The Physics Of Mass Killing

The Physics Of Mass Killing

How can deaths from mass killings be reduced? The math of mass killing is simple: the number of people killed is the number of events multiplied by the number of deaths per event. Mental health monitoring may reduce the number of events but it will not make it zero. Number of deaths per event is a crucial variable.

Firearm Physics

The physics of firearms has three main components: a weapon’s muzzle velocity, the mass of its ammunition and the rate at which ammunition can be fired.

Semi-Automatic Rifles

Semi-automatic rifles fire bullets (or rounds) between 5 and 7 mm in diameter. They have a muzzle velocity of 3200 feet per second (meaning they fire bullets at 2200 miles per hour, almost three times the speed of sound) and, in unmodified US civilian versions, the rate of fire is the same as the rate at which the trigger is pulled, up to a theoretical maximum of 800 rounds per minute, but more likely in practice to be 40 to 100 rounds per minute. Military tacticians call this high rate of fire rapid or intense. Aiming rapid fire is imprecise, even for a professional, because of the instability caused by kinetic energy imparted to the weapon by the rapid fire mechanism, and because there is no time to retarget between rounds. This is less of a problem with multiple targets grouped in a confined space – the rapid fire rate increases the probability that some proportion of rounds will hit and, at this muzzle velocity, bullets have long range and will typically penetrate beyond first targets and most solid cover.

Semi-Automatic Pistols

Semi-automatic pistols come with a broad a range of physical characteristics. To choose one example: the Sig Sauer P228 (also called the M11 in the US) fires 9mm rounds at a muzzle velocity of 1500 feet per second (1000 miles per hour), takes a 15 round magazine and can be used for rapid fire in semi-automatic mode. This empties the magazine quickly, but as pistols are generally light, mobile weapons, reloading is faster than with a semi-automatic rifle, as is retargeting after reloading.


Shotguns fire small multiple projectiles (“shot“) that spread out in flight (sometimes single “slugs” that do not). They have a muzzle velocity of around 1600 feet per second (1100 miles per hour) and a low rate of fire relative to a semi-automatic weapon. The low fire rate is offset in part by the spreading shot hitting multiple grouped targets, but as these projectiles are much smaller than bullets they may wound, not kill, and may not penetrate beyond initial victims or solid cover, despite their initial velocity.

Weapons Physics Most Likely To Cause Death In Mass Killings

Because of these physical characteristics, the mosts deaths in a civilian mass killing are likely to be caused by a semi-automatic rifle set to maximum fire and fitted with a high capacity magazine. This will fire 40 or more rounds a minute with high penetration and long range. A semi-automatic pistol such as a Sig Sauer P226-variant (of which the P228 described above is an example) is a good sidearm: if the semi-automatic rifle  jams or empties, the pistol can be brought to bear quickly with some loss of range but little reduction in rate of fire. A shotgun (or other weapon type, such as a revolver, or bolt action rifle) is less efficient than either the semi-automatic rifle or the semi-automatic pistol. It will kill fewer people per minute because of its lower muzzle velocity and lower rate of fire.

Weapons Used In Recorded Mass Killings

This conclusion is supported by real world data. All perpetrators in the the 30 US mass shootings since 1984 used semi-automatic pistols and / or semi-automatic rifles coupled with high capacity magazines – 15 to 20 rounds in the pistols, 40 or more in the rifles. (Summary data here.)

Effective Rate Of Fire

A weapon’s effective rate of fire is the product of its rate of fire multiplied by its magazine capacity, less reload time. How does effective rate of fire impact the number of deaths in mass killings?

Effective Rate Of Fire In Recorded Mass Killings 

Two extreme rates of fire in the real world data set are the Virginia Tech mass killing by Seung-Hui Cho and the Tucson mass killing by Jared Loughner. Effective rate of fire is important because most rounds fired at rapid or intense rates miss their targets.  Cho used semi-automatic pistols with 15- and 10-round magazines and fired 176 rounds in 11 minutes, an effective rate of fire of 16 rounds a minute . In Tucson, Loughner used a semi-automatic pistol with a 33 round magazine, more than double the capacity of Cho’s. Loughner fired 33 rounds in 15 seconds, an effective rate of fire of 132 rounds per minute, more than 8 times faster than Cho.

