State v. Gondor

{¶ 1} The question presented in these two cases, consolidated sua sponte, is whether the court of appeals applied the correct standard of review when it reviewed de novo the trial court’s judgment granting the appellants’ petitions for postconviction relief. We hold that the court of appeals should have applied an abuse-of-discretion standard rather than a de novo standard in reviewing the trial court’s decisions on the two postconviction petitions. We therefore reverse the judgments of the court of appeals. Because we also hold that the trial court did not abuse its discretion when it granted postconviction relief and ordered new trials for the two appellants, we remand these cases to the trial court for retrial.

Factual Background

{¶ 2} These cases arise from the August 14, 1988 murder of Connie Nardi. She was beaten and then strangled to death; her body was dumped in a pond, where it was discovered the following day. Troy Busta, Robert Gondor, and Randy Resh were charged with Nardi’s murder. In 1989, Busta pleaded guilty; in 1990, Resh and Gondor were convicted of Nardi’s murder in separate trials.

{¶ 3} With some variations, the state presented the same evidence at both Resh’s and Gondor’s trials. Busta was the prosecution’s key witness at both proceedings. He testified to the following events. Around 5:30 p.m., on Sunday, August 14, 1988, Busta went to Ed’s Upper Deck, a bar in Mantua. While at the bar, he met and danced with Nardi. About 30 to 45 minutes later, Resh and Gondor arrived at the bar. At around 7:00 p.m., Busta took Nardi on a motorcycle ride to cool off. They traveled to a washed-out bridge that formerly crossed the Cuyahoga River on Allyn Road. After about 45 minutes, they returned to the bar.

{¶ 4} After returning to the Upper Deck, Busta talked with Resh about Nardi. Busta falsely bragged that he had had sex with Nardi. Resh asked Busta to persuade Nardi to return to the washed-out bridge area and have sex with him. Busta agreed, and the two planned for Busta to buy some beer and take Nardi back to the washed-out bridge. Resh and Gondor were to arrive 15 to 20 minutes later.

*379{¶ 5} Later that evening, Busta bought a “split six” pack of beer — three Coronas and three Pabsts. Busta and Nardi then left the Upper Deck on Busta’s motorcycle and returned to the washed-out bridge. Busta’s stated departure time from the Upper Deck differed at the two trials. At Resh’s trial, Busta testified that he and Nardi had left the Upper Deck between 8:00 and 8:30 p.m. At Gondor’s trial, Busta testified that they had left between 9:00 and 9:30 p.m. At both trials, Busta testified that Resh and Gondor arrived at the bridge in Gondor’s white pickup truck 15 to 20 minutes after Busta and Nardi.

{¶ 6} Resh and Gondor joined Busta and Nardi near the bridge. After a short conversation, Resh asked Nardi to have sex with the three of them. Nardi refused, and Resh became angry. Resh grabbed Nardi and told Gondor and Busta to hold her hands and feet. Resh sat on Nardi’s stomach while she struggled to free herself. Gondor took off Nardi’s shorts and underwear. Nardi freed a leg and kicked Gondor. Resh responded by striking Nardi several times on her head and then strangled her.

{¶ 7} According to Busta, Resh and Gondor put Nardi’s body in the back of Gondor’s truck on top of a pile of two-by-four pieces of wood. Gondor and Resh drove the truck, and Busta followed on his motorcycle, to a parking area near a pond. Gondor and Resh then dumped Nardi’s body into the pond. Busta left Resh and Gondor, and he later discarded Nardi’s purse in a ditch and disposed of the beer cans and bottles near the intersection of Abbot and Allyn Roads.

{¶ 8} At around 11:00 or 11:15 p.m. that evening, Busta returned to the Upper Deck and met Resh and Gondor, who had already arrived. They remained until the bar closed. After leaving the bar, they noticed blood on some of the wood in the back of Gondor’s truck. The three of them started tossing the wood into the parking lot of the bar across the street. Gondor and Resh then followed Busta to his residence. However, Gondor and Resh did not stop at Busta’s house. Other testimony established that Resh and Gondor returned to Resh’s trailer and at 12:42 a.m. on August 15,1988, ordered a pizza from Domino’s.

{¶ 9} During the trials of Resh and Gondor, the state relied on blood evidence and false-alibi evidence. Investigators had examined suspected bloodstains found on the black plastic bed liner of Gondor’s truck because Busta had claimed that Nardi’s bloodstains were left on wood in the back of the truck. Dale Laux, a criminalist with the Ohio Bureau of Criminal Identification and Investigation (“BCI”), testified that the stains were human bloodstains. However, Laux could not determine the blood type because of the age and small size of the samples. During cross-examination at Gondor’s trial, Laux testified that the stains were also sent for testing to the Serological Research Institute (“SERI”). Laux testified that SERI tests to type the stains were unsuccessful.

