Full opinion text
KELLER, P.J., delivered the opinion of the Court in which MEYERS, WOMACK, KEASLER, and HOLCOMB, JJ., joined. Applicant was convicted of aggravated sexual assault and aggravated kidnapping. Some of the evidence supporting the convictions involved DNA testing conducted by the Houston Police Department (HPD) Crime Lab. After widespread problems were discovered with the HPD Crime Lab, the present case was subjected to further investigation, including additional DNA testing. Applicant has filed an application for a writ of habeas corpus based upon this further investigation. He alleges, among other things, that agents of the State consumed the entire DNA sample in bad faith, that a state witness perjured himself or gave false testimony, and that defense counsel was ineffective for failing to discover the problems with the lab’s testing and its analysis of test results. We conclude that, despite problems with the lab, applicant’s claims are without merit. I. BACKGROUND A. Before Trial 1. The Kidnapping On February 11, 2001, six-year-old “E.T.” was kidnapped. One of E.T.’s friends was ten-year-old Remington Allen. While E.T.’s mother worked that day, Remington’s grandmother cared for E.T., his nine-year-old brother “Junior,” and his ten-year-old sister Denetta. Remington, Junior, and their friend Carlos went to a nearby park. Denetta and E.T. walked with them, but crossed the street to go to a store. Denetta went inside the store. E.T. may have accompanied her, but at some point he was outside again. While Remington, Junior, and Carlos played at the park, they saw a man drive up in a car and tell E.T. to “come here.” Remington told E.T. not to get in the car. The trial testimony of the children diverged somewhat at this point. Remington testified that the man got out of the car and acted like he was picking something up. Remington also testified that he saw the man hand E.T. some money when E.T. got into the car. However, on cross-examination, Remington agreed that he never saw the man. Junior testified that he saw the man inside the car and that the man had a mustache and was wearing sweats. On cross-examination, Junior agreed that he did not remember much about the car or the man inside the car because he did not see the man. Junior further testified that he did not know if the man ever got out of the car. E.T. testified that the man tried to get him to take money and then got out and put him in the car. Remington, Junior, and E.T. all testified that the car sped away quickly once E.T. was in it. Junior and Remington chased the car. Remington picked up his scooter and threw it at the car, but the car did not stop. The children remained at the park for “a little bit” to see if the man would bring E.T. back. The children then went to “Momma Ruth’s” place, which was close by, but no adults were home, so they went back up the street to Remington’s grandmother’s house. At first, Remington’s grandmother did not believe the children’s story about E.T. being kidnapped, but once she became convinced, she called the police. At trial, Remington described the car as a “burgundy-like” Oldsmobile with a white top. When asked whether the wheels were shiny, he responded negatively. Junior testified that the car was dark green with a blue top, and scratches at the top. Both Remington and Junior testified that they had picked out a car in a videotape lineup, but they were not asked which car they picked. Houston Police Officer D.D. Thompson was dispatched to the scene at 3:12 p.m. and arrived at 3:19 p.m. According to Officer Thompson, the children described the kidnapper’s vehicle as an Oldsmobile— some said a Monte Carlo. He explained that the “children couldn’t pinpoint exactly what make and model it was.” The descriptions and the colors given by the children were not consistent. The car was variously described to him as a two-door or a four-door, as dark blue or dark green, with a rusty top or a black top or a vinyl top, and with chrome wheels. Officer Thompson dispatched a description that included model years from 1980 to 1990. Sergeant Larry Hoffmaster testified that Junior described the car as a midsized gray car with chrome rims and with damage to the front around the headlights. Sergeant Hoffmaster further testified that the children were taken to a police sketch artist to make a composite drawing of the suspect. The general broadcast for the suspect indicated a “skinny black male,” but Sergeant Hoffmaster acknowledged that applicant was not skinny. When asked if he developed any suspects whose vehicles were black or dark in color, Sergeant Hoffmaster said no. 2. Aftermath E.T. was returned to the neighborhood the next day. He was crying, and his face was bruised and swollen. E.T., who is African-American, described his kidnapper as having skin color that was a little darker than his own. E.T. described the man as having no facial hair, wearing eyeglasses, wearing a black hat, and wearing a purple jacket with green (or a purple and green jacket) and matching purple pants. E.T. described the ear as a dark navy blue in color, like his tennis shoes, with a brown interior, and with a black console between the front two seats. E.T. related that the man offered him money, and when E.T. got closer, the man pulled him into the car. The kidnapper took E.T. to a house that had a brown couch in the front room and a television next to the bed in the bedroom. He tied E.T.’s arms and legs to the corners of the bed. E.T. also said that the man rubbed “orange grease” on his body. The man told E.T. that “if I tell anybody he’s going to kill me.” When asked if he had been touched inappropriately, E.T. “clammed up” and started “tearing up.” When asked by another officer what happened after he was tied to the bed, E.T. became extremely upset, crying and breaking down into hysterics. E.T. was taken to the hospital the day he was returned. Pursuant to instructions from Chemist Reidun Hilleman at the HPD Crime Lab, Officer Lorenzo Verbits-key swabbed E.T.’s face. Verbitskey let the resulting two swabs air dry in his office. He then delivered the swabs to Hilleman. Hilleman also received anal and oral swabs and clothing. Hilleman gave the swabs to Mary Childs-Henry, a forensic biologist at the HPD Crime Lab who analyzed body fluids and conducted serology testing. Childs-Henry extracted two tubes of DNA from each of the face swabs (four tubes in all). For each swab, one of the tubes contained a sperm fraction and the other tube contained an epithelial fraction of the genetic material. She then discarded the original swabs. E.T. subsequently made an outcry to his aunt, Tangela Harding: E.T. described his kidnapper as wearing a baseball hat, wearing eyeglasses, and having a short haircut. The house had two rooms, and they went up some stairs to get into the house. The man said he was going to pull out his thing (Tangela understood E.T. to be referring to his penis) and wanted E.T. to put it in his mouth. When E.T. said no, the man started slapping him. So E.T. complied, his mouth started hurting, and the man “wet his face.” This outcry was relayed to the police on February 14th. That same day, E.T. was taken to have two composite sketches done of his kidnapper — one with eyeglasses and one without. Joseph Chu, a chemist at the HPD Crime Lab, received the four tubes of DNA that had been derived from the face swabs. Chu’s typing of the DNA was completed on February 20th. Chu effectively