Ex Parte Reed

271 S.W.3d 698, 2008 Tex. Crim. App. LEXIS 1569, 2008 WL 5234225

• Court: Court of Criminal Appeals of Texas
• Decided: December 17, 2008
• Docket: AP-75693
• Status: Published

Opinions

OPINION

KEASLER, J.,

delivered the opinion of the Court

in which MEYERS, PRICE, JOHNSON, HERVEY, HOLCOMB, and COCHRAN, JJ., joined.

Rodney Reed was convicted and sentenced to death for the murder of Stacey Lee Stites. In this second subsequent application for a writ of habeas corpus, Reed has failed to prove that the State suppressed evidence in violation of Brady v. Maryland. Reed has also failed to meet the requisite, gateway standard of innocence — showing that it is more likely than not that no reasonable juror would have convicted him in light of the new evidence not presented at trial — under Article 11.071, Section 5(a)(2) of the Texas Code of Criminal Procedure. Relief is therefore denied.

I. Facts

Stacey Lee Stites’s partially clothed body was discovered on the side of a desolate country road in Bastrop County, Texas on April 23,1996.

Stacey and her mother, Carol Stites, moved to Bastrop from Smithville in 1995 after Stacey graduated from high school. After briefly working for a car dealership in Bastrop, Stacey began working at the Bastrop H.E.B., a grocery store, as a cashier and bagger in October 1995. In January 1996, Stacey and her mother moved to the nearby town of Giddings so that Stacey could be with her flaneé, Jimmy Fennell. Fennell, who had completed the police academy at the Capital Area Planning Counsel Organization (CAPCO) in October 1995, was hired as a patrol officer with the Giddings Police Department in December. With a long-term interest in law enforcement, Fennell had previously been employed by the Bastrop County Sheriffs Office as a jailer. Carol described Stacey and Fennell as inseparable since they began dating a few weeks after meeting at the Smithville Jamboree in May 1995. By late December 1995, the two were engaged.

Stacey, Carol, and Fennell moved into an apartment complex just outside Gid-dings. Stacey and Fennell shared an apartment on the second floor of the apartment building, and Carol lived in a separate one-bedroom apartment downstairs.

With a big church wedding planned for May 11, 1996, Stacey transferred into the produce department at H.E.B. to earn more money. The new assignment required her to report to work at 3:30 a.m. to stock produce for the day. Normally, she would wake up between 2:45 to 2:50 a.m. and take anywhere from five to twenty minutes getting ready to leave for work; she would dress in her H.E.B. uniform, which consisted of blue pants and a red shirt with an H.E.B. insignia on the front. Typically, she would wear a white T-shirt and cax-ry the red shirt with her on the way out the dooi’, along with a plastic cup of juice or water. Although Stacey had access to Carol’s white or gray Ford Tempo, she routinely drove Fennell’s red Chevrolet S-10 extended-cab truck to work. Carol’s car was unreliable and had broken down on the road in the past. When commuting to work, Stacey would take Highway 290 to Highway 21 and then Loop 150/Chestnut Street, over the railroad ti'acks into Bastrop. The drive took approximately twenty-five to thii-ty minutes. When she finished her shift in the early afternoon, Stacey would usually go to Carol’s apartment, take a nap, and then get up and prepare things with Carol for the upcoming wedding.

After leaving work on April 22,1996, the day before she died, Stacey arrived at Carol’s apartment early in the afternoon. She ate lunch and took a nap. Fennell came home from work a few hours later, and having boirowed Carol’s Ford Tempo, Fennell returned Carol’s extra set of car keys to Carol by placing them on a shelf in her apartment. Carol designated the extra set as Stacey’s set. The three then briefly talked about their schedules for the following day. Stacey was scheduled to be at woi'k at 3:30 a.m., and Fennell was not scheduled to work. Fennell and Stacey had planned to go to the insurance agent and to pick out flowers for the wedding ceremony after Stacey got off of woi'k. When Fennell suggested driving Stacey to woi’k, Carol offered to drive him to Bas-trop to meet Stacey so that Fennell could sleep in. However, Fennell declined Carol’s offer, stating that he would dx-ive Stacey to work. Fennell then left in his truck to coach a little-league-baseball team with his friend and coworker, Officer David Hall. He returned between 8:00 and 8:30 p.m. Stacey met Fennell outside of Carol’s apartment, and according to Carol, the two then ran upstairs laughing “as hard as they could.”

When Fennell and Stacey returned to their apartment, they showered together. Although Stacey was taking birth-control pills, the two did not have sexual intercourse because, at this point in her prescription cycle, the vitamin pills she was taking allowed for a greater possibility of pregnancy. The two also discussed them plans for the next day for a second time. Abandoning them earlier plan, they agreed that Stacey would take Fennell’s truck to work and that Fennell would arrange to have Carol take him to meet Stacey in Bastrop when she got off of work. Stacey then went to sleep at 9 p.m., while Fennell stayed up and watched the news.

The next morning, April 23rd, Andrew Cardenas, Stacey’s coworker in the produce department, arrived at the Bastrop H.E.B. around 3:30 a.m. and waited for Stacey in the parking lot. Cardenas would usually wait in his car for Stacey to arrive so that they could “keep an eye on each other, to make sure nobody was around and walk inside the store together....” Cardenas regarded Stacey as a punctual employee, and when she failed to show up for work, he became concerned. Cardenas eventually went into work to start his shift, but he kept an eye out for Stacey.

At 5:23 a.m., while on routine patrol, Officer Paul Alexander with the Bastrop Sheriffs Department observed Fennell’s truck parked in the Bastrop High School parking lot. Mindful that the truck had not been parked there during his previous patrol of the area and that there were no other vehicles in the lot, Officer Alexander contacted the dispatcher and requested a stolen-vehicle check. The dispatcher reported that the vehicle was registered to an individual with the last name Fennell. Although Officer Alexander knew Jimmy Fennell, he did not know him well, and it did not enter his mind that the truck belonged to Jimmy Fennell. When Officer Alexander looked inside the cab with his flashlight, he noticed that the driver’s seat was reclined and that there were books and clothing on the seats. Outside the driver’s side door on the ground, Officer Alexander observed a small piece of a broken belt with a buckle. After noting that there was no shattered glass, that the ignition was intact, and that the driver’s side door was locked, Officer Alexander concluded that nothing was out of order and returned to his patrol duties.

Still looking out for Stacey to arrive at work, Cardenas finally decided to call Carol between 6:30 and 7:00 a.m. "When Cardenas told Carol that Stacey failed to show up for work, Carol became upset and immediately yelled out for Fennell. Cardenas then went back to work, and Carol called Fennell on the phone, waking him up. Frantic, Carol told Fennell that H.E.B. called and told her that Stacey did not show up for work. Fennell rushed down the stairs, putting on a shirt on the way down. He told Carol to call authorities and tell them that he and Carol were looking for Stacey. Carol had both sets of keys to her car, so Fennell took Stacey’s set and drove to Bastrop in Carol’s Tempo to look for Stacey. He drove to the H.E.B. and then returned to Carol’s apartment. He did not see any sign of Stacey or the truck. Meanwhile, officers with the Bastrop Police Department were looking for Stacey, and David Board, an investigator with the Department, called Carol to ensure her that they were doing everything possible to locate Stacey.

At approximately 9:00 a.m., after authorities received the missing-persons report, Ed Selmala, an investigator with the Bas-trop Police Department, was dispatched to the Bastrop High School parking lot. Upon arrival, Investigator Selmala notified other law enforcement officers, including Board, of the truck’s location and requested assistance. While numerous investigators from the Bastrop Police and Sheriffs Departments were photographing the truck and other pieces of evidence, Officer Alexander was called back into work to explain why he ran the license plate on the truck earlier that morning and to write a report.

