Serial Killers Are More Likely to Be Victims Than Killers

But let's try not to think serial killers are safe

Do you know this man?

If you don’t recognize him, that’s good. After all, he’s a very disreputable person.

That man is Dr. Harold Shipman, arguably one of the most prolific serial killers in modern history and almost certainly the most prolific British serial killer. The extent of his crimes was documented in a report by the British government that was aptly titled The Shipman Inquiry.

The inquiry into Shipman’s crimes came two years after his 1998 arrest, following finding him guilty of fifteen counts of murder. The scope of the inquiry was extensive: there were 618 deaths that were being investigated as possible Shipman murders. Five years later, when the final report in the inquiry was released, the number of murders Shipman was known to be responsible for had nearly squared to 218. The true number might have exceeded 260.

Shipman was able to kill so many people because his method of murder was hard to detect: he prescribed vulnerable elderly people drugs that would kill them at the doses he gave them. The only reason he was caught was because he raised suspicion by leaving himself £386,000 in a will he forged for a patient he killed.

Shipman was, by all accounts, a prolific murderer, but his methods are not typical of serial killers, either in fact or in the public imagination.

Presumably thanks to popular media, the public’s image of a serial killer is a Bundy, Gacy, or Dahmer. The public thinks serial killers are personally violent psychopaths who have narrowly managed to hide their numerous crimes. In one sense, that isn’t that far from the mark. As of 2016, the Radford University Serial Killer Database showed only 16 serial killer victims in the U.S. who’d been killed by drug overdose, making Shipman atypical. On the other hand, there were 4,259 who were shot, 2,153 who were strangled, 1,471 who were stabbed, 910 who were bludgeoned, 714 who were poisoned, 145 who were axed, and 247 others who were drowned, burned, smothered, run over, or neglected and abused.

Another area where the public image and reality diverge is quantitative: with 3,204 killers and 11,274 victims, there were 3.52 victims per killer. In other words, there haven’t been enough victims to make each killer’s crimes necessarily “numerous”. Serial killers may meet the technical definition of a serial killer (>2 or >3 murders), but the typical one is little more than a deranged person. The modal serial killer did what they did for fun (31.8%), followed by financial gain (30.1%), then anger (18.1%), multiple motives (9.8%), gang activity (6.3%), avoiding arrest (1.4%), being in a cult (0.7%), convenience (0.6%), or for attention (0.5%).

These people are sick, but they’re not typically a Bundy, Gacy, or Dahmer, they’re just some dumb sod.

With that said, how would you like to be alone in a room with one?

---

What’s the Risk?

If you’re a sane, normal person, you probably don’t want to be alone in a room with a serial killer, regardless of their methods or motivations.

But you’re probably wondering why I asked.

I ask because there are many people who shouldn’t mind it if they’re being consistent.

Have you ever heard someone say something like

The mentally ill are more likely to be the victims of crime than the perpetrators.

The phrase is so common that even if you don’t remember hearing it, I’m sure you’ve still been exposed to it. If you read Jonathan Rosen’s The Best Minds, then you at least read it there. But regardless, I doubt you’ve heard the following phrase:

Serial killers are more likely to be killed than they are to kill.

This is technically true, but it’s still understandable to be confused.

The reasoning behind the latter quote is as simple as the reasoning behind the first one. Simply put, if we take the 3,204 American serial killers mentioned in the Serial Killer Database and we take these 89 serial killer executions for granted as the total number of serial killers the U.S. has put to death in the same sample, then almost 3% of serial killers have been executed. Because serial killers are a very small population, that means that they’re at a much greater risk of being killed than killing if you use the numerically intuitive, epistemically empty, and unfortunately common definition of risk behind those quotes up there.

Let’s imagine we have a sample of all of the American serial killers from the 20ᵗʰ century through to 2016 and they’re alive and have done their killings in a single year. So, we have 3,204 killers and 11,274 victims.¹ Now, when all their killing is done, we’re going to execute 89 of the killers, call that victimization of serial killers, and set America’s total population to 300 million. Do note: these numerical choices are generous to the position that serial killers are not disproportionately at-risk.

If 11,274 members of America’s population of 300,000,000 were killed by serial killers and 89 of the 3,204 serial killers out there were killed, then these groups have victimization rates of 3.758 and 2,778 per 100,000. Serial killers are accordingly at 739 times greater risk of being killed by Americans than Americans are at risk of being killed by serial killers. To put this in visual terms, the general population’s victimizations by serial killers are invisible.

Even trebling the number of serial killers and their victims while reducing the size of the general population and the number of serial killers they kill by two thirds, the result would be that serial killers are at nine-times greater risk of being killed by members of the general population than members of the general population are to be killed by serial killers.

So why fear serial killers?

Obviously because they’re violent and being around them increases your risk of being killed. In so far as this “risk” metric tells us they’re the ones who are at risk, it misleads.² The absurdity of this risk metric is incredibly apparent in this example because everyone in the group declared more at risk was a murderer.

So let’s talk about some of the other common uses of this misleading risk metric.

---

Psychosis

Compared to members of the general population, individuals suffering from psychiatric disorders are more likely to victimize others. As this graph shows, they are also more likely to be victimized by others.³