Kennedy, J.
Soon after James Roche and Roy Sweeney were convicted and sentenced for methamphetamine possession, it became public knowledge that Michael Hoover, a chemist at the Washington State Patrol Crime Laboratory, had been self-medicating with heroin sent to the crime lab for testing purposes. Hoover was the chemist who had tested the substances recovered in both Roche’s and Sweeney’s cases and reported them to be methamphetamine. Roche moved for, and was denied, a new trial based on this newly discovered evidence. Sweeney filed a personal restraint petition raising the same issue, after his request for court-appointed counsel to help him move for a new trial was denied by the trial court. Because we agree that this newly discovered evidence of Hoover’s malfeasance broke the chain of custody and tainted the integrity of Roche’s and Sweeney’s trials, we reverse both convictions and remand for new trials, if the State should elect to retry them.
FACTS
Michael Hoover
Michael Hoover began working as a chemist for the Washington State Patrol crime laboratory in 1989. At some [429]*429point in 1998 or 1999, chemists James Boaz and David Northrop, who worked with Hoover at the crime laboratory in Marysville, noticed that Hoover was assigning himself a disproportionately large number of heroin cases. Hoover was also removing heroin cases from other chemist’s file drawers and reassigning them to himself.
Northrop discovered that Hoover had specifically requested that black tar heroin cases be assigned to him. When Northrop asked Hoover why he needed black tar heroin cases, Hoover said that a Washington State Patrol canine officer had requested an extraction of pure heroin for use in training drug dogs. Northrop asked the administrative staff not to comply with Hoover’s request. But Hoover continued to remove heroin cases from the drawer of unassigned cases and self-assign them.
Boaz observed that Hoover’s lab data seemed sloppy. Boaz also suspected that Hoover was reducing his workload by testing a single purified sample and applying the results across a number of cases, a practice known as “dry labbing.” Boaz suspected that Hoover was dry labbing because, in Boaz’s professional opinion, Hoover’s spectra for separate criminal cases were too similar to have come from different samples. Boaz documented at least 14 possible dry labbing incidents from August to October 2000, including several for methamphetamine from Snohomish County cases. Oiie of these 14 cases was Sweeney’s.1
Northrop and Boaz also noticed that Hoover behaved furtively around his workstation, attempted to conceal his activities from them, and kept a large amount of white powder in an evaporating dish on his desk. Boaz would sometimes hear scraping sounds coming from Hoover’s desk area, followed by snorting sounds. Boaz also noticed that Hoover continually turned down the lights in the lab, [430]*430that Hoover had taken to wearing long sleeved shirts and never rolled back his cuffs, that Hoover’s face often appeared flushed at the end of the day, and that he often slipped out early at the end of the day. Northrop and Boaz reported their concerns to Erik Neilson, the supervisor at the lab.
In September 2000, Neilson confronted Hoover about the substances at his workstation. Hoover again said that a canine officer had requested purified heroin for dog training purposes. Neilson told Hoover that no such project had been approved and ordered him to discontinue it. However, Boaz and Northrop noticed that Hoover continued to collect heroin at his workstation.
In November 2000, Neilson asked the Washington State Patrol to investigate Hoover. Detectives placed a hidden video camera above Hoover’s workstation and learned from watching the videotapes that Hoover was scraping the residue from evaporation dishes into small vials and hiding them on his person or around his workstation.
Detectives obtained a search warrant and interviewed Hoover on December 22, 2000, more than two years after Boaz began noticing that Hoover was taking heroin cases that had been assigned to Boaz out of his file drawer. At first, Hoover continued to claim that he was purifying heroin for a canine officer, although he said he could not recall the officer’s name. He also denied that he was using drugs. However, when the detectives revealed that they had been videotaping Hoover at his workstation, he finally acknowledged that he had been taking heroin samples from evidence sent to the lab for testing, purifying it, and using it to self-medicate his back pain. Hoover also admitted that he frequently used heroin at the lab shortly before leaving work. Hoover was given a urine test, which tested positive for heroin.