The Impact of Effective Rate Of Fire On Number Of Hits

Hit efficiency is the number of unique targets wounded or killed for each round fired. Many hits on the same target are counted as one hit; hitting many targets with a single round counts as many hits. At Virgina Tech, Cho killed 32 people and wounded 17 for a hit efficiency of 28% per round discharged. In Tucson, Loughner killed 6 and wounded 13 for a hit efficiency of 58%. Loughner’s higher hit efficiency was almost certainly a result of closer range and less reaction time for his victims. Both gunmen had the same kill efficiency: the number of people they killed was 18% of the number of rounds they fired. This is typical: the average hit efficiency in mass killings is 40%; the average kill efficiency is 21%. The chart below, Shots → Hits → Kills, shows hit efficiency in the Tuscon and Virginia Tech mass killings, the average hit efficiency of all mass killings in the data set, the 5 mass killings with the highest kill efficiency and the 3 mass killings with the lowest kill efficiency.

Shots → Hits → Kills

The Impact of Effective Rate Of Fire On Number Of Deaths

This low hit efficiency is in part a result of the physics of rapid fire. (Other factors include range, how long the incident lasts and the skill and experience of the gunman.) Because such a large proportion of rounds fired miss all targets, firing more rounds per minute increases hits and kills, especially in the first minutes of an attack, when the gunman has the advantage of surprise. This is supported by the data from real incidents.

Rounds Per Minute vs. Victims Per Minute

Rounds Per Minute vs. Kills Per Minute

The mass killing in Tucson is an outlier which creates a stronger dependence between these variables than would exist otherwise. With the Tucson data excluded, there is a positive but less dependent relationship between rate of fire and rate of victims killed.

Rounds Per Minute vs. Kills Per Minute, Excluding Tucson

The Impact Of Reloading On Rate Of Fire

Because kill rates tend to be higher when more rounds are fired per minute the number of times a gunman has to reload per minute and the length of time spent on each reload is a crucial variable affecting kill rate. Time spent on each reload is relatively fixed and dependent on the gunman’s skill, experience and tactical situation. It is independent of the number of rounds in the magazine. The reload time for a skilled civilian using a semi-automatic weapon is probably between 4 to 6 seconds (measured from last bullet on target before reloading to first bullet on target after reloading), possibly longer under pressure. Two hypothetical shooters are modeled below (Magazine Capacity vs. Rounds Per Minute): a “more skilled” gunman who fires 100 rounds a minute and reloads in 5 seconds; and a “less skilled” gunman who fires 50 rounds a minute and reloads in 10 seconds. With a 10 round magazine, the more skilled gunman fires 54 rounds and spends 27 seconds reloading; the less skilled gunman fires 27 rounds and also spends 27 seconds reloading. With a 100 round magazine, the more skilled gunman fires 100 rounds and the less skilled gunman fires 50 rounds, with no time spent reloading.

The Impact Of Reloading On Victim Defense And Escape

Reloading also provides an opportunity for potential victims to escape or defend themselves. The US mass killing data set describes incidents caused by 32 shooters. 25 of these were not apprehended in the act: they either committed suicide, or surrendered or were captured after they finished killing. Two shooters were shot dead and one was wounded by police officers and one was shot dead by a military police officer. Three, including Loughner, were tackled by unarmed civilians during reloading. In summary, the same number of mass killers have been neutralized by unarmed civilians during reloading as have been shot by police. In other incidents, such as the 2008 mass killing at Northern Illinois University, potential victims escaped during reloading.

