Why serial killers go uncaught for years

A case almost never starts as a serial case. A body is found in one county, then 8 months later a similar one in the next, then a third a year later in another state. Each has its own investigator, its own file, its own deadlines, and 20 other cases. No one compares the cases: that requires someone who can see all 3 at once, and often no such role exists in the system. Criminologists have a name for this: linkage blindness.

The second obstacle is the background noise. An unsolved murder doesn't look unusual: there are already so many of them. One extra case in a county's statistics doesn't stand out, so there's no reason to raise the alarm.

Victims aren't chosen at random. More often, they're people whose disappearance will be noticed late: people without a permanent address, runaway teens, women working on the street. A report is filed weeks later, not within a day. By then, the trail has gone cold, witnesses have moved away, and the car she was last seen in has long since been washed.

Then comes the second trap: the suspect has already been found, but not recognized. Gary Ridgway, who killed dozens of women near Seattle, came to the police's attention in the early 1980s. He was questioned, took a polygraph test in 1984, and gave a saliva sample in 1987. The sample sat in storage until 2001 - until DNA analysis became accurate enough to link him to the victims. All that time, the killer wasn't hidden, just untested.

In movies, a psychological profile eliminates the wrong people in one conversation. In reality, it describes a type: approximate age, likely job, habits. That helps prioritize a list, but almost never points to one specific person. When researchers compared completed profiles with killers who had actually been caught, their accuracy was only slightly higher than that of people with no special training who read the same materials. Cases are more often solved by boring things: a database match, a sample tested in time, a parking ticket issued nearby.