Calculating My Odds

Donald Trump and his ilk are fearmongering about terrorism and President Obama wants to calm my nerves. Maybe it’s just that I’m congenitally not a worrier, but in my daily life I’m not in the least worried about being the victim of a terrorist attack or a mass shooting, by ISIS, right-wing extremists, or anyone else. Mass shootings and attacks are horrendous, and I can’t imagine how horrible it must be for the people who survive them and the loved ones of those who don’t, but anyone’s actual chances of being involved in one are extremely slight.

According to shootingtracker.com, there have been at least 462 people killed and 1312 injured in 353 mass shootings this year, which is already more than the 383 people killed last year. They define a mass shooting as one in which four or more people are killed or injured, which is a broader definition than the government has used. (Compare that to the 30K or so people killed in auto accidents each year.) According to the Census Bureau, the U.S. currently has a resident population of 322,367,564, giving me a 0.00000143% chance of getting killed in a mass shooting this year by the broadest definition. That’s almost literally a one in a million chance, the phrase we use when something is so unlikely that we won’t bother to worry about it.

Of those 353 shootings, 2 seemed to be related to Islamist terrorism (Chattanooga and San Bernardino), resulting in 19 deaths and 19 injuries. That’s from a site called thereligionofpeace.com, which is trying to prove how murderous Muslims are. By being so diligent in tracking deaths by Muslims, they ultimately show that Muslim terrorists have been responsible for .006% of the attacks and .041% of deaths in mass shootings this year, making my chances of being killed this year by a such a terrorist infinitesimal. So thanks, anti-Muslim website, for reassuring me.

But since deciding what counts as a mass shooting is difficult, let’s look at the most conservative data. Mother Jones magazine has been tracking mass shootings since 1982, and they use the criteria of four or more people killed, which means they don’t consider the recent Planned Parenthood attack a mass shooting, since “only” three people were killed. According to Mother Jones, there have been four mass shootings in the U.S. this year. According to shootingtracker.com data, there were 41 such shootings so far this year. However, they include domestic shootings such as this one and this one, which, while perhaps indicative of a culture of guns and violence, are in a different category than mass shootings in public places with the intent to kill randomly. One more kink. In 2013 Congress lowered the federal number of deaths required: “the term ‘mass killings’ means 3 or more killings in a single incident.” That would allow us to add the Planned Parenthood attack to the four mass shootings tracked by Mother Jones.

So at a conservative estimate there have been 5 mass shootings in the U.S. this year: San Bernadino (14 dead), Oregon (10 dead), Charleston (9 dead), Chattanooga (5 dead), and Colorado (3 dead). Of the 5 shootings, 2 are related to Islamist terrorism, 2 to what I would consider right-wing extremist terrorism (Colorado and Charleston), and 1 (Oregon) to who knows what craziness. By this reckoning, this year, there was a 40% chance of a mass shooting being related to Islamist terrorism, and 46% of the people killed in mass shootings were killed by Islamist terrorists, but still more people were killed by non-Muslim white American men than were killed by Muslims.

That’s a large percentage, but of a tiny number of events, and by these standards my odds increase significantly and I have a 0.00000013% chance of being killed in a mass shooting, and a 5.89389322e-8% chance of being killed by a Muslim terrorist. So despite all the fearmongering, there have been 2 Islamist terrorism incidents in the U.S. this year, giving us individually an insignificant chance of having been killed in one.

[Update] If we expand this analysis over several years, it looks even better. According to thereligionofpeace website, there have been 89 people killed by Muslim terrorists in the U.S. in the last 14 years. (I’m leaving out 9/11 because that is WAY outside the norm.) That averages to be about 6.4 people killed per year. Dividing that average by an average population of 320 million people each year gives a likelihood of death by Muslim terrorist of 2e-8%. That’s even lower than my previous estimate, because the San Bernadino shooting was uncharacteristic, and the largest number of dead since 2009. That’s a 1 in 50,000,000 or so chance each year that I’ll die in a terrorist attack in the U.S.

Just as who you associate with might lead to a domestic shooting, where you work might as well. I work at a university, where bomb threats seem to be increasing. None have actually been bombed because while any dipshit can buy a gun and start shooting people, making bombs is hard, and none of the shootings seemed related to Islamic terrorism. According to this Wikipedia list of attacks related to post-secondary schools, there have been 6 such attacks this year, leaving 17 people dead (including the 10 dead at Umpqua Community College in Oregon). There are 4,140 public or private colleges in the U.S., enrolling almost 18,000,000 students and having tens of thousands of employees. Again, an infinitesimal chance even within the domain of higher education.

None of this is an argument against fighting terrorism and mass shootings or trying to stop gun violence. However, it is an argument that an abnormal fear of being killed in a terrorist attack or mass shooting in the U.S. is unwarranted, and that the fearmongering is little short of demagoguery. Yes, they could happen anywhere, but they’re extremely unlikely to happen here, wherever here is. If the goal of terrorists is to terrorize me, they haven’t won.

3 thoughts on “Calculating My Odds

  1. A friend commented on Facebook: “It’s kind of a weird calculation. To be fair, you’d probably need to take a period of say 20 years, come up with a single year average either for within the United States or all U.S. citizens killed globally, maybe include a function that factors in a marginal increase projecting into the future and then figure out compound lifetime risk. Your point is sound, but the single year methodology is inherently problematic as evidence.”

    And my response: “I agree completely. This was a passing thought that I decided to spend a few minutes on doing the most basic of calculations. However, I suspect I’d find pretty much the same thing if I analyzed many years of data. There just aren’t that many terrorist attacks in the U.S. in any given year, and if we use the conservative estimation for mass shootings, there aren’t many of those, either. Despite the fear mongering, the fact is that relatively few people in the U.S. have been killed by terrorists since 9/11. The yearly death rate is very small.”

  2. Okay, a bit more. According to thereligionofpeace website, there have been 89 people killed by Muslim terrorists in the U.S. in the last 14 years. (I’m leaving out 9/11 because that is WAY outside the norm.) That averages to be about 6.4 people killed per year. Dividing that average by an average population of 320 million people each year gives a death rate by Muslim terrorist of 2e-8. That’s even lower than my previous estimate, because the San Bernadino shooting was uncharacteristic, and the largest number of dead since 2009.

Leave a Reply

Your email address will not be published. Required fields are marked *