(concurring in part and dissenting in parí). I disagree with the majority that the trial court’s decision not to grant defendant a new trial on the basis of the evidence it considered was an abuse of discretion. However, I agree with the majority that the trial court should have included within its determination defendant’s argument regarding the alleged bad-faith destruction of dna evidence that might have exculpated him. Accordingly, I would remand this case to the trial court for consideration of that issue *160and, if sufficient evidence of bad faith is presented, a redetermination of the motion for a new trial.

As the majority opinion amply demonstrates, defendant produced newly discovered evidence that was not merely cumulative and that could not have been produced with reasonable diligence at the 1985 trial. Accordingly, the only question at issue here is whether that evidence was “such as to render a different result probable on retrial . . . .” People v Lester, 232 Mich App 262, 271; 591 NW2d 267 (1998). The majority correctly states that we may reverse the trial court’s ruling on a motion for a new trial (or for relief from judgment) only if there has been an abuse of discretion. People v Ulman, 244 Mich App 500, 508; 625 NW2d 429 (2001); People v Jones, 236 Mich App 396, 404; 600 NW2d 652 (1999). The question thus before us is whether the trial court abused its discretion in determining that the new evidence was not sufficient to render an acquittal of defendant probable on retrial.

Initially, the trial court decided that the new evidence did suffice in meeting that standard but, after our Court declined to grant an application for leave to appeal that decision, the trial court heard further information regarding the new evidence and reversed its decision. Admittedly, had the trial court maintained its original position and granted defendant a new trial following receipt of the new information, there would have been no abuse of discretion. As the majority opinion shows, the record here would support the conclusion that a new trial was warranted.

That is not to say, however, that the record does not support the contrary result that was, in fact, reached by the trial court here. This was, in other *161words, a “close call” and one ordinarily left to the trial court. The abuse of discretion standard we must employ on appeal is highly deferential, requiring affirmance unless the trial court’s decision here was “ ‘so palpably and grossly violative of fact and logic that it evidences not the exercise of will but perversity of will, not the exercise of judgment but defiance thereof, not the exercise of reason but rather of passion or bias.’ ” People v Babcock, 244 Mich App 64, 76; 624 NW2d 479 (2000), quoting Spalding v Spalding, 355 Mich 382, 384-385; 94 NW2d 810 (1959). Great deference is especially warranted when the lower court’s decision is based on the credibility of witnesses. People v Canter, 197 Mich App 550, 560; 496 NW2d 336 (1992).

Applying these standards here, we should not reverse the trial court’s determination that the new evidence, presented to a new jury, would probably not result in an acquittal for defendant at a new trial. The trial court’s decision here was largely based on an assessment of witness credibility. The trial court apparently determined that, notwithstanding the impeccable credentials of all the experts, the medical testimony that was inconsistent with Ronning’s account of how he killed the victim was more believable than the medical testimony that was not inconsistent with his account. The trial court reviewed Ronning’s videotaped confession to the murder of the victim along with the testimony of four witnesses who stated that Ronning told them that his confession was false, as well as Ronning’s explanation of why that disavowal was itself false. The trial court determined that Ronning was not believable, especially in light of the discrepancies between his *162description of the place where the crime occurred and the actual facts. The trial court assessed the credibility of recantations made by three witnesses who testified against defendant at trial and, while considering the recantations to be significant, concluded that they alone would not warrant a new trial. This is not surprising, especially in light of the trial testimony of four nonrecanting witnesses who recalled how defendant had admitted in graphic terms how he raped and killed the victim. Finally, unlike the majority, I see no mistake of law or other impropriety with the way the trial court analyzed the new evidence that it considered.1 I do not conclude that it was an abuse of discretion to deny defendant a new trial on the basis of that evidence.

Nonetheless, I agree with the majority that the trial court erred in failing to consider defendant’s argument regarding the alleged bad-faith destruction of DNA evidence as part of the motion for a new trial determination. I would remand this matter to the trial court for further factfinding regarding whether there was bad faith. If sufficient evidence is submitted by *163which a factfinder could conclude that there was bad faith,2 a powerful argument that this new evidence would render a different result probable on retrial would be presented. As the majority correctly notes, the appropriate remedy would be an instruction that, upon a finding of bad faith, the jury should infer that the destroyed dna evidence would have been favorable to defendant, had it been preserved. If the trial court determines that the jury at a new trial would be so instructed, the analysis regarding whether a new trial is warranted would likely be greatly affected.

I would remand for further proceedings consistent with this opinion.

Defendant, who in 1985 was convicted of first-degree felony murder, MCL 750.316, appeals by leave granted from the trial court’s order denying his motion for a new trial. We reverse and remand.

i

The victim in this case is Patricia Rosansky, a seventeen-year-old girl who was murdered in early 1983. To summarize the voluminous record generated during defendant’s twenty-six-day jury trial, we reprint the following portion of this Court’s 1988 opinion affirming defendant’s conviction for Rosansky’s murder:

The body of the victim ... was found in a wooded ravine, covered with brush on April 6, 1983, in Bedford Township near Battle Creek. She was identified by dental records. Two pieces of tree limb were found located in her throat, which testimony indicated would have been forcibly put there to obstruct her voice. An autopsy and post-mortem examination revealed that the cause of death was a brain injury due to a blow or blows to the head with a club-like object. There were also defensive injuries on the back of her hands. Several bruises also existed on the back of her *113legs, buttocks and neck. The body was discovered naked from the waist down with panties located at the ankles. There was evidence of forced anal intercourse.

[The victim] had been missing since February 3, 1983. She had left for school that morning at 7:30 a.m. and was last seen by a school friend. Defendant presented one witness who testified that she had seen [the victim] a few days after she was missing when [the victim] and another person drove past her house in a pickup truck. Another witness testified that she had seen [the victim] get into a pickup truck the morning of February 3. The witness also stated that she saw a man other than defendant get out of the same truck at a prior time.

Several witnesses testified that defendant had made statements to them concerning the sexual assault and death of a girl. Defendant lived near [the victim] and had spoken to her. John Moore testified that he lived with defendant and heard defendant state in February 1983 after coming home in the evening, that “he felt a little better because he went and knocked off a piece.” He also heard defendant say that he had killed [the victim].

Terry Moore testified that he lived with defendant and that in July 1983 defendant took Terry, his brother Walter and Cindy Lesley to a wooded area and pointed out where Patty’s body lay (and was later found).

Candy Moore testified that defendant came to her house almost every day in the spring of 1983 and told her on two different occasions that he had killed a girl named Patty and put her in a ditch.

Emery DeBruine testified that in May 1983 defendant saw him in a bar and told DeBruine that defendant had raped and killed a girl because she refused to have sex with him. Defendant also said that it was a perfect crime and that no one would know about it.

Walter Moore testified that defendant had told him that defendant had picked [the victim] up and that they had smoked marijuana. Defendant wanted to have sex and when [the victim] refused, he raped her, killed her and dumped the body in a wooded area. Moore was a convicted *114felon and at the time of confessing his own crimes, he told the police about defendant’s statements.

Defendant allegedly asked several people whether they had ever had sex with a corpse.

Cindy Lesley testified that defendant had taken her out to the ravine where [the victim] was found and told her that he had killed [the victim] and left her body in the ravine after he covered her. Lesley called the police and eventually received a $5,000 reward.

Officers Nick Pestum and Marion Bagent were allowed to testify as to prior consistent statements of Walter Moore, Candy Moore, and Cindy Lesley, for the limited purpose of refuting defendant’s charges that the witnesses were influenced.

Shirley House testified that she was the Moore family’s landlady, and when she was at the house repairing the steps, had heard defendant say, “I cannot believe that I got so hard up I had to kill the bitch for a piece of ass.”

Defendant took the stand in his own defense and denied that he killed [the victim] or that he had told anyone that he did so. He stated that he was delivering papers on February 3, 1983. He presented an alibi witness, Doug Moore, who testified that on February 3, 1983, he and defendant were delivering papers. Defendant did admit to dumping trash illegally at the site where the body was found.