*380{¶ 10} The state also presented testimony that Resh and Gondor had attempted to establish an alibi for Nardi’s murder before they had been identified as suspects. At both trials, the owner of Domino’s Pizza testified that Gondor had contacted him on Monday, August 15, 1988, to make sure that his name was added to a receipt for the pizza that had been ordered by Resh the night before. Brenda Holcomb, another Domino’s employee, testified that both Resh and Gondor went into Domino’s on Tuesday, August 16, 1988, to verify their early Monday morning order because they were being investigated for a murder.

{¶ 11} Resh and Gondor denied any involvement in Nardi’s murders. During his trial, Resh testified that upon arriving at the Upper Deck, he saw Nardi dancing on the bar. Resh claimed that he saw Busta and Nardi leave the bar together and return about 30 to 45 minutes later. Around 9:00 p.m., Resh saw Busta and Nardi leave again. According to Resh, around 10:00 p.m., he and Gondor left the Upper Deck and went to the Village Tavern. They stayed there for about an hour and then returned to the Upper Deck. Resh noticed that Busta had also returned to the Upper Deck and was sitting alone at the bar. Resh testified that he left the bar when it closed and that he, Gondor, and Busta threw some pieces of wood from the bed of Gondor’s truck into the parking lot across the street. However, Resh denied seeing any blood on the wood. Afterwards, according to Resh, he and Gondor drove to Resh’s trailer and ordered the pizza.

{¶ 12} At his trial, Gondor testified that upon arriving at the Upper Deck, he sat with Resh, who told him about a woman who had been dancing on the bar and had left with Busta. Sometime later, Busta and Nardi returned. Around 10:00 p.m., according to Gondor, he and Resh left the Upper Deck and went to the Village Tavern. They remained there for 30 to 40 minutes and then went back to the Upper Deck. After they returned, Gondor noticed that Busta came back to the Upper Deck. When the Upper Deck closed at midnight, Gondor stated, he, Resh, and Busta went out to the parking lot and threw pieces of wood from the bed of his truck across the street. Gondor denied seeing any blood on the wood. Afterwards, Gondor followed Resh to Resh’s trailer, and they ordered the pizza.

{¶ 13} During the defense case at Gondor’s trial, the bartender at the Village Tavern testified that Resh and Gondor were at the Village Tavern for about 30 minutes between 9:45 and 10:45 p.m. on August 14. Resh’s defense team presented similar evidence at his trial.

{¶ 14} Resh was convicted of murder and attempted rape. He was sentenced to five to 15 years for attempted rape and 15 years to life for murder, the sentences to be served consecutively. Gondor was convicted of involuntary manslaughter, kidnapping, and obstructing justice. He was sentenced to ten to *38125 years for both involuntary manslaughter and kidnapping, and 18 months for obstructing justice, all sentences to run consecutively.

Procedural History

{¶ 15} The court of appeals separately affirmed Resh’s and Gondor’s convictions. State v. Resh (Dec. 11, 1992), Portage App. No. 90-P-2256, 1992 WL 366992; State v. Gondor (Dec. 11, 1992), Portage App. No. 90-P-2260, 1992 WL 366988. On May 5, 1993, this court declined to review Resh’s discretionary appeal. State v. Resh (1993), 66 Ohio St.3d 1473, 611 N.E.2d 835. On July 21, 1993, this court likewise declined to review Gondor’s discretionary appeal. State v. Gondor (1993), 67 Ohio St.3d 1408, 615 N.E.2d 1043.

{¶ 16} On September 20, 1996, Resh and Gondor filed separate petitions for postconviction relief. Both alleged that the state had withheld exculpatory material and that they had received ineffective assistance of counsel. On October 10, 1996, the trial court denied both petitions without issuing findings of fact or conclusions of law. On November 12, 1996, Resh and Gondor appealed and also filed with the trial court motions to reconsider, as well as amended petitions for postconviction relief and attached evidentiary materials.

{¶ 17} On January 7, 1997, the court of appeals remanded both cases to the trial court with instructions to enter findings of fact and conclusions of law. On January 30, 1997, the trial court made findings of fact and conclusions of law in both cases and dismissed Resh’s and Gondor’s amended petitions. The court of appeals reversed and remanded for an evidentiary hearing on Gondor’s petition for postconviction relief. State v. Gondor (Dec. 19, 1997), Portage App. Nos. 96-P-0261 and 97-P-0017, 1997 WL 799575. On December 22, 1997, the court of appeals also reversed and remanded for an evidentiary hearing on Resh’s petition for postconviction relief. State v. Resh (1997), 124 Ohio App.3d 694, 706, 707 N.E.2d 531.