Detectives found seven test tubes at the back of Hoover’s workstation that appeared to contain controlled substances. The vials had various labels, including “meth” and [431]*431“heroin.” Testing revealed that six of the test tubes contained heroin, and one contained methamphetamine. Boaz and Northrop thought the quantity of material in the test tubes was larger than normal for “marker samples” commonly kept in the work area.
Hoover was charged with evidence tampering and official misconduct. On July 3, 2001, he pleaded guilty to both charges.
James Roche
In December 1999, detectives from the Narcotics Task Force served a search warrant at the Lake Stevens home of James Roche. The police announced their arrival with a loudspeaker. During the search, police found Roche’s wife, two daughters, and another man on the premises, but not Roche. After entering the house, there was a delay of several minutes before police went upstairs to search. In an upstairs bedroom that was apparently being used as an office, a detective noticed that the window was open, causing him to suspect that Roche may have escaped through the window and into the woods across the street. On the sidewalk below the window, police found a pouch containing a substance that looked like methamphetamine, along with a razor blade and a paper rolled into a device commonly used to ingest drugs.
In the same bedroom, police found a red toolbox containing a gray bag. Inside the gray bag, police found several baggies of a powdery substance that appeared to be methamphetamine, a traffic citation issued to James Roche, a scale, a ledger of drug sales, and $3,000 in cash. The toolbox also contained additional baggies of a substance that appeared to be methamphetamine. In a desk drawer, police discovered two more baggies of powdery substance, plus a rolled up dollar bill. Police took photographs of the scene, showing the open toolbox, baggies, and associated items. These items and photos were made trial exhibits and admitted into evidence.
At trial, Deputy Sheriff Terrence Warren testified to his discovery of trial exhibit 1, consisting of baggies which [432]*432contained a substance that, in his opinion, looked similar to methamphetamine. Deputy Warren testified that the substance was packaged in a manner common in the methamphetamine trade. Deputy Warren field-tested the substance and obtained a positive result for methamphetamine. Deputy Warren and Detective Mike Thompson also testified to finding the other baggies and associated items described above, and identified them in court. The officers testified that the items were packed up and sent to the crime lab for testing.
Michael Hoover testified at trial that he was the state crime lab chemist who tested the substances found at the Roche residence. Hoover testified that his routine for testing a substance is first to put a small amount of the suspected drug on a clean plate and add chemicals to it, and to look for certain color changes indicative of controlled substances. Next, he confirms those preliminary results by placing a small amount of the substance on a clean slide, adding chemical reagents, and examining the sample under a microscope, looking for uniquely shaped crystals that grow only if methamphetamine is present. To further confirm the results, he runs an infrared spectrophotometer test and prints out the resulting spectrum, which shows a “unique fingerprint pattern” for each drug.
Hoover said that after testing, he reseals the evidence with evidence tape and initials it. Hoover identified trial exhibit 1 by the laboratory’s blue evidence tape with his initials on it. Exhibit 1 consisted of 18 plastic baggies containing a chunky tan powder with a total net weight of 61.3 grams. Hoover testified that he used the abovementioned standard testing procedures to test the contents of one of the baggies in exhibit 1, and found it to be methamphetamine. Hoover also identified exhibits 5 and 6 by the blue tape and his initials, and testified that he tested the contents of one of the baggies in each exhibit and found them to contain methamphetamine.
The only defense witness was Samuel Swain, who testified that Roche was at Swain’s house in Wenatchee at the [433]*433time the police searched the Roche residence. Defense counsel argued that Roche’s absence negated the State’s theory that Roche constructively possessed the drugs found at Roche’s house, and that the drugs must have belonged to someone else.
