Geographical Profiling

The Casebook of the Bedroom Killer


When the TV series Numb3rs premiered, a show whose premise was based on the application of mathematics to criminology, its first episode had the math genius used a series of complicated formulas to determine a series of increasingly smaller geographical areas in which their offender either lived or worked in.

With a lot less sophisticated technique, I  used the formula found on page 93 of Mapping Murder: The Secrets of Geographical Profiling by David Canter.


Unscientific speculation based on my research into the case. 

In my speculation based on research, I've felt that if there were two main EARs operating in Sacramento, one (EAR-B), the original offender, lived in the Rancho Cordova Area and that the later attacker (EAR-A) who would also be the DNA-linked killer in Southern California, lived in the general area of Foothill Farms/ Arden. In southern California, I think the DNA-linked killer lived in the general Irvine area, either in the neighborhood of the two murdrs there or somewhere between Irvine and Dana Point. I base this speculation on the attacker's knowledge of the Bedroom Basher (local news), and that after 5 years he returned to Irvine for his last murder. I think that this killer drove up to Ventura to commit that crime in order to get it blamed on the Goleta Killer. I think the Goleta killer lived in the Goleta area, probably north of the crime scenes and near the San Jose Creek or one of its tributaries.


EAR-A: Area between Arden and Foothill Farms

EAR B: Rancho Cordova Area

Unified: Rancho Cordova Area

Southern California DNA Linked: Irvine or area between Irvine and Dana Point

Goleta: Goleta north of the attack scenes

  Map of Unscientific Speculation



Scientific Geographical Profiling

How It Works

The system outlined on page 93 of Mapping Murder, is a simple coordinate system.  The geographical profiler draws intersecting  north-south (NS) and east-west (EW)  coordinate lines on the edges of the map showing the area  of the crimes. Then he/she draws parallel east-west and north-south lines from each crime scene to the EW and NS lines respectively. Then the distances from these lines to the crime scenes are each measured. The sum of each (NS and EW respectively) set of measurements is then divided by the number of crimes. The result gives the average coordinate for the NS and EW axes. The profiler measures along each axis the requisite distance and then draws a perpendicular line across the map at the measurement. Where the two lines meet is the 'center of gravity' of the crime scenes.  An example of how this was done can be found in this pdf file.

A flaw in this system is that if an offender started in one area then branched out (as I think actually happened in this case), this is not taken into acocunt.  Somehow the earliest attacks would have to weighted more than later ones in my opinion. Therefore, I think the results are strongest for the Rancho Cordova and Contra Costa examples and weakest for those involving the whole East Area of sacramento.


Rancho Cordova. I used only the crimes in Rancho Cordova and added one  variant: including or excluding the Maggiore murders

Larger Image

Larger Image

Sacramento. I did two variants on this: all the crimes, and the EAR-A and EAR-B done separately. The results were pretty close in all three cases.


Larger Image


Larger Image

Contra Costa. I did this in two variants: all the supposed EAR crimes in CCC and split into EAR-A and other crimes.




Larger Image

Larger Image

Southern California. I did this in two variants: including Goleta and not including Goleta. 



Larger Image


Orange County.   

 Larger Image


 Complete Case Geo Profiling Charts (including worksheets)