{¶ 18} On March 30, 1999, Busta filed a motion to intervene in both Resh’s and Gondor’s postconviction proceedings and a motion to disqualify counsel for Resh and Gondor. Busta’s motion to disqualify petitioners’ counsel was based on his interest in protecting certain privileged attorney-client matters that might be disclosed in connection with Resh’s and Gondor’s cases. The trial court denied Busta’s motion, and further postconviction proceedings were stayed pending Busta’s appeal. The court of appeals reversed the trial court’s decision denying Busta’s motion to intervene and remanded the matter to the trial court. State v. Resh (June 29, 2001), Portage App. No. 99-P-0035, 2001 WL 735788; State v. Gondor (June 29, 2001), Portage App. No. 99-P-0034, 2001 WL 735781.

*382 Postconviction Evidentiary Hearing

{¶ 19} On May 8, 2002, the trial court began an eight-day joint evidentiary hearing. During the postconviction hearing, James Draper, Resh’s trial counsel, testified that he began representing Resh in April 1990. Draper was in the process of trying a case in federal court at the time he represented Resh. Draper said that he had moved for continuances in the federal case and Resh’s case, but that both motions were denied.

{¶ 20} Draper testified that the prosecutor informed him that discovery in Resh’s case would be conducted though an open-file procedure. Draper admitted reviewing the prosecutor’s file. However, Draper explained that he did not review each page of the prosecutor’s “voluminous files” due to his trial schedule and because he relied on “the things that were pointed out to [him] that went to the theory of the State’s case.”

{¶ 21} Gary Levine, Gondor’s trial counsel, testified that he began representing Gondor before Gondor’s April 5, 1990 indictment. Prosecutors notified Levine “[e]arly on” that discovery would be conducted through an open-file procedure. Nevertheless, Levine testified that he expected a formal response to his discovery requests pursuant to Crim.R. 16. Levine also testified that he personally viewed the prosecutor’s file at the Portage County Prosecutor’s Office and was never refused an appointment to do so.

{¶ 22} During the postconviction hearing, Resh and Gondor presented evidence from the prosecutor’s file that was favorable to the defense but that was never introduced at their trials. First, Resh and Gondor presented a July 6, 1990 report from Brian Wraxall, Chief Forensic Serologist at the Serological Research Institute. Wraxall’s affidavit, dated November 12, 1996, and Wraxall’s April 30, 2001 deposition were also introduced. Wraxall’s July 6, 1990 report set forth the test results of the suspected bloodstains found on Gondor’s bed liner. He wrote that gamma markers and kappa markers “are found in blood serum as well as other body fluids” and that they “are very stable in dried bloodstains.” However, Wraxall reported that his tests detected no gamma markers or kappa markers in the samples submitted. He further reported that DNA “is found in nucleated cells, e.g. white blood cells, spermatozoa, salivary, vaginal and tissue epithelial cells,” but that he detected no DNA in the stains submitted. In his November 12, 1996 affidavit, Wraxall stated, “The body fluid that best fits [the test results] is perspiration.” Wraxall testified at his deposition that he did not remember, and his telephone logs do not show, whether he was contacted by either Resh’s or Gondor’s counsel before the two trials.

{¶ 23} Second, Resh and Gondor presented evidence challenging the prosecution’s theory that they attempted to establish an alibi before they were considered suspects. Resh and Gondor introduced a March 6, 1990 statement of Rick *383Hollibaugh, a Domino’s employee, which established that Hollibaugh and Brenda Holcomb had worked together on the evening that Resh and Gondor returned to confirm their order. Domino’s work records from Monday, August 15, 1988, through Sunday, August 21, 1988, were also introduced. They showed that Holcomb and Hollibaugh first worked together again on Friday, August 19, 1988. This date was after the police had contacted both Resh and Gondor about the Nardi homicide.

{¶ 24} Holcomb’s grand jury testimony and her testimony at the postconviction hearing indicated that it was raining when Resh and Gondor returned to confirm their order. Resh and Gondor presented a climatological report from the Akron-Canton Airport for August 1988, which showed thunderstorms on August 5, 15, 18, 19, 25, and 28. Resh and Gondor claimed that this report, in conjunction with Hollibaugh’s testimony and the work records, showed that Friday, August 19, 1988, was the only date when they could have returned to Domino’s. By that date, the police had identified Resh and Gondor as suspects, and their efforts to establish their whereabouts on the night of the murder would not have been as incriminating as trial testimony that Resh and Gondor had tried to establish an alibi just hours after the crime.

{¶ 25} During the postconviction hearing, the April 26, 1999 testimony of former Domino’s owner Timothy Hayes was presented. Hayes testified that he thought that Gondor had phoned him about his pizza receipt on Monday, August 15, 1988, because Hayes did not have to search for the Sunday receipts and “it would have been right there.” Hayes testified that his testimony about the Monday date was based on the logic of what happened and not on his memory of when the phone call was actually made.