A jury convicted Roche of possession of methamphetamine with intent to deliver; he was given a standard range sentence on November 16, 2000.2 After news of Hoover’s malfeasance became public, Roche moved for a new trial or to arrest judgment, based on newly discovered evidence of Hoover’s malfeasance at the crime lab. Roche also raised the possibility that Hoover was under the influence of controlled substances at the time of trial. At the hearing, a copy of the State’s investigative report in the Hoover matter was admitted into evidence. The State admitted that Hoover had tampered with heroin samples sent to the crime lab for analysis. However, the State argued that Roche was not entitled to a new trial because Roche was convicted of possessing methamphetamine, not heroin, and there was no evidence that Hoover had ever tampered with methamphetamine samples. The State also argued that evidence of Hoover’s misconduct was merely impeaching, and that the new evidence would probably not change the verdict.
On March 28, 2001, the court denied Roche’s motion for a new trial. The court reasoned that the real issue at trial was whether Roche constructively possessed the methamphetamine found at his residence, not whether the substance was in fact methamphetamine; therefore, it was unlikely that evidence of Hoover’s malfeasance would have changed the result of the trial. Furthermore, there was no specific evidence that Hoover was under the influence of heroin at the time he tested the substance recovered from Roche’s house. Finally, because this was a methampheta[434]*434mine case, not a heroin case, there was no problem with the chain of custody. Roche now appeals from the trial court’s ruling.
Roy Sweeney
On October 20, 1999, Deputy Stemme of the Snohomish County Sheriff’s Office observed Sweeney driving a motor vehicle on State Road 530. The deputy had contacted Sweeney earlier that day regarding a trespassing complaint and learned that Sweeney’s license to drive was suspended. The deputy, who was in uniform in a fully marked police car, turned on his blue lights and tried to stop Sweeney. However, Sweeney did not stop. As the deputy pursued him, Sweeney reached speeds of up to 100 mph. At an intersection, Sweeney locked his brakes and slid sideways across two lanes of traffic. He then proceeded about one quarter to one half mile, with smoke billowing from his car, until his car finally stopped.
Deputy Stemme placed Sweeney under arrest and called a tow truck for Sweeney’s car. Sweeney asked Deputy Stemme to get Sweeney’s belongings out of the trunk of the car. The deputy complied, removing two bags and a jacket from the car. The deputy searched the jacket pockets and found drug paraphernalia and a chunky powdery substance. The deputy also searched the bags and found a scale with residue in one of them. As Deputy Stemme, carrying the bags and jacket, approached the patrol car where Sweeney was seated, Sweeney said, “Those drugs aren’t mine and I didn’t know those cell phones were stolen!” The powdery substance and the residue on the scale were sent to the crime lab, and tested by Michael Hoover, who reported them to be methamphetamine. Sweeney stipulated at his trial that the substances were in fact methamphetamine.
On October 17, 2000, a jury convicted Sweeney of possession of methamphetamine and attempting to elude a pursuing police vehicle. Sweeney was sentenced on November 16, 2000. Less than two months later, it became public that Hoover had been using heroin at the crime lab. On May 8, [435]*4352001, Sweeney requested that counsel be appointed to represent him on a motion for a new trial. On September 4, 2001, the trial court denied the motion.
Sweeney then filed a pro se personal restraint petition, seeking to have his conviction reversed. We granted Sweeney’s request for appointed counsel under RCW 10.73.150(4) to assist him in pursuing his petition.
DISCUSSION
We have linked Roche and Sweeney’s cases so that we need issue only one opinion concerning the effect of the newly discovered evidence of Hoover’s malfeasance. However, because each case presents a different procedural posture, we must carefully distinguish and apply the appropriate standards of review.
Roche appeals from the trial court’s denial of his motion for a new trial based on newly discovered evidence. Atrial court’s decision regarding a motion for new trial will not be disturbed on appeal, absent an abuse of discretion. State v. Williams, 96 Wn.2d 215, 221, 634 P.2d 868 (1981). A court abuses its discretion where the decision was manifestly unreasonable or based on untenable grounds or reasons. Moreman v. Butcher, 126 Wn.2d 36, 40, 891 P.2d 725 (1995).