consumed the entire sample in all four tubes during testing. He determined that the sample was a mixed sample and that DNA patterns within the sample showed that a portion of the sample was consistent with E.T.’s DNA. He did not at that time have any information regarding the identity of any other contributor. On February 23rd, E.T. told Officer Shawn Valenta that the kidnapper referred to his house as “his friend’s house.” E.T. described the house as wood frame, with about four concrete steps, and a number “2” on the front of the building. Applicant was not in custody at the time E.T. communicated this information. Later that day, police received a Crime Stoppers tip implicating applicant. Sergeant Hoffmaster obtained a photograph of applicant and compared it to the composite; he could not eliminate applicant as a suspect based upon a comparison of the two. Sergeant Hoffmaster discovered that applicant had an outstanding parole warrant and went to applicant’s house at 8602 Alsuma. The house matched the description given by E.T.: a white wood frame home with a concrete sidewalk leading to two concrete steps in the front and the number 2 in the address. The address was on the mailbox and above the door to the house. The car parked in front of the house was a 1982 Buick Regal, with white vinyl over dark gray. The interior of the car was brown with gray. There was no grille in the front and there was damage around the headlights. Applicant was arrested on the parole warrant at 10:00 p.m. that day. On February 24th, the police obtained a search warrant for applicant’s house, car, and body. Police discovered that applicant’s house had two rooms that were in use: a living room and a bedroom. The living room contained only one piece of furniture: a brown sofa. In the bedroom was a fourposter bed with a headboard and a foot-board, a television on a stand beside the bed, and a nightstand beside the bed. On the nightstand was a bottle of orange cocoa butter lotion. A Houston Oilers can-nister and purple pants were also found in the house. A dark baseball cap and a multicolored windbreaker with dark sleeves and a purple inlay was recovered from the vehicle. The next day, Sergeant Hoffmaster recorded a videotaped lineup that included applicant. He also recorded a video lineup that included applicant’s car and four other cars. Before the video lineup of applicant was shown, witnesses were told to view the entire videotape and see if they recognized anyone, “and if they didn’t recognize anyone they should tell us that, that it was okay to do that.” In addition, they were told to view the entire tape before saying anything or making any decisions. When the suspect-lineup video panned on applicant (in the number three position), a big smile came upon E.T.’s face, but E.T. continued to watch the rest of video. As soon as he was asked if he recognized anyone, E.T. immediately pointed to applicant and said, “That’s the man that took me to his friend’s house.” E.T. indicated that he was positive in his identification. E.T. also identified applicant’s car in the vehicle lineup. On February 28th, Chu received applicant’s reference sample for DNA testing. It was found to be consistent with the remaining portion of the DNA from the face swabs. On March 15th, E.T. told Officer Valen-ta that applicant had placed his private part in his mouth and peed in his mouth. When asked what he meant by private part, E.T. pointed to his penis. 3. HPD’s DNA Test Results The following table summarizes the results obtained from Chu’s testing of the face-swab DNA contained in the four tubes and from the victim’s and applicant’s reference samples: With respect to the results for the four swab samples displayed in the table, the alleles unique to applicant (possessed by applicant but not by the victim) are displayed in bold. As can be seen from the table, the epithelial samples did not yield particularly useful results. One of the epithelial samples yielded results at all ten of the loci covered by the test, matched the victim’s DNA profile at all ten loci, and matched ten alleles from applicant that were distributed across seven loci. But only one of the alleles matching applicant was foreign to the victim. The other epithelial sample yielded results at only seven of the ten loci, matched the victim’s profile at those seven loci, and matched seven alleles from applicant that were distributed across six loci. But none of the alleles in the second epithelial sample that matched applicant were foreign to the victim. The sperm samples yielded far better results. Both samples yielded results at all ten loci. The first sample matched the victim’s DNA profile at nine of the ten loci, matched all of applicant’s alleles except one, and matched eight alleles from applicant (distributed across six loci) that were foreign to the victim. The second sample matched the victim’s applicant’s profiles in their entirety and matched nine alleles from applicant (distributed across six loci) that were foreign to the victim. With respect to all of the samples, all of the alleles foreign to the victim were alleles that matched applicant’s profile. 4. Trial Preparation On April 23, 2001, defense counsel Steven Greenlee filed a motion to preserve/motion for independent analysis of semen and fibers — which included a request to hire a defense DNA expert at county expense to conduct the independent analysis and to testify at trial. That motion was granted. The prosecutors, Di Glaeser and Denise Nassar, had an open file policy and were prepared to fully cooperate with the defense regarding matters of discovery. Glaeser called the HPD Crime Lab to arrange for the delivery of DNA material for independent testing by the defense. Chu told her that there was no DNA material left because it had all been used up in the HPD Crime Lab’s analysis. Glaeser conveyed this information to Greenlee. After discovering that there was no DNA material left to test, Greenlee decided not to hire a DNA expert. At a pretrial hearing, applicant complained that his attorney had not done any investigation. He claimed that Greenlee had not disclosed to him copies of files, DNA, police reports, warrants, or anything else. Applicant complained that Greenlee “has no experience or knowledge about DNA to question any DNA expert officially.” Applicant made numerous other complaints about his attorney that we need not detail here. The trial judge explained that applicant was free to hire his own attorney if he could do so, but he was not free to choose his court-appointed counsel. B. Trial 1. State’s Case The evidence outlined in part A, sub-parts 1 and 2, of this opinion was presented at trial. In addition, testifying by closed-circuit television, E.T. related the events of his kidnapping. He testified that his kidnapper wore a purple jacket, purple sweat pants, eyeglasses, and a blue hat. E.T. described the car that he was pulled into as being light brown inside. E.T. said that he was taken to a white, wood house with the numbers 2, 1, and 6 on it. He further testified that he went up concrete stairs to get into the house. Inside the house was a brown couch and a black television. E.T. said that the kidnapper tied him to the bed, with his “feet back” and his “hands in the front.” When asked, “Did he tie you to parts of the bed with clothes?” E.T. answered, “Yes.” E.T. testified that the television was on the side by the bed. With the help of anatomically correct