Magazine Capacity vs. Rounds Per Minute

The Impact Of Carrying A Concealed Weapon (CCW) On Victim Defense And Escape

Data about the impact of legal concealed carry weapons on general crime rates has been used to show both that civilians carrying a concealed weapon, or CCW, reduces crime and that it has no impact at all. The number of homicides committed by people with concealed carry weapons, as well as the number of people whose concealed weapon has prevented a crime (know as defensive gun use, or DGU)  is also a matter of debate. But, as described above, none of the incidents in the data set were ended by the use of firearms by an armed civilian. There are at least two cases where a civilian with a concealed weapon was involved in a mass killing or potential mass killing. In Tucson, CCW holder Joseph Zamudio arrived after the killing had stopped and Loughner had been disarmed and at least partially neutralized. Zamudio did not use his weapon. In another incident, not listed in the data set because it does not meet the FBI’s mass killing threshold of at least four fatalities, CCW holder Nick Meli was present during an attempted mass killing in Clackamas Town Center in Oregon. Meli says he drew his semi-automatic pistol when Jacob Tyler Roberts started firing an AR-15 semi-automatic rifle. Meli took aim but did not fire as he was concerned about the safety of people behind Roberts. Meli believes that Roberts saw him and that his presence was a factor in Roberts’ decision to flee the scene after shooting three victims, two fatally, and commit suicide. Two parts of Meli’s report are relevant here: first, reloading was significant – Meli drew his weapon and took aim while Roberts was fumbling a reload. Second, Meli was unable to fire due to the possibility that his rounds hit might other people, either because they missed the target or penetrated beyond Roberts. In Tucson, Loughner, like Roberts, fumbled his reload and created a window of opportunity for defensive intervention. 4 to 6 seconds, the fastest likely civilian reload, is a long time. A National Football League quarterback typically takes 3 seconds to pass the ball after receiving the snap.

While reloading provides the best opportunity for defensive action, it is very difficult to shoot a gunman firing at rapid to intense rates, even for professionals. Rapid fire is used by military tacticians to achieve fire superiority and is effective at doing so, even when the shooter is an amateur. This is partly because of the physics. An armed responder would almost certainly have to be in the line of sight without being in the line of fire and must be aware of the risks of rounds missing or penetrating beyond the first target, as Meli was in Clackamas. High velocity rounds also penetrate most forms of cover, so a good defensive position may be difficult to find. At the mass killing in Fort Hood, Nidal Malik Hasan discharged 214 rounds for 43 hits in 4 minutes from a twenty round pistol before being shot and wounded while he was distracted “stalking” a victim. Hasan reloaded ten times before being shot despite the presence of many armed professionals. During the mass killing at Columbine, Sheriff’s Deputy Neil Gardner exchanged fire with a gunman after 2 people had been killed and 10 wounded. Gardner fired 4 rounds with no hits. 11 more people were killed and 14 more were wounded after this exchange. During the 2012 mass killing in Wisconsin, armed police Lieutenant Brian Murphy was shot and wounded 15 times.

How To Reduce Deaths In Mass Killings

Reducing magazine capacity and therefore effective rate of fire is unlikely to have an impact on more common gun homicides, with the important exception that it may reduce the risk armed criminals pose to the police and other law enforcement officials. It is, however, a practical way to begin to reduce the number of deaths from mass killings. While it is not possible to reduce the rate of fire of weapons that already exist, it is possible to restrict the manufacture and sales of high capacity magazines so they eventually become more difficult to obtain. This will not stop mass killing but, over time, it may reduce the average number of people hit and killed in each incident, both by reducing rounds fired per minute and increasing the opportunity for potential victims to escape or defend themselves during reloading. The conclusions that being able to fire more rounds into a crowd is likely to result in more deaths and that having to reload more often provides more opportunities for escape and defense and also reduces the time a gunman spends firing may seem obvious and trivial, but it has been strongly argued that reducing magazine capacity will have no impact on mass killing. (See, for example, LaPierre, December 23, 2012: “It’s not gonna work.” ) The data suggests that these two variables are correlated and that assertions to the contrary are incorrect. According to the math, physics and evidence from actual mass killings, reducing the capacity of the magazine reduces the capability of the killer.