After closing arguments by both counsel, the court instructed the jury on first-degree felony murder, with the predicate felony being first-degree criminal sexual conduct, second-degree murder and voluntary manslaughter. The jury returned a verdict of guilty of first-degree felony murder. [People v Cress, unpublished opinion per curiam of the Court of Appeals, issued February 4, 1988 (Docket No. 86748).]

This Court rejected defendant’s claims on appeal that the prosecutor improperly introduced prior consistent witness statements, that the trial court erred in failing to order a corporeal lineup before one witness’ in-court identification of him, and that the court improp*115erly admitted photographs of the victim’s body. The Supreme Court subsequently denied defendant’s application for leave to appeal this Court’s decision. People v Cress, 431 Mich 856 (1988).

In 1997, defendant filed a motion for a new trial on the basis of newly discovered evidence. Defendant argued that the following three grounds entitled him to a new trial: (1) his trial counsel rendered ineffective assistance by failing to introduce at trial and losing an audiotape recording of a telephone conversation in which prosecution witness Walter Moore allegedly admitted that he and other prosecution witnesses set up defendant by testifying that he acknowledged committing the victim’s murder, (2) prosecution witnesses Walter Moore, Candy Moore,1 and Cindy Leslie had recanted their trial testimony, and (3) Michael Ronning recently admitted murdering the victim.

Battle Creek Police Detective Dennis Mullen ascertained Ronning’s potential involvement in the victim’s murder while investigating the August 1982 murder of twenty-year-old Margaret Hume. Hume had been killed inside her Battle Creek apartment. In 1986, the Arkansas State Police apprised Mullen that Ronning, whom they had arrested for an Arkansas murder, had lived in the same Battle Creek apartment complex in which Hume resided at the time of her murder. Mullen’s investigation of Ronning and his travels led Mullen to form the opinion that Ronning likely had committed murders of numerous young women in Arkansas, California, Michigan, and Texas. Mullen re*116counted that by 1991 he had concluded with certainty that Ronning murdered Hume, and believed that a high degree of probability existed that Ronning also murdered several other young women, including the victim, in and around Calhoun County during the early 1980s.

Mullen initially visited Ronning in an Arkansas prison in 1987 to seek information regarding Hume’s murder, but Ronning refused to cooperate. Mullen returned to Arkansas in early 1992 to inquire whether Ronning had any interest in confessing to the murders he had committed in exchange for being able to serve a life sentence of imprisonment in Michigan where members of Ronning’s family resided. During the 1992 visit, Mullen advised Ronning that someone was in prison for one of the Battle Creek area murders that Mullen believed Ronning committed. Mullen also mentioned to Ronning his opinion that Ronning had committed a third murder in the area.

In late summer 1996, the Governors of Michigan and Arkansas entered into an executive agreement that Ronning would be transferred to the Michigan Department of Corrections to serve life without parole provided that Ronning offered “Mirandized confessions” of the details of all his Michigan murders, pleaded guilty of these murders, and waived appellate review of his convictions.2 After arriving in Michigan in the fall of 1996, Ronning took a polygraph examination. Ronning responded affirmatively *117when questioned whether he had committed three homicides in Michigan, and the polygraph examiner determined that Ronning had responded truthfully. In the spring of 1997, Ronning made three “Mirandized,” videotaped confessions of committing the victim’s murder that provided varying levels of detail regarding the crime. Ronning also was videotaped during two unsuccessful attempts to locate the exact location where the police discovered the victim’s body. Ronning later signed an affidavit attesting that he alone had murdered the victim.

On November 5, 1997, the trial court held an evidentiary hearing regarding defendant’s motion for a new trial. In December 1997, the trial court granted defendant’s request for a new trial. The trial court’s opinion did not rely on the missing audiotape recording; the court did not consider the tape as newly discovered evidence but instead found that the tape could have been discovered and produced at the time of defendant’s trial. With respect to recantation, the court found evidence that could impeach the testimony of several of the prosecutor’s trial witnesses, but ultimately was “not convinced that this evidence standing alone would justify ordering a new trial.” The court most heavily relied on Ronning’s confession in concluding that defendant should receive a new trial. The court explained as follows:

It is important to note some observations concerning the trial testimony. There were no eyewitnesses to the murder of Patricia Rosansky. . . . There was absolutely no physical evidence linking the Defendant, Mr. Cress, to this crime. The only evidence connecting him to the crime was the testimony of several witnesses ... all of whom testified that Mr. Cress had admitted to each of them his involvement in Ms. Rosansky’s murder.

*118* * *

This Court has had the opportunity to review the videotaped statements of Michael Ronning in which he confesses to the murder of Patricia Rosansky. Parts of his statements agree with the established facts in this case, and parts of his statements may not agree with the established facts.

. . . [I]t appears to this Court that to deny the Motion for a New Trial in this case, one must be able to conclude that Mr. Ronning’s confession is incredible, unbelievable, or simply unsubstantiated by the established facts. This I cannot do for several reasons.

First, there are portions of Mr. Ronning’s statements which do conform to the established facts in this case. Second, although there are parts of his statements which may not be in conformity with the established facts, it must be noted that we are dealing with events which occurred 14 years ago. Given that lapse of time, it is possible that one’s memory of some of the specific details may be sketchy. And finally, there is the testimony at the hearing of Battle Creek Police Detective Dennis Mullen.

Detective Mullen testified that he has been working on this murder case and two others since the 1980’s. He stated under oath at the Hearing that he encouraged the Prosecutor’s office to issue an arrest warrant against Michael Ronning for the murder of Patricia Rosansky. The testimony clearly indicates that Detective Mullen, based upon his knowledge of the circumstances surrounding Patricia Rosansky’s murder and his subsequent investigation, believes Mr. Ronning’s confession is true.

It is obvious that Detective Mullen and the Prosecutor’s Office have a difference of opinion concerning the believability of Michael Ronning’s confession. That difference simply indicates to this Court that the Ronning confession cannot be summarily dismissed. Ultimately, at a new trial, the jury may believe Mr. Ronning and acquit Thomas Cress. On the other hand, the jury may totally reject Ronning’s confession and convict Mr. Cress of Murder.

*119Considering the fact that at Mr. Cress’ trial, there was no physical evidence connecting him to the crime; that his conviction was based solely upon the statements attributed to him by several prosecution witnesses; that some of those witnesses may have recanted their trial testimony; and that Mr. Ronning’s confession cannot be deemed incredible or unbelievable, I believe that the Defendant has met his burden of establishing the four factors . .. required for granting a new trial. . . .

... It will be up to a new jury to weigh all the evidence presented, including Mr. Ronning’s confession, and then determine whether there is evidence beyond a reasonable doubt that Thomas Cress committed the murder of Patricia Rosansky.

In January 1998, the prosecutor filed in this Court an application for leave to appeal. Shortly thereafter, defendant filed in the trial court a motion “for evidentiary hearing and dismissal of the charges.” Defendant alleged that his discovery of more than seven hundred pages of the prosecutor’s investigative materials revealed that prosecutor John Sahli had knowledge by January 1992 that Ronning had indicated his responsibility for the victim’s murder, but that Sahli nonetheless in May 1992 authorized the state police to destroy all evidence accumulated during the investigation of the victim’s murder, including a sanitary napkin pad containing spermatozoa and a foreign pubic hair sample recovered at the crime scene. Defendant urged that the charges be dismissed because the prosecutor’s bad-faith destruction of evidence had deprived defendant of his due process right to conduct scientific testing of the crime scene evidence to assist in clearing his name.

On August 5, 1998, the prosecutor’s application for leave to appeal was denied by this Court on the basis *120that the trial court had not abused its discretion in granting defendant a new trial. Referring to evidentiary materials developed after the trial court’s ruling, however, the Court noted that “denial of leave to appeal does not preclude a party from asking the trial court to revisit the merits of its order . . . based on information developed subsequent to such order,” including evidence derived from the forensic testing of the decedent’s remains following exhumation. Consequently, this Court declined to affirm the trial court’s order.