dolls, E.T. testified that the kidnapper put his “private” in E.T.’s mouth. The kidnapper subsequently gave E.T. a bath, but did not wash his face. The next day, the man took E.T. back to the store where he had been abducted, and E.T. then went to Remington’s grandmother’s house. E.T. testified that he had previously pointed out the man who kidnapped him on a video lineup. When shown photographs of applicant’s purple jacket and pants, E.T. testified that they looked like the ones his kidnapper had worn. E.T. also testified that applicant’s house and concrete steps, depicted in photographs, looked like the place the kidnapper had taken him to, and E.T. pointed to circles on the sidewalk that he remembered seeing. E.T. also testified that a brown couch and a Houston Oilers cannister that were depicted in photographs were items he had seen in the kidnapper’s house. However, when the camera panned around the courtroom and E.T. was asked to identify his assailant, he did not do so. The State questioned Childs-Henry and Chu regarding their roles in subjecting the face-swab samples to DNA analysis. The focus of the State’s questioning was on the semen fraction of the DNA samples. Childs-Henry testified regarding her role in extracting DNA material from the face swabs for analysis. Although the quantity of the sample was small, she determined that enough DNA was present for analysis. On cross-examination, defense counsel questioned Childs-Henry concerning the DNA extraction protocol. Childs-Henry could not say how many steps were in the extraction protocol that she followed. When asked whether an entire DNA sample could be consumed by testing, she responded that such a thing could happen “if you are a sloppy chemist” or “if there’s not enough sample.” When asked whether the sample was large enough not to have been completely consumed by testing, Childs-Henry responded affirmatively, and she responded that the sample in this case was not entirely consumed. “So there’s a DNA sample left?” defense counsel asked. ‘Wes, there is,” she responded. “And this was after Mr. Chu had done his analysis?” defense counsel queried further. ‘Yes,” Childs-Henry replied. Chu testified regarding his role in analyzing the DNA sample after it had been extracted. Chu explained that he had a Master’s degree in Chemistry, that he completed in-house training as well as outside agency training on DNA analysis, and that he had to pass a proficiency test twice each year on federal guidelines. Chu had been working eight years in the DNA section of the HPD Crime Lab. Chu testified that he conducted the STR method of DNA analysis. He determined that the DNA was a mixture of two individuals. One half was consistent with E.T.’s DNA, he explained, while the other half was consistent with applicant’s. “Statistically,” the DNA results from the face swab material “matched” applicant. The odds of a random match with an individual other than applicant was about the human population of the planet. Chu later testified that he could be about 99.999 percent sure DNA from that swab sample was applicant’s, or about a one-in-1.3 trillion possibility of a random match with another individual. Chu also testified that he conducted a cross analysis of the DNA from both swabs to make sure that they were consistent with each other. On cross-examination, Greenlee questioned Chu regarding whether any DNA material was left for independent analysis. Chu said that no evidence was left for reanalysis. “Very unfortunately, this case it’s very small amounts,” he said, “I have to consume.” Chu also testified that there were ways to determine whether the sample was contaminated. There were strict guidelines in the laboratory to monitor contamination; controls in the DNA test itself would indicate whether contamination was present. So, if a mistake were made, Chu testified, it would show up. “In this case we don’t have any contamination in the laboratory to cause misleading results,” he explained. Greenlee’s cross-examination of Chu covered approximately six pages of the trial record. 2. Defense’s Case Reverend James Ginns testified that applicant was his nephew and that he, applicant, and applicant’s brother had keys to applicant’s residence, but applicant was the only person who lived there on February 11th. Although this evidence suggested that two other persons had access to applicant’s home, no attempt was made to implicate either of these two individuals in the crime. David Savage, applicant’s cousin, testified that applicant came to Savage’s grandmother’s house on February 11th to have Savage work on the stereo in applicant’s car. Applicant arrived at the house between 10:30 and 11:00 a.m. Savage’s grandmother lived on Newbury Street, which was in a part of town that was close to the Southmore location from which E.T. was abducted. Savage testified that applicant left between 2:30 and 3:00 p.m. But Parole Officer Joel Butler later testified that Savage told him that applicant left at about 1:30 p.m. Savage testified that he remembered that day because he saw applicant on television the day after E.T. was taken. A prosecutor would later argue to the jury that, because applicant was not a suspect until February 23rd, Savage must have recognized applicant on February 12th from the composite drawing. Darlene Savage, David Savage’s mother, said applicant left around 3:00 or 3:30 p.m., but she later said she thought he left before she did, and she left between 2:30 and 3:30 p.m. On cross-examination, she said that applicant left after 2:00 p.m. and then said she did not really remember. Applicant’s brother, Ronnie, testified that applicant came by his house between 3:30 and 4:00 p.m. Ronnie further testified that applicant was alone, and did not stay long, because he had to go back home. Ronnie stated that he went over to applicant’s house about five or ten minutes later and stayed thirty to forty-five minutes. Ronnie did not see a child in the house or in applicant’s car. Ronnie returned to applicant’s house at around 6:00 p.m., stayed for about ten minutes, and still did not see a child in applicant’s house or car, despite the fact that he walked through the entire house. On cross-examination, Ronnie said that his second visit lasted thirty to forty minutes. But Ronnie told Officer Valenta, in a February 27th conversation, that he went to applicant’s house at 5:00 p.m. on February 11th, as was his usual routine, and that he went there only once. Ronnie told Parole Officer Paul Ford that, when he left applicant’s house, applicant was asleep. Ronnie testified that he saw his brother’s car when he left for work, between 5:00 and 6:00 a.m. on February 12th, but he had told Officer Valenta that he had seen the car at around 6:30 or 6:35 a.m. Against his professional judgment, Greenlee called applicant’s parole officers, Butler and Ford, to the witness stand because applicant insisted that he do so. Greenlee discussed his reservations with the trial court, but applicant insisted that calling these witnesses was essential to establishing an alibi by showing that he was on electronic monitoring. Applicant also expressed the opinion that he would have to “come behind” these witnesses and testify, even though he knew that doing so would enable the jury to learn his prior criminal record. Applicant reiterated his prior frustrations with his attorney, referred to what he considered to be an earlier request to represent himself, and