“What inevitably stops these horrific crimes: Armed good guys stopping armed bad guys.”
When Brandon Tsay disarmed a gunman who had opened fire at a Monterey Park, Calif., ballroom on Jan. 21, killing 11 people, he did not have his own weapon. Security footage shows him struggling with the gunman and wresting a semiautomatic pistol away from the shooter before he had an opportunity to attack another dance hall, in nearby Alhambra, where Tsay was working.
Last summer, Elisjsha Dicken ended a rampage at a shopping mall in Indiana by pulling his 9mm pistol and killing a man who had already killed three people and injured two with high-powered weapons.
Two narratives: A “good guy” without a gun. A “good guy” with a gun.
Whether one is more common than the other depends on the data you use — though among all such shooting attacks, neither is the norm. The vast majority of more than 430 “active-shooter” incidents catalogued by the FBI since 2000 ended when the shooter fled, when law enforcement killed or apprehended the shooter, or when the shooter died by suicide.
Diving into the data to determine how often citizens come to the rescue is a difficult exercise. By law, the FBI has been publishing data on active-shooter incidents since 2000. The FBI reports suggest that about three times as many citizens without a gun have ended an active-shooting incident as citizens with a gun.
But the FBI reports, though often cited by the news media, have numerous errors and, by the FBI’s own admission, are not necessarily complete or even consistent in how certain criteria are applied. “Incidents identified in this study do not encompass all gun-related situations; therefore caution should be taken when using this information without placing it in context,” the FBI said when it issued its first report.
In compiling the reports, the FBI works with the Advanced Law Enforcement Rapid Response Training (ALERRT) Center at Texas State University. ALERRT has its own data set that is slightly different from the FBI numbers. This data shows that since 2000, citizens without a gun have halted nearly twice as many incidents (42) as citizens with guns who were not commissioned law enforcement officers (22). If security personnel or off-duty officers are excluded from the list, the number with a gun drops to 12.
John R. Lott Jr., a gun rights researcher, is skeptical of the FBI data. He has compiled his own tally of “good guy with a gun” incidents that he says the FBI has missed. Including all of Lott’s incidents would significantly change the result — he has a list of more than 100 instances between 2014 and 2021, linked to news reports, when a citizen with a lawful firearm ended an active-shooting situation.
Toward the end of the Trump administration, Lott, the author of the 2010 book “More Guns, Less Crime,” was a senior adviser for research and statistics at the Justice Department’s Office of Legal Policy. In a paper he wrote for that office, he said the office found many missed cases in the FBI reports, including 20 multiple-victim shootings involving at least two people killed between 2000 and 2013. In the 2018 and 2019 reports, he concluded, the FBI missed seven shootings, including six when a concealed-handgun permit holder stopped the attack.
“The FBI reports keep excluding cases where shooting attacks have been stopped by concealed handgun permit holders,” Lott’s Justice Department report said. He argues that a more complete data set would show that a higher percentage of shootings are stopped in places that do not prevent people from having concealed weapons.
The FBI brushed aside repeated efforts by The Fact Checker to discuss its reports and the questions raised by Lott. “We have no additional information to provide other than what is provided within the active shooter reports on our website,” the agency said in an emailed statement.
Of course, Lott has his own bias. He keeps track only of the active-shooter incidents that someone with a weapon has ended. So a list including instances when a person without a firearm ended the shooting could also be higher.
Errors in the FBI database
Whether a database is expansive or limited often depends on the definitions used to compile the statistics. In assembling active-shooter incidents, the FBI relies on a definition that covers an individual or individuals actively engaged in killing or attempting to kill people in a confined and populated area with firearms.
“This is not a study of mass killings or mass shootings, but rather a study of a specific type of shooting situation law enforcement and the public may face,” the FBI said in its first report. “Shootings that resulted from gang or drug violence — pervasive, long-tracked, criminal acts that could also affect the public — were not included in this study. In addition, other gun-related shootings were not included when those incidents appeared generally not to have put others in peril,” such as an accidental discharge of a firearm or suicides in public areas. The FBI also does not include shootings in homes, but it keeps the focus on public places.
Shootings that result in homicide are widely reported. But not every active-shooter incident results in deaths, making it difficult to ensure a complete accounting because there is little news coverage. For instance, in 2011, at Deer Creek Middle School in Colorado, a shooter opened fire on eighth-graders leaving for the day, until two teachers tackled the shooter. Two students were wounded but none was killed.