On August 6, 1998, the prosecutor filed in the trial court a motion to reopen the proofs relevant to defendant’s motion for a new trial. The prosecutor desired to present new evidence attacking the veracity of Ronning’s confession and more evidence regarding the allegedly recanting prosecution witnesses. The trial court granted the prosecutor’s motion.

During December 1998 and March 1999, eight more days of evidentiary hearing occurred regarding the motion for a new trial. The bulk of the new evidence constituted expert medical testimony regarding the manner of the victim’s death, which testimony the medical experts based in part on examinations of the victim’s skeletal remains. Several prosecution witnesses also testified that Ronning had expressed to them that he falsely confessed to the victim’s murder.

The trial court ultimately reversed its December 1997 decision and denied defendant’s motion for a new trial. The court explained that neither the missing audiotape nor witness recantations warranted a new trial and that it no longer found Ronning’s confession persuasive.

*121The evidence presented since the Court granted the Prosecution’s Motion to Re-open Proofs has established overwhelmingly and convincingly that Michael Ronning is in fact a false confessor to the Patricia Rosansky murder. The primary reasons for this conclusion are as follows:

1. Mr. Ronning stated in his confession that he strangled Ms. Rosansky, and he demonstrated how he struck her one time with a rock to the back of her head. This Court heard from four expert witnesses concerning the blow(s) to her head: two Forensic Anthropologists . . . and two Forensic Pathologists .... Some of the professional opinions of these witnesses are contradictory. When weighing this evidence, one must consider not only the expert’s qualifications (all of which are impeccable), but one must also consider the underlying facts and circumstances giving rise to those opinions. After considering the expert testimony presented in this matter, this Court is convinced that there were in fact multiple blows to the head and neck of Patricia Rosansky. That fact finding is important because although Mr. Ronning is vague and claimed a lack of memory about many details in his description of the murder, he consistently claimed striking her in the head only one time. The expert testimony, whether it be the number of blows to Ms. Rosansky’s head, or the presence of defensive wounds, or the lack of any evidence of strangulation, or the linear, rod-like shape of the object used to strike Ms. Rosansky, all rebut Mr. Ronning’s version of the manner of Ms. Rosansky’s death.

2. There were four people who testified in December, 1998, that at various times over the course of the last several years, Mr. Ronning confided in each of them that he was falsely confessing to this murder in order to do his prison time in Michigan. . . .

It was an acknowledged fact from the outset that Mr. Ronning had a motive to confess to the Rosansky murder

Of those four witnesses, Melissa Meyer was particularly persuasive. Mr. Ronning had been her guardian in 1983-84, and she had a close relationship with him. She testified that Mr. Ronning admitted to her that he had committed the *122murder in Arkansas. He also told her that his goal was to do his time in Michigan and that he had not committed the murder of Ms. Rosansky. She also testified that Mr. Ronning told her he had obtained information from the secretary of his Michigan attorney, had read some transcripts of the court proceedings in this matter, and had attempted to memorize the facts contained therein. She also testified, based upon her prior relationship with him, that Mr. Ronning is a very intelligent and a very manipulative person.

The testimony of these four witnesses is a direct attack on Michael Ronning’s believability. It consistently establishes that Mr. Ronning’s confession is self-motivated and untrue. After considering the testimony of these four witnesses, their demeanor while testifying, and any motives which may have influenced their testimony, this Court finds that this evidence is credible and believable.

3. Perhaps the most compelling evidence which causes this Court to now conclude that Mr. Ronning is a false confessor comes from Mr. Ronning himself. In April, 1997, Detective Mullen and others had Mr. Ronning attempt to show them where the scene of the crime was. This was videotaped and admitted as Exhibit 54. Although there was evidence that Detective Mullen may have caused some confusion by using the wrong two-track to enter the area, eventually Mr. Ronning did come to an area where he believes the murder occurred. He stated on that videotape that there was a clearing where he could turn his car around. He described where the car would have been, where the body was placed after he strangled her, from which direction he would have thrown the rock, and how far the rock would have gone “with the roll.”

The area Mr. Ronning stated “may very well be the place” is shown on the videotape. Although Mr. Ronning qualified his identification of the crime scene by saying “this could be it” and “this has to be it, but I don’t really recognize it per se,” he nonetheless was firm and definite in stating that if the particular clearing they were in wasn’t it, it nevertheless “was a place like this.” The area where Mr. Ronning *123believes the murder occurred is a flat piece of ground, a clearing next to a two-track. There are no man-made landmarks in the immediate vicinity.

At the hearing in December, 1998, numerous photographs were admitted into evidence of the scene of the crime taken in 1983. Those photographs clearly show that Ms. Rosansky’s body was not found in a flat, open area as described by Mr. Ronning. Rather, her body was found in a ravine. This ravine was not just a slight indentation in the ground. Each side rose to a height of seven or eight feet, according to the testimony of Trooper [Harry] Zimmerman. The body was found at the bottom of the ravine, within view of a concrete well station. Mr. Zimmerman testified that the ravine and well station look similar in appearance today, compared to 1983. Indeed, Mr. Zimmerman testified that a metal roof vent shown in the 1983 crime scene photographs is still there. He had no difficulty locating the area where Ms. Rosansky’s body was found.

When one compares the videotape of the area Mr. Ronning concludes was the scene of the crime (or as he said, “it was a place like this”) to the photographs of the scene of the crime, the difference in topography and terrain is dramatic. This is not a situation where Mr. Ronning’s recollection is clouded due to a lapse in time. On the 1997 videotape, Mr. Ronning describes the crime scene based on his recollection. When one compares his description of the crime scene to the actual crime scene, the only reasonable conclusion one can draw is that Mr. Ronning didn’t know where the crime scene was because he did not commit the crime. Indeed, Mr. Ronning was shown the cement well station which is located at the beginning of the ravine about 40 feet from where the body was found. Mr. Ronning said he would have remembered that well station if it had been visible from the scene of the murder. Mr. Zimmerman testified it is easily observable.

The court further rejected its prior reliance on Detective Mullen’s opinion that Ronning killed the victim. The court explained that other police agencies and detectives disagreed with Mullen that Ronning killed *124three young women in Michigan, including the victim, and that Mullen did not investigate Rosansky’s murder, speak with the state police who initially investigated the victim’s murder, read defendant’s trial transcript, or speak with witnesses from defendant’s trial or with defendant himself. The court concluded that “it would be inappropriate in effect to enhance the credibility of Michael Ronning based upon one investigating officer’s professional opinion” because apart from Ronning’s statement, “Detective Mullen’s opinion that Mr. Ronning committed the Rosansky murder is based primarily upon his professional opinion and instinct, as opposed to any newly-discovered facts or evidence obtained during the course of his investigation.” The court lastly noted that Ronning “had the ability and opportunity over the years to obtain information from various sources . . . about relevant facts and circumstances surrounding the Rosansky murder.” The court denied defendant’s motion for a new trial because it “no longer believes that a different result at a re-trial is probable.”

n

Defendant filed his motion “for new trial” approximately twelve years after his initial conviction and many years after exhausting his appeal of right to this Court and denial of his application for leave to appeal to the Supreme Court. Consequently, his postappeal motion seeking a new trial correctly should have been deemed a motion for relief from judgment. MCR 6.431(A)(4), 6.501; People v Kincade (On Remand), 206 Mich App 477, 482; 522 NW2d 880 (1994). This Court reviews for an abuse of discretion a trial court’s *125ultimate ruling on a motion for relief from judgment.3 People v Ulman, 244 Mich App 500, 508; 625 NW2d 429 (2001). We review for clear error the trial court’s findings of fact supporting its ruling. MCR 2.613(C); People v Crear, 242 Mich App 158, 167; 618 NW2d 91 (2000).

