Suspect Based Investigation

“A wrongful conviction is an organizational accident. Many small failures, no one of them independently sufficient to cause the event, combine and cascade, and only then produce a tragedy”  - Doyle, J. M.

 Looking Beyond the ‘White Bears’ in Criminal Justice   The Crime Report (June 6, 2017).

•Within days of Janet’s disappearance the investigation had taken so many missteps it started to look like a case study in how to create a wrongful conviction. Standing in 2024, and having access to a recent scientific study that examined 50 cases of actual innocence that juries got wrong, it’s shocking to see how predictable it was that Auburn PD and PCSO would prosecute the wrong man, and let the real killer go free to murder more innocent women.

Case Deconstruction of Criminal Investigative Failures

Prematurely shifting from an evidence-based investigation to a suspect-based investigation can shut down evidence collection efforts. Confirmation bias distorts the evaluation of evidence reliability, alters probability assessments, and confuses logical evidence analysis.

The most common organizational problem was lack of proper supervision and management. This void enabled a number of errors, including confirmation bias and incompetent investigations. In certain cases, police management ignored (and perhaps unofficially encouraged) misfeasance and noble cause corruption. A police agency should have the necessary procedures and regulations in place to make sure basic investigative steps regarding evidence are followed. Supervisors can control risky investigative practices and monitor illogical investigative conclusions, while managers can prevent the development of noble cause corruption.

•Chief Nick Willick hired, promoted and assigned Auburn PD Detective Danny Boon—and Willick was supposed to direct his actions, and keep his cases on track. Sheriff Nunes was responsible for supervising PCSO Inspector Johnnie Smith. CalDOJ Criminalists were called in to act as disinterested experts with specialized knowledge in evidence collection, preservation, examination and testing. Boon had zero experience in homicide investigations, and Smith had a pile of unsolved cases—including two double murders. The forensic search failed to collect the documented evidence pointing to a stranger kidnapping, and didn't check any of the adjoining properties for signs of a disturbance or footprints. Nobody was looking for, or following the evidence. 

Case Deconstruction of Criminal Investigative Failures

Personal factors were the most frequent cause of criminal investigative failures, particularly wrongful convictions. Premature judgment often led to tunnel vision and confirmation bias. Confirmation bias then produced problems of poor thinking, illogic, witness reliability misjudgment, and flawed evidence assessments. Tunnel vision creates an overly narrow focus on a particular investigative theory and leads to the unconscious filtering of information.

Confirmation bias is a type of selective thinking. Once a hypothesis has been formed, our inclination is to confirm rather than refute; we tend to look for supporting information, interpret ambiguous information as consistent with our hypothesis, and minimize any inconsistent evidence. Types of confirmation bias include: (1) the biased search for evidence; (2) the biased interpretation of information; and (3) biased memory (selective recall).

Confirmation bias can cause a detective, prosecutor, or forensic scientist to focus on evidence confirming the investigative theory, while ignoring or refusing to look for contradicting evidence. 

• Failure to preserve the evidence — Despite being notified within 24 hours of Janet’s disappearance, Auburn PD waited four days to conduct any investigation inside the residence where Janet was last seen, and six days before calling in CalDOJ to conduct a full field investigation. 

• Lack of Neighborhood Canvass —Auburn PD reports, witness interviews, and trial testimony prove that law enforcement only made contact with three of the thirteen homes on the Forest Court Cul-de-sac, the two homes immediately next to the Kovacichs, and one home on the opposite end with whom Chief Willick was close friends. There were multiple properties surrounding the backyard, and the large, rural convent property owned and operated by the Sisters of Mercy that were completely ignored. Nobody on the streets leading into Forest Court were asked if they had seen any people or vehicles, including Paul, Janet, or the Bronco.  

•  Failure to Communicate —Even after the CalDOJ field investigations on September 14th, Auburn PD failed to use any of the existing statewide systems to communicate the information. While it was not yet mandatory, missing person cases in the same jurisdiction used CLETS and CII to alert other agencies, including dental records and descriptions. The result was an absence of other clues, and no possibility of an MO match to other cases, or known offenders.

• No Investigation of Paul’s Alibi — APD first learned of Paul's alibi information the day after Janet was last seen. The same information was repeated again in multiple interviews. Neither APD nor PCSO investigators ever bothered to interview any witnesses at those locations, or request video.  As a law enforcement officer, Paul wrongly believed that once the investigators had the alibi information, they would take these necessary steps to verify his whereabouts between 10:00 am and 12:00 pm, so they could “get him out of the way,” and look for Janet. 

•Tunnel Vision — The investigators intentionally avoided confirming any information that could harm their “theory” of the case. They didn’t look at the newspaper for the sale on Valvoline. They didn't go to K-Mart, Payless, Rock Creek Plaza security, Beverly Brunkhorst, or the gym. They didn’t take fingerprint lifts at the house, or collect the broken window and pried screens. They didn’t have Fuzz autopsied. They didn’t contact Sacramento city or county authorities to ask for matches to known offenders. 

 •Order not to obstruct — Because of Paul’s status as police officer, he was instructed not to contact any witnesses, confirm his own alibi, search for Janet, or interfere in any way. This prevented him from obtaining any corroboration while still available. Worse yet, the prosecutor later painted all of that as proof that Paul didn't care about Janet, and it prejudiced his available defense.

• Incompetent Investigation —Following the interview of Jean Gregoire (Janet’s mother) on September 12, 1982, Auburn PD Boon and PCSO Johnnie Smith had decided to take everything Jean told them as direct evidence rather than opinion, and they treated it as proven fact that Paul had killed Janet. 