This incident ended up in the FBI statistics. But a 2015 shooting in Ohio did not. Two people were arguing outside an apartment in a Cincinnati neighborhood when the man pulled a gun and began firing at the woman. The woman’s brother, who had a concealed-carry permit, shot the gunman in the leg. The gunman then ran into his home and came out with a gun in each hand, firing at the woman and her 1-year-old child.
The FBI told Lott that this was not included because it was a “domestic dispute,” according to his Justice Department paper. Lott notes that the shooting took place on a public street and that multiple people were shot — just like another 2015 incident that was included in the database. In that shooting, police killed the shooter.
The FBI acknowledged that it has missed some active-shooter incidents. In 2015, a man killed a clerk and a customer at a liquor store in Conyer, Ga., after an argument some hours earlier. The man had returned to the store with a handgun and begun firing. One customer, hailed as a hero by police, returned fire and the shooter fled.
In an email shared by Lott, an FBI official acknowledged that “the FBI did not come across this incident during its research in 2015, but it does meet the FBI’s active-shooter definition.” The official noted that the active-shooter reports “are limited in scope” and cases will be missed. In any case, the incident has not yet been added to the FBI database.
Lott also identified how the FBI, in its reports, sometimes misidentifies citizens as security personnel. In 2019, for instance, a man with a shotgun began shooting inside a church in White Settlement, Tex. He killed two parishioners before a church member drew a firearm and killed him. The FBI report inaccurately said “two members of the security team were killed before they had the chance to engage the shooter,” and the incident was not recorded as an example of citizen engagement. As it turned out, 19 or 20 parishioners attending services were carrying concealed weapons at the time of the attack.
Deciding what to count
While the FBI declined to comment, an official at ALERRT said its database does not factor in how a shooting incident ends.
“I can confidently say that the event resolution is not an inclusion or exclusion criteria in the discussion,” said M. Hunter Martaindale, research assistant professor at Texas State University. “Events are excluded for a variety of reasons (e.g., gang motivated, targeted or domestic only events, attacks in the commission of another crime, etc.) but none are excluded based on the manner in which they end. We have courses dedicated to civilian response and firmly believe that citizens can play a vital role in ending these tragic events.”
Looking at some of the incidents that Lott says should be counted, Martaindale indicated why they would not merit inclusion. In 2019, a man opened fire in a dental office in Colonial Heights, Tenn., killing a woman. A patient drew his concealed weapon, shot the gunman and held him at gunpoint until police arrived. “It’s a citizen, a concealed-carry permit holder that saw a threat, eliminated that threat and stood by until law enforcement arrived, and did a really good job,” Sullivan County Sheriff Jeff Cassidy told reporters.
It turns out, the gunman shot his wife, who worked there. So this would be considered a domestic dispute, Martaindale said.
Another incident, in 2014, involved a shooter who opened fire in a strip club in Portland, Ore., hitting three people, including a bouncer who was critically wounded. Another bouncer, with a valid concealed-handgun permit, followed the gunman outside and fatally shot him in the back. But Martaindale said the case sounded like a retaliation murder. The bouncer who was shot had refused entry to the gunman a half-hour earlier, and the gunman returned, masked and armed. The two other people who were shot appeared to have been struck by errant rounds.
The FBI made a similar observation to Lott in correspondence with him. But Lott says this case is a good example of the FBI’s inconsistency in its reporting. He said the strip club shooting was similar to incidents that the FBI said qualified for inclusion.
The FBI database includes a 2009 case in which a man was being escorted out of a bar by security when he began firing. The database also included a 2008 case in which a man got into an argument with his supervisor at a plastics plant. He pulled a gun as he was being escorted out of the plant and opened fire, shooting the supervisor and several employees. There is also a 2010 case in which a female employee was suspended and lost her company ID at a Kraft plant — and she retrieved a .357 Magnum from her car and returned to shoot the people with whom she had quarreled. The FBI also lists a 2011 case in which a man fired 70 to 90 rounds into a courthouse, intending to kill the judge who presided over his divorce.
The Bottom Line
Deciding what to include as active-shooter incidents is obviously a judgment call. But it’s worth remembering that the FBI does not claim that its database is comprehensive, making it difficult to make an easy determination about whether good guys with guns or good guys without guns are more common.
Send us facts to check by filling out this form
Sign up for The Fact Checker weekly newsletter
The Fact Checker is a verified signatory to the International Fact-Checking Network code of principles