The truck was later taken to a local tow shop and held until it could be transported to Austin so that members of the Texas Department of Public Safety Crime Laboratory (DPS Crime Lab) could process it for evidence. While the truck was at the tow shop in Bastrop, authorities requested Fennell’s presence to identify items found in and outside of the truck. Fennell was specifically instructed not to touch anything and to peer into the cab and identify anything that was not supposed to be in the vehicle. Fennell observed several things in the truck that were “out of the ordinary.” First, one of the tennis shoes that Stacey normally wore to work was on the floorboard of the passenger’s side of the truck. Second, there was a foamy substance resembling saliva on the carpet covering the hump over the truck’s transmission. Third, there were broken pieces of green plastic in the console from the type of cup that Stacey usually took with her in the truck. Fourth, the driver’s seat was laid back at a forty-five-degree angle. Fifth, the driver’s seatbelt was still buckled. And sixth, there was a large smudge on the back window on the passenger’s side. Fennell also identified several items found outside the truck. First, there were carbon copies of checks from his checkbook. And second, regarding the piece of the belt with a buckle attached, Fennell told investigators that it was part of the belt that Stacey normally wore to work. After this, Fennell returned to his apartment complex in Giddings.

When the truck was delivered to the DPS garage in Austin, a crime-scene team began to process it for evidence. The team stopped them initial overview of the truck when Stacey’s body was discovered by Kenneth Osborn shortly before 3:00 p.m. on Bluebonnet Drive, located off of FM 1141. Osborn, a real estate appraiser, was early for a 3:00 o’clock appointment and decided to drive on Bluebonnet Drive to pick some flowers for his wife. He spotted Stacey’s body among some thorny brush in a ditch on the side of the road. When Osborn approached Stacey’s body, he realized that she was dead. He got back into his car, stopped at a house nearby, called the police, and then went back to Bluebonnet Road to wait for the police.

John Barton, an investigator with the Bastrop County Sheriffs Department, was one of the first law-enforcement officers to arrive at the scene. He covered Stacey’s body with a green blanket to prevent the media, circling above in a helicopter, from taking photographs. He also closed off the crime scene and began to photograph the area and Stacey’s body. Shortly thereafter, Bastrop authorities, joined by Texas Ranger L.T. Wardlow, who became the designated lead investigator assigned to work with both the Bastrop Police and Sheriffs Departments, decided to call in DPS Crime Lab members to process the scene.

The DPS crime-scene team arrived in Bastrop from Austin at approximately 5:15 p.m. Karen Blakley, who specialized in DNA and serology, was designated the team leader by her coworker, Wilson Young. Other members of the team, led by Blakley, included a trace analyst, a photographer, a latent-print examiner, and a trainee in serology and DNA. Detailing the condition of Stacey’s body, Blakley noted that Stacey was missing a shoe and that her white sock was clean, indicating that she had not likely walked on an outside surface. An H.E.B. name tag with the name “Stacey” written on it was found in the crook of Stacey’s leg, and a white T-shirt, which Fennell later identified as belonging to him, was strewn over some brush near Stacey’s body. Stacey was clothed in a black bra and a pair of blue pants with a broken zipper. Her visible green underwear was wet in the crotch and bunched around her hips. Viewing this as indicative of a sexual assault, Blak-ley tested for the presence of semen, and the initial test yielded a positive result. Blakley then collected additional swab samples from Stacey’s vagina and breasts. Because rigor mortis had set in, Blakley could not determine if Stacey had been anally sodomized. “She was already very stiff, and in order for me to try to get to the anal area I could possibly cause injury or further damage and make it look like she had suffered something that she didn’t.”

According to Blakley, it “looked like a great force had been applied [to Stacey’s neck] ... because it was like an indentation but red, like it had cut into her skin.” Blakley concluded that the injury was caused by a piece of webbed belt that was located near Stacey’s body on the side of the dirt road “[b]ecause it matched the pattern that was on [Stacey’s] neck.” And when the piece of belt with a buckle found near Fennell’s truck at the high school was brought to the scene, Blakley compared the two and concluded that they matched. Another criminalist on the team designated to search for trace evidence concurred with Blakley’s determination, concluding that the pieces matched. Going a step farther, he also concluded that the belt had been torn not cut.

Documenting other injuries to Stacey’s body, Blakley observed that there were scratches on her abdomen and arms, a burn from a cigarette on her arm, and shallow wounds on her wrists and back that looked like they were caused by fire-ant bites. Blakley also documented a large amount of mucus that ran from Stacey’s nose, down the side of her face, and into her ham.

Terry Sandifer, the latent-fingerprint examiner, collected two Busch beer cans that were located across the road from where Stacey’s body was discovered. When Sandifer processed the cans for fingerprints at the lab, she discovered no suitable fingerprints to analyze.

After processing the scene, Blakley returned to the lab that evening around 11:00 p.m. so that she could look at the substance on the vaginal swabs under a microscope. She discovered intact sperm — sperm heads with the tails still attached — that, in her opinion, indicated that the sexual activity was recent. Her conclusion was based on a published study finding that “26 hours is about the outside length of time that tails will remain on a sperm head inside the vaginal tract of a female.” She immediately reported her finding to Ranger Wardlow. Ranger Wardlow viewed the presence of semen as a “smoking gun,” surmising that the evidence of sexual assault gave the perpetrator a motive to kill. Ranger Wardlow theorized that identifying the man who left the semen would lead to the discovery of Stacey’s killer.

Dr. Robert Bayardo, the Travis County Medical Examiner, conducted an autopsy on Stacey’s body the following afternoon at 1:50. He estimated that Stacey died on the 23rd of April at 3:00 a.m., give or take a few hours, based on changes that occur in the body after death. Dr. Bayardo noted that Stacey had pre- and post-mor-tem injuries. He differentiated between the two based on the absence of bleeding; once the heart stops beating, there is no more bleeding and no more bruising. The burn, which Blakley believed was caused by a cigarette, occurred after Stacey died, as did several scratches, in Bayardo’s opinion. Although Stacey’s skull showed no outward signs of injury, when Dr. Bayardo looked inside the skull, he documented multiple bruises that “had the appearance of injuries sustained by being struck on the head with the finger knuckles with a closed hand.” Comparing the injury pattern on Stacey’s neck with the pieces of webbed belt collected by authorities, Dr. Bayardo concluded that the belt was the murder weapon and that Stacey died as a result of asphyxiation caused by strangulation. He estimated that asphyxiation takes approximately three to four minutes and that a person becomes unconscious within one to two minutes.

Because of evidence indicating sexual assault, Dr. Bayardo took vaginal swabs. Viewing the swabs under a microscope, he observed the presence of sperm with both heads and tails. This, according to Dr. Bayardo, indicated that the sperm had been introduced into Stacey’s vagina “quite recently.” Continuing the sexual-assault exam, Dr. Bayardo took rectal swabs. Viewed under a microscope, he identified several sperm heads without any visible tails, which led him to report the result of the test as negative. Sperm, according Dr. Bayardo, breaks down much faster in the rectum than it does in the vagina because of the presence of other bacteria in the rectum. When conducting a visual exam of Stacey’s rectal area, Dr. Bayardo noticed that her anus was dilated and that there were some superficial lacerations on the posterior margin. In his opinion, this was consistent with penile penetration, even though he did not entirely rule out the possibility that the presence of sperm in the anus was the result of seepage from the vagina. Utilizing his education and experience about determining whether a particular injury occurred before or after death, Dr. Bayardo concluded that Stacey sustained the injury to her anus at or around the time of her death and that the penetration was therefore not consensual.

Because Blakley had prior commitments, Young took over the serological duties on the 24th. Young conducted two types of Polymerase Chain Reaction (PCR) DNA testing, DQ-Alpha and D1SS0, on Stacey’s blood, the vaginal swabs taken by Blakley and Dr. Bayardo, and the substance found on the crotch of Stacey’s underwear. Young conducted only one type of PCR DNA testing, DQ-Alpha testing, on the anal swabs taken by Dr. Bayardo because the quantity of sample was limited.