Defendant’s motion argued that newly discovered evidence entitled him to a new trial. A defendant seeking a new trial on the basis of newly discovered evidence must demonstrate that “(1) the evidence itself, not merely its materiality, is newly discovered, (2) the evidence is not merely cumulative, (3) the evidence is such as to render a different result probable on retrial, and (4) the defendant could not with reasonable diligence have produced it at trial.” People v Lester, 232 Mich App 262, 271; 591 NW2d 267 (1998).

A

The trial court rejected the notion that Ronning’s confession constituted newly discovered evidence that entitled defendant to a new trial. Instead, the court entirely dismissed it in part on the basis that the medical testimony presented demonstrated that Ronning falsely confessed to the victim’s murder. After carefully reviewing the abundant record in this case, we have reached the conclusion that the trial court clearly erred in finding that Ronning’s admissions of guilt lacked any probative value in establishing defendant’s right to a new trial. The medical testimony did not preclude the possibility that the victim’s *126murder occurred as Ronning recounted, specifically by his strangulation of the victim and subsequent dropping of a rock onto the back of the victim’s skull while she lay facedown on the ground. During one of his videotaped confessions, Ronning demonstrated by lifting both his hands above his head and then forcefully descending them onto the back of the victim’s skull.

With respect to strangulation, the trial court appeared to conclude that the available evidence prevented a finding that the victim was strangled. While the trial court correctly observed that none of the medical experts testified about observing evidence that they could directly relate to strangulation, the record was inconclusive whether strangulation actually occurred. Dr. Bader Cassin, a forensic pathologist who performed a second postmortem examination of the victim’s body on April 8, 1983, testified at defendant’s trial that he looked for but did not find evidence of strangulation. Cassin explained, however, that “evidence of strangulation is very difficult to observe in some cases. And it is important to observe them [sic] before any significant manipulation of skin, soft tissue. Upon my examination, that manipulation had already occurred.”4 According to Cassin, the initial autopsy of the body made “it . . . impossible to make a determination” whether strangulation occurred, and the decomposition of the victim’s body might also have obscured evidence of strangulation.

Dr. Laurence Simson, Jr., a forensic pathologist who examined the victim’s remains in 1998 and testi*127fled at the evidentiary hearing after the trial court reopened proofs concerning defendant’s motion for a new trial, likewise could not point out evidence of strangulation, but explained that depending on the manner of strangulation either minimal or no visible injury might result to the victim’s neck. Simson opined that strangulation could also have caused “a small amount of hemorrhage in the soft tissues of the neck that certainly may not have been visible in . . . the body at the time of the original autopsy because of decomposition.” Simson was not asked to detect evidence of strangulation and did not possess enough information to conclude with certainty whether evidence of strangulation existed. No expert testimony precluded the possibility that strangulation had occurred.5

With respect to the fractures of the victim’s skull, significant expert testimony suggested that these skull fractures could have resulted from one blow by a rock to the back of the victim’s head. At defendant’s trial, pathologist Dr. William Walters opined that the victim died because of a massive skull fracture resulting from a blunt force impact. Walters believed that “[i]t could have been one massive blow or several *128submassive blows.” Walters reviewed x-rays he had taken of the victim’s entire body and observed no fractures other than the massive skull fracture. By the time of the evidentiary hearings regarding defendant’s request for a new trial, these x-rays were missing.

Cassin also testified at defendant’s trial regarding the “very broad and large depressed fracture opening in the [victim’s] skull.” Cassin repositioned within the area of the fracture four or five loose pieces of the victim’s skull and concluded that a large force had struck the left rear area of the victim’s skull. Cassin found it “certainly possible” that the force could have been delivered in one blow, but stated that more than one blow to the same area likewise could have created the fractures. Cassin stated that the force could have arrived in the form of “the drop of a heavy rock to the head from a height of several feet, such as 6 feet, or it could be a swing with perhaps a smaller object, or even a small rod” or a tire iron.

The issue regarding the specific number of blows to the victim’s head received extensive attention during the evidentiary hearings regarding defendant’s request for a new trial. Doctors Simson and Sundick and Dr. Norman J. Sauer, a forensic anthropologist, all testified at the reopened evidentiary hearing regarding their reviews of the victim’s exhumed skull and vertebrae. Each of these experts to varying degrees also reviewed different portions of previous autopsy reports and photographs and statements by examining Doctors Walters and Cassin. Several pieces of the victim’s skull in the left rear area of the large fracture were missing, however, thus preventing Sauer, Simson, or Sundick from definitively determining the number and order of blows the victim’s skull *129suffered. As Simson indicated, “that’s the very thing that makes this case controversial, is we don’t have those pieces.” The experts testifying at the evidentiary hearing agreed with the notion that Cassin, who had the opportunity to replace all the skull fragments now missing, was in the best position to make a definitive determination regarding impacts to the victim’s skull. No photograph existed showing the victim’s skull with all the fragments, including those later missing, in place.

Sauer testified that in light of the large fracture to the left rear of the victim’s skull and the fracturing of the bones forming the base of the victim’s skull, he believed that the victim likely suffered at least two blows to her skull—one delivered in the left rear in the area of the missing bone fragments and one in the area where the base of the victim’s skull met the top of her neck.6 Sauer could not rule out, however, the possibility that one massive blow from an angular *130rock to the back of the head of an individual lying facedown on the ground (as the victim was found) could have produced all the fractures present in the victim’s skull.7

Simson opined that the victim experienced between three and four distinct strikes to her skull.8 Simson believed that a strike to the base of the victim’s skull and upper neck created the open wound in the back of the victim’s neck that exposed her vertebrae, some of which apparently became stained from blood. Simson suspected that the neck wound was inflicted by “an edge” of a heavy object, but could “[n]ot 100 percent” rule out that the open wound in the back of the victim’s neck resulted from a surgical incision during one of the two postmortem examinations by Walters and Cassin.9 Irrespective of how the victim’s neck *131skin was split, Simson believed that the basilar skull fractures and bruise on the back of the victim’s neck evidenced a blunt force strike. Simson concluded that “a single blow with a rock isn’t going to make multiple impacts from different directions,” but agreed that fractures to the base of a skull could occur from a massive blow to the back of someone’s head.10

Cassin testified at the reopened evidentiary hearing that he had no independent recollection of performing the April 8, 1983, examination of the victim’s body. After reviewing his prior testimony at defendant’s trial and the reports summarizing the postmortem examinations of the victim, Cassin restated his conclusion that the victim’s skull fractures could have resulted from either one or multiple blows by a rock or a tire iron.* 11 If one blow occurred, Cassin characterized it as “broad-based” and enormous, similar to the amount of force that would be involved if a head *132struck a hard surface “after jumping or falling from a high level; several stories high up in a building, for instance.”12 Because of some important, unknown variables, Cassin could not state whether a more likely scenario of the skull injuries would involve at least two skull strikes by a tire iron or a single blow by a skull-sized rock with an uneven surface.

After reviewing the extent of damage to the victim’s skull, Sundick opined that all the fracturing most likely resulted from one blow to the left rear portion of the victim’s skull.13 One of the other areas fractured included the fragile basilar skull bone. When presented with a hypothetical question referencing Ronning’s description of his infliction of the victim’s skull fracture by dropping a rock onto the back of the victim’s head, Sundick confirmed that the victim’s skull fractures could have so occurred. The blow could have been delivered by a rock approximately two feet in diameter or a large crowbar or many other objects harder than the victim’s skull bones. Sundick conceded that the victim’s skull fractures might possibly have been inflicted by more than one blow if the additional blow or blows struck the same left rear area of the victim’s skull.

*133Our review of the expert medical testimony regarding the cause of the victim’s death and the potential sources of her fatal skull fractures leaves us with the definite and firm conviction that the trial court made a mistake when it found that the expert testimony precluded the possibility that Ronning damaged the victim’s skull by dropping a large rock on it. People v Thenghkam, 240 Mich App 29, 43, 45-47; 610 NW2d 571 (2000). We reiterate that most of the experts could not describe precisely the number of blows the victim’s skull sustained because of the missing skull fragments.14 Four of the five experts who testified, either during defendant’s original trial or during the evidentiary hearings, opined that the victim’s skull fractures could have resulted from a single massive blow to the back of her head: of these four, Walters and Cassin apparently believed that one blow to the victim’s head was at least as likely as two, and Sun-dick opined that most likely one strike caused the skull damage. Three of the four experts acknowledged that the single blow could have been delivered with a rock.