Image: “Bullet Through Banana“, Harold Edgerton, 1964

Data Table, Physics Of Mass Killing

34 thoughts on “The Physics Of Mass Killing

  1. Based on the total number of deaths from 1984, 281 people lost their lives in mass shootings. That number is slightly more then half that are killed each year in the city of Chicago. I believe our focus on specific weapons and magazines is misguided.

  2. Dear Anthony,

    Thank you for you comment. I agree. Limiting magazine size does not solve the whole gun violence problem, just a very small part of it, and even then incompletely. I focused on mass killing because it is something everybody wants to reduce, is simple to analyze and the subject is timely. Additional policies would also help reduce deaths from mass killings and others might reduce other kinds of violent crime. I may try to write about some of those in the future.

    Best regards


    • Thank you for your comment. There were some rumors that semi-automatic pistols were used in the Sandy Hook mass killing. Connecticut State Police Department issued a press release on January 18 to address these rumors. Three weapons were seized inside Sandy Hook Elementary School: a Bushmaster .223 caliber model XM15-E2S rifle with high capacity 30 round magazine, a Glock 10 mm handgun and a Sig-Sauer P226 9mm handgun. An Izhmash Canta-12 12 gauge Shotgun was seized from a car in the parking lot. Lt. J. Paul Vance of Connecticut State Police has also confirmed verbally that the weapon used to kill all the victims in the school was the Bushmaster rifle.

      Best regards


    • Thank you for your comment. I had not seen the National Review article and read it with interest. I will also read the work by Lott and Landes which it cites. I do not know if it is true that “the majority of mass shootings occur in places where law-abiding citizens are not allowed to bring guns,” and I will look into it. I do know that all the States in the mass killings data set I used now issue Concealed Carry permits. I do not know whether these laws came before or after the mass killings, and will study that and write about it in the future. As described in the paper, I am aware of two mass killing incidents where citizens had concealed weapons, and these were not discharged in either case. There are also two killings described in some detail where the shooters knew armed people were present – Columbine and Fort Hood – and this did not act as a deterrent. At Ford Hood, the shooter was eventually wounded by an armed police officer, at Columbine the presence of an armed Deputy Sheriff does not appear to have affected the outcome, even though he exchanged fire with one of the shooters. There are several other incidents where armed police officers have been hit multiple times by mass killers. This suggests that, at the very least, the presence of “good guys with guns” does not always make a difference. Another point, especially with respect to Aurora, which is referenced in the article you cite, is whether armed civilians would have been effective had they been present. Rapid or intense fire is a tactic for suppressing return fire and gaining fire superiority. In the face of a surprise attack with high velocity, high rate of fire rounds, a civilian would need to find effective cover (semi-automatic rounds travel easily through movie seats and other people), draw a weapon, find an unobstructed line of fire, aim and shoot. This would be very difficult to do in the dark, and in chaos. Note I do not propose that concealed carry weapons should be banned, nor do I make any comment about gun free zones – I did not have enough data to form a conclusion. I can say that if magazine capacity were reduced, this would increase the chances for a civilian with a concealed weapon to intervene: the rate of fire would be less intense and / or there would be more reloads, which provide the best opportunity to move from a covered position to one which gives a clear shot. This was the case with Nick Meli in Oregon. Last, and I will find time to look at this too, we must balance the probability of good things happening because we increase the number of places where civilians can carry concealed weapons against the probability of bad things happening for the same reason, and look at the net effect. Some concealed carry weapons holders will do harm, some may do good. We should not consider one without the other. Again, thank you for your comment. I will take time to examine the data you pointed me to and will use it in a future post.

