Preface: I'm a shooting enthusiast and competitor who recognizes we have some issues related to them in our country. As a professional marketing exec, I'm very data driven and make decisions based on facts. Finally, I'm a marginally talented blogger with some shit to say. So, digging in to gun related data and writing about guns in general is a natural extension for my blog. I write about things that interest me and do so for my own enjoyment. So, I don't have the first clue how to monetize what I've written and have no interest in doing so. My only goal is those who read my blog feel as though their time was well spent or at least not wasted.
A word on suicides: In my opinion, suicides are completely different animals than homicides, because they're self-inflicted. The motivation for taking your own life is dramatically different from that of taking another's, so I don't include them in my numbers.
Gun Homicides: The Numbers
Like most other folks, I hear about the gun violence problem in the media and have become concerned. We're bombarded by the message that thousands of people are being gunned down in mass shootings every day. I wanted to understand if this was true or bullshit, so I spent a lot of time slicing, dicing, and validating gun related homicide numbers. Some of my results can be found in several previous entries, but it's become so spread out, I have difficulty remembering where or if I shared a certain tidbit. I've decided the stats deserve their own entry that can be referred to as needed and updated as I receive new data.
Like most other folks, I hear about the gun violence problem in the media and have become concerned. We're bombarded by the message that thousands of people are being gunned down in mass shootings every day. I wanted to understand if this was true or bullshit, so I spent a lot of time slicing, dicing, and validating gun related homicide numbers. Some of my results can be found in several previous entries, but it's become so spread out, I have difficulty remembering where or if I shared a certain tidbit. I've decided the stats deserve their own entry that can be referred to as needed and updated as I receive new data.
I'll begin with some basic data, followed by info on mass events, and finish up with semiautomatic rifles. I've tried to be light on commentary and heavy on charts and graphs; it's still long.
Sources and methodology are cited at the end to avoid cluttering the thing up. I apologize that some of the visuals may run large; there's a lot of info. Also, while I may have a certain position on guns, data doesn't lie. The material below hasn't been massaged to affect a desired outcome; I've been willing to let it take me wherever the truth lie.
Sources and methodology are cited at the end to avoid cluttering the thing up. I apologize that some of the visuals may run large; there's a lot of info. Also, while I may have a certain position on guns, data doesn't lie. The material below hasn't been massaged to affect a desired outcome; I've been willing to let it take me wherever the truth lie.
Chart 1 shows gun related homicides by year, from 1988 through 2021, represented in both absolute numbers as well as per 100k. You can see the murder rate was much higher in the early 90's, then dropped precipitously until it settled at its lowest between 2009-2014. From that point onward, it's been on an upward climb.
 |
| Chart 1 |
Chart 2 represents gun ownership versus gun homicides per 100k during the same period. Gun ownership has only varied by about 24% whereas homicides per 100k has swung more than 50%. Since 2017, the murder rate has increased 27%, whereas gun ownership is unchanged, save the small dip in 2019.
 |
| Chart 2 |
Gun Related Homicides by Firearm Type
Table 1 shows gun related homicides by year from 2012 through 2021. The raw data that came from the FBI included a significant number of "Firearm Not Specified". This spreadsheet shows how I allocated that chunk across the Handguns, Rifles, and Shotguns (complete methodology at the end). The FBI has indicated the 2021 figures were understated; apparently, they changed reporting methods and not all LE organizations were signed up yet.
Table 1 shows gun related homicides by year from 2012 through 2021. The raw data that came from the FBI included a significant number of "Firearm Not Specified". This spreadsheet shows how I allocated that chunk across the Handguns, Rifles, and Shotguns (complete methodology at the end). The FBI has indicated the 2021 figures were understated; apparently, they changed reporting methods and not all LE organizations were signed up yet.
 |
| Table 1 |
Chart 3 is a visual representation of the data. NOTE: Rifles means ALL rifles; assault rifles will be a subset of that number and addressed below.
 |
| Chart 3 |
Table 1.5 is a clean version of the calculations in Table 1, with the addition of homicides committed by all other weapons, for context. Apologies if it's a bit of an eye chart.
 |
| Table 1.5 |
Mass Murders / Mass Homicides and Mass Shootings
The figures below use the FBI's definition for mass homicides and the generally accepted definition of mass shooting. I previously devoted an entry to the topic which you can find HERE. As you'll see, the number of fatalities from mass events is a pretty small chunk of the total. They fluctuate between 1.6 - 4.2% of total gun related homicides. Mass murders run between 0% and 0.7% of gun related homicides.
Table 2 shows the stack up of mass shootings, mass murders, and others by year. Chart 4 shows the same data in, you guessed it, in a graphical format.
I must admit to being surprised, when I put this data together for the first time. We've been conditioned to believe a false reality.
 |
| Table 2 |
 |
| Chart 4 |
AR-15's
Everyone wants to see the data on AR-15's because it's the evil black rifle and kills a small town every week. I have a feeling some are in for a bit of disappointment. The data for mass murders is known, whereas the percentage of the whole has been calculated, using the best logic I could conjure while on the throne. I outline the methodology below.
NOTE: For mass homicides, I've lumped all semiautomatic rifles together in the AR-15 bucket. This consolidates a bit of a spread out category, so I'm not forced to break out AK47's, SKS's, AR-10's, etc. Also, it would be intellectually dishonest not to do so. For reference, the number of incidents since 2000 would have dropped from 35 to 25, with the number of fatalities going from 413 to 295.
Since we're talking about mass events, I'll start there.
