Here’s What the Data Says About Shooting Sprees | David Thornton

With all the talk about guns in the wake of the Uvalde school shooting – and in the wake of several discussions online where we often couldn’t agree on basic facts – I thought it would be interesting to take a look at what the data says about the rampage murders that seem to be plaguing America these days. We should all be interested in what the data shows because if we act without regard for the facts, we are unlikely to resolve the problem and may end up making things worse. 

After reading about spree killings in the past, I can tell you that the data is confusing for several reasons. First, even though it seems like there have been a lot of these incidents, they are still relatively rare from the perspective of collecting data. Sadly, with every shooting spree, more data is available.

Second, there is no single definition for spree killings so looking at different sources is often comparing apples to oranges. Federal law defines a “mass killing” as “three or more killings in a single incident,” but doesn’t differentiate between a killing spree with random victims and a mass murder in which the three or more victims were specifically targeted. 

FBI crime data is the gold standard for most crime statistics, and it can tell us many things about homicide in the US. For example, the data shows that homicide offenders and victims are both most likely to be in their 20s, male, and black. The most likely location for a murder is in a home, and when a relationship is reported, the perpetrator is mostly likely an acquaintance. Statistically, the most likely weapon to be used in a homicide is a handgun. Rifles rank below knives and blunt objects on the list. 

Already, we can speculate that spree killing statistics are not going to match the overall homicide statistics. The problem is that the FBI report doesn’t break down the data into mass shootings, which wouldn’t really be the same as killing sprees anyway. 

An FBI resource page defines “active shooter” as simply “an individual actively engaged in killing or attempting to kill people in a populated area” with no regard for body count. Therefore, an “active shooter” can be involved in “mass killing,” but the two terms can also be mutually exclusive. An “active shooter” doesn’t have to successfully kill anyone and a “mass killing” can be perpetrated with poison, an automobile, or any number of possible weapons in addition to guns.

I was considering scrolling through the lists of killings available online and making a spreadsheet. Thankfully, before I took on that task, I found two good online resources that were already put together. First, the FBI has a “quick look” page of active shooter statistics from 2000 through 2018 as well as an updated report for 2020 and 2021. Second and even more helpful, the nonpartisan Violence Project, has a searchable database that is current through March 2021. Voice of America used this data for a recent article that focused on school shootings. Much of the information revealed by the data is not what you would expect from the news coverage and political debates. Objectivity is needed and that’s what the resources provide.

First, while both active shooter incidents and mass killings seem to be increasing, this has been a long-term trend. A Violence Project graph shows a steady rise in the number of mass killings since the 1980s while the FBI data shows a rise in the average number of active shooter incidents since the year 2000. It is interesting to note that many other forms of violent crime decreased over the same periods. 

The FBI’s update is more alarming. It shows a 33 percent increase in active shooters from 2019 to 2020 and another 52.5 percent increase from 2020 to 2021. However, the VOA article points out that school mass shootings are not necessarily increasing in frequency. The highest concentration of school rampages occurred from 1998-99. 

The increasing trends are disturbing but considering the relative recency of the surge, we can say that there is no firm link between the expiration of the Clinton-era “assault weapons” ban, which expired in 2004, and the rising number of mass shootings. Fact checkers from both and the Washington Post dispute claims that the ban’s lapse is responsible for today’s epidemic of spree killings.

Also interesting is that the FBI data shows only a modest increase in the average number of victims of mass shooters. In this data set, 2017 was a particularly bad year with a high of 30 active shooters and 729 victims. The Sutherland Springs church shooting in Texas and the Las Vegas concert attack were among the killing sprees that year. The number of victims has declined from the 2017 peak.

Before we delve too deeply into the Violence Project database, you may wonder what their statistics include. The site uses a Congressional Research Service definition for a “mass shooting,” which is:

“A multiple homicide incident in which four or more victims are murdered with firearms—not including the offender(s)—within one event, and at least some of the murders occurred in a public location or locations in close geographical proximity (e.g., a workplace, school, restaurant, or other public settings), and the murders are not attributable to any other underlying criminal activity or commonplace circumstance (armed robbery, criminal competition, insurance fraud, argument, or romantic triangle).”

In keeping with this definition, the database excludes domestic murders as well as shootings which are related to criminal activity or gang violence. The resulting database includes 176 incidents and 180 shooters over nearly 60 years.

Shooting sprees go by many names, but since there is no consistent definition, I refer to these cases as active shooters, mass shootings, rampage killings, shooting sprees, and murder or killing sprees. These all mean the same thing and I’m not splitting hairs. I use the terms interchangeably. 

One of the things I was most curious about was the prevalent type of weapon. Despite the perception that “assault rifles,” a term that only has meaning when defined in specific legislation or research, handguns are the most common type of weapon to be used in spree killings. Seventy-eight percent of shooters used a handgun while 10 percent used an extended magazine. “Semiautomatic assault weapons” were the second-most common choice but were used by only 28 percent of spree killers. About three-quarters of those who carried an “assault” weapon also used a pistol and the data doesn’t break down how many people were killed by which weapon.

We can also narrow the search to K-12 school shooters, which represent 7.6 percent of the perpetrators in the database. (Workplace attacks were most common at 31 percent.) Handguns were also most common in school shootings, with nine of the 13 school shooters (69 percent) choosing a pistol. Only five (38 percent), used an “assault” weapon. 