indicated that he personally needed to question some of the witnesses “because all the facts of this case is not being brought out by my attorney, and he will not do it himself.” The parole officers testified that applicant was on parole for rape and aggravated rape. Applicant was on the highest level of supervision — the Super Intensive Supervision Program (SISP). This program included electronic monitoring. For a while, applicant was on GPS monitoring. At that time, only twenty-five people in the entire state were on that kind monitoring. Applicant was taken off GPS monitoring on December 12, 2000, and placed on a more traditional form of electronic monitoring, which recorded only whether applicant was at home. Applicant was required to report to his parole officer once a week, every Wednesday. Under SISP, restrictions were placed on where applicant could go and when he could be there, and curfews were imposed. Where applicant could go and when he could be there depended upon the weekly schedule made out by his parole officer. On February 11th, applicant was allowed to leave the house at 10:00 a.m. to go to church until noon, and then he was allowed to go to his friend Ben’s until 5:30 p.m. Newbury Street was not a place that he was allowed to go that day. On February 18, 2001, a parole warrant issued against applicant for a curfew violation. Applicant told his parole officer that he was delayed in getting home that day because of a flat tire, and he showed a tire in his car with a screwdriver sticking out of it as proof. The warrant was in the TCIC system, but applicant’s parole officers chose to investigate the warrant, and a decision was later made to withdraw it, but applicant was arrested on the warrant before it was withdrawn. Finally, applicant testified. Applicant denied having anything to do with E.T.’s abduction and sexual assault. Applicant confirmed that he was on parole for aggravated rape. Applicant said that he received permission from his parole officer to go to Newbury Street, but the parole officer just would not admit it. Defense counsel asked, “You heard Mr. Butler testify; is that correct?” Applicant responded, “I heard him lie, yes.” When asked by defense counsel when he informed Butler that he would be visiting the Newbury residence on February 11th, applicant first responded that he would have told him on the 21st, when he visited the parole office. When Greenlee explained that the time frame had to be before Sunday, February 11th, applicant responded that he gave notice the Friday before, February 9th. Applicant further testified that he had no regularly scheduled reporting day — he just had to report every week. When confronted with the fact that Ford and Butler said it was every Wednesday, applicant retracted his earlier statement and confirmed that he was required to report every Wednesday. Then applicant claimed that he told Butler about going to the Newbury address on February 7th. When asked what his schedule was like on that Sunday, applicant replied, “Oh, it’s different times, different Sundays that I am doing different things on different days. What particular Sunday are you asking me about?” Greenlee further asked, “You heard [Butler] testify that your initial movement on Sunday was restricted to going to church, did you hear that?” Applicant responded, ‘Yeah, I heard him say it, but he’s lying. He’s lying. He knew I was going over there.” Defense counsel later asked, “It’s your testimony you had authority or permission to go to Newbury?” Applicant replied, “That’s true.” Applicant also testified that there was no damage to his car, and that the police department tore it up because they were the only ones that had access to the car. He testified that he left the Newbury address at around 2:30 to 2:35. He said that he did not go straight home because he stopped to get some gas, and then ran by Ronnie’s house. When asked whether he was familiar with the Southmore and Live Oak, South-more-288 area, applicant responded that he was, because he had been there in November or December looking for his “brother’s sister” who was on drugs. Applicant agreed that it was a restricted area and that he was in violation of his parole. But then applicant claimed that it was not a parole violation to drive through the area so long as he was on his way to go somewhere he was allowed to go. Applicant stated that he went to the area many times looking for his brother’s sister. On cross-examination, applicant conceded that he was in the area in October, November, and December. When confronted with GPS records, applicant said that he did not know but may have been in the Southmore area on multiple occasions at times referred to in the records, with the last time being on December 10th. He said that he was taken off GPS two days later. The prosecutor then asked, “I’m sure your parole officer will come in and tell us that you had permission to cruise whatever area you wanted, is that correct?” Applicant responded, “He probably won’t. He won’t tell you nothing else that was the truth.” Applicant said that he told his parole officer that every Sunday from the middle of December through his arrest he was going to his friend Ben’s on Sunday after church to work on his car. Later, the prosecutor asked, “Would you agree with me, Mr. Napper, that there were many occasions when you would deviate from your weekly schedule and do whatever you felt like doing?” Applicant replied, “Well, what’s your definition of deviate, going another direction, or is that what you’re saying?” Upon further questioning, applicant agreed that he could go only to church or to Ben’s on Sunday unless he told his parole officer ahead of time or there was an emergency. Applicant admitted to normally wearing a black baseball cap. When asked if he was taken from his residence on February 23rd, applicant responded, “I was kidnapped. They didn’t have a warrant.” With respect to whether there was body lotion in his bedroom, applicant said that his brother had a lot of stuff there and applicant “might have run in his room, grabbed it, put it in there.” When asked who purchased the lotion, applicant said his brother’s sister purchased it. Applicant admitted that he had a prior aggravated rape conviction dated August 25, 1981, but he claimed that he did not do it, that his attorney sold him out just like in the present case. Applicant also admitted that he pled no contest to rape on September 4, 1981, but he claimed that he did not commit that crime either. When asked about a rape conviction on September 13, 1979, applicant claimed that he knew about only two rapes. When asked, “You remember being placed on probation one time before for rape?” applicant responded, “I was told that was dismissed or dropped.” Applicant admitted being convicted of indecent exposure on September 21, 1993. He claimed that he was sitting inside his car peeing in a can, and he said he did not think he was guilty of that offense either. Applicant acknowledged that, according to his electronic monitor, he arrived home at 3:07 p.m. on February 11, 2001. Applicant protested that this effectively gave him an alibi because the victim was picked up at 3:10 p.m. Applicant continued, “Parole officers know that. They won’t tell you all of that. They told the news media that. I don’t know why they won’t tell the jury. That’s my reason for testifying.” Applicant also testified that the children did not describe his car: “It’s not