While Sauer and Simson found more likely that the victim’s killer used a heavy linear object to strike the victim, their determinations rested on their opinion that a lateral open wound to the victim’s neck similarly had been inflicted by a linear object. We observe, however, that the nature and origin of the neck wound was unclear. Autopsy photographs and Cassin’s and Walters’ testimony all tended to establish that the victim’s neck skin was intact until incised *134during a postmortem examination. Walters’, Cassin’s, and Sundick’s testimony all disagreed regarding even the existence of a bruise on the victim’s neck. We note that the trial court did not credit any particular expert testimony over any other, but merely commented that all the experts’ qualifications were impeccable.15 Thus, while the expert testimony did include the possibility that a linear object inflicted the victim’s wounds, this fact was not established with any high degree of certainty.

Furthermore, the trial court clearly erred in interpreting the medical testimony as precluding the possibility of strangulation. While the experts did not determine that evidence of strangulation existed, the testimony of Cassin and Simson establishes that such evidence was not likely to be ascertained under the circumstances,16 and none of the experts could rule out that strangulation occurred.17

*135Accordingly, we conclude that the trial court clearly erred to the extent that it relied on the expert medical testimony to establish that Ronning falsely confessed to murdering the victim.

B

Our review of the instant record also leaves us troubled that in completely rejecting Ronning’s confession to the victim’s murder the trial court failed even to address the fact that Ronning passed a polygraph examination during which he confessed having committed several murders in Michigan. Nearly a quarter century ago the Michigan Supreme Court held that “the judge in a post-conviction hearing on a motion for a new trial based on newly found evidence may in his or her discretion consider the results of a polygraph examination.” People v Barbara, 400 Mich 352, 412; 255 NW2d 171 (1977). Polygraph test results may be considered in deciding a motion for a new trial under the following circumstances: (1) they are offered on behalf of the defendant, (2) the test was taken voluntarily, (3) the professional qualifications and the quality of the polygraph equipment meet with the approval of the court, (4) either the prosecutor or the court is able to obtain an independent examination of the subject or of the test results by an operator of the court’s choice, and (5) the results are considered only with regard to the general credibility of *136the subject. People v Mechura, 205 Mich App 481, 484; 517 NW2d 797 (1994).

Evidence presented at the original and reopened evidentiary hearings established that Ronning took a polygraph examination after arriving in Michigan in the fall of 1996. The Calhoun County Prosecutor’s Office required that Ronning submit to and pass a polygraph examination regarding the number of Michigan murders he had committed before the prosecutor would elicit Ronning’s confessions and further investigate his crimes. The prosecutor apparently arranged the polygraph examination of Ronning, during which the examiner asked Ronning whether he had truthfully indicated that he had committed three murders in Michigan. The examiner’s written report concluded that Ronning truthfully acknowledged committing three Michigan murders.

Under these circumstances, Ronning’s polygraph results were admissible at the evidentiary hearings to prove his general credibility. Mechura, supra. Defendant offered the results on his own behalf, and Ronning testified that he voluntarily participated in the examination. We further note that the prosecutor made no objection to defendant’s elicitation of evidence regarding the results of Ronning’s polygraph examination, and the trial court at no time questioned the professional qualifications of the examiner or the quality of the polygraph equipment employed.

Although the trial court had discretion to consider the results of Ronning’s polygraph examination, Barbara, supra, the record does not reflect that the trial court exercised its discretion in this case. The results of Ronning’s polygraph examination plainly had relevance to the central issue of the evidentiary hearings: *137whether Rorming’s confession consisted of truth or, as the trial court found, fiction. Consequently, we find that the court committed an error of law in this case in failing to even address that Ronning passed a polygraph examination.18 See People v Lukity, 460 Mich 484, 488; 596 NW2d 607 (1999) (noting that a trial court abuses its discretion when it admits evidence that is inadmissible as a matter of law); Thenghkam, supra at 50 (finding that a misapplication of the relevant law results in an abuse of discretion).

To the extent that defendant asserts that the trial court erroneously refused to consider evidence that defendant also passed a polygraph examination during which he denied killing the victim, we conclude that the trial court properly declined to consider the results of defendant’s examination. The polygraph test of defendant, because it serves only to, in effect, duplicate evidence the jury already has heard, i.e., defendant’s testimony at his trial denying his guilt, is merely cumulative, and does not properly fall within the purview of newly discovered evidence. Barbara, supra at 363, n 2.

c

Defendant also argues that in denying his motion for relief the trial court erred in failing to even consider evidence that several of the witnesses who testified at defendant’s trial regarding his incriminating statements perjured themselves. The discovery that *138testimony introduced at trial was perjured may constitute grounds for a new trial, provided that the evidence of perjury otherwise qualifies as newly discovered evidence. Mechura, supra at 483. Where newly discovered evidence takes the form of recantation testimony, however, Michigan courts traditionally regard it as suspect and untrustworthy and only reluctantly grant new trials on the basis of recanting testimony. People v Canter, 197 Mich App 550, 559-560; 496 NW2d 336 (1992). The trial court has discretion whether to grant a new trial on the basis of recanting testimony, and due regard must be given to the trial court’s superior opportunity to appraise the credibility of the recanting witnesses and other trial witnesses. Id. at 560.

In its initial opinion granting defendant a new trial, the court noted the evidence presented that Candy Moore recanted her testimony at defendant’s trial that he had implicated himself in the victim’s murder and that Curtis Feltner, a third cousin of defendant, averred that Cindy Lesley informed him that she had lied when she testified at defendant’s trial that defendant had admitted killing the victim. In its discretion, the trial court characterized the evidence of recantation as significant.19 Although the court did not believe that the recantation evidence alone warranted a new trial, the court did consider the evidence as one applicable factor, together with the fact that Ronning’s confession could not be deemed incredible, in reaching its conclusion that defendant had estab*139lished his right to a new trial on the basis of newly discovered evidence.

After the reopened evidentiary hearings concluded, the trial court reiterated its earlier conclusion that the evidence of recantation alone would not justify defendant’s receipt of a new trial. Because the court totally rejected Ronning’s confession, however, it denied defendant’s request for a new trial. Contrary to defendant’s suggestion, the trial court did contemplate within its second opinion the fact that Candy Moore and Cindy Lesley recanted their original trial testimony incriminating defendant.

The trial court’s opinions indicate that the court did not reject the recantation evidence on the basis of its appraisal of the credibility of the witnesses who testified regarding recantation. Canter, supra at 560. We agree with the trial court that the evidence of recantation alone would not entitle defendant to a new trial because other witnesses who testified at defendant’s trial regarding his incriminating statements had not recanted their trial testimony. Id. at 561. In light of our determination that the trial court mistakenly concluded that Ronning’s confession contravened the medical evidence presented in this case, however, we conclude, as did the trial court in its initial opinion, that the significant evidence of recantation constitutes an additional factor tending to establish that a different result would be probable on defendant’s retrial. Id. at 559.

D

Defendant further suggests that in denying him a new trial the trial court erred in dismissing Detective *140Mullen’s firm belief that Ronning killed the victim. We find that the record does not support the trial court’s conclusion that beyond Ronning’s implication of himself in the victim’s murder, Mullen’s belief in Ronning’s guilt was supported by nothing more than “Mullen’s professional opinion and instinct, as opposed to any newly-discovered facts or evidence obtained during the course of his investigation.”