Every person receives one DQ-Alpha allele and one D1S80 allele from each parent; therefore, every person possesses two DQ-Alpha alleles and two D1S80 alleles. Stacey’s blood possessed the DQ-Alpha alleles of 1.2 and 4 and the D1S80 allele of 24, which meant that each of her parents contributed a 24 D1S80 allele to her genetic makeup. On the male portion of the vaginal swabs taken by Dr. Bayardo, the results showed DQ-Alpha alleles 1.2, 3, and 4 and D1S80 alleles of 22 and 24. The presence of three DQ-Alpha alleles, according to Young, is a common occurrence when there is carryover of DNA from either of the two donors that cannot be entirely eliminated during the testing process and does not affect the validity of the results. The 22 D1S80 allele was foreign to Stacey. Regarding the vaginal swab taken by Blakley, the male portion showed DQ-Alpha alleles of 1.2 and 3 and D1S80 alleles of 22 and 24. This signified no carryover from Stacey and indicated that the semen donor possessed the DQ-Alpha alleles of 1.2 and 3 and the D1S80 alleles of 22 and 24. Testing on the male portion from the rectal swabs indicated the presence of DQ-Alpha alleles 1.2, 3, and 4. While there was carryover, the 3 DQ-Alpha allele was foreign to Stacey. Testing of the male potion of DNA from the crotch of Stacey’s underwear showed the presence of DQ-Alpha alleles 1.2 and 3 and D1S80 alleles 22 and 24, indicating the absence of any carryover. Finally, testing on the swabs from Stacey’s breasts showed the presence of DQ-Alpha alleles 1.2, 3, and 4 and D1S80 alleles of 22 and 24. The 3 DQ-Alpha allele and the 22 D1S80 allele were foreign to Stacey, even though there was carryover. Given the results, Young concluded that there was a single semen donor.

Young also participated in processing the truck on the 25th, accompanied by Sandifer, the latent-print examiner, and Ranger Wardlow. Blakley joined them the next day when she returned to work. In processing the truck and the carbon copies of Fennell’s checks found outside the truck for prints, Sandifer did not discover anything remarkable. Sandifer could find only a few items with suitable prints. When she examined the prints, she was either unable to make a match or identified the prints as belonging to either Stacey or Fennell. Young focused on looking for the presence of blood or semen but discovered none. And although Young collected other items, including a portion of the saliva or mucus substance that Fen-nell previously noticed on the carpet over the transmission hump, he did not discover anything significant that would help in identifying the perpetrator. Blakley, having observed Stacey’s body, noted that the substance on the transmission hump looked similar to the mucus that had flowed out of Stacey’s nose.

Young, Ranger Wardlow, and Blakley all took note of the reclined position of the driver’s seat and that the driver’s seatbelt was fastened. Ranger Wardlow specifically noted that the lap portion of the belt looked like someone sat on it because it was in a downward bow. The three then tested whether it was possible to pull a person from the vehicle while the seatbelt was fastened. Putting Blakley, who was similar in height and weight to Stacey, in the driver’s seat with and without the lap belt on, Ranger Wardlow and Young took turns pulling her from the vehicle by either the feet or the shoulders. In each instance, Ranger Wardlow and Young were able to remove Blakley from the truck. Further, when Young, who was six-foot-two, sat in the reclined driver’s seat, he noticed that he had a clear view out of the back window of the truck in the rear-view mirror. When DPS completed processing the truck, it was returned to Fen-nell. Fennell immediately transported it to the dealership and traded it in.

Over the course of the next eleven months, authorities focused then- investigation on people that Stacey knew, and with a $50,000 reward offered by H.E.B., numerous leads and information poured in. For instance, a newspaper-delivery person reported that Stacey’s body was not on Bluebonnet Drive when he drove by the site where her body was found at 4:00 a.m. In all, officials interviewed hundreds of people, including former classmates, boyfriends, and coworkers, as well as Stacey’s friends and coworkers at H.E.B. Over twenty-eight male suspects were identified, some immediately and some during the ensuing investigation. Each suspect was asked to consent to give blood, hail', and saliva samples. With the exception of one, Brian Haynes, all of the suspects offered their consent and provided the samples. Although Haynes refused to consent, he was compelled to provide samples after authorities obtained a search warrant. Authorities also requested and obtained samples from Officer Hall. Because of his friendship with Fen-nell, Officer Hall was viewed as a suspect. Upon request, he voluntarily provided samples.

Hall, who lived approximately one block away from Fennell’s apartment, had an alibi — that he was home with his wife, Carla Hall, when Stacey disappeared. When investigating Officer Hall, Ranger Wardlow found no evidence refuting Officer Hall’s alibi. The alibi, coupled with DNA testing excluding Officer Hall, led Ranger Wardlow to conclude that Officer Hall had not been involved in Stacey’s death.

As the last known person to see Stacey alive, Fennell was deemed a suspect from the outset. Despite this, authorities never made an effort to search Fennell’s apartment. Fennell, however, was vigorously interrogated on several occasions by Ranger Wardlow, who was, at various times, joined by Investigators Selmala, Barton, or Board. Fennell also voluntarily provided authorities with a blood sample, and even though DNA testing excluded him as the donor of the semen, authorities tried to make a case against him anyway. Ruling out the possibility that Fennell used Carol’s Ford Tempo during the commission of the offense because Fennell had to retrieve the keys from Carol on the morning of the 23rd before he went looking for Stacey, Ranger Wardlow investigated alternative methods of transportation that Fennell could have used. Toward that end, Ranger Wardlow examined taxi records and the vehicle mileage on all of the cars belonging to the Giddings Police Department. This investigation revealed nothing, and officials believed that Fennell could not have walked the thirty-five miles from Bastrop to Giddings between 3:00 a.m. and 6:45 a.m. Authorities also canvassed Fennell’s apartment complex, looking for anyone that could shed some light on anything relating to Stacey or Fennell on the morning of the 23rd. No one reported being awake and about that morning. Finding no evidence to support Fennell’s involvement in the crime, authorities eventually eliminated him as a suspect.

David Lawhon, Brian Haynes’s brother, emerged as a viable suspect shortly after Stacey was killed when authorities discovered that he murdered a woman named Mary Ann Arldt in Elgin. Arldt was murdered by Lawhon a few weeks after Stacey was killed, and officials learned that La-whon had bragged about killing Stacey. Because the two cases bore some similarities, authorities homed in on Lawhon in investigating Stacey’s case. A few people informed authorities that there had been a relationship between Lawhon and Stacey, but authorities were unable to confirm any connection between the two. Indeed, a mutual friend never had any indication from either Lawhon or Stacey that they knew one another. Like Fennell, Lawhon was excluded as the donor of the semen through DNA analysis and was later eliminated as a suspect.

Investigator Selmala also became a suspect in August 1996 after he committed suicide in his home. Ranger Wardlow investigated his death. A note written by Investigator Selmala’s girlfriend was found by his body. The note revealed that he was distraught over his relationship with his girlfriend. Taking into account his knowledge about Investigator Selmala, which included the note and the investigation into Stacey’s death, Ranger Wardlow found no reason to conclude that Investigator Selmala had any involvement in Stacey’s death. Indeed, the investigation into Stacey’s death revealed no connection between Investigator Selmala and Fennell or Investigator Selmala and Officer Hall. The only common thread between Investigator Selmala and the other two was that all three were law-enforcement officers. Nevertheless, Ranger Wardlow directed that a blood sample be drawn from Selma-la during Selmala’s autopsy and submitted to DPS for DNA testing. Ranger Ward-low made this decision anticipating that someone might try to link Investigator Selmala’s suicide to Stacey’s murder. If such an allegation ever arose, Ranger Wardlow would then be able to give an answer — DNA testing cleared Investigator Selmala as a suspect.

All of the other potential suspects that were investigated were excluded as a result of DNA testing.

Eventually, officials received information that led them to look into Reed, an African-American who was approximately the same height as Young, as a suspect. Throughout their investigation, officials found nothing that indicated that Stacey knew Reed. Reed lived in the City of Bastrop on Martin Luther King Drive near the railroad tracks. Several of Reed’s family members and friends, as well as his girlfriend, lived nearby. Bas-trop High School is also located near the railroad tracks, about sixth-tenths of a mile from Reed’s house. The location of Reed’s home was significant to authorities because Fennell’s truck was found nearby at the Bastrop High School. Authorities had, early on in the investigation, theorized that the location was convenient for the perpetrator.

