The FBI’s active-shooter database routinely misses cases where an armed civilian stops an active-shooter attack, according to a new report from the Crime Prevention Research Center.
The center said more than a third of attacks from 2014 to 2025 were stopped by “good guys with guns,” or nearly 10 times the rate suggested by the FBI’s data.
Looking only at incidents outside “gun-free zones,” where laws and ordinances preclude civilians from carrying weapons, the rate tops 50%, the center calculated.
In some cases, the FBI’s data did not include the shootings. Less frequently, the FBI misclassified the shootings.
The result, said John R. Lott, the center’s president, is headlines that declare civilians stopping active shooters to be “rare,” when he said the data indicates it is almost routine.
The FBI’s database lists 408 active shooter incidents from 2014 through 2025 and says 16 of them were stopped by armed civilians.
The Crime Prevention Research Center’s count found 615 incidents in total and said armed citizens stopped 225 of them.
The center also identified 35 other cases where an armed civilian likely headed off a gun attack because the potential shooter never fired a shot.
“Law-abiding citizens stopping these attacks are not rare. What is rare is national news coverage of those incidents,” the center concluded.
The country got a firsthand look at one situation last month, when a 24-year-old man began shooting at an In-N-Out Burger restaurant and across a wider commercial area in Twin Falls, Idaho. He killed three and wounded seven others before taking his own life.
Authorities said an armed civilian, Jordan Salinas, joined an off-duty state trooper in confronting the attacker and helped prevent more casualties.
That incident drew extensive coverage, likely because of the multiple deaths, but Mr. Lott’s data suggests far more incidents that did not draw national headlines.
In December, a man opened fire on customers at a restaurant in Conyers, Georgia, but two armed patrons shot back, forcing the gunman to flee before anyone was killed. Police credited the two patrons with saving lives.
In August 2025, a gunman in Glenwood, Iowa, killed two people, set his own house on fire and shot into nearby homes before being shot by Harrison Friar, one of those whose homes were struck. The gunman retreated to his home, which caught fire. He died days later from burn injuries. Authorities said he had hundreds more unspent rounds that he had intended to spray into the neighborhood.
The FBI defines active shooters as cases where one or more people either kill or attempt to kill others in a public place. Gang violence, robberies, domestic situations and incidents where the suspect goes after a particular individual at a workplace are not generally included.
One case that Mr. Lott says the FBI misclassified was a June 22, 2025, shooting at CrossPointe Community Church in Wayne, Michigan. The FBI database lists the intervention under formal security personnel, but Mr. Lott noted, and church leaders confirmed, that the responders were actually volunteer parishioners serving on the church’s internal safety team.
The FBI did not respond to a request for comment for this report.
Previously, the bureau said it tried to spot all incidents that should be included, but acknowledged limitations. It said it makes “every effort” to include cases that meet its definitions.
However, Mr. Lott said he has been raising cases for years and the FBI still does not add them.
“Is it possible that you can miss cases? Yeah, of course. But when somebody points out that you’ve missed the cases and you don’t fix it, that is concerning to me,” he told The Washington Times.
Mr. Lott worked at the Justice Department as a senior adviser for research and statistics at the Office of Justice Programs from October 2020 to January 2021, and said he raised concerns about missing cases directly with the FBI. The cases still were not added.
“This is the frustrating thing to me, and that is, these reports are used by the media, they’re used in legislative debates, they’re used in court cases, they’re used in academic research. And they don’t fix these things,” he said.
Mr. Lott said he understands that judgment calls are involved in which cases to include, but it appears the data disproportionately misses cases where civilians used guns to stop attacks.
The issue of civilians with guns raises big questions about who should be counted on to stop an attack, particularly in a reality where guns are already so prevalent.
Those who favor stricter rules on access and carrying firearms say it is the job of police to respond, while Second Amendment advocates say police usually arrive well after the attack.
One worry of gun control advocates is the armed civilian harming bystanders in an exchange of fire.
Mr. Lott’s database, though, has yet to find an instance of a civilian accidentally shooting a bystander. In only one instance in the data did police shoot the “hero citizen.” That was in Arvada, Colorado, in 2021.
Among Mr. Lott’s more striking findings is the difference in attacks that are stopped by civilians in areas where they are allowed to carry concealed weapons, as opposed to “gun-free zones,” where civilians are restricted.
When his data is limited to areas where civilians are allowed to carry firearms, he calculates that they stopped 52.9% of attacks during the study period.
In 2025 specifically, his study identified 36 incidents in places where civilians were allowed to legally carry guns, and in 21 of them an armed civilian stopped the attack — a 58.3% rate.
Mr. Lott said some mass shooters know this.
Their manifestos show that they picked areas where guns would be restricted, figuring they could get a higher kill total.
“They’ve figured out the more people they kill, the more media coverage,” Mr. Lott said. “These guys may be crazy in some sense, but they’re not stupid. They explicitly say they want to go to a place where guns are banned because they know it’ll be easier for them to kill more people and get more coverage.”
Contact the author
Please read our comment policy before commenting.