To obtain a new trial based on newly discovered evidence, a defendant must demonstrate that the evidence: (1) will probably change the result of the trial, (2) was discovered after the trial, (3) could not have been discovered before trial by the exercise of due diligence, (4) is material, and (5) is not merely cumulative or impeaching. State v. Swan, 114 Wn.2d 613, 641-42, 790 P.2d 610 (1990). The absence of any one of these five factors is grounds to deny a new trial. Williams, 96 Wn.2d at 223.
The State concedes factors (2), (3), and (4), i.e., that the evidence was discovered after the trial, could not have been discovered before trial by the exercise of due diligence, and [436]*436is material. However, the State argues that Roche has failed to establish that the evidence would probably change the result of the trial, and the State also contends that the evidence is merely impeaching.
Roche contends that the evidence would probably change the result of the trial because Hoover’s malfeasance broke the chain of custody and so devastated Hoover’s credibility as to undermine the State’s ability to prove by his testimony that the substance recovered from his residence was methamphetamine.
“Before a physical object connected with the commission of a crime may properly be admitted into evidence, it must be satisfactorily identified and shown to be in substantially the same condition as when the crime was committed.” State v. Campbell, 103 Wn.2d 1, 21, 691 P.2d 929 (1984). Evidence that is unique and readily identifiable may be identified by a witness who can state that the item is what it purports to be. 5 Karl B. Tegland, Washington Practice § 402.31 (1999). However, where evidence is not readily identifiable and is susceptible to alteration by tampering or contamination, it is customarily identified by the testimony of each custodian in the chain of custody from the time the evidence was acquired. Id. This more stringent test requires the proponent to establish a chain of custody “with sufficient completeness to render it improbable that the original item has either been exchanged with another or been contaminated or tampered with.” United States v. Cardenas, 864 F.2d 1528, 1531 (10th Cir. 1989). Factors to be considered include the nature of the item, the circumstances surrounding the preservation and custody, and the likelihood of tampering or alteration. Campbell, 103 Wn.2d at 21. The proponent need not identify the evidence with absolute certainty and eliminate every possibility of alteration or substitution. Campbell, 103 Wn.2d at 21. “[M]inor discrepancies or uncertainty on the part of the witness will affect only the weight of the evidence, not its admissibility.” Id.
[437]*437The State contends that there is no chain of custody problem here because Hoover admitted to stealing only heroin from the crime lab, and there is no specific evidence that Hoover or anyone else tampered with the evidence in Roche’s case or in any other methamphetamine case. We agree that the evidence shows that Hoover diverted and ingested heroin, not methamphetamine. But we do not agree that the chain of custody is thereby rendered intact in methamphetamine cases that Hoover handled. Hoover’s credibility has been totally devastated by his malfeasance. Not only did Hoover steal heroin from the crime lab, he also admitted that he regularly used heroin on the job. He repeatedly lied about his activities until he was finally confronted with the fact that he had been videotaped. Even then, he maintained that it all started when an officer asked him to purify heroin for a drug-dog training project, although he could not provide the name of the officer who allegedly made this request. Furthermore, Hoover’s coworkers thought that his work seemed sloppy and even suspected, with some scientific basis to support their suspicions, that he might have been dry labbing some methamphetamine cases.3 These events are serious enough that a rational trier of fact could reasonably doubt Hoover’s credibility regarding his testing of any alleged controlled substances, not just heroin, and regarding his preservation of the chain of custody during the relevant time period.
It is true, as the State argues, that photos taken at the crime scene in Roche’s case demonstrate that the exhibits were in substantially the same condition as before Hoover handled them. And it is also true that the substances found in Roche’s home were field-tested by one of the investigat[438]*438ing officers before they were sent to the state crime lab, thus providing some corroboration for Hoover’s testimony regarding the nature of the controlled substances he reportedly tested. But one cannot ascertain whether a baggie filled with a chunky powdery substance has been tampered with or properly tested from looking at a photograph. That is precisely why a chain of custody must be laid for evidence that is not readily identifiable.