Reed was frequently seen by Bastrop patrol officers walking in the area near his home late at night. When he worked the night shift in 1995 through the early part of 1997, Officer Michael Bowen would see Reed almost every night between 9:00 p.m. to 3:00 a.m. or 4:00 a.m. When Officer Bowen saw Reed, Reed was usually at Long’s Star Mart, located near Reed’s house on Loop 150/Chestnut Street and Haysel Street. Bowen also saw Reed walking along the railroad tracks on more than one occasion. Officer Steven Spencer reported seeing Reed in the early morning hours walking near Long’s Star Mart and the All Star Grocery, which was located at Loop 150/Chestnut and Pecan Street.

Officials contacted DPS to inquire about whether Reed had a DNA sample on file with the state database, which includes compiled DNA from convicted sexual offenders. When they learned that there was a sample, they requested a comparison between Reed’s DNA and the DNA from the vaginal swab taken by Blakley. Michelle Lockhoof, a specialist in DNA and serology with DPS, conducted DQ-Alpha and D1S80 PCR testing on the two samples. Reed’s DQ-Alpha alleles were identified as 1.2 and 3 and his D1S80 alleles were identified as 22 and 24. When compared with the sample taken from Stacey, Reed could not be excluded as the donor of the semen. In Young’s opinion, 99.8% of the Caucasian population, 99.8% of the African-American population, and 99.92% of the Hispanic population would be excluded as the donor of the semen.

Investigator Board interviewed Reed after learning that the preliminary DNA results could not exclude him as a suspect. Investigator Board withheld the results of the DNA testing and Mirandized Reed. Reed waived his rights and gave a written statement. In it, he stated, “I don’t know Stacey Stites, never seen her other than what was on the news. The only thing that I do know is what was said on the news is that she was murdered.” Pursuant to a search warrant, blood was drawn from Reed and turned over to the DPS lab.

Lockhoff subjected the sample to a more discriminating type of DNA testing, Restriction Fragment Length Polymorphism (RFLP). Once again, Reed could not be excluded as the donor of the semen when four individual sites were tested. Regarding the statistical frequency in which Reed’s RFLP profile would appear in the population, Lockhoff calculated that it would be one in 590 million for the Caucasian population, one in 330 million for the African-American population, and one in 3 billion for the Hispanic population. Combining the results of the PCR and RFLP testing, the frequency in which Reed’s genetic profile would be present in the world’s population is one in 5.5 billion for the Caucasian, African-American, and Hispanic populations.

Reed’s father and three brothers were then excluded as possible donors through DQ-Alpha and D1S80 DNA testing.

Because the testing conducted by DPS could not exclude Reed, DPS sought the assistance of LabCorp, an independent lab, to conduct additional testing. Meghan Clement, the director for the forensic-identity-testing department, received DNA samples from Stacey and Reed and conducted PCR testing, which included testing on genetic sites of the DNA strand that are distinct from those considered during DQ-Alpha and D1S80 testing. Looking at ten different sites on the male ¿•action of the substance on the vaginal swab taken from Stacey, Clement could not exclude Reed as the contributor of the semen; in fact, the sample matched Reed’s genetic profile. The probability of randomly selecting an unrelated individual with this profile is approximately one in 449,000,000 for the Caucasian population, one in 46,800,000 for the African-American population, and one in greater than 5,500,-000,000 for the Hispanic population. Combining some of the additional PCR testing with the previous DQ-Alpha and D1S80 results, only one person in the world’s population would have this particular genetic profile. Testing on the male portion of the substance from the rectal swab revealed DQ-Alpha alleles of 1.2 and 3 and, therefore, matched Reed’s DQ-Alpha profile. Recalling her prior experience working on sexual-assault cases for ten-and-a-half years, Clement noted that she never found intact sperm more than twenty-four hours after commission of a vaginal-sexual assault and that sperm breaks down faster in the rectal area than in the vaginal vault.

Reed was charged with capital murder in May 1997. At trial, to raise reasonable doubt during the guilt phase, Reed mounted a two-prong challenge to the State’s evidence. First, Reed pointed to the possibility that another person, particularly Fennell and Lawhon, had committed the offense. And as a secondary theory, Reed focused on showing that he had a romantic relationship with Stacey and that his semen was therefore present in Stacey’s body because of consensual intercourse.

To prove a romantic relationship between Stacey and Reed, Reed’s defense team called Iris Lindley, a longtime friend of Reed’s parents, to testify. In early 1996, Lindley was sitting on the porch of Reed’s house visiting with Reed’s mother. A young woman with brown hair pulled in front of the house in a gray truck, walked up to the porch, and asked if Reed was home. When Reed’s mother told the young woman that Reed was not home, the young woman asked Reed’s mother to tell Reed that “Stephanie” had come by. Clarifying the name, Lindley said that it was either “Stacey or Stephanie.” When Lind-ley was shown a picture of Stacey, she stated that Stacey looked like the young woman who had come by Reed’s house that day. While Lindley first testified that she formulated the impression that Stacey and Reed were dating, she conceded on cross-examination that she had no such knowledge.

To establish that Lawhon knew Stacey, Reed’s attorneys called Jose Coronado, who had worked with Lawhon at Walmart and with Stacey in the produce department at the H.E.B., to testify. Coronado stated that he once saw Stacey and La-whon talking in the Walmart parking lot and that later, when he and Stacey worked together at H.E.B., Stacey told him that she and Lawhon had dated and that La-whon was “sort of a player.” On cross-examination, the State asked Coronado whether it would surprise him to know that Lawhon was dating a woman named Christie Macy and that she would frequently meet him in the Walmart parking lot. Coronado stated that he did not know about Macy or that she met Lawhon in the parking lot.

Supporting Coronado’s testimony, Cynthia Jones, a friend of Lawhon’s, testified that she and her boyfriend were with La-whon and Stacey at a party in Elgin and then again at Smithville Jamboree in 1995. Jones said that Lawhon introduced Stacey as “his girl” for the first time at the Jamboree.

Scott Parnell furthered the defense’s strategy to implicate Lawhon when he testified that Lawhon confessed to killing Stacey. While drinking at a bar one night in 1996, Lawhon told Parnell that he strangled Stacey with either his or her belt and that Stacey had pretty blue eyes before she closed them. On cross-examination, the prosecution questioned Parnell about a signed written statement that Parnell made at the Sheriffs Department in which Parnell stated that Brian Haynes made the confession. Explaining the evident discrepancy, Parnell testified that both La-whon and Haynes had confessed. Additionally, when the prosecution inquired about the motive behind his testimony, Parnell admitted that he knew about the $50,000 reward offered by H.E.B.

To rebut the evidence supporting any relationship between Stacey and Lawhon, the State called two of Stacey’s best friends from high school to testify. Cathy Vacek went to the Jamboree with Stacey in 1995 and stated that she would have known if Stacey dated Lawhon and had gone with him to the Jamboree. Sherry Lastovica went to the Jamboree with Stacey on Friday night in 1995 and stated that after Stacey attended the Jamboree for a second time the following day, Stacey told Lastovica that she had met Fennell. Neither woman knew anything about a relationship between Stacey and Lawhon.

The State also offered testimony from Lawhon’s wife. She specifically remembered the night that her husband murdered Arldt. On that night, when Lawhon failed to come home, she locked the screen door, which did not have a key, so that she would know when he got home. When he finally returned home, the two then argued about it. She recalled that the argument ensued because it was unusual for him to come home so late. When asked whether anything like that happened on April 23rd, Lawhon’s wife remembered the day because it was her son’s first birthday, and she stated that nothing unusual happened.

Turning their attention to Fennell, Reed’s defense team devoted a considerable amount of time highlighting the shortcomings of the investigation into Fennell by officials. Specifically, they were able to call the jury’s attention to the fact that the lion’s share of information provided to officials about Stacey’s whereabouts before she died, Stacey’s routine and habits, and the items in Fennell’s truck was given by Fennell himself. They also emphasized that officials did not search Fennell’s home, thereby precluding the possibility of ever discovering evidence that may have implicated Fennell.

Tami Renee Hannath, Stacey’s high-school friend, cast Fennell as controlling and possessive. She testified that when she and Stacey were on the phone, making arrangements for Stacey to come to Smith-ville for a visit, Fennell came home. Stacey then told him about the upcoming plans while Hannath remained on the phone and then the phone was disconnected.

