Figures lie and liars figure.
It is easy to twist data or limit data ranges or dates to show a different result than what a broader data analysis would show.
That's true of any given dataset.
What we see here is a complete redefinition of that dataset. Instead of including only those Mass Shootings as most people understand them (4 or more deaths in a single incident), they also included gunshot injuries so the number of mass shootings is multiple times larger. The FBI definition includes ALL mass murder events regardless of the means used (explosives, knives, automobiles, etc.). I guess that FBI number isn't large enough to push the anti-2A agenda, so Gun Violence Archive (GVA) had to expand the criteria -- even though nobody in government uses those stats for analysis since you're now mixing apples and oranges.
So, if a lone victim is shot by a home invasion crew, and the VICTIM manages to shoot and wound/kill 4 of the home invaders before ending the break-in, GVA would classify this as a mass shooting: "an incident in which four or more people are shot or killed, not including the shooter."
By any definition, when a good guy with a gun stops an intended mass shooting (only 2-3 people actually killed/shot), that isn't classified a mass shooting at all. It's pure speculation as to whether or not the gunman would have succeeded in shooting or killing more than 4 people as he stood in the middle of a shopping mall. Duh!
We are funneling more of our tax dollars to the CDC to study gun-related violence. Couldn't they spend a week's worth of this time and money hashing out standard definitions and criteria for a variety of incidents to track? Otherwise, we're playing a game of "Telephone" with players who all speak a different language.