Moreover, the evidence of Hoover’s malfeasance is more than “merely” impeaching; it is critical, with respect to Hoover’s own credibility, the validity of his testing, and the chain of custody. See State v. Savaria, 82 Wn. App. 832, 838, 919 P.2d 1263 (1996) (“[IJmpeaching evidence can warrant a new trial if it devastates a witness’s uncorroborated testimony establishing an element of the offense. In such cases the new evidence is not merely impeaching, but critical.”). The record establishes that after Hoover’s malfeasance became known, the State dismissed dozens of pending “Hoover cases” involving drugs other than heroin, including methamphetamine cases, because of the devastating damage to Hoover’s credibility and to the chain of custody.
In denying Roche’s motion for a new trial, the court noted that the main issue at trial was whether Roche constructively possessed the substances found at his residence, not whether the substances were in fact methamphetamine. But Roche had no reason to challenge Hoover’s testimony at his trial because evidence of Hoover’s malfeasance had not yet come to light. As far as the defense bar knew at that time, Hoover was a respected and reputable chemist whose integrity and scientific methodology were above reproach. There can be no doubt, however, that if evidence of Hoover’s theft of heroin, use of heroin at work, sloppy work habits, and the factually supportable suspicion of his fellow chemists that he was dry labbing had come to light during Roche’s trial, the admissibility of the trial exhibits would have been vigorously challenged — and probably the exhibits would not have been admitted into evidence at all.
[439]*439Indeed, the record reflects that if the newly discovered evidence had been known to the State at the time of Roche’s trial, Hoover never would have been called as a witness. On January 12, 2001, shortly after the evidence came to light, Assistant Chief Criminal Deputy Michael Downes of the Snohomish County Prosecutor’s Office issued an internal memorandum to all the trial deputies in the office who were handling drug prosecutions, outlining the office policy for handling pending drug cases where Hoover had tested the substances, and telling them, “Do NOT call Mike Hoover as a witness.”
The trial court also reasoned, in denying Roche’s motion for a new trial, that there was no problem with the chain of custody because Hoover stole heroin from the crime lab, not methamphetamine. But in the same memorandum, Downes told the trial deputies, “Do NOT have the controlled substance retested. Our problem is chain of custody. Having the substance retested does not solve the problem and causes more (and wasted) work for the crime lab.” Downes’ memorandum applied to all pending drug cases where Hoover had done the testing, not just to heroin cases.
In the same memorandum, Downes outlined the office policy for handling possession with intent and delivery cases that had not yet gone to trial, or that had not yet gone to sentencing: “Any possession with intent or delivery case in which we do not have both a good confession to the identity of the substance (we do not need to have a confession re: intent or delivery) and a positive field test shall be dismissed.” And for cases in which the defendant had already been found guilty by plea or trial, but had not yet been sentenced, Downes wrote:
Cases in which we do not have both a good confession to the identity of the substance and a positive field test shall be dismissed. In cases [for] which we have the field test and the confession, we will argue to maintain the conviction on grounds that there is sufficient independent evidence to support the conviction.
App. A to State’s Resp.
[440]*440In Roche’s case, the State had a positive field test; it did not have a confession from Roche as to the identity of the substance. We thus conclude that but for the fact that Roche was tried, convicted, and sentenced before the evidence of Hoover’s malfeasance came to light, he likely would not have been tried at all. As Mr. Downes also wrote in the memorandum above mentioned:
The most important consideration for us now is the preservation of the integrity of the criminal justice system. We must handle these cases in such a fashion that the public, the defense bar, and the courts clearly know that we will not base criminal convictions on tainted evidence but will insist upon proper standards of conduct and procedure.
App. A to State’s Resp. We wholeheartedly agree. We also conclude that in a case such as this where the defendant would not have been tried or sentenced at all if the newly discovered evidence had come to light before he was tried, convicted, or sentenced, the question of whether the result at trial would have been different if the evidence had come to light earlier becomes effectively moot.
