In Melton v. Klee, 2019 WL 1315723, the United States District Court, Eastern District of Michigan, Southern Division, considered a pro se petition for a writ of habeas corpus following the petitioner’s conviction for assault with intent to rob while armed. 

The petitioner’s conviction arose from the attempted robbery of the victim outside her home in Michigan.  During the attempted robbery, the would-be robber panicked, ran to his vehicle and drove away.  A police officer observed a vehicle matching the description provided over the radio and followed it until it was abandoned and the driver fled on foot.  The officer gave chase but eventually lost the driver.  The officer found a BB gun in the car and an identification card belonging to the petitioner.  The dashcam video from the officer’s patrol car was played for the jury.  The video player used in court did not have the capability to play the video frame by frame.  The officer testified that he had previously viewed the video frame by frame and was able to see an individual in the backseat of the vehicle.  It was the prosecution’s theory that the petitioner had been in the backseat of the vehicle while his brother had been the driver. 

The petitioner’s brother testified that he was the actual would-be robber and that the petitioner was not present.  The petitioner testified that he had lent his borrowed vehicle to his brother and was not present during the attempted robbery, nor was he in the backseat when the car was abandoned.  The trial judge disbelieved both the petitioner and his brother and convicted the petitioner, finding him to be the would-be robber. 

The petitioner exhausted all avenues of appeal in the state courts and thereafter filed a petition for habeas corpus in federal court.  One of his claims for relief was ineffective assistance of counsel, arguing that his trial counsel should have and failed to have the dashcam video examined by an expert in forensic video analysis.  Had he done so, he argued, it would have been established that the petitioner’s brother was the sole occupant of the vehicle.

During the post-conviction state court proceedings, the trial court held an evidentiary hearing on this issue and the petitioner’s trial counsel testified.  He said that he reviewed the dashcam video on a frame by frame basis and several frames showed someone in the backseat of the vehicle.  The individual appeared to be a light-skinned black male.  Trial counsel testified that he did not want to retain an expert because he believed such analysis would reveal information detrimental to the defense.  The trial court found that trial counsel was not ineffective in failing to retain a video expert since having seen the second individual in the vehicle, he reasonably concluded that an expert would only serve to support the prosecution’s theory that the petitioner and his brother acted in concert to rob the victim.

The Expert Evidence

In the federal court habeas corpus proceeding, the petitioner argued that a report by a [LEVA] certified forensic video analyst proved that trial counsel was ineffective in failing to retain an expert.  This report was first presented in state court on appeal from the trial court’s denial of his motion for relief from judgment.  The federal court examined the report of the analyst, who concluded that no one was in the backseat of the vehicle.  The basis for the analyst’s conclusion was described by the Court as follows: 

[The analyst] characterizes the experiment upon which he relied in preparing his report to be “not a scientifically accurate experiment.” Id. at PageID.112. [The analyst] isolated four consecutive frames from the video. The isolated frames revealed “an object that could be identified as a human face,” but not a particular human face. Id. at PageID.101. [The analyst] conducted an experiment to determine whether the “face” seen in the video could have belonged to a person seated in the back seat of the car (the location Officer Butler testified the second person was seated). Id. [The analyst] could not obtain a Chrysler Sebring, so he conducted the experiment [using] a Chevrolet Malibu, a vehicle of “similar proportions.” Id. at PageID.112. [The analyst], who is the same height as Petitioner, apparently then attempted to assume a position in the backseat of the vehicle which would have placed his face in the same location as the face seen on the video. [The analyst] “found it impossible to assume the required position.” Id. Without accounting for differences in flexibility or mobility, [the analyst] concluded his experiment demonstrated “the improbability that a person would be so positioned.” Id. Based in part on this “improbability” and the assumption that someone in the backseat would try to hide his face from police not place it in the window, [the analyst] concluded that the “alleged face” is not a face at all, but an “unfortunate combination of light and shadow on the driver’s side head-red [sic] of the vehicle that gives the appearance of being a face.” Id.

[NOTE: The name of the analyst has been removed from this case study as the purpose of this case study is not to draw attention to the particular analyst but rather to highlight the concerns raised by the Court as to the soundness of the expert’s methodology and opinion.]

The Court noted that no curriculum vitae or other evidence of the analyst’s qualifications or expertise were provided to the Court.  Allowing leeway on that point, the Court stated, “Even assuming [the analyst] is qualified as an expert in video analysis, the report is largely based on conjecture and speculation.”

In rejecting the habeas corpus claim, the Court stated:

[The analyst’s] report fails to show that counsel was ineffective in failing to retain an expert witness or that Petitioner suffered prejudice. First, counsel articulated a reasonable strategy for not calling an expert. Second, the expert report is wholly unpersuasive and does not offer exculpatory evidence. Third, expert testimony that the face was actually a shadow, even if it was deemed credible, would not have outweighed the evidence incriminating Petitioner, because the victim’s testimony identifying him as the gunman, Petitioner’s wallet was in the vehicle, and Petitioner was previously acquainted with the victim.

Petitioner has not demonstrated that counsel’s strategic decision not to retain an expert was unreasonable or that he was prejudiced by counsel’s decision; nor has he shown that the state court’s holding that counsel was not ineffective was contrary to or an unreasonable application of Strickland. The Court, therefore, denies this claim.

Analysis

Forensic video analysis is defined as the scientific examination, comparison and/or evaluation of video in legal matters.  Conducting a technical evaluation of video images for the purpose of interpreting what is observable is properly part of the role of a forensic video analyst.  The exercise of the analyst positioning himself in the backseat of a different model of vehicle and then offering an opinion as to whether it was probable that the petitioner could have been so positioned in the crime vehicle is not scientific at all.  When the analyst himself states that the exercise is “not a scientifically accurate experiment”, that is an indication that the opinion proffered is subjective in nature and perhaps ought not to have been given.  Opinions offered by an expert should be objectively verifiable as that is a hallmark of scientific reliability.  The further subjective opinion that someone trying to hide from the police would keep his face out of camera view has no place in forensic video analysis.  Experts owe a singular critical duty to the court to assist in the discovery of the truth, even if doing so does not further the interests of their client.  These issues understandably caused concern for the federal court and led to the rejection of the opinions tendered.