light green or red. My car is dark. It’s gray. It’s not even black.” After admitting that several of E.T.’s descriptions matched applicant’s house, applicant said, “You didn’t even put that in the indictment, that it was at my house. [E.T.] was taken to an apartment before he come to my house. Y’all ain’t telling the jury that.” Applicant denied telling Butler that Ronnie spent the night at his house on February 11th. 3. State’s Rebuttal On rebuttal, Sergeant Hoffinaster testified that applicant looked different at trial than he did when Hoffinaster first encountered him. Butler testified that applicant told him on February 27th that Ronnie Napper came over to applicant’s house at 4:30 p.m. on February 11th and stayed all night. Butler also testified that applicant never told him that he was not going to church that day or that he wanted to go to the Savage residence. Butler further testified that applicant did not have permission to be at the Newbury address, and he testified that David Savage told him that applicant left Savage’s grandmother’s house at 1:30 p.m. Butler also testified that applicant had been instructed to take the most direct route to his destination and was not authorized to travel around the Live Oak and Calumet area for 30 minutes. Finally, Butler testified that applicant told him that he did not do it. Butler replied that if applicant did not do it, then DNA ought to show that he did not do it. Applicant responded, “Well if it does, they are just setting me up.” 4. Verdict and Judgment The jury found applicant guilty of aggravated sexual assault and aggravated kidnapping. At punishment, applicant testified and continued to maintain his innocence. After defense counsel completed his questioning of applicant, applicant stated, “You sold me out Greenlee.” Greenlee then rested the defense, and applicant stated, “I didn’t do it. I swear I didn’t do this.” While the jury was in recess, applicant stated, “They framed me in this case. I will fight it the rest of my life. They framed me.” When the jury was brought in, applicant further stated, ‘Y’all going to pay me. Fed investigation, as soon as I get a chance y’all going to pay me. As soon as I can there’s going to be a fed investigation.” Finding that applicant had committed two prior sex-related felonies, the jury sentenced applicant to an automatic life sentence. When asked whether he had anything to say before the trial judge pronounced sentence, applicant replied, “When they frame your family, remember me. That’s all I have to say.” C. After Trial 1. Motion for New Trial Applicant gave notice of appeal, and the trial court appointed Bob Wicoff as appellate counsel. Wicoff filed a motion for new trial on applicant’s behalf, claiming that applicant was denied his right to represent himself at trial and that he received ineffective assistance of counsel. In connection with the motion, applicant submitted an unsworn declaration in compliance with the Texas Civil Practices and Remedies Code. Among other things, applicant pointed out that, although Green-lee elicited testimony from applicant’s parole officers that applicant was on electronic monitoring on February 11, 2001, Greenlee failed to elicit any testimony from them that the electronic monitoring confirmed applicant’s presence at home at 3:07 p.m. Applicant stated that this was the reason that he had wanted to call these witnesses and that the failure to elicit this information forced him to take the stand to testify about the matter. Wicoff also submitted an affidavit from Butler, who expressed surprise at the fact that Greenlee failed to elicit this information from him— leading Butler to wonder why defense counsel chose to call him as a witness. Wicoff also filed a motion to hire a DNA expert at county expense, which was granted. Wicoff later submitted an affidavit from Dr. Elizabeth Johnson, a forensic DNA expert who had developed some widely utilized DNA analysis techniques. Dr. Johnson concluded that the testimony of Childs-Henry and Chu reflected “a lack of understanding” on their part “as to some of the fundamental aspects of body fluid identification, DNA extraction and typing as well as interpretation of results.” Dr. Johnson further stated that she had reviewed numerous HPD Crime Lab cases in which these individuals performed anal-yses and she had detected serious errors in most of these cases, including the “unnecessary consumption of the evidence due to poor extraction protocols and techniques ... and erroneous interpretation of data.” She also stated that re-testing was a critical part of the evaluation of a criminal case involving DNA evidence and a necessary-part of preparing an adequate defense. “Because DNA evidence can have a tremendous impact on the outcome of a criminal case,” she explained, “it is generally accepted in the relevant scientific community that re-testing needs to be performed whenever the evidence permits this.” Dr. Johnson cited a National Research Council (NRC) recommendation that “whenever feasible, investigative agencies and testing laboratories should provide for repeat testing.” Dr. Johnson further observed that, in this case, “as in many other cases I have reviewed involving the Houston PD lab, the lab apparently consumed all of the critical evidence ... without notifying the prosecution of their intent to do so.” She explained that “[m]any laboratories” with which she was familiar would “not consume limited samples without notification so that defense counsel can have the opportunity to have the testing observed if duplicate testing is not possible.” Although re-testing was not possible, Dr. Johnson stated that a qualified expert should have reviewed the bench notes from the HPD Crime Lab to determine whether “an error in the analysis was documented or if the data obtained were interpreted and testified to correctly.” Dr. Johnson identified numerous questions that defense counsel should have asked Childs-Henry and Chu, including “what were the relative proportions of major and minor DNA contributors and the corresponding peak heights” and whether there could be an “alternative interpretation of the mixed DNA profiles other than the approach utilized by Mr. Chu in which he determines the DNA profiles in the evidence based on his knowledge of the victim’s and defendant’s reference DNA profiles” and “whether it was appropriate of Mr. Chu to isolate the defendant’s profile out of a mixture and report the frequency of the defendant’s DNA profile rather than calculating the combined frequency of all possible donors to the mixed DNA profile obtained from the evidence.” Based upon a study conducted in Connecticut, Dr. Johnson further explained that mixed samples were especially difficult to interpret and that “often an analyst will ‘see’ a particular individual’s profile in a mixture by comparing it to a defendant’s or victim’s profile and perform a biased analysis. This type of biased analysis does not consider that the evidence profile in a mixed sample could be the result of the profiles of individuals other than the defendant and the victim.” Dr. Johnson stated that this information was a subject that should have been brought out in cross-examination. Dr. Johnson also criticized defense counsel for failing to cross-examine the witnesses regarding the NRC recommendation or the practice of many laboratories to provide for repeat testing. Dr. Johnson concluded that the expert testimony from Childs-Henry