{¶ 26} Third, Resh and Gondor presented evidence about Busta’s credibility that was not presented at their trials. The petitioners introduced a transcript of a September 14, 1988 interview between Busta, his lawyers, and his investigator that took place before Busta’s trial. Busta stated during that 1988 interview that Lieutenant David Easthon, the police investigator, had told him that Busta’s assistance was needed to implicate Resh and Gondor in Nardi’s murder. During the interview, Busta denied seeing anyone strike Nardi. Busta also said that he did not know why Gondor and Resh were throwing wood from Gondor’s truck, but he indicated that it was “good wood.”

{¶ 27} As the interview progressed, Busta’s attorneys asked him whether he would be willing to set up Resh and Gondor for Lieutenant Easthon. Busta replied, “I’d do anything I have to do to get myself out of this. I ain’t sitting in no chair for somebody else. I’m not going to sit in jail for somebody else.” Busta also said, “I’m not going down for it. If I do when I do get out, I’m going to find out who did it. * * * If they did do it 100%, I’m going to prove it some *384way. I don’t know how.” Busta also asked, “Has anything been checked on her husband or * * * anything like that? Would he have anything to do with this?”

(¶ 28} During postconviction proceedings, Resh and Gondor also presented a police report dated September 11, 1988, concerning an attempt by Busta’s mother, Kay Busta, to smuggle a knife into jail to her son. Charges against his mother were dropped as part of the plea agreement with Busta. Neither Resh’s nor Gondor’s counsel used this incident during his cross-examination of Busta at the defendants’ trials.

{¶ 29} Finally, Resh and Gondor presented postconviction evidence that was not used during their trials, showing that others may have been involved in Nardi’s homicide. First, Donna Sabarese provided police with a statement about an encounter that she had with three men in the general vicinity of Nardi’s body. Early Monday, August 15, 1988, Sabarese was delivering newspapers when she was approached by three men in their twenties who told her that they had been fishing and that “a girl was missing * * * [and she] was a friend of theirs.” Sabarese said that the men appeared agitated, and she told them to stay away from her. Sabarese thought they were carrying pipes because she heard a “tinny” sound when one of them dropped something on the pavement. A month later, during a search of the area of the encounter, the police found a metal strut and a broom handle.

{¶ 30} Second, Resh and Gondor presented a police report containing statements from Pauline and Fred Green. The Greens lived on the corner of Allyn and Abbot Roads. The Greens said that around 6:00 or 7:00 p.m. on August 14, 1988, they saw a man and a woman on a motorcycle, a man on a dirt bike, and a man on another motorcycle down by the river. Around 11:30 p.m. that evening, Pauline saw two people on a motorcycle that she believed was the same motorcycle she had seen earlier, and they threw something into the weeds near her home. The next day, she retrieved two full cans and one empty can of Pabst beer, and one full bottle and one empty bottle of Corona beer.

{¶ 31} Third, Resh and Gondor presented the transcript of a September 20, 1988 telephone interview conducted by Lieutenant Easthon with Ed Douglas, the owner of the Upper Deck bar. Douglas reported during that telephone interview that he received information that Busta and Nardi had gone to a party the night of August 14, 1988, where two other men, Eddie Price and Bennett Nelson, were present. Easthon replied, “Well we haven’t spoken to these two guys. Yeah, we’re beyond that. We’re in more critical stages now.”

{¶ 32} Fourth, Resh and Gondor presented a statement that Mr. and Mrs. Thomas provided to the police on October 18, 1988. They resided on the first piece of property north of the alleged location of the homicide. The Thomases *385reported that on Sunday, August 14, 1988, they never saw Busta on his motorcycle and never saw a white pickup truck.

{¶ 33} During the posteonviction hearing, James Draper, Resh’s trial attorney, testified that before Resh’s trial, he did not see or receive Wraxall’s report, the Domino’s work records, the Hollibaugh interview, or Holcomb’s grand jury testimony. Additionally, Draper did not see or receive the transcript of Busta’s interview with his attorneys and did not remember receiving the police report about Busta’s mother. According to Draper, he also did not see or receive the police investigative reports containing Donna Sabarese’s statements, Douglas’s telephone interview, or the Green and Thomas statements.

{¶ 34} Gary Levine, Gondor’s trial attorney, testified at the postconviction hearing that before Gondor’s trial, he did not see or receive the transcript of Busta’s interview with his attorneys or the police report concerning Busta’s mother. Further, Levine did not see or receive police investigative reports relating to Sabarese’s statements, Douglas’s telephone interview, or the Green and Thomas statements. Levine also did not see or receive Domino’s work records or Holcomb’s grand jury testimony, and he did not recall seeing Hollibaugh’s interview.

{¶ 35} Levine testified that he first learned about Wraxall’s report during his cross-examination of Dale Laux. Levine stated at the postconviction hearing that he was “astounded” at trial when Laux testified that another laboratory had conducted testing on the stains. Although Laux discussed the results of those tests during further cross-examination, Levine explained at the postconviction hearing that he did not see or receive a copy of Wraxall’s report before the end of Gondor’s trial.