Finally, Reed’s defense team presented its own DNA expert, Dr. Elizabeth Ann Johnson from Technical Associates Incorporated. Dr. Johnson’s DQ-Alpha and D1S80 DNA test results on the vaginal swabs taken by Blakley and the fluid found in Stacey’s underwear were consistent with those obtained by DPS. And although Dr. Johnson attempted to test the rectal swab, she determined that there was not enough DNA to conduct accurate testing. Dr. Johnson’s DQ-Alpha testing on the saliva from breast swabs taken by Blakley yielded the same results as the previous testing conducted by DPS. On the swab taken from Stacey’s left breast, testing indicated 1.2, 4.1, and 3 alleles, and on the swab taken from Stacey right breast, testing indicated 1.2, 3, and 4.1 alleles. Dr. Johnson conceded that in all of the sixteen sites tested in this case, Reed could not be excluded as the donor of the semen and saliva found on Stacey’s body. Further, Dr. Johnson did not dispute the statistics that Lockhoff devised as a result of her testing.

To quell the prosecution’s theory that Stacey had been anally sodomized before her death, Dr. Johnson was questioned about vaginal drainage. Dr. Johnson testified that vaginal drainage, which allows semen to be deposited in surrounding areas, may occur when a body is moved around after intercourse. She opined that when there has been an ejaculation in the rectal area, there should be a lot of sperm because a full ejaculate contains hundreds of millions of sperm. And regarding the decomposition of sperm, Dr. Johnson stated that she was unaware of any difference in the rate of decomposition of sperm in the vagina versus that in the rectum. In her experience, she obtained better sperm samples from rectal swabs. On cross-examination, Dr. Johnson admitted that a male can deposit a small amount of sperm without ejaculating when there is penetration and that trauma to the anal area should be considered when determining whether there has been penetration.

After weighing the evidence, a jury found Reed guilty of capital murder. And following a separate punishment hearing, Reed was sentenced to death.

II. Post-trial Background

A. Reed’s Direct Appeal

Reed appealed, claiming, among other things, that the evidence was factually insufficient to support his conviction for capital murder.¹ We rejected Reed’s sufficiency claim, holding, “Given the strength of the DNA evidence connecting [Reed] to the sexual assault on [Stacey] and the forensic evidence indicating that the person who sexually assaulted [Stacey] was the person who killed her, a reasonable jury could find that [Reed] is guilty of the offense of capital murder.”² And concluding that Reed’s other claims were without merit, we affirmed Reed’s conviction and sentence.³

B. Reed’s First and Second State Applications for a Writ of Habeas Corpus

Reed also sought habeas relief under Article 11.071, Texas Code of Criminal Procedure. Regarding Reed’s original application, based on the trial judge’s recommended findings and conclusions and our own review of the record, we denied relief in a written order.⁴ While Reed’s original application was pending, Reed filed a supplemental claim for relief, which we later construed as a subsequent application under Section 5, Article 11.071, Texas Code of Criminal Procedure.

Relying on Brady v. Maryland,⁵ Reed claimed in the subsequent application that the prosecution failed to give his defense attorneys a letter from Young dated May 13, 1998. The letter was addressed to the lead prosecutor, Lisa Tanner, an Assistant Attorney General whom Bastrop District Attorney Charles Penick had called in to prosecute the case. In the letter, Young acknowledged a request for DNA analysis on the beer cans recovered from the scene where Stacey’s body was found and a request for a comparison of the results to samples of Stacey’s DNA as well as samples from other individuals that had been submitted throughout the course of the investigation. Young subjected the samples to DQ-Alpha DNA testing and documented the results. Testing on one of the cans, identified by officials as item number 24, revealed the presence of DQ-Alpha alleles 1.3 and 4. A possible 1.2 DQ-Alpha allele was potentially masked but was not specifically detected. Testing on the other can yielded no DQ-Alpha results. Based on the results, Young concluded that Reed was excluded as a possible source of the DNA. Young, however, could not exclude Stacey if the source of DNA compromised a mixture of DNA but could exclude her as a donor if the DNA was provided by single source. Officer Hall and Investigator Sel-mala could not be excluded as possible sources of the DNA. According to Reed, the State failed to make Young’s letter available to him until the State attached it as an exhibit to its response to the allegations raised in his original habeas application.

The trial judge held a live evidentiary hearing on this claim and, after evaluating its merits, recommended that we deny relief. Testimony from the hearing supplied additional insight into the DNA testing conducted on the beer cans.

On May 13th, when Young documented the results from the DQ-Alpha DNA testing on the beer cans, the guilt phase of the trial was underway and the defense was in the process of presenting its case-in-chief. On that particular day, the court was in recess because Dr. Johnson was the defense’s next witness and she was not available to testify until the following day. Initially, the prosecution did not request that testing be conducted on the beer cans, having concluded that they were a non-issue. According to Ranger Wardlow, the cans, which had some pine needles on top of them and compressed needles below, appeared to have been there longer than Stacey’s body. Conversely, proceeding under the theory that everything should be tested, Reed’s defense team ordered testing on the cans. As a result, when Dr. Johnson went to the DPS lab and met with Young on April 15th, Young swabbed the lips and sides of the cans for saliva in Dr. Johnson’s presence and split the swabs with Dr. Johnson. Dr. Johnson later used her portion of the swabs to conduct DQ-Alpha and Polymarker testing. On May 5th, Reed’s defense attorneys requested that blood samples from the other suspects, including Officer Hall and Investigator Selmala, that had been collected by DPS, be made available to Dr. Johnson. The trial judge granted this request. Alerted to the defense’s decision to test the beer cans, Tanner requested that Young test the portions of the swabs that he had retained. When Dr. Johnson testified at trial, she did not testify about the results that she obtained from the DNA testing conducted on the beer cans.

Answering Reed’s allegation that she failed to disclose Young’s report to his defense team at the evidentiary hearing, Tanner began by testifying that she learned of Young’s test results on May 13th through Missy Wolfe, an investigator with the Attorney General’s Office, who was assigned to work on Reed’s case with Tanner. Because the trial court was in recess that day, Wolfe called Tanner at home and told her about the results. Young faxed the report on the 14th when the court was back in session, and Wolfe received the report and gave it to Tanner. Tanner stated that, upon receipt of the report, it would have been stamped and given to Reed’s defense attorneys. Prior to trial, the prosecution instituted a Bates stamping system; each page of each document subject to disclosure was assigned a sequential number. Four copies of each document were then made. One copy would be placed in the district clerk’s file, one would be retained by the prosecution in its discovery file, and the remaining two would be given to Reed’s defense attorneys. Young’s May 13th letter, which consisted of two pages, was numbered 3,183 and 3,184. Under the hectic conditions of the trial, the standard procedure began to break down and the prosecution dispensed with providing copies of discovery materials to the district clerk. Tanner stated that she believed she gave a copy of Young’s letter to Reed’s defense team because, when she reviewed her file, she found three stamped copies of the letter and the district clerk’s file did not contain a copy. Based on the usual policy and practice of disclosure in this case, Tanner was convinced that the fourth copy had been given to Reed’s defense team. Wolfe testified similarly. However, neither Tanner nor Wolfe had any independent recollection of specifically providing the report to Reed’s attorneys.

Tanner also testified that she considered Young’s results to be exculpatory when she first received them. Therefore, on May 13th, she directed Wolfe to have Young forward the DNA samples to Lab-Corp via FedEx for additional, more discriminating D1S80 and Polymarker DNA testing. However, Tanner cancelled the testing the next day when she reviewed Dr. Johnson’s report and notes during the lunch break before Dr. Johnson was set to testify. Dr. Johnson’s DQ-Alpha testing yielded the same results as Young’s. But through Polymarker' testing, Dr. Johnson excluded Stacey, Officer Hall, and Investigator Selmala as contributors. Still firm in her belief that she had given Young’s report to the defense, Tanner stated that the exculpatory value of Young’s report was negated when she learned about Dr. Johnson’s exclusion of Stacey, Officer Hall, and Investigator Selmala. At that point, Tanner believed that the issue with the beer cans “had been put to bed” and directed Wolfe to cancel the additional testing with LabCorp.