and Chu was “some of the most poorly presented and cross-examined testimony that I have ever reviewed.” She found that Greenlee’s “lack of knowledge ... in the area of serology and DNA typing is apparent through his extremely cursory and superficial cross examination.” The State submitted an affidavit from Greenlee. After Greenlee became aware that there was no DNA left to test, and after consulting with other attorneys, he decided “that the only purpose that an expert would serve would be to review the procedures and methodologies used in the analysis by the State’s experts.” Greenlee believed that he “could effectively argue the unfairness of no sample being available to the defense to analyze” without the help of an expert. He further believed that, “if there were some issues with regard to methodologies and procedures,” he “could get more mileage out of this on cross-examination.” He also feared “the distinct possibility that our expert would confirm the propriety of the State’s methodologies and procedures during the laboratory testing, thereby reinforcing the validity of the DNA results.” Greenlee believed that he had effectively cross-examined the witnesses involved in the DNA collection and testing. With respect to the testimony of the parole officers, Greenlee said that applicant believed that they would testify that there was not a valid parole warrant in effect at the time of his arrest. Greenlee had counseled against calling the parole officers, but applicant had insisted that they be called. Greenlee further stated that he did not spend much time on applicant’s time of arrival at home on February 11th because “it would not eliminate [applicant] as being capable of committing the offenses alleged.” This was so because applicant was “in an area not too far from the kidnapping” that day and because intervening circumstances that took place prior to the police being called would have given applicant “more than enough time to travel from the kidnapping scene to his residence.” Defense counsel believed that the weakest areas of the State’s case were the DNA testing procedures and the victim’s inability to identify applicant in the courtroom. Greenlee believed that, if jury were inclined to find guilt despite those weaknesses in the State’s case, then they were not going to find the time of applicant’s arrival at home to be significant. Greenlee also stated that applicant was very demanding and uncooperative, and he repeatedly failed to take Greenlee’s advice. Finally, Greenlee pointed out that he used an investigator, filed many motions on applicant’s behalf, reviewed reports, and spoke to witnesses. 2. Appeal The trial court denied the motion for new trial, and applicant appealed. The sole issues on appeal were claims of ineffective assistance of counsel. Applicant claimed that Greenlee was ineffective because: (1) he did not elicit testimony from applicant’s parole officers regarding the fact that the electronic monitoring system had indicated that applicant had returned home at 3:07 p.m., (2) he elicited damaging testimony from his parole officers regarding his prior convictions and details of his previous level of surveillance while on parole, and (3) trial counsel’s performance was deficient with respect to the DNA evidence. The court of appeals affirmed. With respect to the electronic monitoring allegation, the court responded that applicant was not prejudiced because the State elicited the 3:07 p.m. — arrival information during its cross-examination of applicant. The court of appeals also found that applicant’s 3:07 p.m. arrival time did not necessarily exclude him from being the kidnapper. Regarding the allegation about the testimony of the parole officers, the court of appeals found that applicant had insisted on calling these witnesses after being warned of the potential dangers and that defense counsel was not unreasonable in addressing, on direct examination, unfavorable facts that “most likely would have been addressed by the prosecutors on cross-examination.” Finally, with respect to the allegations regarding DNA evidence, the court of appeals concluded that Dr. Johnson’s affidavit did “not allege any actual errors in Childs-Henry’s or Chu’s work which trial counsel failed to discover or develop.” The appellate court also pointed to defense counsel’s cross-examination of police officers regarding how the facial swabs were handled prior to the HPD Crime Lab’s receipt of the evidence and to defense counsel’s emphasis on the fact that no sample remained for applicant to test independently. A petition for discretionary review was subsequently filed and refused. 3. Additional Testing In November of 2002, adverse publicity began to arise about the condition and practices of the HPD Crime Lab. Within a month, the acting Chief of Police commissioned an outside review of the lab’s DNA/Serology section. Based upon a preliminary oral report of the auditors, HPD suspended the performance of all DNA analysis at the Crime Lab. In 2003, HPD and the Harris County District Attorney’s Office began identifying cases in which some form of DNA analysis had been performed by the lab. “This process evolved into a long-term re-testing project coordinated among HPD, the District Attorney’s Office, and outside DNA laboratories.” One of the cases identified in this process was applicant’s. The seemingly empty tubes that had contained DNA from the face swabs were forwarded by the HPD Crime Lab to ReliaGene Technologies. The lab also forwarded what purported to be reference samples from E.T. and applicant. ReliaGene used a buffer solution to re-suspend any DNA residue that might be left in the tubes, and it tested the DNA at fourteen genetic loci, including the original ten loci tested by Chu. Of the four tubes of DNA from the face swabs, only one tube — containing the epithelial fraction from one of the face swabs — produced any results. In May of 2004, ReliaGene reported that the major component of the DNA was not consistent with E.T. or applicant, while the minor component contained two weak alleles that were consistent "with applicant. These two weak alleles were found in the additional loci not tested by Chu. After these results, it was determined that the tube purporting to contain E.T.’s reference sample was not the correct tube. On June 30, 2004, after a new reference sample was procured from E.T. and analyzed, Relia-Gene concluded that the major component of the face-swab DNA was indeed consistent with E.T.’s DNA. The conclusion with respect to the two weak alleles matching applicant was unchanged. No other DNA alleles foreign to E.T. were detected in the sample. In an affidavit later obtained in habeas corpus proceedings, Gina Pineda, Assistant Director at ReliaGene, explained that she did not believe that the DNA extracts received by ReliaGene were quantitatively the same as those tested by the HPD Crime Lab. Consequently, she would expect ReliaGene to detect fewer alleles in its analysis, but for the results to remain consistent with what the HPD Crime Lab had found. Her examination of the data indicated that, indeed, the results obtained by the two testing laboratories were “not discrepant.” She also expected that, if a third round of testing were performed, there would be a further progressive loss of alleles. Some of the victim’s alleles might also become lost and “some allele drop-in may also be observed.” In an August 2004 newspaper article, assistant