Graph A shows the breakdown of fatalities in mass murders by use of AR's versus other guns. The AR number is much smaller than many think it is. Note that it didn't become popular until very recently; many believe its use began to spike immediately following the Assault Weapons Ban's sunset. You can clearly see the impact of the Pulse Nightclub in 2016. In 2017, the numbers were particularly high as well, due to the Vegas Strip and Texas First Baptist killings.
Table A provides the detail, except I limited it to the last ten years.
 | ||
Graph A
|
School Mass Homicides using AR-15's, while horrifying, aren't as common as some would lead you to believe. There have been five of them.
1989 Stockton Schoolyard Shooting 6 fatalities
2012 Sandy Hook Elementary 27 fatalities
2018 Marjory Stoneman Douglas High School 17 fatalities
2022 Robb Elementary, Uvalde 21 fatalities
2018 Marjory Stoneman Douglas High School 17 fatalities
2022 Robb Elementary, Uvalde 21 fatalities
2023 Nashville Covenant School 6 fatalities
There have been 15 school mass homicides where an AR-15 (or similar) wasn't used, with a total of 126 fatalities.
(updated 3/27/23)
AR-15: Overall Impact
Finally, let's circle back to where we started and revisit the overall numbers. Table B below has the same data as Table 1.5, except I've broken the rifle numbers into AR and non-AR homicides. The breakout numbers are calculated; methodology at the bottom. The numbers are inflated, in my opinion, but I wanted to avoid claims I was working an agenda. Graph B is the visual representation of the chart. Note how other homicide methods compare to the AR-15. More people are killed with knives and blunt objects per year than semiautomatic rifles. On a positive note, fewer people are being strangled to death every year. Again, apologies for the eyechart.
 |
| Table B |
 |
| Graph B |
One omission some may identify in these numbers is the category of school shootings. While I think school shootings are particularly heinous, there's no solid data on them, beyond the mass murders I mention above. The listings of school shootings I've reviewed are rather broad in their definition of what a school shooting is, perhaps to boost the numbers in support of a particular narrative. To me, a school shooting is when an outsider or student enters the school with the intent of killing others. Someone being shot in a school parking lot at midnight or a student accidentally firing the gun in their backpack are not school shootings. Unless a solid data source comes my way, I prefer to avoid the topic altogether, from an analysis perspective.
Sources:
Chart 1 & 2: CDC data, Violence Policy Center: Gun Ownership in America, November 2022
Chart 1 & 2: CDC data, Violence Policy Center: Gun Ownership in America, November 2022
Table 1 & 1.5 and Chart 3: Homicide data - FBI
Table 2 and Chart 4: CDC, Gun Violence Archive, Mother Jones mass murder database
If you pick through the various FBI pages, you can find the old spreadsheets, such as the one I used as my base. However, a few years ago, they migrated to a crime explorer that sucks, because you can't mass download data. https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/home
My data is a combination of spreadsheets I'd downloaded as well as hand copying data from the explorer.
Mass murder data - Mother Jones Database
Note: MJ includes mass events with 3 fatalities to be a mass murder versus the FBI's definition of 4 fatalities. My version of the database has been scrubbed accordingly.
Mass shooting data - Gun Violence Archive
Methodology:
Table 1 - You can see where I tried to integrate what was in the "Firearms, type not stated" bucket, which represented a not insignificant number. For example, it was 3,300 in 2019. The logic I used was if handguns represent X percent of the grand total, then I'll consider handguns to be X percent of the Not Stated as well. I did the same with rifles and shotguns. There are still a lot of "other" guns unaccounted for, but I didn't want to run the risk of skewing the data in such a way to make it not meaningful, if my logic was wrong.
Table 1.5 includes my numbers calculated in Table 1, along with a pure extract of straight FBI data.
Graph 2 - CDC data
Table 2 - Because GVA's mass shooting number includes the MJ mass homicide number:
Total homicides - mass shootings = non mass events (other)
Total homicides - mass shootings = non mass events (other)
Mass shootings (GVA) - mass homicides (MJ) = true mass shootings
Chart 4 - data came directly from Mother Jones Mass Shooting database.
Table B - AR-15 related fatalities were calculated in the following manner:
AR Fatalities =
Total Rifle fatalities - 100 to account for hunting accidents; AR's are rarely used in hunting
Subtotal x .7 (which I think is being incredibly generous)
+ Fatalities from mass murders where AR-15 was present
4/23/23 Update
A note on datasets: I've been relying exclusively on data from the FBI for overall numbers as well as breakdowns by weapon type. However, they freely admit to screwing the pooch on implementation of a new reporting platform in 2021, noting the numbers for that year are understated. Furthermore, they don't make their historical data available. The dataset that everyone uses for their analysis comes from the CDC, but their numbers are 25% higher than the FBI's, until 2019. Seriously, CDC's top line is exactly 25% higher than the FBI's from 2012 through 2018. In 2018, the gap jumps to 29%, increasing to 33% in 2020. A gap pegged at 25% says to me that one of the sources is bullshit and fudged upward or downward, but neither shares their methodology, so identifying the bullshit source is beyond my capabilities. Therefore, going forward, I'll be using CDC data for top line numbers and FBI stats for their breakout of gun type and consider both to be bullshit, but they're what's available, kids.
This entry is part of my "Gun Series" that focuses on providing insight into the gun debate and gun violence. You can find the other entries in the series HERE.
About the author: Sean R is a recovering conservative who owns a consulting firm specializing in strategic marketing. He's been a competitive shooter since the early 90's and holds a High Master classification in PPC and a Master classification in USPSA. As an instructor, he taught courses in gun safety and competition. He lives in Raleigh, North Carolina with his overly vocal dog, Sadie.
No comments:
Post a Comment