I've written about the issues with defining a "mass shooting" in another answer previously. I'm going to copy and paste the pertinent parts here.
The main problem with statistics like these is that there is no one definition of "mass shooting" and thus, by adjusting your parameters, can make the number be whatever you want to fit a narrative. Depending on the definition you wish to choose for a "mass shooting", the percentage for the quoted statistic can range from 2.6 every 100 days to 97.4 per every 100 days.
What is a mass shooting? There are some events, like the Stoneman Douglas High School shooting, the 2017 Las Vegas shooting, and the Columbine High School massacre, where everyone can agree that it should be considered a mass shooting.
But what about this incident in Saginaw, Michigan, where 5 people were shot at a house party and no one was killed? Or this event in Madison, Alabama, where 4 people were shot and no one was killed?
The Wikipedia article on Mass Shooting points out the huge issue in the definition of Mass Shootings.
The term was originally defined as the murder of four or more people with no cooling-off period but redefined by Congress in 2013 as being murder of three or more people. According to CNN, a mass shooting is defined as having four or more fatalities, not including gang killings or slayings that involve the death of multiple family members. In "Behind the Bloodshed", a report by USAToday, a mass killing is defined as any incident in which four or more were killed and also includes family killings. A crowdsourced data site cited by CNN, MSNBC, The New York Times, The Washington Post, The Economist, the BBC, etc., Mass Shooting Tracker, defines a mass shooting as any incident in which four or more people are shot, whether injured or killed. As of November 2017, the FBI defines a mass shooting as an incident involving "four or more people shot at once." A noteworthy connection has been reported in the U.S. between mass shootings and domestic or family violence, with a current or former intimate partner or family member killed in 76 of 133 cases (57%), and a perpetrator having previously been charged with domestic violence in 21.
The lack of a single definition can lead to alarmism in the news media, with some reports conflating categories of crimes.
PolitiFact agrees that there's a huge issue with that definition. In a 4 October 2017 article titled "How is a ‘mass shooting’ defined?", PolitiFact refuses to rate a claim by House Minority Leader Nancy Pelosi where she claimed
"273 mass shootings in 2017 -- one for each day of the year."
PolitiFact, before getting into the meat of the article, includes this sentence.
We are not rating her claim on our Truth-O-Meter, however, because our current research indicates the debate over the definition of mass shootings is more unsettled than ever.
They continue along in the article as if it was another fact check, although without an actual rating.
Pelosi’s spokesman told us the congresswoman’s number comes from the Gun Violence Archive, a crowdsourcing website that tracks gun deaths using media reports. It defines a mass shooting as one in which at least four people are injured or killed in one location, not including the suspect.
The Gun Violence Archive listed 273 mass shootings as of Oct. 1, 2017, the day of the Las Vegas massacre.
(...)
Pelosi’s claim and the mass shooting definition it rests on ignores strict criteria developed in a July 2015 Congressional Research Service report. It defines a mass shooting as "a multiple homicide incident in which four or more victims are murdered with firearms, within one event, and in one or more locations in close proximity." The report has been frequently cited by some criminologists and gun rights advocates.
Criminologists who spoke with PolitiFact in the past say this definition was more in line with events such as the shootings at Columbine, Virginia Tech or in San Bernardino.
They also told PolitiFact that the Gun Violence Archive’s tally includes far more than what would be considered mass shootings because it lumps in gang shootings and home invasion robberies.
As PolitiFact Florida reported in 2015, "Politicians or others who want to make a point about guns choose a set of data and a definition that reinforces the point they want to make. People who want more gun control tend to choose more expansive definitions." (emphasis mine)
PolitiFact continues by explaining that many criminal experts have their concerns even with these definitions.
Gun control groups say it’s arbitrary to distinguish between a death and an injury. They point out a significant problem: Some shootings that injure a dozen or more people but don’t kill four people would not be considered a mass shooting under the more restrictive definition.
"I would submit that sometimes the only difference between a shooting and a murder could be a centimeter, an inch, an unlikely ricochet, whatever," Bueermann, who now is president of the Police Foundation, which researches law enforcement practices, told the Post. "If we're trying to capture true gun violence in our country, a broader definition [of mass shooting] is probably more useful than a narrow one."
(Frederic) Lemieux went on to say the broad definitions of mass shooting remain problematic, citing the inclusion of crimes such as familicide, where "victims are exclusively family members and not random bystanders."
"Gang murders are usually crime for profit or a punishment for rival gangs or a member of the gang who is an informer. Such homicides don’t belong in the analysis of mass shootings," he added.
Gun Violence Archives brings this into greater clarity with their statistics for 2018. They use the "4 or more injuries" criteria to define a mass shooting. By this count, they have 103 mass shootings as of 24 May 2018. However, by using the more strict definition of "four or more victims murdered with firearms, within one event, and in one or more locations in close proximity", the count is at most 10. However, of the 10 you have
- 2 school shootings (Parkland - (17 Dead, 17 injured) and Santa Fe - (10 Dead 13 Injured))
- An event where three people were shot in a gas station, a fourth some time later, before the killer committed suicide - Five Dead including shooter
- An event where two women were killed before the shooter returned home, killed two of his own family members, and committed suicide - Five Dead including shooter
- An event where a man kills his three children, his ex wife's boyfriend, and wounds his ex-wife, before committing suicide - Five Dead including shooter, one wounded
- A deliberate mass shooting at a car wash (body armor, multiple weapons including an AR-15, indiscriminate) that left 5 dead, including the shooter - Five Dead including shooter
- The Waffle House Shooting - 4 Dead, 3 Injured
- An event involving two shooters barricading themselves after killing 3 - Four dead including one shooter, three wounded
- An event where three were found dead and a fourth died in the hospital after being found in a house, shooter unknown - 4 Dead
- An event where a man shot a woman and her children, before killing himself, possibly domestic violence - 4 Dead including shooter, 3 injured
With the exception of the two school shootings, the Waffle House shooting, and the Car Wash shooting, many of these wouldn't even be considered mass shootings under this more restrictive definition. Some only hit the 4 deaths mark due to including the shooter, some take place in multiple locations, some involve family members, and some are possibly gang-related.
In terms of this claim by the actor, if we were to use the loosest definition of "mass shooting" (4 or more victims, not necessarily killed, and can be of any relation to the shooter), the statistics are actually worse for 2018. Gun Violence Archive has 227 mass shootings for 2018 as of 21 August 2018. As 21 August 2018 is the 233rd day of the year, the odds of there being a "mass shooting", using the most liberal definition of "mass shooting", is actually
227 / 233 = 97.4%.
However, we can also use the much more restrictive definition of "mass shooting" (4+ deceased victims excluding the subject/suspect/perpetrator, one location). Gun violence archive uses the tag "Mass Murder" for these types of events. Using this more restrictive definition, we get a total of 12 events, and the odds now become
12 / 233 = 5.2%
If you also require it to be one location, we can exclude 2 of the 12 events due to them being in multiple locations.
10 / 233 = 4.3%
And by requiring the even more restrictive definition of it not being gang/drug/family related, you can exclude 3 (and possibly 4) additional events.
7 / 233 = 3.0%
6 / 233 = 2.6%
Because the definition is so unsettled and no one can seem to agree on what the definition should be (or even if there should be a set definition in the first place, as seen in this event which is not included in the restrictive statistics in spite of the fact that 16 people were shot), you can settle on whichever definition fits your worldview and claim any statistic that you want.
Now, it is impossible to know exactly what methodology they are using to calculate the "9 out of every 10 days" number, but by changing your definitions, you can certainly create that number.