Calvin Garvie and Lydia Clay-Jackson, Reed’s trial attorneys, testified that they had not seen Young’s May 13th report until Reed’s habeas counsel gave them a copy. Both attorneys also recalled that the prosecution gave each of them their own copies of all discovery materials before and during the trial. Garvie, who took the responsibility of dealing with the DNA evidence in the case, stated that he remembered someone was excluded and was certain that, if Dr. Johnson’s results had included any of the other suspects, he would have had Dr. Johnson testify to that fact. Game further stated that, had he received Young’s report, it could have affected the jury’s verdict because when there is evidence from the State suggesting innocence and showing the presence of other individuals at the scene, it is “huge.” Clay-Jackson agreed, stating it would have helped advance them theory of the case by giving an explanation of how Fen-nell could have traveled to Giddings from Bastrop and back to Giddings. When asked whether he would have used Young’s report, Garvie stated that he would have used it to consult with Dr. Johnson. Garvie further stated that he would hesitate in using the report because of Dr. Johnson’s exclusion. Clay-Jackson expressed a similar sentiment, stating that Young’s report would not have given her a reason to “exhale” during the trial if she would have known that Dr. Johnson’s testing refuted Young’s results.

Regarding Officer Hall, Carla Hall testified, verifying her husband’s alibi. She stated that, in the early morning hours on April 23rd, her husband was at home with her. She remembered that night because her two-month-old daughter woke up with a “bloodcurdling scream” at 3:30 a.m. While her husband held them baby, she went to fix a bottle. Her husband then left for work at 5:35 a.m. because he was scheduled to be on duty at 5:45. Officer Hall testified consistently with his wife. And when asked if he had any involvement with Stacey’s death, Officer Hall stated that he did not.

Recommending that Reed’s Brady claim be denied, the trial judge adopted the State’s proposed findings of fact and conclusions of law. In doing so, the trial judge entered findings of fact consistent with the testimony given at the hearing and found Tanner, Garvie, Clay-Jackson, and Officer Hall and his wife, Carla, to be credible. The trial judge also adopted the following conclusions of law:

• The State did not intentionally suppress the May 13,1998 DPS lab report in violation of Brady v. Maryland.

• The State provided the May 13, 1998 DPS lab report to Applicant’s attorneys.'

• There remains a legitimate fact issue as to whether Applicant’s trial counsel actually received a copy of the May 13, 1998 DPS lab report during Applicant’s trial.

• If the May 13, 1998 DPS lab report was disclosed and used by the defense effectively, it would not have made a difference between conviction and acquittal, since the defense’s own expert has already reached the same conclusion as that reflected in the report.

• Because the DNA results reflected in the May 13, 1998 DPS lab report were previously refuted by Applicant’s own expert, they are not material because they do not create a probability sufficient to undermine the confidence in the outcome of Applicant’s trial.

Considering Reed’s Brady claim as raised in a subsequent application for a writ of habeas corpus, we held that Reed failed to show that his claim meets any of the exceptions outlined in Article 11.071, Texas Code of Criminal Procedure.⁶ As a result, we dismissed Reed’s subsequent application as an abuse of the writ.⁷

C. Reed’s Federal Petition for a Writ of Habeas Corpus

Reed then sought federal habeas corpus relief under Title 28, United States Code, Section 2254. Although the magistrate judge permitted discovery and ordered several depositions, the United States District Court for the Western District of Texas determined that several of Reed’s claims were unexhausted because Reed had failed to present them to this Court before pursuing federal habeas corpus relief.⁸ As a result, the District Court entered a stay in March 2004 allowing Reed to exhaust his state-court remedies.⁹

D. Reed’s Second Subsequent State Application for a Writ of Habeas Corpus

In March 2005, Reed filed a second subsequent state application for a writ of ha-beas corpus under Article 11.071, Texas Code of Criminal Procedure. In it, Reed claimed, among other things, that he is actually innocent under Herrera v. Collins¹⁰ and that the State suppressed exculpatory evidence in violation of Brady. Contending that the State violated Brady, Reed maintained that the State suppressed the following evidence:

• DNA evidence linking the beer cans found near Stacey’s body to Officer Hall.

• Eyewitness information from Martha Barnett that she had seen Stacey and Fennell the morning that Stacey was murdered.

• Reports from family members Jennifer and Brenda Prater that Stacey had been seen early in the morning on April 23rd with a man who was not Reed and who had a dark complexion.

• Longstanding information that Fennell and other Giddings officers engaged in a pattern of brutality against suspects.

Reed also claimed that the State suppressed information from Mary Blackwell, formerly known as Mary Best. Blackwell is a former classmate of Fennell’s at CAPCO, and she states that she overheard a conversation in which Fennell stated that he would strangle his girlfriend with a belt if he ever caught her cheating on him. While this claim was originally filed under seal, it was made a matter of public record at the evidentiary hearing when Blackwell testified in open court.

In October 2005, we determined that Reed’s Brady claims concerning Barnett and Blackwell satisfied the requirements of Article 11.071, Section 5(a) and remanded the claims to the trial judge for a live evidentiary hearing and ordered the trial judge to enter findings of fact and conclusions of law. With respect to Reed’s remaining grounds for relief, we held that Reed failed to satisfy Section 5(a) and dismissed those claims as an abuse of the writ.

We now turn to the details of Reed’s Brady claims concerning Barnett and Blackwell.

1. Barnett

To support his claim concerning the nondisclosure of eyewitness information from Barnett, Reed attached an affidavit from Barnett.

On the morning of April 23rd, 1996 at approximately 5:00 to 5:30 AM I was on my way to work. I pulled into the parking lot of the Old Frontier. At that time I saw Stacy [sic] Stites and a man I recognized as Jimmie [sic] Fennell standing in front of a red pickup on the side walk. I got out of my vehicle and approached the soda machine. I got my coke, turned and got into my vehicle. There was a 4 door car leaving the parking lot as I turned in. I presumed it was the newspaper deliveries’ [sic] people because the newspaper rack was full. I recognized Stacy [sic] because I always went thru her line at H.E.B. I worked at a restaurant in front of H.E.B. I found out about 2 weeks later that the man with her that morning in front of the Frontier was Jimmie [sic] Fennell because his picture was run in the Giddings Times and News and that’s when I recognized him.

Reed also attached affidavits from Barnett’s attorney, Steven Keng. Keng was formerly the Lee County Attorney, a county adjacent to Bastrop County. Keng stated that Barnett told him about seeing Stacey and Fennell at Old Frontier in Paige, Texas, a town between Giddings and Bastrop, on the morning of April 23rd. Barnett relayed this information to Keng sometime in late 1997 or early 1998 when Keng was representing her in Lee County. Keng felt that the information was important because of newspaper reports stating that Fennell was excluded as a suspect because officials could not explain how he committed the crime. When Keng was at the Bastrop County Courthouse a few weeks later, he approached District Attorney Penick and relayed Barnett’s disclosure without specifically identifying Barnett by name, referring to her only as a client. Keng was under the impression that Reed had not yet been tried and approached District Attorney Penick with the information, knowing that a prosecutor has a duty to explore all of the evidence and to see that justice is done. Keng was surprised by Penick’s reaction to the disclosure.

He laughed and told me that he had all of the evidence that he needed, and he did not want to hear anymore about the case. He did not indicate that the case was over, and a conviction secured, (which I would have expected if the case had already been tried), only that he did not need anymore evidence.

When Keng returned to his office, he told his wife, who assisted him at the office, of Penick’s response.

During 2001 and 2002, Smithville newspapers reported that the Bastrop District Attorney’s Office had engaged in prosecu-torial misconduct in Reed’s case. Believing that the allegations of misconduct were defamatory, Penick filed a civil suit against the papers.¹¹ When pressed during a deposition taken as part of the civil suit in August 2001, Penick stated that he remembered Keng approaching him at the Bastrop County Law Enforcement Center and stating that he had a client who knew something about Reed’s case. Penick recalled telling Keng that he had all of the information that he needed. Penick believed that Keng was making a joke because Keng never stated anything about having exculpatory evidence.