It was undisputed that Mullen did not participate in the original investigation of the victim’s death. In January 1983, Mullen was assigned to investigate the murder of Margaret Hume, who was approximately twenty years old when she was found murdered in mid-August 1982 in the closet of her apartment, her body covered. The Arkansas State Police notified Mullen in January 1986 that they had arrested Ronning for murdering a young woman between the ages of eighteen and twenty who had been abducted and stabbed, and whose body was found on the ground in a wooded area, covered with branches and debris. After the Arkansas police advised Mullen that Ronning had resided in the same apartment complex as Hume at the time of her murder, Mullen began further investigation into Ronning’s background to eliminate him as a suspect in Hume’s murder.

Mullen’s investigation confirmed that Ronning had lived in the apartment immediately below Hume’s at the time of her murder. Mullen discovered that Ronning had visited his cousin near Dallas, Texas, from approximately August 1 through August 14, 1982, when he returned to Battle Creek. On approximately August 19, 1982, the day after Hume’s murder, Ronning left Battle Creek and returned to Dallas where he again stayed with his cousin.

*141Mullen contacted Ronning’s cousin, who confirmed Ronning’s visits and informed Mullen that his girlfriend had been murdered and suggested to Mullen his belief that Ronning had killed her. Mullen spoke with Bedford, Texas, police who told him that Ronning’s cousin’s girlfriend had indeed been murdered; she was found lying on the ground in a wooded area near a dump, her body covered with shingles. The Texas police further advised Mullen that a second body had been discovered in the same area. The second body was a young female between eighteen and twenty years of age, who was found lying on the ground and covered with plywood. The Texas police believed that the second victim was killed during the summer of 1984. Mullen later discovered that during the summer of 1984 Ronning had worked in Dallas as a subcontractor installing plywood.

Mullen learned from speaking with Ronning’s wife that Ronning had California arrests for kidnapping and rape. The California State Police also wanted to question Ronning in connection with a strangulation murder that occurred in a motel where Ronning had worked. Ronning’s wife additionally assisted Mullen by recounting her travels with Ronning and advising Mullen that at the time of Hume’s murder Ronning wore the same style shoes as those that left a footprint at the scene of Hume’s murder. Ronning’s wife recalled that one night while they lived in Dallas (from September 1982 until January 1983) Ronning arrived home with blood on his clothes, which he later discarded.

Mullen was aware of and considered the circumstances surrounding several murders of young women in and around Calhoun County, including that of the *142victim, which involved similarities to Ronning’s other murders and possible connections to Ronning. The victim was a young female in her late teens who disappeared in February 1983 and in April 1983 was found lying on the ground in a wooded area, covered by a refrigerator door and other debris. Mullen learned that Ronning lived in Battle Creek at the time of the victim’s disappearance and that the victim’s body was discovered within two miles of the apartments where Ronning lived at the time of her murder. Ronning subsequently relocated to a rural area of nearby Bellevue, Michigan, by March 1983. On March 12 or 13, 1983, Carrie Evans, another young female between approximately eighteen and twenty years of age, disappeared from Bellevue and her body was later found lying on the ground and covered with branches and debris. Mullen found that Ronning had enrolled his sister in school in Bellevue and frequently drove there. Evans’ body was discovered within IV2 miles of Ronning’s residence. Before the murders of Hume, the victim, and Evans, Sherrie Edwards had disappeared from Battle Creek in late July 1982 and was last seen entering a two-tone vehicle matching the description of the vehicle that Ronning drove at the time. Edwards was found in some weeds, her body covered. By late 1991, Mullen believed that Ronning had committed all the similar murders in and around Calhoun County, all of which besides the victim’s remained unsolved.

In response to inquiries by the prosecutor during the reopened evidentiary hearing, Mullen indicated that he had only read parts of the transcripts of defendant’s trial and had not spoken with the Michigan State Police involved in the original investigation *143of the victim’s murder or reviewed all evidence the prosecutor presented during defendant’s trial at the time he formed his opinion of Ronning’s guilt of the murder. The record indicates, however, that Mullen had reviewed all the Michigan State Police reports involving the victim’s murder and that Mullen clearly had knowledge that no eyewitness testimony or physical evidence linked defendant to the victim’s murder and that several witnesses had testified that defendant admitted the crime to them.

Mullen also testified at some length regarding the many details Ronning provided in his account of the victim’s murder that were consistent with the known facts of the victim’s murder. Furthermore, contrary to the trial court’s findings regarding Mullen’s investigation, Mullen did in fact interview some witnesses from defendant’s trial, including Candy Moore, Thomas Clark, and defendant himself. Mullen also spoke with defendant’s trial counsel, who told Mullen that defendant always denied involvement in the victim’s murder. Mullen met with prosecution expert Sauer and several experts in the field of serial killers regarding Ronning’s potential involvement in the Calhoun County area murders.

Mullen concluded that on the basis of all that information, all the information he had learned from the evidentiary hearings, and his twenty-nine years of experience as a police officer and detective, he believed that Ronning killed the victim. We note that the trial court did not reject Mullen’s opinion because it found him unworthy of belief, but instead because “but for Michael Ronning’s statement, Detective Mullen’s opinion that Mr. Ronning committed the Rosansky murder is based primarily upon his professional *144opinion and instinct, as opposed to any newly-discovered facts or evidence.” In light of the foregoing, lengthy testimony regarding Mullen’s efforts, investigation, and discoveries, which included both information directly pertaining to the victim’s murder and circumstantial evidence of Ronning’s guilt of other, similar crimes, we are left with the definite and firm conviction that the trial court made a mistake in attributing Mullen’s belief in Ronning’s guilt to nothing more than his opinion and instinct regarding the veracity of Ronning’s confession.20 Thenghkam, supra.

E

In light of the trial court’s (1) reliance on its miscomprehension of the expert medical testimony regarding the victim’s injuries, (2) mistake of law in failing to even address the fact that Ronning passed a polygraph examination, and (3) rejection of Detective Mullen’s belief in Ronning’s guilt of the victim’s murder because of its mischaracterization or misunderstanding of the bases for Mullen’s determination, we conclude that the court abused its discretion in deny*145ing defendant’s motion for relief from judgment and request for a new trial. Ulman, supra. The trial court clearly erred in finding that the newly discovered evidence of Ronning’s confession did not make a different result probable on a retrial of defendant. Thenghkam, supra.

The trial court correctly observed that Ronning’s account of the victim’s murder did not match all the known details of the victim’s murder. For example, the trial court relied heavily on the videotape recording, which we reviewed also, of Ronning’s failed attempt to locate the exact scene where the crime occurred. The trial court did not consider, however, the unrebutted testimony that Ronning heavily used illegal drugs during the early 1980s at the time of the victim’s murder or Ronning’s suggestion of the possibility that he might have confused certain details of the victim’s murder with the several other murders of young women that he allegedly committed. While it is undisputed, as retired state trooper Harry Zimmerman testified, that the ravine and well station were present both at the time of the victim’s murder and when Ronning sought to relocate the murder scene more than fourteen years later, the record also established undisputedly that the wooded area itself, the pathways into the wooded area, and the buildings and roads surrounding the area had changed dramatically.21 We note that on viewing the April 1997 videotaped excursion into the woods we found it somewhat challenging to even locate the ravine, the top of *146which was covered to a significant extent by fallen trees and overgrowth.

In rejecting Ronning’s confession, the trial court also observed that “Ronning had a motive to confess to the Rosansky murder.” The court cited the testimony of family members of Ronning, whom the court found credible, and the testimony of a jail inmate who had spoken to Ronning, all of whom asserted that Ronning told them that he had not committed the Michigan murders but would falsely confess to them so that he could serve his lifetime imprisonment in Michigan. Ronning himself admitted at the initial evidentiary hearing in November 1997 that he wanted to serve his prison term in Michigan, primarily so that he would reside in the same state where members of his family lived, but also to avoid being extradited to a death penalty state in the event that law enforcement authorities elsewhere ever obtained enough information to charge Ronning with one of his several murders. The trial court did not consider, however, Ronning’s own acknowledgment at the initial evidentiary hearing that he lied to his family members and other inmates regarding his commission of the Michigan murders22 because he found it difficult to face them when they would despise him for committing the murders.