We thus conclude that the trial court should have granted Roche’s motion for a new trial, leaving it up to the State to decide whether to retry Roche or to dismiss the charges. Accordingly, we reverse Roche’s conviction and remand so that the State can decide whether to retry Roche or to dismiss the charges against him.
We now turn to Sweeney’s personal restraint petition. Generally, a personal restraint petition alleging a constitutional error must show “actual and substantial prejudice,” while a petition alleging nonconstitutional error must show “ ‘a fundamental defect which inherently results in a complete miscarriage of justice.’ ” In re Pers. Restraint of Cook, 114 Wn.2d 802, 810, 811, 792 P.2d 506 (1990) (quoting Hill v. United States, 368 U.S. 424, 428, 82 S. Ct. 468, 7 L. Ed. 2d 417 (1962)). However, these threshold requirements do not apply “when the challenge is to a [441]*441decision . . . from which the inmate generally has had no previous or alternative avenue for obtaining state judicial review.” In re Pers. Restraint of Cashaw, 123 Wn.2d 138, 148-49, 866 P.2d 8 (1994). Where that is so, the appellate court reviews the petition by examining only the requirements of RAP 16.4. Id. at 149. Under that rule, petitioners must show they have been restrained under the provisions of RAP 16.4(b) and that the restraint is unlawful under the provisions of RAP 16.4(c). Restraint is unlawful where material facts exist which have not been previously presented and heard, which, in the interest of justice, require vacation of the conviction. RAP 16.4(c)(3). The appellate court will grant relief by a personal restraint petition only if other remedies that may be available to the petitioner are inadequate under the circumstances and such relief may be granted under chapter 10.73 RCW. RAP 16.4(d).
The petitioner must state with particularity facts that, if proven, would entitle him to a hearing:
If the petitioner’s allegations are based on matters outside the existing record, the petitioner must demonstrate that he has competent, admissible evidence to establish the facts that entitle him to relief. If the petitioner’s evidence is based on knowledge in the possession of others, he may not simply state what he thinks those others would say, but must present their affidavits or other corroborative evidence. The affidavits, in turn, must contain matters to which the affiants may competently testify. In short, the petitioner must present evidence showing that his factual allegations are based on more than speculation, conjecture, or inadmissible hearsay.
In re Pers. Restraint of Rice, 118 Wn.2d 876, 886, 828 P.2d 1086 (1992). On motion of the opposing party, the appellate court will strike incompetent and inadmissible evidence offered by the petitioner in support of the personal restraint petition. In re Pers. Restraint of Pirtle, 136 Wn.2d 467, 473, 965 P.2d 593 (1998).
The trial court denied Sweeney’s request for appointed counsel to assist him in moving for a new trial, making it very difficult if not impossible for him to make the [442]*442record necessary to introduce the evidence of Hoover’s malfeasance. A personal restraint petition was his only realistic opportunity to obtain court-appointed counsel, and thus his only realistic means of seeking state judicial review. Therefore, the Cook standards do not apply, and we evaluate Sweeney’s personal restraint petition under the standards of RAP 16.4.
Sweeney’s argument centers on allegations that Hoover’s test results were completely unreliable. Sweeney points to evidence that Hoover used heroin at work, that he may have been “dry labbing,” that his work seemed sloppy, that he took samples of evidence home with him, and that he engaged in an ongoing pattern of deceit and denial about his activities. Counsel for Sweeney at the trial court level has stated that, had she known of the allegations against Hoover, she would have cross-examined Hoover, and she would not have stipulated that the substance seized from Sweeney’s car was methamphetamine.4
The State argues that Sweeney has failed to present sufficient competent, admissible evidence to sustain his petition. According to the State, Sweeney’s petition relies on mere speculation regarding Hoover’s possible; misconduct in methamphetamine cases, and Sweeney did not submit any evidence that Hoover committed misconduct in any methamphetamine cases, including Sweeney’s.