      Best regards


  3. Thanks for your thoughts. The issue of magazine size has been a topic of (mostly anecdotal and theoretical) debate but until now your table is the first set of data I have seen. I was curious to do some crude analysis on it, so I looked for a correlation between magazine size (X), shots fired (Y1, calculated from hit efficiency), hits (Y2), and kills (Y3). As a check I found a strong positive correlation between Y1, Y2, and Y3 as expected. When compared to magazine size X, I found no correlation or negative correlation between magazine size and shots fired or kills, and a weakly positive but statistically not-significant correlation between magazine size and hits. This was due to one data point from Aurora that was an outlier on the X vs Y2 scatterplot; if this data point was removed then there was even stronger positive correlation among Y1, Y2, and Y3 but no correlation or negative correlation between magazine size and shots fired, hits, or kills.
    Now admittedly, there is at least one flaw in my logic. It does not make sense to have a magazine size of zero but if X were 0, then Y1, Y2, and Y3 would have to be zero as well. Depending on your point of view, there are either zero or infinite data points at the origin of any X vs Y scatterplot. That there could be no X vs Y correlation is believeable to me, but a negative correlation (negative slope on best fit line) seems odd. Should that line be forced through the origin? I don’t know.
    Anyway I am not a statistician, and this is an admittedly simpleminded thing to do, but it is striking because it does not agree with your analysis and conclusion. Hopefully we will never be able to obtain more data on this.

    • Good question. Short answer: increased magazine capacity enables increased rate of fire which enables increased hits and kills. Mass killings take difference lengths of time, from a few seconds to many hours, which is why rate of fire, which normalizes length of time as a variable, is so important. There’s a strong positive correlation if Tucson (which had an exceptionally high rate of fire) is included, and a weaker but still significant correlation if Tucson is excluded (and even if we do not exclude mass killings with exceptionally low rates of fire as well). Every incident is different, and data is limited, but we can draw some general conclusions: all mass killers use semi-automatics with high capacity magazines as primary weapons; most rounds miss or hit people that have already been hit; more people tend to be wounded and killed if more rounds are fired per minute.

      Best regards


  4. I found this site via a comment that was left at a NY Times Krugman column. This work is very interesting, and makes very salient points about magazine sizes. I’ve heard recently on the radio that the assault weapons ban has been removed from the Senate gun crime bill. On the one hand that is disappointing, but on the other there were so many exceptions to the ban that it seemed almost pointless. I also think I heard a commentator (David Brooks, maybe, on NPR’s All Things Considered on Friday 3.22.2013: Week In Politics: Obama’s Mideast Trip, Immigration, Gun Control.) point out – like your analysis – that reloading is a crucial act in mass shootings, so a ban on large capacity magazines might be far more effective than a ban on a few types of assault weapons.

    • Thanks for your comments. Yes, it should be obvious that the point of high capacity magazines is to reduce the number of reloads and increase the potential rate of fire, but sadly it is not. Reload is the time people can run or fight back – whether they are armed or not. If we reduce magazine size we will — eventually — reduce the number of deaths in mass killings.

  5. Thank you for your story. I am a mom with no prior experience with Bushmasters or Sigs etc. I read your data with interest and I am going to use the science of it to back up my arguments for smaller magazines. I belong to Mothers Demand Action and I have traveled to Washington DC to lobby for our cause. I constantly have to argue with pro gun politicians that magazine size makes no difference. One of the Sandy Hook parents told me children escaped when he reloaded. Now I better understand.

    • Thanks for you comments. Many people claim they can “reload in an instant,” and therefore magazine size makes no difference. This is easily tested. For an experienced professional, it takes about four seconds. That is the fatest time likely. For amateurs, especially in high pressure situations, it takes much longer, and there are several instances in the data set where shooters fumbled reloads completely, allowing their targets to fight back. It would be interesting to take some politicians to a range and test their claims about how quickly they can change magazines. If I can be of any help with your work, just ask.

  6. Thank you! This is what I’ve been arguing for months. They may make up a small percentage, but the number of our firearms homicides are so high that even a small percentage is still an unacceptable number. And yes, these are especially deadly to law enforcement and the unlucky victims of mass shooters. Thank you again.