district attorney Marie Munier was quoted as saying about the ReliaGene results, “You can’t even do a statistical analysis (on the probability that the DNA was Napper’s). I don’t know what it would be, but it wouldn’t be very much.” In 2005, during the instant habeas proceedings, the parties agreed to more testing, by Orchid Cellmark. They acknowledged in writing that the testing was expected to consume all of the evidence and agreed to waive the opportunity to have a representative observe or participate in the testing. In June of 2005, Orchid Cellmark reported that it was able to obtain results for the face-swab tubes containing the epithelial fractions of DNA but was unable to obtain results for the tubes containing the sperm fractions. The samples were analyzed at sixteen different loci, including the ten loci tested by Chu and the four additional loci tested by ReliaGene. The samples for both epithelial-fraction tubes were mixtures. For one of the epithelial samples, Orchid Cellmark determined that the major profile matched the victim, and the suspect could not be excluded as a possible donor of the minor alleles in the mixture. Four alleles foreign to the victim were found (across four different loci), all of which matched applicant’s DNA profile. One of these alleles was found at a locus that had been tested by both HPD and ReliaGene but had not previously been found there. One of the alleles was confirmed at one of the new loci that had been tested by Relia-Gene. The other two alleles were found at new loci tested only by Orchid Cellmark. Orchid Cellmark determined that the other epithelial sample was a mixture of DNA from the victim and an unknown individual. Five alleles foreign to both the victim and applicant were found, and two loci yielded inconclusive results. In subsequent testing, Orchid Cellmark was able to exclude as possible contributors to the foreign alleles A.G. Riddle, C. West, J. Chu, J.W. Belk, L.R. Verbitskey, M. Childs-Henry, and R. Hilleman. Orchid Cellmark provided statistical frequency calculations for the epithelial sample that contained the four alleles that matched applicant’s DNA profile. According to Orchid Cellmark, the approximate frequencies for unrelated individuals for all possible types included in the mixture, making no assumptions regarding the number of DNA sources, are 1 in 255 for African-Americans, 1 in 3,427 for Caucasians, and 1 in 585 for Hispanics. 4. Bromwich Report As a result of problems discovered with the HPD Crime Lab, a team headed by Michael R. Bromwich was chosen to further investigate and evaluate the Crime Lab’s practices, both past and present. The report, published on June 13, 2007, contained numerous-scathing criticisms of the Crime Lab’s serology/DNA section. “On the whole,” the serology/DNA section’s “work did not meet the generally accepted forensic science principles that existed at the time and posed major risks of contributing to miscarriages of justice in extremely significant cases.” Bromwich attributed this failure to budgetary problems, incompetent leadership, and lack of training for the DNA analysts. “[Ujntil the public crisis engulfed the Crime Lab, it was never provided adequate financial support to hire and train the number of criminalists necessary to handle the Lab’s ever-increasing workload.” The DNA section was in “shambles — plagued by a leaky roof, operating for years without a line supervisor, overseen by a technical leader who had no personal experience performing DNA analysis and who lacked the qualifications required under the applicable Federal Bureau of Investigation (‘FBI’) standards, [and] staffed by underpaid and under-trained analysts.” Training was one of the first areas in which funding was cut when the HPD Crime Lab’s budget became tight. The lab was populated by civilian, as opposed to law enforcement, employees, and as such, was marginalized within HPD. Until the 2002 audit, the HPD Crime Lab did not submit to reviews by outside agencies and never sought to achieve accreditation. As a result, HPD’s analysts became isolated from the rest of the forensic science community. Although the person with supervisory control over the DNA section certified that internal audits conforming to FBI quality assurance standards found that those standards were met, the outside audit found to the contrary. Bromwich characterized the HPD DNA analysts as “woefully undertrained.” “The problems we observed in the historical DNA cases,” he further explained, “are not attributable to individual rogue analysts who departed from the Crime Lab’s approved practices. On the contrary, the widespread and serious deficiencies in the historical Crime Lab were consistent with the Crime Lab’s accepted and understood practices.” Consequently, “[fjlawed practices and embedded misunderstandings ... became accepted by analysts within the DNA/Serology Section as the correct way of doing things. These misunderstandings infected the work of the Section’s analysts from the analysis through the trial testimony.” Bromwich found “the same types of major issues across all the Crime Lab’s DNA work, regardless of the analyst or the DNA typing system used.” One of the cases that Bromwich reviewed was applicant’s. Bromwich criticized Childs-Henry for failing to properly document the procedures she used and for failing to use some of the more definitive procedures for detecting and confirming the presence of semen. With respect to Chu’s testing, Bromwich found that the “original DNA testing appears to have generated good quality and clear results from potentially very difficult forensic evidence samples.” The “overall assessment” of Chu’s testing was that “he developed clean, interpretable profiles from the four evidence samples he tested.” But, according to Bromwich, applicant’s case “illustrates two significant problems with the Lab’s historical DNA work. First, the Crime Lab analysts utilized all of the readily testable sample in this case,” and second, Chu used an inappropriate statistical analysis of the random match probability. Regarding the first problem, Bromwich found that “it was unnecessary and inappropriate for Mr. Chu to have tested the extracts from both swabs, thereby consuming the available sample in this case.” Although Chu stated in an affidavit in the instant habeas proceedings that he did not “consume all of the extract in this case out of carelessness or out of a desire to prevent additional testing,” Bromwich concluded that “there was no need, and it was a mistake, for him to consume both of the redundant ‘face-cheek’ swabs.” Doing so was “the product of very poor laboratory practice.” This “misstep” was compounded by the fact that Childs-Henry had “discarded the tubes containing the raw evidence (the swabs) after she performed the DNA extractions.” And while “outside laboratories have been able to obtain some results” — “mixed results” — from testing the unseen residue in the tubes, those results “do not approach the strength” of “Chu’s original results.” Notably, the Bromwich Report observed, “Neither laboratory has been able to confirm the alleles consistent with Mr. Napper’s DNA profile that Mr. Chu detected in the sperm fractions of each of the swabs, which captured most of the consistency between evidence profiles and Mr. Napper’s DNA profiles.” Turning to Chu’s statistical analysis, Bromwich initially criticized statements made by Chu in a March 30, 2001 report. There, Chu characterized