As a preliminary matter, we address the State’s motion to strike certain statements in the record. The State, citing ER 602,5 argues that statements of James Boaz regarding his suspicion that Hoover was dry labbing cases are inadmissible because Boaz did not have personal [443]*443knowledge whether or not Hoover was dry labbing, and the suspected dry labbing cases were not related in any way to Sweeney’s case. We disagree. Boaz is a professional chemist at the same crime lab that employed Hoover. His work station was located near Hoover’s. Boaz concluded that Hoover may have dry labbed on at least 14 occasions based on his professional judgment that the spectra in those 14 cases appeared so similar that, in his opinion, they all came from the same sample. One of the 14 cases is, in fact, Sweeney’s — it bears the same agency case number as Sweeney’s. Boaz’s suspicions have a reasonable scientific basis. The substances in six of these cases, including several methamphetamine cases from Snohomish County prosecutions, were retested on September 17, 2001. Sweeney’s was not one of the six. All six proved to be the same controlled substances as reported by Hoover, but the lab reports neither confirm nor rule out dry labbing, as they do not include comparisons of spectra with those reported by Hoover. The samples from Sweeney’s case were retested by Boaz on March 12, 2002, and found to contain methamphetamine, but as before, the retesting report neither confirms nor rules out dry labbing by Hoover on the earlier test.
This evidence is relevant because it raises a reasonable inference that Hoover could have dry labbed the substance found in Sweeney’s case that is not rebutted by the result of the retesting. We deny the motion to strike this evidence.
Next, although the State provides no argument to support the motion, the State moves to strike Boaz’s statements regarding Hoover’s habit of keeping the lights low and having a flushed face in the office. Boaz personally observed Hoover’s habits and physical appearance at the lab, and explained that these observations, along with Hoover’s habit of hoarding heroin cases, made him suspect that Hoover was using heroin at the lab. Those statements are admissible because they are based on personal observation. They are relevant because they raise a reasonable inference that Hoover was using drugs on the job. We deny this motion to strike.
[444]*444Next, the State moves to strike statements from James Boaz, David Northrop, and Erik Neilson that were summarized in police reports, because the detective who wrote the reports would not be allowed to testify to the hearsay statements contained in the reports in order to prove the truth of the matters asserted. We agree and grant this motion to strike except insofar as the statements could potentially be admissible for nonhearsay purposes, such as impeachment or the thoroughness of the police investigation of Hoover’s malfeasance.
Insofar as Sweeney’s personal restraint petition is concerned, though not as a reflection of the validity of its own prosecution of Hoover, the State argues that the evidence regarding Hoover’s activities is purely speculative. But the State completely ignores Hoover’s confession, in which he admitted diverting heroin from evidence sent to the lab for testing, and to self-medicating with it on the job, on almost a daily basis. Furthermore, Hoover pleaded guilty to evidence tampering and official misconduct. These statements remove the allegations against Hoover from the realm of conjecture. Boaz’s allegations that Hoover’s work was chronically sloppy and that he may have been dry labbing are also significant because they lead to an inference that Hoover may not have properly tested the substance in Sweeney’s case, regardless of the result of the retesting.
The next question is whether Sweeney has established enough material facts to vacate his conviction pursuant to RAP 16.4(c)(3). Sweeney’s petition is required to meet the same standard as a motion for a new trial pursuant to CrR 7.5(a). State v. Harper, 64 Wn. App. 283, 292, 823 P.2d 1137 (1992). The standard is identical to that previously discussed for Roche’s case: (1) the evidence must be such that the results will probably change if a new trial were granted; (2) the evidence must have been discovered since the trial; (3) the evidence could not have been discovered before the trial by exercising due diligence; (4) the evidence must be material; and (5) the evidence cannot merely be cumulative or impeaching. Id. at 291-92.