Regarding the trial court’s last observation that “Ronning had the ability and opportunity over the years to obtain information from various sources, including the Battle Creek Enquirer, about relevant *147facts and circumstances surrounding the Rosansky murder,” we failed to ascertain within the instant record from what sources Ronning had the opportunity to glean the level of detail he provided during his videotaped confessions to the victim’s murder. While Ronning advised his sister that he had received transcripts of defendant’s trial from a secretary of one of his Arkansas attorneys, these statements came within the context of Ronning’s acknowledged lies to his family members regarding his guilt of the Michigan murders and explained how he expected to confess to crimes of which he was not guilty. Regarding the Battle Creek Enquirer, it was established that Ronning subscribed to this newspaper after arriving in Michigan during the fall of 1996 and that the Enquirer had printed articles concerning Ronning’s confession to the victim’s murder. The prosecutor did not demonstrate, however, that any newspaper articles to which Ronning might have had access, either articles printed at the time of the victim’s murder or articles printed during 1996 and 1997, could have contained all the consistent information that Ronning provided regarding his commission of the victim’s murder.

While we would not conclude that Ronning’s guilt of the victim’s murder has been established on the instant record beyond any reasonable doubt, the standard governing whether defendant should receive a new trial requires only that the newly discovered evidence “is such as to render a different result probable on retrial . . . .” Canter, supra at 559. Whatever doubts may linger regarding the veracity of Ronning’s confession to the victim’s murder, in light of (1) the newly discovered evidence of Ronning’s confession, *148which, contrary to the court’s findings, was consistent with both the medical testimony presented and Mullen’s investigation, (2) the newly discovered evidence that Ronning passed a polygraph examination, which the trial court did not consider, and (3) the newly discovered witness recantation evidence that the trial court deemed significant, we find that a different result is probable on defendant’s retrial.

We reject the dissent’s suggestions that we have not appropriately deferred to the trial court’s superior ability to render a determination regarding Ronning’s credibility and that we employed insufficient deference to the trial court in applying the abuse of discretion standard of review. We remain cognizant of the well-established principle that “regard shall be given to the special opportunity of the trial court to judge the credibility of the witnesses who appeared before it.” MCR 2.613(C). The central issue of Ronning’s credibility, however, does not involve a traditional credibility determination that rested on the trial court’s special opportunity to view the witness while testifying. We note that the trial court did not once express that its opinion of Ronning’s credibility rested on Ronning’s demeanor on the witness stand or otherwise on its superior opportunity to observe Ronning’s testimony. To the contrary, as the dissent appears to recognize, the trial court’s opinion plainly set out several distinct factual findings leading it to conclude that Ronning falsely confessed. As our opinion explains at length, our review of the record leads us to conclude that the trial court clearly erred in making the factual findings that supported its ultimate conclusion that Ronning lied. Our roles as appellate judges clearly encompass this review of the trial *149court’s findings of fact in light of the trial court record.

With respect to the appropriate standard of review, the dissent correctly sets forth the deferential abuse of discretion standard applicable when “the exercise of discretion turns upon a factual determination made by the trier of the facts,” specifically that under these circumstances an abuse of discretion occurs when “the result [is] so palpably and grossly violative of fact and logic that it evidences not the exercise of will but perversity of will, not the exercise of judgment but defiance thereof, not the exercise of reason but rather of passion or bias.” Spalding v Spalding, 355 Mich 382, 384-385; 94 NW2d 810 (1959). The dissent neglects to mention, however, that a trial court’s misapplication or misunderstanding of the law in reaching its decision also may constitute an abuse of discretion. See Thenghkam, supra at 50 (concluding that “[w]hen the trial court gives unwarranted weight to one or more findings on the statutory factors, contradicting the language of the statute, the balance emphasized in case law, or the true circumstances as expressed in the record, we find an abuse of discretion”); Stepp v Dep’t of Natural Resources, 157 Mich App 774, 779; 404 NW2d 665 (1987) (finding that the trial court’s misunderstanding of the legal basis for its ruling resulted “in an incorrect conclusion that was violative of logic and, therefore, an abuse of discretion”). As our opinion attempts to explain, the trial court in ruling on defendant’s request for a new trial made multiple mistakes of law and contradicted “the true circumstances as expressed in the record.” Thenghkam, supra at 50.

*150m

Defendant also contends that in ruling on his motion the trial court erroneously failed to consider that the prosecutor destroyed potentially exculpatory physical evidence. The prosecutor responds that this Court should not consider the alleged destruction of evidence because defendant’s motion for relief from judgment seeking a new trial did not rely on evidence destruction and the trial court did not address the issue of evidence destruction.

We initially note that defendant’s motion for relief did not mention evidence destruction because defendant apparently had no knowledge in 1997 that the prosecutor had authorized the destruction of all the evidence discovered during the investigation of the victim’s murder. Defendant first decried the prosecutor’s destruction of evidence as a due process violation in his motion to dismiss the charges against him, which defendant filed in early February 1998. Defendant’s motion explained that after the trial court initially ordered a new trial in December 1997, he obtained discovery of all the prosecutor’s investigative materials regarding the victim’s murder in anticipation of the new trial. The materials turned over by the prosecutor included prosecuting attorney John Sahli’s May 14, 1992, authorization permitting the Michigan State Police to destroy “ [a]ll evidence taken in on [the victim’s murder] complaint.” Although defendant requested that the trial court address the prosecutor’s destruction of evidence, the motion went unaddressed because on March 17, 1998, this Court stayed further trial court proceedings during the pendency of the prosecutor’s application for leave to *151appeal the trial court’s order granting defendant a new trial. The stay continued until this Court issued its August 5, 1998, order denying the prosecutor’s application for leave to appeal and vacating the stay. Immediately thereafter, the prosecutor moved to reopen the proofs and the trial court granted the prosecutor’s motion.

At the commencement of the hearing on the first day of reopened proofs, December 2, 1998, defendant again urged the trial court to consider the issue of evidence destruction when ruling on his motion for relief from judgment. Defendant suggested that scientific testing of the destroyed hair and sperm samples taken from the victim would have constituted further new evidence that would have assisted the court’s decision whether to grant a new trial, but that the prosecutor’s knowing destruction of the evidence prevented defendant from effectively presenting his case. The prosecutor disputed defendant’s characterization that the evidence was destroyed in bad faith, but did not oppose the presentation of documents and testimony regarding the destruction of the evidence. The trial court declined to consider evidence destruction in the context of the request for a new trial, explaining merely that “the Motion to Dismiss, it seems to me, would be the appropriate setting for that ... if my decision continues as is and I grant a new trial.”

Nonetheless, testimony was elicited at the reopened evidentiary hearing regarding the circumstances surrounding the destruction of the evidence. Although the destruction of evidence issue technically qualifies as unpreserved because the trial court did not consider it, we will address the issue because the record reflects that defendant diligently presented the *152issue to the court to the extent possible under the circumstances, the issue involves an important constitutional question, and the record contains facts sufficient to permit our review. People v Matelic, 249 Mich App 1; 641 NW2d 252 (2001).

The facts concerning the prosecutor’s destruction of evidence were contained within notes, a police report, and the testimony of Detective Mullen. By 1992, Mullen had become convinced that Ronning committed four murders in Michigan, including the victim’s. Mullen testified that in early 1992 he advised prosecutor John Sahli and then Detective Sergeant Joe Newman of the Battle Creek Police Department that Ronning may have committed several minders, including the victim’s murder for which defendant was imprisoned.23 Mullen, Newman, and Sahli agreed that Mullen should travel to Arkansas to determine whether Ronning would entertain the notion of discussing all the murders he had committed in return for serving his prison time in Michigan. Mullen received authorization to visit with Ronning in Arkansas in January 1992. Mullen recalled that on returning from his January 1992 visit to Arkansas, he met with Sahli in the prosecutor’s office and advised Sahli of Ronning’s perceived willingness to confess to Calhoun County murders if he had representation by counsel. Mullen also specifically advised Sahli of his belief that Ronning would implicate himself in the victim’s murder and that Ronning could clear defen*153dant, who was in prison for the victim’s murder. Mullen understood that the prosecutor’s office intended to conduct research into the matter of making a deal with Ronning.