{¶ 36} Both Draper and Levine testified at the postconviction hearing that all of these reports would have assisted the defense in refuting Busta’s testimony at trial and could have helped to suggest to the juries that other individuals might have murdered Nardi. Draper and Levine also testified that there would have been no reasonable strategic reason for them to fail to use such reports, and they believe that they would have rendered ineffective assistance to their clients had they known about the evidence and failed to present it.

{¶ 37} Ron Craig, an investigator with the Portage County Prosecutor’s Office, testified that he was responsible for gathering and organizing all the evidence pertaining to the Nardi homicide at the time of the investigation and the trials. Craig testified that the prosecutor’s office had an open-file discovery policy. According to Craig, Draper and Levine were informed that the open-file discovery applied to discovery in the Resh and Gondor cases, which meant that the defense was going to receive everything the prosecutor’s office had in its files. Craig testified at the postconviction hearing that Draper and Levine came to the *386prosecutor’s office on several occasions to review the prosecutor’s master file before the trials.

{¶ 38} Craig organized the evidence in the Resh and Gondor cases into a master file before the trials. He also prepared notebook binders containing all the information from the master file. Craig testified that Draper and Levine were each provided a copy of these notebooks before trial. As the cases progressed, Draper and Levine were also provided with supplemental evidence that was not included in the original notebooks.

{¶ 39} During postconviction proceedings, Craig remembered much of the evidence that he had included in the master file and the notebooks. Craig testified that Wraxall’s July 6, 1990 report, Holcomb’s grand jury testimony, Domino’s work records, and the climatological data report from the Akron-Canton Airport for August 1988 were made part of the master file and were included in the notebooks. He also testified that the police report containing the Greens’ statements, the Douglas telephone interview, and the police report relating to Busta’s mother were in the master file and the notebooks. Craig also recalled that the transcript of Busta’s interview with his attorneys was supplementary evidence that was driven by Ted Hornyak, an investigator, to Draper’s Cleveland office. According to Craig, the transcript of Busta’s interview was also provided to Levine before Gondor’s trial.

{¶ 40} David Norris, the Portage County Prosecutor at the time of the trials, testified that he had informed Draper and Levine that open-file discovery would be in effect for the Resh and Gondor cases. Norris testified that “open-file discovery” meant, “Our files were open for review, for copying. Our files, our preparation, our work product, other than some of our handwritten notes, but other than that, just pretty much * * * everything else” was accessible. According to Norris, both defense attorneys were also provided with trial notebooks that contained the “[pjolice reports, witness statements, scientific reports, scientific lab work, Grand Jury testimony, [and] just pretty much anything” needed before trial.

{¶ 41} Norris testified that Wraxall’s July 6, 1990 report was received by the prosecutor’s office shortly before Resh’s trial. Norris remembered providing Draper with the report during jury selection. Norris also testified that he personally provided Levine with a copy of Wraxall’s report before Gondor’s trial.

{¶ 42} On June 14, 2002, after the postconviction hearing had concluded, the trial court denied Resh’s and Gondor’s Brady claim, finding that the prosecution did not withhold material, exculpatory evidence. See Brady v. Maryland (1963), 373 U.S. 83, 83 S.Ct. 1194, 10 L.Ed.2d 215. However, the trial court found that both Resh and Gondor had received ineffective assistance of counsel because their attorneys had failed to properly examine the prosecutor’s file and present *387evidence in it to the trial juries. The trial court concluded that “had such evidence been disclosed to the juries that decided these cases, there is a reasonable probability that the result of the proceedings would have been different.” The trial court therefore vacated Resh’s and Gondor’s convictions and ordered new trials.

{¶ 43} On appeal, the court of appeals conducted a de novo review of the trial court’s postconviction findings of fact and conclusions of law. State v. Resh, Portage App. No. 2002-P-0074, 2004-Ohio-7220, 2004 WL 3090208, ¶ 23; State v. Gondor, Portage App. No. 2002-P-0073, 2004-Ohio-7219, 2004 WL 3090209, ¶ 25. The court of appeals reversed, holding that neither Resh nor Gondor had received ineffective assistance of counsel. Resh, 2004-Ohio-7220, 2004 WL 3090208, ¶ 121; Gondor, 2004-Ohio-7219, 2004 WL 3090209, ¶ 115.

{¶ 44} The cause is now before this court upon the acceptance of discretionary appeals filed by both Gondor and Resh.

Standard of Review

{¶ 45} In proposition of law II, Gondor argues that the Court of Appeals for the Eleventh District erred when it reviewed de novo the trial court’s postconviction findings. Gondor argues that the court of appeals should have reviewed the trial court’s findings under an abuse-of-discretion standard. We agree.