  7. Your section about the firearms themselves is full of inaccuracies. You really should verify your information before you present it as fact. One: semi automatic rifles fire a much broader range of bullets then diameters between 5 and 7 mm. There are larger calibers for semi auto rifles then that a few smaller ones too. Your FPS is a general statement and you should have been specific on what caliber it in reference to as not all of the calibers will perform the same. Two: the ammo type and caliber is what determines, muzzle velocity, energy of the round not the type of action. The barrel length can play a part but a bolt action versus a semi auto will not change the fps much and if anything it would be lower on the semi auto as it needs some of the energy to cycle the action. Three: You can not set a semi automatic to a maximum rate of fire. There is no such setting. It can only fire as fast as the action will cycle and the trigger is pulled. The only firearms that can be set to a rate of fire are select fire, restricted and not offered for sale to civilians unless it is manufactured before a certain date and they have a special federal license for it. Four: The speed at which a skilled user can cycle a manual action pump firearm versus a semi auto isn’t very different. You can get a high rate of fire out of manual action firearm the only difference is you won’t keep it up as long (get tired) and typically those firearms usually have lower capacity or fixed magazines though some pump action hunting rifles you can get larger magazine for. Five: Shotguns come in manual actions and semi auto so again you need to check your information before you present it as fact. Shotgun shells contain typically either shot or slugs, Some is smaller then bullets and some of it is as big or bigger then bullets as is the slugs. Course the number of shot in each shell is determine by it’s size and the gauge of the shotgun. .410 is a small shot gun while a 12 gauge is much larger. I do get you are trying to prove a point that the reloading time gives others a chance but by having so many errors they are presented in a way to be taken as fact makes it harder to take anything else in your article as fact. Just FYI some skilled shooters can reload faster then 4-6 seconds but those types of shooters have been practicing years and very unlikely to be the typical mass shooter anyways. There was one guy I can’t remember his name but he could shoot a single action revolver (meaning you had to cock the hammer back for each shot) 6 times more accurately faster then most people could shoot one shot out of any pistol. He was on super-humans.

    • Thanks for your comments. Everything you say is correct, and yes, I know it all already. There is only so much detail that fits in a post of reasonable length, and as you point out, none of this detail changes any of the facts. To cover every caliber of weapon would create a piece of encyclopedic length. If you have any evidence of anyone who can reload a magazine-based weapon faster than four seconds, measured from last shot on target to first shot on target, I would like to see it. It is hard to do prone on a range, and very difficult in a tactical situation. Your point about shotguns is incorrect. An AR-15 type weapon is capable of firing 800 rounds a second and it is trivial to modify one legally to simulate fully automatic fire. There is no way to cycle any shotgun at that rate: the mere fact you have to load the shells makes it impossible. This is why all mass killings have a semi-automatic as primary weapon. Higher magazine capacity enables a higher rate of fire which enables a higher kill rate. Avoiding reloading and achieving more intense rates of fire is the purpose of high capacity magazines. You can do a simple test: fire a semi-automatic weapon with a high capacity magazine as fast as possible for a minute and count the rounds discharged, then do the same with the same weapon loaded with a ten round magazine. Do the same with a shotgun. The semi-automatic with the high capacity magazine will achieve the highest rate of fire. The evidence for the effectiveness of high capacity magazines is the data from the last few decades of mass killings in the United States. Higher capacity magazines achieve higher rates of fire that kill more people.

  8. I’ll reread your and my original comment as I’m a bit rushed right now. I understand to include every caliber would be crazy to include but at least linking the data to a specific caliber would be clearer. There are shotguns with large capacity magazines by the way. But I was referring to someone firing off a set number of rounds say 8 with a manual action will not be much slower then a semi auto. The difference is as you point out is magazine capacity. Which by the way some manual action rifles also can be equipped with large capacity magazines. As for reloading magazines or more accurately switching magazines just search you tube and you will find quite a few videos of it being done fast. In reality if they truly want to do something about it in relation to the guns is ban large capacity magazines and require either new firearms come with a fixed magazine or incompatible with existing large magazines or make it harder to rapidly swap out.