-As if on cue, prior to the CalDOJ field evidence searches, Boon and Smith held a group meeting, and “briefed” the criminalists on their conclusion that Paul had killed Janet in the Forest Court home, and transported her body in the Bronco. They made Jean Gregoire’s guesses sound like fact, and referred to Paul as “suspect” on all of the forensic search documents. Boon and Smith not only introduced bias into the DOJ teams, but they personally directed them at Forest Court, and told them to ignore any evidence that pointed to a stranger abduction—and they did. A broken window, and bent screens turned into “no sign of forced entry.” 

-Supervisor Lou Maucieri ignored the field manual he wrote, and failed to collect the broken glass, window, and screens. His fingerprint specialist, Marty Collins, didn't collect even one print lift because if he found an unidentified fingerprint, it would hurt the “case” Boon and Smith were building against Paul. All of the evidence collected and tested by CalDOJ was from the back of the Bronco because that fit the theory they were trying to prove. 

•Chief Willick and Sheriff Nunes should have told Boon and Smith not to refer to Paul as “the suspect,” or give the forensics team any information beyond the fact that Janet disappeared from the home with her purse within a certain two hour span on the 8th. Any Auburn PD or PCSO officers on the scene should have stayed out front to handle the neighbors, and should not have entered the property beyond the driveway so as not to contaminate the scene, or distract the forensic officers. That was a failure of command and leadership. Wilick and Nunes were in charge of making sure they followed the evidence, not a suspect theory, and did not stray from procedure in order to “make their case.”

 •These investigative traps were taught to properly trained homicide detectives, but neither Boon nor Smith had received that type of instruction—and relied on their “guts,” and “instincts.”

Case Deconstruction of Criminal Investigative Failures

Existing evidence is interpreted in a biased manner; evidence that supports the investigative theory is taken at face value, while contradicting evidence is skeptically scrutinized. Other manifestations of confirmation bias include the failure to search for evidence that might prove a suspect’s alibi, not utilizing such evidence if found, and refusing to consider alternative hypotheses. It often leads to logic failures, which are closely tied to probability errors (e.g., believing something is likely when it is not, or vice versa). Intuition, rush to judgment, tunnel vision, and groupthink all pose risks to objective and accurate evidence evaluation and analysis; probability errors and faulty logic led detectives to defective conclusions. 

• No examination of alternate theories — Because investigators failed to collect any other evidence, they developed a severe case of “tunnel vision,” focused entirely on Paul. From the entire period of 1982 to 2008, over multiple investigations...  not one other suspect was ever questioned or checked. Even though Chief Willick had been stalked, and his own family targeted at their nearby home (1,400 feet away), no connection was explored. Auburn PD also had a witness who said that Janet reported being followed and stalked at the same time as Chief Willick in October/November of 1979. Was Paul the target of some displaced anger against a police officer? Was Fuzz poisoned to get him out of the way? Who let Adolph out of the secure area on the night before Janet was last seen?  Was Janet another victim of the “Cul De Sac Killer”? 

• Ignoring Conflicting Evidence —The broken window noted by CalDOJ Criminalist Lou Maucieri was a missed opportunity. The manual he helped write provided clear protocol on collecting and examination of that evidence, which he failed to follow. Two investigators (APD Odin & PCSO Landry, in the company of CalDOJ Robertson & Ito) exhumed Fuzz on 9/14/82--and simply noted: “All appeared to be in order.” If the investigators wanted to conduct an autopsy or examination, that was their opportunity to do so. In early 2000, when the Placer Coroner was conducting an examination of human remains, why was the male femur not tested for DNA? Why has that same leg bone still not been tested using forensic genetic genealogy, and identified? Why was the skull found with dentures that did not belong to Janet?

•Linkage Blindness—No link to other Placer/Sacramento crimes or offenders was ever investigated. Because Boon and Smith failed to upload the details of Janet’s case to CLETS or CII to look for an MO match or information from other jurisdictions, they didn’t see where the local sex offenders or parolees were on September 8th. Within a few days, they simply stopped considering the possibility that anyone other than Paul was responsible for Janet's disappearance, despite obvious abductions with matching methods and victimology over an extended period of time. As noted in the Case Deconstruction study:  “Linkage blindness was an issue for serial offenses, as the failure to connect the offender’s crimes prevented the development of a complete picture of the series and undermined potential alibis of innocent suspects.” 

Using the Case Deconstruction analysis, this is a view of the Initial Investigation (1982-1983)

Using the Case Deconstruction analysis, this is a view of the Eason Investigation (1996-1997)

Using the Case Deconstruction analysis, this is a view of the 

Jerry Johnson Investigation (2004-2007)

Case Deconstruction of Criminal Investigative Failures

The study identified the causal factors associated with wrongful convictions, and ranked the factors based on frequency. The number of factors associated with this investigation (especially dominant factors like cognitive bias, evidence failures, and external issues) is significant.   

•The delayed investigation made it impossible to answer the most basic question in the case:  Was Paul Kovacich at Rock Creek Plaza, the PCSO office, and the gym from 10:30-11:45 am on September 8, 1982?

When the Placer DA discovered that Paul’s alibi was never checked, he should have told law enforcement that the case was unprosecutable due to their own failure to conduct a minimally competent investigation. This is especially true since Sheriff Nunes ordered Paul not to contact witnesses, look for Janet, make public statements, or ask the investigators any questions about Janet’s case. The DA’s decision to prosecute Paul knowing that he very well may have been exactly where he said he was—making it impossible for him to have killed Janet—was a clear violation of Paul’s due process rights, and fundamentally unfair. 

•Critical witness like original case investigators, Jean Gregoire, and Forest Court neighbors were dead, and couldn't be interviewed by the defense, or have their prior statements cross-examined for inconsistencies and biases.