the DNA material as a “mixture” consistent with the victim and with applicant. Chu went on to state, “Given the population on the face of the earth and eliminating the probability of identical twins, the DNA profile statistically matches only Lawrence Napper.” Bromwich criticized these statements as “contradictory” and called Chu’s statistical statement “incorrect and misleading.” Further, Bromwich’s examination of Chu’s file revealed that Chu had based his statistical calculation of the random match probability upon applicant’s known profile, “which was the usual — and extremely flawed — practice in the Crime Lab.” In another portion of his report, Bromwich explained that, while calculating the random match probability of a single DNA profile was “relatively simple and straightforward” and can “provide the most discriminating information about whether a particular individual could be the source of the biological evidence,” the same could not be said of a mixed sample containing DNA from more than one person. When a DNA profile contains DNA from more than one person, “it is much more difficult to provide compelling statistical evidence that a particular person’s DNA was found in an evidence sample.” Random match probabilities related to a mixture “may result in frequency estimates that indicate that a relatively large proportion of the human population could have contributed to the biological evidence.” A frequency estimate based upon a suspect’s known reference sample “is completely irrelevant to the strength of the DNA evidence when the DNA profile is a mixture.” Nevertheless, despite the inappropriateness of calculating a random match probability using a defendant’s known profile, “the Crime Lab virtually always calculated its reported frequency estimates” in this fashion. When setting out an allelic table, Brom-wich bolded the alleles that were unique to applicant’s profile (matching applicant but foreign to the victim), indicating that it was those unique alleles from which a random match probability must be calculated. Based on Chu’s STR data, Bromwich calculated the accurate random match probability as “1 in 232,000 in the African American population, 1 in 1,920,000 in the Caucasian population, and 1 in 7,430,000 in the Hispanic population.” He characterized these statistics as “relatively strong results given the nature of the sample in this case,” but “a far cry from the sole source, unique match that Mr. Chu reported.” 5. Habeas Wicoff filed a habeas application on applicant’s behalf. In addition to the evidence above, the habeas court obtained affidavits from Glaeser, Greenlee, Childs-Henry, Chu, and Johnson. Applicant submitted affidavits from two attorneys— Cunningham and Downey. The court conducted an evidentiary hearing, at which Greenlee testified. And the habeas court received various pieces of documentary evidence. The facts conveyed by Glaeser’s affidavit have already been discussed. In his affidavit, Greenlee stated that applicant’s response to the criminal charges “was an adamant denial of the charge.” Before trial, Greenlee believed there would be three main pieces of evidence at trial: DNA evidence, evidence relating to the bottle of orange lotion, and evidence relating to applicant’s electronic monitoring. Greenlee believed he had effectively shown on cross-examination that there was not a clear set of procedures, methodologies, and practices regarding how the DNA material was analyzed. In her affidavit, Childs-Henry stated that she “did not observe the vials of extract after they were transferred to Chu” and had “no way of knowing what amount of liquid, if any, actually remained in the vials once Chu had completed his analysis.” Realizing that this statement conflicted with her trial testimony, Childs-Henry explained that she had trained under the RFLP variant of DNA analysis and was unfamiliar with the STR procedure at the time of trial. She had just assumed there would be sample left based on her RFLP training because there was generally sample left after RFLP testing. After reviewing her trial testimony, Childs-Henry stated that she realized that the defense attorney was asking whether she knew for a fact that liquid DNA extract still existed and that, in reading over her answers, she could only assume that, while she was being cross-examined by defense counsel, she “did not understand the intent behind his questioning.” She further stated that it was not her intent to be misleading or untruthful but that her statement that there was DNA sample left was “unintentionally incorrect.” Chu’s affidavit contained his statement, recounted above, that his consumption of all of the DNA material was not from carelessness or a desire to prevent additional testing. He said he did not expect to be able to develop a profile from the sample because the concentration of DNA was so low. He also stated that in such cases, it is not uncommon to consume all of the liquid extract in testing. He also said that, when asked whether there was any DNA left to test, he interpreted the question, under “the common language and words of art used in our lab,” to mean whether there was any liquid extract remaining — not whether it was possible to obtain DNA test results from residue that might have been left in the tubes. His statement to Glaeser and his subsequent testimony at trial accorded with that understanding and was not an attempt to hide or withhold anything from the defense. Dr. Johnson’s affidavit responded to the DNA test results from ReliaGene. She believed that the two alleles that matched applicant’s DNA had to be analyzed separately for the purpose of establishing a random match probability. One of the alleles was present in 14.5 percent of African-Americans, 16.5 percent of Caucasians, and 10.29 percent of Hispanics. The other allele was present in 30 percent of African-Americans, 32.5 percent of Caucasians, and 39.23 percent of Hispanics. She also expressed dismay at HPD’s initial error in forwarding an incorrect reference sample to ReliaGene. “This type of serious error is typical of the work performed at the HPD lab based on my review of many cases,” she said, “and it calls into question the validity and reliability of retesting any DNA that was originally extracted by the HPD lab.” She stated that it was “highly likely given the sloppy work performed by HPD, that even if the foreign alleles detected in the face swab extract originated from Mr. Napper’s DNA, that it could be a result of contamination within the HPD lab.” Dr. Johnson also criticized the statistical frequency estimate offered by Chu at trial (that the DNA from the swabs statistically matched applicant’s profile). She contended that the estimate was “completely unsupported scientifically” and was “patently false.” She further stated that a “mixture of DNA cannot possibly statistically match only one person” and “attempts to discern the individual profiles in a mixture of DNA is very difficult.” In his affidavit, Cunningham opined that Greenlee’s cross-examination of Chu was “inadequate” and “demonstrated the dangers of trying to question an expert if you have no training in his area of expertise.” Cunningham faulted Greenlee for failing to hire a consulting expert. Cunningham believed that a consulting expert was necessary to help defense counsel understand the subject better, to shed light on the statistical calculation made by Chu, and to enable defense counsel to know wha