Granted, this data set goes back to a time before the ready availability of “assault” rifles, so let’s confine the set to sprees from 2000 on. In that case, the database yields 95 incidents. Of these, only 30 (32 percent) involve an “assault” weapon. 

I should point out also that while AR-15s are “military-style” rifles, they are not military rifles. They are semiautomatic, one-shot-per-trigger pull, guns, not fully automatic machine guns. On the other hand, “assault” rifles of various types seem to have been used in shooting sprees since 1977 when Frederick Cowan, a neo-Nazi, went on a rampage and killed five people in New Rochelle, N.Y. after being suspended from his job at a moving company. It isn’t clear whether Cowan’s gun was fully automatic.

The misconception that “assault” rifles are the weapon of choice for spree killers probably stems both from high- profile attacks like the one in Uvalde as well as the focus of media and politicians on the popular rifles. It is true that these rifles are used more frequently in spree killings than in other crimes but handguns are still far more popular.

Demographically, some of the stuff we all know about spree killers is true, while some is not. For example, it is true that virtually all (98 percent) active shooters are male. However, only 52 percent are white. Blacks make up the second-largest category (20.9 percent), followed by Latinos (8.1 percent), Asians (6.4 percent), Middle Easterners (4.1 percent), and Native Americans (1.7 percent). These statistics track relatively closely with overall national demographics, especially when we consider that the data set goes back through six decades of shifting national ethnicities. 

Other factors are more common than race or ethnicity. Most shooters have a criminal record (65 percent) and/or a record of violence (63 percent). Employment problems (51 percent), childhood trauma (including parental suicide, abuse, neglect, and severe bullying) (42 percent), and domestic abuse (36 percent) were also common. 

Another thing that does not seem to make a big difference is gun control. The map of murder sprees dots the country. Incidents seem to track more closely with population than anything else, with the gun-friendly red states of the Dakotas, Montana, and Wyoming the only areas not showing an attack. In 2021, California, a highly restrictive state, had the most active shooters and California also has the dubious honor of having the highest number of school mass shootings

An oft-discussed factor that is borne out by the statistics is mental illness. Sixty-nine percent of active shooters had a history of mental health problems. 

As is also often observed, spree killers often presented warning signs before their attack. In fact, more than 80 percent showed “red flags” such as “exaggerated emotional responses, an increased interest in violence, and signs of hopelessness.” Some even discussed their plans with others.

The Violence Project database also looks to motives to help us understand why these murderers do what they do. The most common reason was psychosis, which the Project defines as “a mental condition that makes it difficult for a person to recognize what is real and what isn’t.” Thirty percent of active shooters were classified as psychosis sufferers.

Following that mental motive were employment problems, which were a motive for 23 percent of shooters and nondomestic conflicts with coworkers, friends, and/or family at 20 percent. 

I was surprised that terrorism and radicalism accounted for such a small percentage of attacks. My perception was that these politically and religiously motivated murders were a larger share than the 9.3 percent attributed to racism or the 5.2 percent classified as “religious hate.” However, the Violence Project does note that hate-motivated and fame-seeking attacks have increased markedly since 2015. 

When it comes to their weapons, almost half of active shooters (46.5 percent) acquired all of their guns legally. Of these, most came from private dealers with only 4.1 percent obtaining guns through unregulated private sales. About 29 percent of shooters obtained their guns illegally. Their methods included theft, lying on applications, system failure, straw purchases, and street sales. A caveat here is that in about a third of these cases, it couldn’t be determined where the shooter got his guns.

If we are going to reduce the instances of shooting rampages, we should begin to attack the problem by looking at the data. And the data tells us that banning AR-15s or any specific type or class of weapon won’t solve the problem.

By far, the most common threads among rampage killers are mental illnesses and psychoses. The most effective strategy then would seem to be intervening to help those with mental illnesses, especially in cases where these mental health issues make the person prone to violence. 

Preventing these individuals from obtaining guns should also be a priority. This could take a variety of forms including red flag laws, better background checks, mandating secure storage of weapons, and even including biometric sensors in guns. 

For that matter, it seems that we also need to do a better job of making sure that violent criminals don’t get guns. The database indicates that more than 20 percent of mass shooters had a criminal record and a history of violence, yet still managed to get their guns legally. Quite a few of these cases are recent.

Interestingly, what Drs. Jillian Peterson and James Densley, the founders of the Violence Project, do not include in their list of data-driven policy recommendations an “assault” weapons ban. The data just doesn’t support the conventional wisdom that AR-15s are to blame or that banning them would save lives.

I’m a pro-Second Amendment guy, but I’m also a guy who acknowledges that business-as-usual is not working when it comes to gun violence and mass shootings. As I wrote last week, it isn’t the prospect of some common-sense changes to gun laws that present the biggest threat to the Second Amendment, it’s the fact that scores of people are being gunned down seemingly at random and with alarming frequency. 

We do need to make changes, but we need to take action that will be effective at actually helping to solve the problem at hand. There is the emotional appeal of banning “assault” rifles on the one hand and the more clinical and focused approach of red flag laws on the other. 

The data is clear on which path should be our focus.

The First TV contributor network is a place for vibrant thought and ideas. Opinions expressed here do not necessarily reflect those of The First or The First TV. We want to foster dialogue, create conversation, and debate ideas. See something you like or don’t like? Reach out to the author or to us at

Follow David Thornton on Twitter (@captainkudzu) and Facebook