    • Understood, but it’s all a little off-topic. It is hard to dispute the math: rate-of-fire has been a factor in the number of people killed and injured in all US mass killings since 1984, which have all used semi-automatic weapons as a primary firearm. Reducing magazine capacity will reduce rate of fire and therefore reduce deaths. Two other points:

      1. You can see video of a typical fast reload here. The time is 2.4 seconds, in a non-tactical target situation. There are two catches. The shooter, a firearms instructor, does not take any time to realize his magazine is empty. He fires a single shot and immediately release the magazine and replaces it. In a tactical situation, especially a mass killing, a shooter is unlikely to be able to keep track of rounds discharged and will likely pull the trigger, realize the magazine is empty, then reload. That would add at least half a second to the reload time. You will notice on this best case video, it takes 0.46 seconds for the shooter to discharge the first round after he is given a “go.” That immediately takes his best case reload time up to 3 seconds. The other catch is he knows where his next magazine is. That works well for a second magazine, but locating additional magazines after that is harder — you cannot put multiple magazines in exactly the same place. So, if we allow another half second to locate magazines 3, 4 and 5, we see that even in this very best of scenarios, expert reload is 3.5 seconds on a range, and without taking cover.

      2. Even if we calculate reload speeds of 2 or 3 seconds, reducing magazine capacity will reduce rate of fire. It does not matter how fast you are, you cannot fire while reloading, and you cannot reload faster than a semi-automatic can chamber another round. If the instructor in the above video has 10 round magazines and fires a round a second, he would still have to reload at least four times in a minute. That would be about 15 seconds when he is not firing, about 4 times longer than it takes an elite NFL quarterback to pass after receiving a snap. It’s not a lot, but better than nothing, and very few if any people can reload at this speed in a tactical situation.

  9. What I was trying to point out is instead of them banning based on the action (semi auto versus bolt, pump or lever action. They should just go after the magazines and nothing else. It is really the only feature that allows for a Sustained high rate of fire. All the other features you find on firearms including how it looks will not affect the functionality or firing rate (the semis would only be slightly faster then any other style assuming same number of rounds). If they truly want to do it right they would ban manufacture of any firearm that allows the use of an older large capacity magazine, along with the and also the sale of new large capacity magazines or adapters that would allow the use of old magazines. Basically force any firearm that currently has large capacity magazines manufactured for them change their magazine well to only accept new specific magazines. You could also ban those stupid slide fire stocks.

    • I agree completely. The UK banned all magazine-type weapons after a school shooting in Dunblane, Scotland that was not that different to Sandy Hook. There has not been a mass killing using firearms since. This a much clearer regulation than an “Assault Weapons” ban which is too vague to be useful — any weapon, including a fork, could be used to “assault” — whereas banning anything that took a magazine would limit us pretty much to shotguns, revolvers and standard action rifles. I agree on slide fire stocks, too. Tactically, the difference between semi-automatic and automatic is pretty meaningless for an unmounted weapon, and these devices make it irrelevant.

  10. Very well researched and articulately written, Kevin. I reference much of the conclusions you’ve dawn during testimony to legislative efforts limiting the size of magazines to 10 rounds. I’d encourage you to continue researching other areas of US gun policy.

  11. Kevin, this is the most cogent, polite, well researched site I have found. I appreciate your answering the comments and the respect that is evidenced on both sides. I appreciate this site and will share with friends on both (all) sides of the gun violence issue.

  12. Hi Kevin,

    I read this post with interest. I really appreciate you breaking things down in scientific as opposed to dogmatic terms.

