The True Odds of Shooting a Bad Guy With a Gun

Who we actually shoot is crazy.

Posted Jan 12, 2017

More than 30,000 people are and shot and killed in the U.S. each year.  About two-thirds of these, 20,000 people, are suicides.  In other words, two out of three people who use a gun to kill a person use it on themselves.

If people want to kill themselves, we should grant them this right.  Some of the happiest places do, like Switzerland and Oregon. But if suicide is what we're going to use guns for, then bullets should be sold in the value-added packaging that is being used to package peanuts and lego figures: in packages of one at a time and one per customer. Guns should also be limited to Derringers or single-shot rifles. This has the added value that if someone looney shoots one in a crowded space, people then get to beat the salt out of him.

Just as an aside, using a gun to kill yourself is selfish. Guns are messy. Gun wounds ruin furniture, wallpaper, and appliances. They also show a general disrespect for other people around you. Someone is going to have to clean up your mess. Cleaning up your life is the better strategy, and it's way more fun.  Just go ask someone who's done it. Seriously.

A bit less than a third of all gun deaths are homicides. That's about 10,000 people a year or roughly twice as many pedestrians as get hit by cars. In America, you're more likely to get hit by a bullet than a car. 

Homicides are often classified as one individual intentionally killing another. Walking into a room with a handgun you think isn't loaded and accidentally shooting your old girlfriend's boyfriend (who used to be your best friend) as he tries to jump out of the way is considered an accident by most law enforcement officials.  Remember that. This is a case where guns kill people and people just stand around and watch. There are about 500 gun accidents of this kind a year. One out of 10 of them will be a child.  

Homicides are against the law and this really is a law worth enforcing.  Statistics show that most people think about killing someone else at least every once in a while.  Statistics also show that the people we tend to think about killing most are the people we actually hang out with.  Et tu, Brute, and that kind of thing. You can see how this could turn bad pretty quickly. A law against homicides is a law that protects potential murderers from being lonely because they killed all of their friends.   

People like to argue that they need guns to protect themselves.  About 250 Americans manage to shoot a bad guy per year.  Granted many more of them may protect themselves without shooting the bad guy, so we have to take what follows with a big grain of salt.  It's impossible to estimate this protective effect without making a lot of assumptions.  For example, it's hard to know how often that "This property is protected by Smith & Wesson" sign works, but I imagine it occasionally does. 

Given that important caveat, let's do the math on what remains. After all, we're interested in how often bullets wind up in bad guys, not how often guns protect us.

What we find is this:

One justified gun death per 2 accidental deaths.

Five justified deaths per accidental child death.  

About 1000 illegal deaths per justified death.

About 100 suicides per justified death.  

About 100 homicides per 2 justified deaths.

Those odds are way worse than flipping a coin. At least then a gun would have a 50% chance of killing a bad guy. But no. A person pulling the trigger on a gun is most likely to be shooting themselves, then their family, then commit a felony, then way, way, way down the line, if they're lucky, they hit a bad guy.   

To add some more painful perspective, you are about 1500 times more likely to be shot by a true American than by a terrorist. Hunters, true Americans according to many, accidentally shoot about 1000 people a year, roughly 50 times more than terrorists. About 500 people a year are killed by mass shooters, still putting terrorists to shame.

It's totally reasonable to argue that the odds of shooting a bad guy without a gun are zero.  But it's a bit like drinking a steady regimen of Drano to prevent parasites that may not ever actually appear. It's not wise, but it could work. 

The reality seems to be that when our bullets don't wind up in non-human animals or street signs, then when you use one to shoot someone, about 99 times out of a 100, you will commit a felony, shoot yourself, or shoot someone by accident.  In one out of a hundred cases, you shoot the bad guy. 

my twitter

and If Norwegians had guns when Breivik showed up, then what? 


Gun data is from

The stats here are from various sources, which don't always say the same thing. Many of those statistics require assumptions, like what is a "terrorist" death and when is a death "justified."   That's why I use the word "about" a lot. If you think you found the holy grail of statistical truth, please put it in the comments, and I'll take a look.  

More Posts