There's now an algorithm which catches serial killers, but police aren't listening
If there's one thing that never fails to make me paranoid in the dead of night, it's the thought of serial killers prowling the streets. If you're a true-crime enthusiast with an overactive imagination, it's impossible after a while to shake the chilling suspicion that a killer could be on the loose. The problem is that, even today, with all our sophisticated technology and investigative methods, finding and capturing a serial killer is still an incredibly challenging task.
Worse, there are so many deaths and disappearances around the world, missing people or cadavers recovered where the cause of death is uncertain, there's no guarantee that a potential serial murderer isn't going to slip through the cracks. A psychopathic killer could hunt for decades with impunity and, if they were lucky enough, the authorities would be none the wiser.
Serial killers often feel like a thing of the past; a relic of more dangerous and less-informed decades like the 70s and 80s, and few modern killers have managed to achieve the level of notoriety that made men like Ted Bundy, Jeffery Dahmer, Dennis Rader, David Berkowitz and Edmund Kemper so rightly feared in their time. The acclaimed FBI criminal profiler John Douglas, (the basis for the characters of Jack Crawford and Will Graham in Thomas Harris' Hannibal Lector novels) estimated in his non-fiction book Mindhunter: Inside the FBI Elite Serial Crime Unit that approximately 25 to 35 serial killers are active in the United States at any given time.
But hard data and computer science appear to have contradicted Douglas' experience - with figures that will disturb even the most desensitised individual. Yes, criminal profiling is about to take a great leap forward thanks to remarkable new programming, and according to new figures determined algorithmically, approximately 2,000 murders recorded between 1983-2016 can actually be attributed to the work of serial killers.
Former journalist-turned-homicide-archivist, Thomas Hargrove, has come up with an algorithm which he believes can not only identify serial killers, it can also assist investigators in apprehending them before they kill again. The problem is, most police precincts aren't interested in the project. Hargove's magnus opus is his Murder Accountability Project, now the biggest archive of American homicides available, which identifies 27,00 more criminal cases than the FBI has a record of.
Hargrove believes that his programming has given him the power to pour over staggering volumes of data and stats and see connections via commonalities which human analysis would miss. The Murder Accountability Project crunches data provided by police reports and then identifies clusters of unsolved murders which correlate by modus operandi, geography, time of murder, the victim’s gender and other environmental/circumstantial factors.
The now-retired news reporter first made a name for himself when he applied his skills in data analysis comparing rates of Sudden Infant Death Syndrome (although known as "crib death" or "SIDS") between hospitals in the states of Florida and California. Hargrove found that, according to numbers provided by the Centres for Disease Control, far more babies in Florida died from accidental suffocation than children in California. Hargrove discovered that SIDS wasn't a disease per say, but a way for doctors to write off cases of suffocation when babies were swaddled under blankets. Hargrove was praised for his investigative work, but he wanted to move on to bigger and better things; "the first thing I thought was, 'I wonder if it’s possible to teach a computer to spot serial victims,'" he stated in a recent interview.
Hargrove started his project by asking for homicide reports dated between 1980 to 2008, which comprised of approximately 500,000 murders. First he sorted these killings by their type. Murders where the victims had been strangled or bludgeoned were statistically more likely to be serial killings than cases where the victim had been stabbed or shot. Serial killers prefer to sadistically draw out the act, so there is a clear commonality between people killed in this fashion. Approximately 70 per cent of serial-killer victims are female, compared with the majority of conventional victims of homicide, which are mostly male. This data, working in conjunction with the MO data, revealed more similarities. He did this, day by day, for over a year, until more and more connections gradually revealed themselves.
The algorithm also bears certain limitations. Since it depends upon geographic location as a search term, killers who choose to murder further away than adjacent counties are lost in the data. Furthermore, police departments have been slow to pick up on it. Recently, Hargrove used the Murder Accountability Project to predict that a serial killer was loose in Gary, Indiana. Police were informed about the 15 eerily similar unsolved strangulations in the area, yet Hargrove stated that the response was: "absolute radio silence. They would not talk about the possibility that there was a serial killer active." It took another seven victims before Gary's local police department took notice. Hargrove called this "the most frustrating experience of my professional life."
Hargrove is also concerned that if his algorithm makes any kind of error, his project will be shut down. "What if they arrest the wrong guy, and he sues?" he stated, "I contacted a bunch of police departments in 2010, when I was a reporter, because I wanted to see if the algorithm worked. Now I know it works—there’s no question in my mind. In certain places, we can say, ‘These victims have an elevated probability of having a common killer.’ In 2010, though, I had a big media company behind me, with lawyers and media insurance. Now I’m a guy with a nonprofit that has fourteen hundred dollars in the bank and a board of nine directors and no insurance."
Hopefully, police will begin to implement Hargrove's revolutionary work into their investigative procedures in the near future. It seems as though it is capable of making great strides in criminal profiling, and as we have learned ourselves, there are a hell of a lot of bad people out there. Check out our article all about the five notorious serial killers who have never been caught if you don't believe me.