    I just wanted to point out two underlying issues. First, the data set you drew on is skewed. The Citizens Crime Commission of New York City (CCC-NYC) purposely selected high-profile mass shootings “all of which involved large capacity ammunition magazines.” You wrote, “All perpetrators in the the 30 US mass shootings since 1984 …” implying that there were only 30 mass shootings in the US since 1984, and that they all used semi-automatic firearms. You repeated this claim in your comment from April 18: “rate-of-fire has been a factor in the number of people killed and injured in all US mass killings since 1984, which have all used semi-automatic weapons as a primary firearm.” The problem, however, is that there have far many more mass shootings than just 30 since 1984. Mother Jones documented many of them (other people posting comments provided links to that data set). In my own research, I have found 68 such shootings wherein 6 or more people have been killed (not including the perpetrator) since 1984. Several of these shootings did not involve semi-automatic firearms.

    Second, the data collected by the CCC-NYC contains errors. For example, it implies that James Holmes in Aurora discharged 100 bullets. He didn’t. He discharged 76 rounds: 65 from his .223 semi-auto rifle, 5 from his .40 caliber semi-automatic Glock, and 6 from his Remington 870 shotgun. This of course does not undermine your argument. It might even arguably bolster it, as he managed to hit 70 people with 76 rounds. (Also, as recorded on a 27 second 911 call, he managed to shot 30 rounds out of his AR-15 — for a rate of roughly one projectile per second. That means he emptied two-thirds of his 100-round drum in approximately one minute. [Then it jammed].)

    A source on Aurora shooting information:

    I just wanted to share this information with you. I figured you would find it helpful.



    • Thanks Lou. I had the Aurora event as 70 rounds fired — I got most of that type of information from news and police reports — but I did assume the data set covered all mass killings and appreciate you pointing out that it does not. Out of interest, how many of the 68 mass killings you found did NOT use semi-automatics as the primary weapon? Also were they all single events (one location, continuous attack) or were some of them “spree” killings (several locations, killer spends time moving from one place to another without shooting). A couple of other comments, which I suspect are obvious to you: the benefit of semi-automatics is that they reduce number of reloads and increase rounds fired between reloads, because reloading is a moment of vulnerability; it doesn’t matter how fast a shooter can reload, no one can shoot and reload at the same time, which is why it is a moment of vulnerability. Having done the math on this, I find the argument that “magazine size doesn’t matter” kind of baffling. A better conclusion for this piece might have been, “well, d’uh.”

  13. Hi Kevin,

    I don’t have the numbers in front of me right now. Moreover, I am still breaking down my older cases (I have dug up all mass shootings involving six or more fatalities not including the perpetrator since 1964 — 101 in total). What I can say from memory is that most high-fatality gun rampages involve semi-automatic weapons. I don’t remember how many involve extended capacity magazines though. Also, most are single location rampages, although a few are sprees. (I did not include shootings that lasted more than 36 hours. I gave myself a larger window of up to 36 hours though to allow for stand-offs that extended into the next day.)

    I can tell you this though: there have been 16 double-digit fatality mass shootings in the U.S. in the past 50 years (i.e., since 1964). Of those, 12 involved a semi-automatic firearm with an extended capacity magazine. Two more involved a semi-automatic firearm (but I can’t verify if they also involved extended capacity magazines). Two others did not involve semi-automatic firearms as far as I can discern.

    I agree with your conclusions about extended capacity magazines — and plan to suggest proposals for reducing rampage violence by focusing on such magazines in my forthcoming book.

    Again, I appreciate your work here!



    • I took a quick look at the Mother Jones dataset, which is excellent, up to date and I think was released after I wrote this (or I missed it). It does include spree killings, and I have not taken those out, but a quick summary is:

      12 of the 67 cases (18%) did not involve semi-automatic weapons. (I counted everything listed as “rifle,” “shotgun,” or “revolver” as not semi-automatic. I included one case where the shooter took a handgun from one of of his victims.

      The base size of the cases that did not involve semi-automatics is a little small, but on average, these cases caused 25% fewer deaths and injuries between 4 and 5 fewer victims per case.

      This is the best data I have seen. It is preposterous that there is no FBI database or other rigorous official record of these events. We have much better data on automobile safety, and probably dog bites and soda recalls.

      Good luck with the book.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s