In October 2003, Penick, who was by then retired, elaborated on his conversation with Keng during a deposition ordered by the federal magistrate judge. Penick was certain that the conversation took place after Reed was convicted and sentenced. He asserted that Keng approached him in January or October of 2002 while Keng was at the Law Enforcement Center during one of the arraignments on a murder case involving Amanda Sykes. Penick reached this conclusion when he called up the District Attorney’s Office following the deposition in the civil case and requested that they pull the dates involving Sykes’s case. Penick stated that his response to Keng was likely prompted by the news articles; he was “ill-tempered” at the time and perceived Keng’s statement as a “jab” at him. Penick further claimed that in the twenty years he has known Keng, he has never known when to take him seriously and stated that Keng “didn’t seem to sound serious about this.”

In response to Penick’s claim that Keng’s disclosure took place long after Reed’s trial, Keng, in his affidavit, steadfastly maintained that the conversation took place prior to Reed’s trial. Keng’s awareness that Reed had yet to be tried prompted him to believe that the information would be important to the State. Keng reviewed his appointment book to identify the dates that he would have been in the Bastrop County representing clients. His review showed that he had been in Bastrop between March 1998 and April 1998. Keng also recalled that the conversation did not take place during any of the pretrial proceedings held on the Sykes case. Keng was dealing with an assistant district attorney on that case and was informed that Penick was on vacation during at least one of the pretrial settings. Keng claimed that Penick did not participate in the trial of Sykes.

2. Blackwell

In support of his Brady claim concerning the State’s failure to disclose Fennell’s statement in which he threatened to strangle Stacey if he ever caught her cheating on him, Reed attached an affidavit from Blackwell. In the affidavit, Blackwell states that she is a licensed Texas peace officer and that she attended a training class at' CAPCO with Fennell in 1995. During the class, Fennell sat behind her with some of his friends. Continuing, Blackwell stated:

I also knew who Jimmy’s girlfriend was. One day after training class, I met a woman in the parking lot who asked for Jimmy. I told her he was inside and volunteered to get him. As I went in, Jimmy met me coming out of the building. Jimmy looked at us and said, “What are you telling my girlfriend? Keep away from her.” Earlier that day, Jimmy and others in our class were learning self defense tactics. Jimmy’s friend had broken my hand during one of the exercises. After Jimmy passed me in the parking lot I saw him go up to his girlfriend and could hear him telling her in a commanding voice what to do.

The men from Bastrop that were taking the CAPCO class would talk about Jimmy’s girlfriend. They said she was nice, but that Jimmy talked down to her in an abusive way — in a demanding kind of way.

Towards the end of the CAPCO course, instructors had passed out photographs from real suicides and murders. Each student was supposed to say whether their group’s photograph depicted a suicide or murder. The class had to break because one of the students had a relative who had committed suicide and that relative’s suicide was depicted in one of the photographs.

During the break, I overheard Jimmy talking to this other guy in class. He said, “If I ever find my girlfriend cheating on me, I’ll strangle her.” I told him that if he did that he would be caught because he would leave fingerprints. Jimmy then said, “That just goes to show you’ll never know shit; I won’t leave any prints because I’ll use a belt.” I didn’t think much about Jimmy’s comments until I heard Captain John Vasquez discussing the murder of Jimmy’s girlfriend. Captain Vasquez was in my office and seemed to know a lot about the murder scene. He had worked as an investigator in connection with the murder. He told me the details from the murder scene and seemed to indicate that Jimmy’s girlfriend knew her attacker. I then told him the details of what Jimmy said.

John Vasquez, who had retired from the Austin Police Department before Reed’s trial after twenty-six years of service, documented Blackwell’s recollection of Fen-nell’s threat in an affidavit:

After I retired, I became a private investigator and investigated several homicides. In 1998, I was appointed to assist the defense team of Rodney Reed. I was not the original investigator on the case and was only able to do a limited amount of work shortly before his trial began. I continued my involvement with his case until his conviction and sentence was [sic] decided.

Sometime after Mr. Reed had been sentenced to death, I happened to speak to Travis County Deputy Constable Mary Best. We talked about the Stacey Stites case and she shared with me that she had been in a training class with Stites’ fiancé, Jimmy Fennell. Mary recalled that they were talking about domestic problems. Fennell made the remark that if he ever caught Stacey cheating on him he would choke her to death. He then laughed and said he was joking. After I learned this information, I checked the police training academy and confirmed that Mary Best and Jimmy Fennell had indeed been in a joint training session. I told the Bastrop District Attorney’s Office about the information that I learned. I believe that I spoke directly to District Attorney Charles Penick. I never heard anything more about it and I do not know what the District Attorney did with the information I gave them.

3. Live Evidentiary Hearing

At the direction of this Court, limited to the Brady issues concerning Barnett and Blackwell, the trial judge held a live evi-dentiary hearing in March 2006.

a. Barnett

Barnett expanded on the facts surrounding her sighting of Stacey and Fennell at the Old Frontier store on the morning of April 23, 1996. At that time, Barnett was working at Papa’s Catfish restaui-ant in Bastrop, located in front of the Bastrop H.E.B. She knew Stacey because she shopped at the H.E.B., and Stacey had checked her out.

Barnett lived in Paige and took Highway 290 to Highway 21 into Bastrop. Her commute took approximately twenty-five to thirty minutes. Although Barnett normally worked the 2:00 p.m. to 10:00 p.m. shift or 4:00 p.m. to 11:00 p.m. shift at the restaurant, on April 23rd, she was scheduled to cover a coworker’s shift at 6:00 a.m. Barnett woke up at 4:00 a.m., got her children ready for the day, and dropped them off at her mother’s house, located approximately thirty to forty-five seconds from her house. She stayed at her mother’s house for twenty to twenty-five minutes. Before heading to Bastrop, Barnett drove east on Highway 290 to the Old Frontier store. When she arrived at the store at approximately 4:45 a.m., she observed a man and a woman standing in front of the store making hand gestures that indicated to her that the man and woman were involved in some type of conflict. The man and the woman then got into a red pickup when Barnett opened her car door. After getting a soda from the machine, Barnett saw the head of the woman sitting in the passenger’s seat of the red truck go down and back up. When she got back into her car, Barnett heard elevated, muffled voices from the truck, even though the windows of the truck and her car were closed at the time. Barnett left the store while the truck was still parked in the lot and began her commute to work. Barnett did not recall whether she saw the red truck on her way into Bastrop, and she estimated that she arrived at work early at 5:30 a.m.

Believing that she saw Stacey on the day she was murdered, Barnett reported what she saw to her parents in January 1997. Her mother, Marjorie Cowan, advised her to talk to Keng. Cowan stated that Barnett had told her that she knew Stacey from the H.E.B. and that “the young man that was with her was very— looked like he was angry.” Cowan could not remember if Barnett identified the young man by name. However, she recalled urging her daughter to talk to Keng because she knew Keng’s father and had used Keng as a lawyer several times.

On direct-examination by Reed’s habeas counsel, Barnett testified that a year later, in January 1998, she met with Keng when he was representing her on a charge of driving while intoxicated (DWI) and told him about what she saw. Barnett stated that she realized sometime after the 23rd that the man with Stacey was Fennell when she saw his photograph in the paper.

On direct-examination, Barnett also revealed that she had prior misdemeanor convictions for theft by check.

On cross-examination, Barnett acknowledged that news of Stacey’s murder was a big deal and that she failed to report what she saw to law-enforcement officials, the Bastrop District Attorney, or the Attorney General’s Office, even though she was aware that authorities would have been interested in having the information. Confronted with her prior sworn statement, in which she claimed that she saw Stacey and Fennell at the Old Frontier store between 5:00 and 5:30 a.m. on the 23rd, Barnett was questioned about whether anyone had told her that Fennell’s truck had been located at the high school at 5:23 a.m., making it impossible for Barnett to have seen Fennell and Stacey with the red truck twenty to twenty-five minutes away at the store in Paige between 5:00 and 5:30 a.m. In response, Barnett stated that no one had talked to her about the time frame. The State continued to question Barnett about her time line:

[State] Q. So it would have taken you only 45 minutes to get four kids ready, get them out of the house, get them dressed, get them to somebody else’s house, visit with them for 20 to 30 minutes and then get to that store; is that your testimony?