[445]*445As in Roche’s case, the State contends that the evidence would probably not change the result of the trial. The State points out that Sweeney spontaneously declared “those drugs aren’t mine” as the deputy brought the jacket and bags to the patrol car. Although Sweeney’s statement does provide evidence that he knowingly possessed a controlled substance, it does nothing to prove what that substance actually was. More importantly, counsel for Sweeney undoubtedly would not have stipulated that the substance was methamphetamine if she had been aware of the evidence of Hoover’s malfeasance. The fact that the substance retested positive for methamphetamine does not remove the taint of Hoover’s malfeasance or solve the same chain of custody issue that caused us to reverse Roche’s conviction. Moreover, there was no field-test performed on the substances found in Sweeney’s car. According to the Downes memorandum aforementioned, “Any felony possession case pending trial which does not have both a positive field test and a good confession/admission shall be dismissed.” And even in possession cases having both a positive field test and a good confession, the policy was to offer the defendant a gross misdemeanor, and to proceed to trial on the felony only if the defendant were to reject the offer.6
The State also argues that Sweeney is not entitled to a new trial because the new evidence is merely impeaching. The analysis discussed above in the Roche case applies equally to Sweeney. The evidence of Hoover’s malfeasance is so corrupting that Snohomish County instructed its prosecutors not to call Hoover as a witness to testify in court and not to try to reconstruct the chain of custody by having the substances retested. This evidence is critical and not “merely” impeaching.
Although the State in Roche’s case conceded that the evidence was material, the State in Sweeney’s case argues [446]*446that the new evidence is neither material nor admissible. First, the State argues that Hoover’s improprieties with heroin do not lead to a logical inference that he improperly handled the methamphetamine in Sweeney’s case. The State also dismisses the dry labbing allegations as inadmissible and irrelevant, notwithstanding the fact that Sweeney’s was one of the 14 cases in which dry labbing was specifically suspected. For the reasons we have already discussed, we disagree. Even though there is no evidence that Hoover consumed methamphetamine, there is evidence that he kept more of it at his desk than was needed for legitimate laboratory purposes. And since he admittedly consumed heroin on the job almost every day, all of his drug testing work during the relevant time period must be called into serious question.
Sweeney has shown that admissible material facts exist which have not been previously presented and heard, which, in the interest of justice, require vacation of the conviction. RAP 16.4(c)(3). Therefore, we grant Sweeney’s personal restraint petition, and vacate his conviction. We do so for the same reasons that we have reversed Roche’s conviction, as well as the reasons specific to Sweeney’s petition, and remand for a new trial if the State should elect to retry him.
This ruling makes it unnecessary for us to address Sweeney’s contention that his restraint is unlawful on grounds that the State committed a Brady7 violation by failing to sooner disclose the evidence of Hoover’s malfeasance. Neither do we need to address Sweeney’s argument that the State violated his right to equal protection on grounds that he was not given the benefit of the “Hoover” policy because he had been convicted and sentenced before evidence of Hoover’s malfeasance came to light, and because prosecutors in other counties decided to dismiss all controlled substances cases where they decided they would not prevail on motions to withdraw guilty pleas or for a new trial, even where sentencing had already taken place.
[447]*447CONCLUSION
We adopt as our own the reasoning of the Snohomish County Prosecutor’s Office as stated in the Downes memorandum, and restate it for purposes of this opinion as follows: The most important consideration for us now is the preservation of the integrity of the criminal justice system. We must handle these two cases now before us in such a fashion that the public, the defense bar, the prosecuting attorneys, and the courts of Washington will clearly understand that we will not tolerate criminal convictions based on tainted evidence, but will insist upon proper standards of conduct and procedure.
We vacate Roche’s conviction for possession of methamphetamine with intent to deliver and remand for a new trial if the State should elect to retry him. We grant Sweeney’s personal restraint petition insofar as it relates to his possession of methamphetamine conviction, vacate that conviction, and remand for a new trial if the State should elect to retry him. We deny Sweeney’s personal restraint petition insofar as it relates to his conviction for eluding.
Coleman and Appelwick, JJ., concur.