{¶ 46} The appellate court wrote, “It is important to note the standard of review we apply in determining this appeal, i.e., we review the trial court’s judgment granting [the defendant’s] petition for post conviction relief de novo. State v. Braden, 10th Dist. No. 02AP-954, 2003-Ohio-2949 [2003 WL 21321457], ¶ 13.” Gondor, 2004-Ohio-7219, 2004 WL 3090209, ¶ 25; Resh, 2004-Ohio-7220, 2004 WL 3090208, ¶ 23. However, Braden does not stand for the proposition that all appeals from postconviction relief hearings are to be reviewed de novo. Braden merely states — as an aside — that the standard of appellate review for postconviction proceedings brought pursuant to R.C. 2953.08, which involves sentencing matters, is de novo. Braden added that for postconviction relief brought pursuant to R.C. 2953.21, the standard of review is different:

{¶ 47} “[I]n the interest of providing finality to judgments of conviction, courts construe the post-conviction relief allowed under R.C. 2953.21(A)(1) narrowly. See State v. Steffen (1994), 70 Ohio St.3d 399, 410, 639 N.E.2d 67. Further, when a trial court rules on a petition for post-conviction relief after a hearing, an appellate court will give deference to the trial court’s findings of fact.” Braden, 2003-Ohio-2949, 2003 WL 21321457, ¶ 13.

{¶ 48} Braden’s statements regarding the standard of review in postconviction claims brought pursuant to R.C. 2953.21 are correct. A postconviction claim is not an ordinary appeal: “A postconviction proceeding is not an appeal of a *388criminal conviction, but, rather, a collateral civil attack on the judgment.” Steffen, 70 Ohio St.3d at 410, 639 N.E.2d 67.

{¶ 49} Further, the Eleventh District Court of Appeals has itself applied an abuse-of-discretion standard of review in a postconviction case where the defendant alleged ineffective assistance of counsel. In State v. Pierce (Dec. 22, 2000), Lake App. No. 98-L-232, 2000 WL 1876592, *5, the court held that “[a]s a postconviction relief proceeding is a collateral civil attack on a judgment, the judgment of the trial court is reviewed under the abuse of discretion standard.” The court concluded in Pierce that the trial court did not abuse its discretion in finding that there was a lack of credible evidence establishing that the defendant’s trial counsel’s performance had been deficient. Id. at *8.

{¶ 50} We also find persuasive the ruling of the court in State v. Mitchell (1988), 53 Ohio App.3d 117, 119, 559 N.E.2d 1370, in which the court held in a postconviction case involving a claim of ineffective assistance of trial counsel that “[ajbsent a showing of abuse of discretion, a reviewing court will not overrule the trial court’s finding on a petition for post-conviction relief which is supported by competent and credible evidence.” See, also, State v. Mullins (June 25, 1996), Franklin App. No. 96APA01-32, 1996 WL 362073, *3.

{¶ 51} In postconviction cases, a trial court has a gatekeeping role as to whether a defendant will even receive a hearing. In State v. Calhoun (1999), 86 Ohio St.3d 279, 714 N.E.2d 905, paragraph two of the syllabus, this court held that a trial court could dismiss a petition for postconviction relief without a hearing “where the petition, the supporting affidavits, the documentary evidence, the files, and the records do not demonstrate that petitioner set forth sufficient operative facts to establish substantive grounds for relief.” This court reversed the judgment of the appellate court in Calhoun, holding that “the trial court did not abuse its discretion in dismissing the credibility of [the] affidavits,” which served as the basis for his petition. (Emphasis added.) Id. at 286, 714 N.E.2d 905.

{¶ 52} We thus determined in Calhoun that the trial court’s gatekeeping function in the postconviction relief process is entitled to deference, including the court’s decision regarding the sufficiency of the facts set forth by the petitioner and the credibility of the affidavits submitted. We established in Calhoun that a court reviewing the trial court’s decision in regard to its gatekeeping function should apply an abuse-of-discretion standard. The consistent approach is to grant that same level of deference to the trial court in regard to its posthearing decision.

{¶ 53} Upon direct appeal, appellate courts generally review claims of ineffective assistance of counsel on a de novo basis, simply because the issue originates at the appellate level; no trial court has ruled on the issue. Appellate courts *389review the trial record and are left to judge from the bare record whether the assistance was effective.

{¶ 54} On postconviction petitions, however, the trial judge holds a hearing and receives testimony on the very issue of ineffective assistance. The trial judge can delve into the motivation or reasoning of trial counsel through trial counsel’s testimony. The court can hear the testimony of witnesses that were never called to testify at the original trial and can determine the worth of their testimony as well as the witnesses’ credibility. The trial judge can ask what the counsel knew, when he knew it, and whether a mistake was not strategic, but was instead careless. As here, in a postconviction hearing, a judge can hear testimony about what evidence was made available to trial counsel and when it was made available. A trial court in a postconviction proceeding thus plays a unique role in the consideration of claims of ineffective assistance of counsel. It is the only court that actually hears testimony on that issue.