Mullen’s notes from January 27, 1992, indicated that he, Newman, and Salih again met and discussed the Ronning situation. The notes specified that Sahli wished to arrange Ronning’s transfer to Michigan before authorizing warrants against him, that Sahli seemed reluctant to discuss the authorization of warrants, and that Sahli advised Mullen he would be contacted within a couple of days. Mullen’s notes later indicated that on February 14, 1992, Newman again spoke with Sahli, who stated that he would assign the matter to an assistant. By May and June 1992, the assistant had not reported any progress to Mullen. Mullen’s notes indicated that on April 28, 1992, the Battle Creek police chief spoke with Sahli again, and Sahli advised him that “something would be done.”

On May 14, 1992, Sahli authorized destruction of all physical evidence on the basis that the case was closed and defendant had exhausted his appeals. According to Mullen’s notes regarding August 22, 1992, on that date Keith Hall, Ronning’s attorney from Arkansas, Mullen, Newman, and Sahli and his assistant held a meeting in Battle Creek, during which the victim’s case was mentioned. The Michigan State Police apparently destroyed the evidence of the victim’s murder investigation in October 1992.

We find that the trial court erred as a matter of law in declining to consider in ruling on defendant’s motion for relief from judgment the prosecutor’s destruction of hair, sperm, and other physical evi*154dence from the scene of the victim’s murder.24 Contrary to the trial court’s finding, proofs regarding the prosecutor’s destruction of evidence were highly relevant to the issue whether defendant should receive a new trial.

As we have noted, defendant’s postconviction motion for relief from judgment sought to establish his entitlement to a new trial. MCR 6.431(A)(4), 6.502. The Legislature has codified a criminal defendant’s right to a new trial under certain circumstances.

The judge of a court in which the trial of an offense is held may grant a new trial to the defendant, for any cause for which by law a new trial may be granted, or when it appears to the court that justice has not been done, and on the terms or conditions as the court directs. [MCL 770.1 (emphasis added).]

See also MCR 6.431(B) (also authorizing “a new trial on any ground that would support appellate reversal of the conviction or because [the court] believes that the verdict has resulted in a miscarriage of justice”).

In this case, the prosecutor’s destruction of evidence clearly qualifies as a “cause for which by law a new trial may be granted . . . .” MCL 770.1. The United States Supreme Court has found that a violation of *155the Fourteenth Amendment Due Process Clause of the United States Constitution occurs when a prosecutor in bad faith fails to preserve potentially exculpatory evidence.

The Due Process Clause of the Fourteenth Amendment, as interpreted in Brady [v Maryland, 373 US 83; 83 S Ct 1194; 10 L Ed 2d 215 (1963)], makes the good or bad faith of the State irrelevant when the State fails to disclose to the defendant material exculpatory evidence. But we think the Due Process Clause requires a different result when we deal with the failure of the State to preserve evidentiary material of which no more can be said than that it could have been subjected to tests, the results of which might have exonerated the defendant. ... We think that requiring a defendant to show bad faith on the part of the police both limits the extent of the police’s obligation to preserve evidence to reasonable bounds and confines it to that class of cases where the interests of justice most clearly require it, i.e., those cases in which the police themselves by their conduct indicate that the evidence could form a basis for exonerating the defendant. We therefore hold that unless a criminal defendant can show bad faith on the part of the police, failure to preserve potentially useful evidence does not constitute a denial of due process of law. [Arizona v Youngblood, 488 US 51, 57-58; 109 S Ct 333; 102 L Ed 2d 281 (1988).]

The exculpatory value of the hair and sperm evidence destroyed in this case is not readily apparent. As the parties acknowledge, the state of available technology existing in 1984 and 1985 at the time of defendant’s initial trial did not permit deoxyribonucleic acid (dna) typing or testing of the hair and sperm samples. Over the next seven years, however, rehable DNA typing procedures developed. People v Lee, 212 Mich App 228, 264, 281-283; 537 NW2d 233 (1995); People v Adams, 195 Mich App 267, 273-*156277; 489 NW2d 192 (1992), mod on other grounds 441 Mich 916 (1993).

According to the existing record, defendant had no reason to suspect that in 1992 the prosecutor had authorized the destruction of all evidence collected during the investigation of the victim’s murder. Only after Ronning had come forward and confessed to the victim’s murder, the trial court initially had ordered a new trial, and defendant in anticipation of the new trial received discovery documents from the prosecutor did defendant become aware that potentially the most definitive evidence of his guilt or innocence was forever beyond his reach. Mullen’s testimony and documentation indicating that by January 1992 the prosecutor’s office had awareness of Ronning’s possible involvement in the victim’s murder, and consequently defendant’s wrongful imprisonment for at least seven years, renders the prosecutor’s authorization to destroy the material evidence shortly thereafter in May 1992 deeply disturbing. An inference of the prosecutor’s bad faith arises under these circumstances.25 People v Albert, 89 Mich App 350, 354; 280 NW2d 523 (1979).

We are aware, however, that the prosecutor contests that the evidence was destroyed in bad faith and that on the basis of the trial court’s decision to ignore the evidence destruction issue the prosecution determined that it would not call certain witnesses who could have testified at the reopened evidentiary hear*157ing with respect to the destruction issue. In light of our determination that defendant is entitled to a new trial on the basis of newly discovered evidence, we direct that the prosecution present its witnesses who might testify regarding the circumstances surrounding the evidence destruction during the course of defendant’s retrial so that the jury may resolve the contested issue whether the prosecutor intentionally or in bad faith authorized the destruction of the material, potentially exculpatory hair and sperm evidence.26

Consistent with Youngblood, supra at 57-58, and Michigan law, the trial court shall instruct the jury that if it determines that the prosecutor acted in bad faith it may infer that the destroyed, potentially exculpatory evidence would have been favorable to defendant.27 Although the trial court need not employ any *158particular wording in phrasing its instruction, we suggest that the trial court incorporate within the instruction the elements of the Youngblood analysis as restated in United States v Jobson, 102 F3d 214, 218 (CA 6, 1996), that the jury may infer that the destroyed evidence would have been adverse to the prosecution if it finds that (1) the government acted in bad faith in failing to preserve the evidence, (2) the exculpatory value of the evidence was apparent before its destruction, and (3) the nature of the evidence was such that the defendant would be unable to obtain comparable evidence by other reasonably available means.28

We conclude that the trial court abused its discretion in failing to consider the alleged bad-faith destruction of material, potentially exculpatory evidence in ruling on the propriety of a new trial for defendant.

IV

We scarcely can imagine a situation that would constitute a more appropriate basis for an order granting a new trial. As the trial court noted in its initial opinion granting defendant a new trial, no physical evidence connected defendant to the victim’s murder and no eyewitnesses observed any involvement by defendant in the victim’s murder. Ronning’s subsequent confession, containing many consistencies with the existing evidence of the victim’s murder and *159believed by Mullen after his extensive investigation, the ignored results of Ronning’s polygraph, the recantations of several prosecution witnesses and the possible bad-faith destruction of physical evidence that might conclusively have determined the guilt or innocence of defendant or Ronning, all weigh heavily in favor of our decision to reverse the trial court’s order denying defendant’s request for a new trial. To refuse defendant a new trial under these circumstances, from which “it appears to the [C]ourt that justice has not been done,” MCL 770.1, would represent a miscarriage of justice. MCR 6.431(B). We emphasize that we neither harbor nor intend to express any convictions regarding the ultimate guilt or innocence of defendant or Ronning, but believe that under the circumstances that have developed since defendant’s original trial he is entitled to have a jury consider anew his alleged commission of the victim’s murder.

We reverse and remand for further proceedings consistent with this opinion. We do not retain jurisdiction.