Derrick Todd Lee, Baton Rouge Serial Killer

While media agencies are comparing the FBI Behavioral Analysis Unit's early profile of the traits of the Baton Rouge serial killer to the primary suspect, Derrick Todd Lee, several things must be kept in mind.

First, Lee has not yet been tried or convicted for these crimes, so measuring him against the profile makes certain assumptions and relies on hastily gathered material about his life. We don't yet see the entire picture.

Second, despite heavy criticisms by outside investigators during the investigation, each case is unique and the decisions made about whether and how to use a profile are generally unique to that case. Mistakes do get made, but outsiders do not have privileged access to how the decisions were made, or why.

Third, a behavioral profile is just one of many tools in use when trying to solve a crime. Its primary function is to help prioritize leads. It's not based on psychic insight but on what is actually known about the behavior involved in a specific crime or series of crimes, compared with information gathered on other similar cases. The profile usually states that suspects should not be eliminated just because they don't match some specific trait.

Gregg McCrary's The
Unknown Darkness

"A criminal profile details the characteristics and traits of an unknown offender," says Gregg McCrary, a former FBI profiler and author of The Unknown Darkness. "It's based on inferences we can draw about the choices the person made during the crimes. Behavior reflects personality, but you have to be careful that the information you're using is accurate and complete. If there's information out there that hasn't been collected, that can affect the accuracy.

A profile is based on the available evidence, and faulty information can guide the profile in the wrong direction. In this case, people supplied information about white trucks and white male drivers associated with several of the crime scenes, and that was necessarily taken into consideration. In retrospect, that looks faulty, but at the time no one could say if it was or was not relevant. (For example, in the case of the Canadian killing team of Karla Homolka and Paul Bernardo, several witnesses "saw" two men in a Camaro. All of them were mistaken, but until the police realized that, the eyewitness information affected how the profile was devised.)

Fourth, there are anomalies in human behavior and while profiles are based on what occurs more often than not, those who use this method know that they cannot predict every possible human behavior. For example, there are more lone white snipers than black sniper teams, so it made sense to suggest that the Beltway Sniper might be a lone, white male. Yet with additional crime scenes and more behavioral evidence, the profile evolved. It did in this case as well. Before Lee was arrested, the police had already indicated they could be looking for a black man, based on "critical new information" that they had learned about the fifth case.

Fifth, it's important to note that the initial profile in this case was based only on the first three incidents, with only the barest amount of forensic evidence. Then with a fourth murder, there was an addendum, which indicated that the offender was familiar with the area where he dumped the fourth body and that he might live with others, forcing him to take the victims to remote locations.