{¶ 55} In many cases, the judge presiding over the postconviction hearing will be the same judge that presided over the underlying trial and stands in an especially strong position to determine the significance of a trial counsel’s errors in the original trial. The state contends that the de novo standard of review was appropriate here because a visiting judge had presided over the evidentiary hearing. The state argues that the hearing court possessed only secondhand knowledge of the original trial from reading the trial transcripts and stood in no better a position than a court of appeals in assessing the credibility of the witnesses. We reject this argument. Such a rule, if adopted, would require an appellate court to apply two different standards of review, depending upon the trial judge’s prior experience in the case. Moreover, a visiting judge at a postconviction hearing is in a totally different position from the appellate judges. The postconviction judge sees and hears the live postconviction witnesses, and he or she is therefore in a much better position to weigh their credibility than are the appellate judges.

{¶ 56} A de novo review by appellate courts would relegate the postconviction trial court to a mere testimony-gathering apparatus. Nothing in R.C. 2953.21 indicates that that should be the case. Further, a clean slate on appeal would encourage every petitioner to seek another day in court, which would not benefit the overall administration of justice in Ohio:

{¶ 57} “[C]ases of postconviction relief pose difficult problems for courts, petitioners, defense counsel and prosecuting attorneys alike. Cases long considered to be fully adjudicated are reopened, although memories may be dim and proof difficult. The courts justifiably fear frivolous and interminable appeals from prisoners who have their freedom to gain and comparatively little to lose.” State v. Milanovich (1975), 42 Ohio St.2d 46, 51, 71 O.O.2d 26, 325 N.E.2d 540.

*390{¶ 58} We thus conclude that proposition of law II has merit. The court of appeals erred by using a de novo standard of review in reversing the trial court’s findings. We hold that a trial court’s decision granting or denying a postconviction petition filed pursuant to R.C. 2953.21 should be upheld absent an abuse of discretion; a reviewing court should not overrule the trial court’s finding on a petition for postconviction relief that is supported by competent and credible evidence.

Trial Court’s Findings of Fact and Conclusions of Law

{¶ 59} In his propositions of law I and III, Gondor argues that the trial court did not abuse its discretion when it held that he received ineffective assistance of counsel. In his sole proposition of law, Resh makes a similar argument.

{¶ 60} As stated above, a trial court’s decision regarding a postconviction petition filed pursuant to R.C. 2953.21 will be upheld absent an abuse of discretion when the trial court’s finding is supported by competent and credible evidence. “The term ‘abuse of discretion’ connotes more than an error of law or of judgment; it implies that the court’s attitude is unreasonable, arbitrary or unconscionable.” State v. Adams (1980), 62 Ohio St.2d 151, 157, 16 O.O.3d 169, 404 N.E.2d 144; State v. Keenan (1998), 81 Ohio St.3d 133, 137, 689 N.E.2d 929.

{¶ 61} The trial court employed the correct legal standard in resolving the defendants’ claims of ineffective assistance of counsel. In evaluating claims of ineffective assistance of counsel, a two-step process is used. “First, the defendant must show that counsel’s performance was deficient. This requires showing that counsel made errors so serious that counsel was not functioning as the ‘counsel’ guaranteed the defendant by the Sixth Amendment. Second, the defendant must show that the deficient performance prejudiced the defense. This requires showing that counsel’s errors were so serious as to deprive the defendant of a fair trial * * Strickland v. Washington (1984), 466 U.S. 668, 687, 104 S.Ct. 2052, 80 L.Ed.2d 674; State v. Bradley (1989), 42 Ohio St.3d 136, 538 N.E.2d 373, paragraph two of the syllabus.

{¶ 62} On the issue of counsel’s ineffectiveness, the petitioner has the burden of proof because in Ohio, a properly licensed attorney is presumed competent. Calhoun, 86 Ohio St.3d at 289, 714 N.E.2d 905, citing Vaughn v. Maxwell (1965), 2 Ohio St.2d 299, 31 O.O.2d 567, 209 N.E.2d 164. In order to overcome this presumption, the petitioner must submit sufficient operative facts or evidentiary documents that demonstrate that the petitioner was prejudiced by the ineffective assistance. State v. Davis (1999), 133 Ohio App.3d 511, 516, 728 N.E.2d 1111. To demonstrate prejudice, “[t]he defendant must show that there is a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different. A reasonable probability is a probability sufficient *391to undermine confidence in the outcome.” Strickland, 466 U.S. at 694, 104 S.Ct. 2052, 80 L.Ed.2d 674.

{¶ 63} Here, the trial court found that the assistance of Draper and Levine was ineffective because they both failed to use evidence available to them at the 1990 trials. The trial court concluded:

{¶ 64} “[H]ad such evidence been disclosed to the juries that decided these cases, there is a reasonable probability that the result of the proceedings would have been different. Certainly the juries should have been given this evidence for their consideration for their verdicts to be worthy of confidence. Without the benefit of such evidence this Court finds that the verdicts are not worthy of confidence.

