep up the research, we need real data and answers. We need to stop making laws based on the feelings of people uneducated about firearms and their use.

    • avatarLarry says:

      I hear you! Let’s start by grinding the bayonet lug off an AR, or removing the flash suppressor. I can’t recall anyone using a mounted bayonet, and who needs a flash suppressor in broad daylight when you’re guaranteed to be the only one armed? But there was the Clinton AWB, worried about that terrible bayonet lug.

  23. avatarEd Rogers says:

    If you are able to create another group of scenarios, have more cameras and have the operators plan their vantage points. It wouldn’t hurt for them to plan their editing as well. You wasted some valuable camera (and therefore training) time.
    Nonetheless, thanks for making a great effort!

  24. avatarBen says:

    With the conversation going on about letting college students with CCWs carry on campus, I’d love to see another scenario:

    Classroom of 20-25 students, 1 or 2 carrying concealed, professor unarmed. Have normal traffic of students through the classroom. Without announcement or prior warning, have a shooter enter the classroom and open fire.

    Seeing the effectiveness (or lack thereof) of randomly armed students in a chaotic situation such as this would be very interesting, and could help convince colleges to allow concealed carry.

  25. avatarMark says:

    ++++Will we never learn from history? I’m 62, when I was 13 or so there were headlines every couple of weeks about Palestinian gunmen taking over an Israeli school and murdering the teacher and children. After 6 months of this, the Israeli government put every teacher through military training and gave every teacher an Uzi. They did this very publicly so everyone, including the Palestinians, knew about it. The result, no more school takeovers and no children killed with any teachers gun. Amazing, dozens of fully automatic firearms in each school and not one casualty.

  26. avatarJoe Kuhle says:

    This “demonstration” was so flawed, I don’t know where to begin. All one really needs to read here are two things: “Nick Leghorn is a gun nerd” and “Ted Nugent reads TTAG.”

    The next time you want to “demonstrate” how guns in schools will make a difference, use a Real School, (when school is in progress), use Real teachers teaching Real students. Give the school administrator a simulated gun or paint gun. Tell him or her that an “assault” will happen on his/her school sometime in the next two weeks, but do not give the exact day or time. Adam Lanza didn’t phone ahead and tell them he was coming! I’ll bet my results would show something entirely different than that of TTAG and King 33 Training.

    • avatarQuiet Professional says:

      You ‘ve got to be kidding. A real attack could happen during your weeks long test period. You’d have a test subject engaging a real attacker with a paintball gun.

    • avatarAntiCitizenOne says:

      If you think your method is better, why haven’t LEO Departments tried that kind of thing when school actually WAS in session?

  27. avatarQuiet Professional says:

    One problem with this study: no control scenario. One scenario should have required NO armed response on the part of a teacher or guard.

    • avatarLarry says:

      We have had many of those, Sandy Hook was an excellent example, and was pretty realistic. The “test shooter” killed and killed and killed until he got bored. What else is there to test?

  28. Pingback: UPDATE: Arapahoe HS Armed School Resource Officer Confronted Shooter | Families Fighting Cancer

  29. Pingback: UPDATE: Arapahoe HS Armed School Resource Officer Confronted Shooter

  30. Pingback: Study Shows Teachers with Guns Keep Students Safe | Mississippi Gun News

  31. Pingback: It Should Have Been A Defensive Gun Use: Mary Sherlach Edition | The Truth About Guns

Leave a Reply

Please use your real name instead of you company name or keyword spam.