{¶ 65} “It is therefore determined that the Petitioners herein were denied their Constitutional right to the effective assistance of counsel at their trials.”

{¶ 66} In finding that Draper and Levine were ineffective, the trial court examined four categories of evidence that neither counsel had used during the trials. First, the trial court stated that counsel were ineffective because they had failed to utilize the Domino’s records and the weather reports. The trial court found that this evidence was “clearly exculpatory” because it showed that Brenda Holcomb was wrong in testifying that Gondor and Resh had contacted Domino’s on Tuesday, August 16, before they became suspects in Nardi’s murder.

{¶ 67} Second, the trial court found that Draper and Levine were ineffective because they had failed to utilize the SERI report to cast doubt on Laux’s testimony that bloodstains were found on the bed liner of Gondor’s truck. In reaching this conclusion, the trial court noted, “This report was not used at either trial, although the testimony established that both defense counsel, by phone call from Ron Craig * * * and in person by David Norris, * * * were notified of its existence.” Although the trial court incorrectly wrote that the 1990 SERI report “stated that the stains were ‘most likely perspiration,’ ” the report did present findings that were inconsistent with blood. Further, when Brian Wraxall was asked to interpret his 1990 report, he wrote in his 1996 affidavit that “[t]he body fluid that best fits this scenario is perspiration. In summary, the interpretation of this evidence should be approached with extensive caution.” The court correctly found that the SERI report would have cast doubt on Laux’s testimony.

{¶ 68} Third, the trial court found that Draper and Levine were ineffective because they failed to use several statements showing that other people might have been involved in the murder. This evidence included Sabarese’s statements, Douglas’s phone interview, Pauline Green’s and Fred Green’s statements, and the Thomases’ statements.

*392{¶ 69} Fourth, the trial court found that both counsel were ineffective because they had failed to use the police report showing that Busta’s mother had tried to smuggle a knife into the jail where Busta was confined. This report was completed shortly before Busta implicated Resh and Gondor. Thus, the trial court found that the defense attorneys could have used this report to show that Busta’s statements implicating Gondor and Resh were motivated by his concern that his mother might be prosecuted for smuggling contraband into the jail.

{¶ 70} Finally, the trial court found that Draper and Levine were ineffective because they had failed to use the transcript of Busta’s pretrial interview with his lawyers and defense investigator. This interview revealed that Busta said that he was uncertain about the facts of Nardi’s murder and that pressure that was being exerted on him prior to his own trial. In emphasizing the importance of this information, the trial court pointed out, “Draper testified that, in his opinion, it ‘would have changed the outcome of the trial.’ ”

{¶ 71} In an 11-page opinion prepared after the postconviction hearing, the trial court set forth the appropriate legal standards and addressed Resh’s and Gondor’s claims. Having reviewed the petition and supporting documentation, the trial court concluded that Resh and Gondor had received ineffective assistance of counsel at their trials in 1990.

{¶ 72} The trial court properly considered the cumulative effect of trial counsels’ errors. Strickland directs us to look at the “totality of the evidence before the judge or jury,” keeping in mind that “[sjome errors will have had a pervasive effect on the inferences to be drawn from the evidence, altering the entire evidentiary picture * * *.” Strickland, 466 U.S. at 695-696, 104 S.Ct. 2052, 80 L.Ed.2d 674. We therefore consider these errors in the aggregate. State v. DeMarco (1987), 31 Ohio St.3d 191, 196, 31 OBR 390, 509 N.E.2d 1256. See, also, Moore v. Johnson (C.A.5, 1999), 194 F.3d 586, 619 (court should examine cumulative effect of errors committed during both trial and sentencing); Stouffer v. Reynolds (C.A.10, 1999), 168 F.3d 1155, 1163-1164 (“Taken alone, no one instance establishes deficient representation. However, cumulatively, each failure underscores a fundamental lack of formulation and direction in presenting a coherent defense”).

{¶ 73} Gondor’s first and third propositions of law and Resh’s proposition of law have merit. The trial court properly weighed the credibility of the evidence, properly found that the defendants had set forth sufficient operative facts to obtain postconviction relief, and properly issued sufficient findings of fact and conclusions of law. Accordingly, we conclude that the trial court did not abuse its discretion when it found that Resh and Gondor received ineffective assistance of counsel, vacated the convictions and sentences, and ordered new trials. Therefore, we reverse the judgment of the court of appeals.

Disposition

{¶ 74} In case No. 2005-0329 (Robert Gondor) and case No. 2005-0336 (Randy Resh), the judgments of the Portage County Court of Appeals are reversed. Gondor and Resh are remanded to the trial court for retrial.

Judgments accordingly.