A. I [sic] mother doesn’t live very far from where I lived at the time.

Q. And then you told us that it would have taken you about 30 minutes to get in to work from there; right?

A. About, yes.

Q. Okay. So you certainly wouldn’t have gotten to work at like 5 o’clock or 5:15 would you, because that would have been just crazy; right?

A. I don’t recall that it was that early.

Q. In fact, it was closer to six, wasn’t it?

A. No, I wasn’t that late.

When the State asked Barnett if she had omitted any information in her affidavit about Stacey and Fennell gesturing as if they were involved in a conflict, Barnett admitted that she had.

Also, raising an issue left unaddressed on direct examination, the State questioned Barnett about her arrest for DWI, which had occurred before Barnett told Keng that she had seen Stacey on the morning she was murdered. Fennell and Officer Hall arrested Barnett for DWI, and Fennell cited her with failing to maintain a proper lane on November 5, 1997. Fennell also executed a prior arrest warrant in a separate case in which Barnett was charged with theft. Barnett consulted Keng on the DWI ease and told him about seeing Stacey and Fennell two months later. Barnett stated that she had not been happy about the arrest and acknowledged that she had been aware that Fennell would have testified against her if the case had gone to trial.

On redirect, Barnett admitted that she was drunk when arrested and testified that her intoxication at the time of her arrest affected her ability to recognize Fennell. She further stated that she did not execute the affidavit to retaliate against Fennell; she made the connection between Fennell and Stacey when she saw Fennell’s picture and name in the paper.

Testing whether Barnett had actually recognized Fennell from a photograph in the Giddings newspaper, as she had claimed in her affidavit, the State presented Barnett with copies of the weekly paper issued from April 25, 1996, to May 28, 1998, and asked her to identify which paper included a picture of Fennell. After looking through the papers, Barnett conceded that Fennell’s picture did not appear in any of the papers. Barnett then stated that she must have seen it in another paper.

To further undermine Barnett’s testimony about recognizing Fennell from a photograph in the paper, the State called Emanuel Miranda, an investigator with the Postconviction Litigation Division at the Office of the Attorney General, to testify. Preparing for the hearing, Miranda was tasked with finding any articles in the Bastrop, Austin, and Giddings newspapers from April 23, 1996, the date Stacey was murdered and her body was found, to May 30, 1998, the day after Reed was sentenced, that were related to Stacey’s murder or Reed’s trial. Miranda was ordered to look for a picture of Fennell that appeared with any of the articles. Miranda’s research revealed that Fennell’s picture did not accompany any of the articles relating to Stacey’s murder or Reed’s trial from April 23, 1996, to May 30,1998.

Reed’s trial attorneys, Garvie and Clay-Jackson, both testified that they had not been aware of Barnett’s sighting of Fen-nell and Stacey on the morning of April 23rd before or during trial. Both attorneys agreed that, had such information been disclosed, it would have significantly altered their trial strategy. Because Fen-nell was the only source to verify his whereabouts on April 23rd, Garvie states that he would have used Barnett’s sighting to establish reasonable doubt. First, Fen-nell’s testimony that he had been at home sleeping when Stacey left for work would have been impeached, exposing Fennell as a liar. Next, Garvie believed that the defense could have challenged the State’s time line by showing that Stacey was alive between 4:30 and 5:30 a.m. on the morning of the 23rd. Also, in Game’s opinion, with Fennell in Paige, the distance that Fennell would have had to travel to get back to Giddings would not have been as great as that theorized during the trial. Clay-Jackson testified that, if she had known Fennell had been in Paige, she would have investigated Fennell’s history for violence and his associates more closely to determine if someone else drove Fennell from the Bastrop High School to Giddings. Garvie testified that he would have taken another look into Dr. Bayardo’s time of death, and Clay-Jackson stated that she would have had them medical experts explore the forensic evidence for anything that could place Fennell at the scene of the offense. With Barnett’s sighting, Garvie believed that Fennell’s exclusion as a suspect through DNA would not have made a difference because Fennell had no motive to rape Stacey if he was just going to kill her. Clay-Jackson stated that, had she known of the information in Barnett’s affidavit, she would have encouraged Reed to testify in his defense. However, on cross-examination, Clay-Jackson stated that she previously admitted that she did not want Reed to testify because he had several prior sexual-assault offenses that the State could have used against him.

Parroting his prior statements, Keng testified about the facts surrounding Barnett’s disclosure, his subsequent attempt to inform Penick of the disclosure before Reed’s trial, and Penick’s reaction. He recalled that Barnett had come to see him about the DWI and a family-law matter and that he did not end up representing her on the DWI. Regarding his attempt to inform Penick of Barnett’s disclosure, Keng added that he told Penick that he could give him his client’s name and telephone number. He chose to tell Penick about Barnett’s disclosure in person instead of calling him because the information was important, he knew Penick, and Penick was in charge of the prosecution. Keng stated that he failed to convey the information to anyone else in the District Attorney’s Office or any law-enforcement officials, even after he was rebuffed by Penick. Keng testified that Barnett never gave him the impression that she had told anyone else about the sighting and that, if he thought that Barnett was lying, he would not have given the information to Penick.

With the timing of Keng’s disclosure to Penick as a hotly contested issue, both parties attempted to nail down exactly when Keng spoke to Penick. Confident that he spoke to Penick before Reed’s trial, Keng stated that he informed Penick about Barnett’s sighting in February or March of 1998. On cross-examination, the State challenged whether Keng spoke to Penick before Reed’s trial by noting that Keng had previously stated in an affidavit that, according to his schedule, he had been at the Bastrop County Courthouse representing other clients in March and April 1998, immediately before, and during, Reed’s trial. When cross-examining Keng about his prior statement in which he stated that he recognized the importance .of the information provided by Barnett because newspapers reported that Fennell had been eliminated as a suspect, the State presented Keng with the first news articles from the Austin and Gid-dings papers to report this particular information from mid-May 1998. The State then asked Keng whether it is possible that he did not disclose Barnett’s sighting until after Reed’s trial. In response, Keng stated that he believes that his statement suggests that memory is cumulative in people and that when he wrote the affidavit, the information about Fennell having been excluded as a suspect was published. He added that Penick never told him that Reed’s trial was over and that he believed if it had been over, Penick would have told him that.

Keng was also questioned about his participation in the film “State versus Reed.” While a graduate student at the University of Texas, Ryan Polomski made the film for his thesis project. He interviewed Keng for the film and videotaped the interview. Polomski maintained that the video camera was visible in the room when he conducted his interview with Keng. Polomski testified that he was uncertain whether he informed Keng that the thesis project would possibly become a documentary film that would be shown to the public. And Polomski did not inform Keng when the film was later shown to the public. When Keng was questioned by the State about his appearance in the documentary film, Keng stated that he was unfamiliar with the film and that he did not know who interviewed him or where the camera was located.

Penick testified that when he receives exculpatory information, he turns it over to the defense. During Reed’s trial, he was approached by a woman named Elizabeth Keehner, and she told him that she did not believe that Reed was guilty of murdering Stacey. Viewing Keehner’s statement as exculpatory information, Penick stated that he told Tanner or someone in law enforcement about the statement and that someone working for the State took a statement from her. The statement was then turned over to the defense. Wolfe corroborated Penick’s testimony, stating that she interviewed Keehner and obtained an exculpatory statement that was later disclosed to the State. She also stated that Penick never expressed “an attitude that we’ve got everything we need.”

Regai'ding Barnett, Penick testified that he did not receive any information involving Barnett before or during Reed’s trial. In fact, Penick stated that he did not learn that the client Keng referred to was Barnett until he read Keng’s affidavit. Penick maintained that, had he received information about Barnett’s sighting, he would have investigated it and shared the information with the defense if the investigation revealed that it was exculpatory. Penick recalled that Keng relayed the information from Barnett in passing as he and Keng were leaving the courtroom after a docket call approximately four years after Reed’s trial. On cross-examination, Penick was asked what he thought of Keng’s allegation that he had disclosed the information about Barnett’s sighting to him in 1998. Penick said that Keng was “telling a big lie.” Penick then acknowledged that, in supporting Keng’s reelection hi 1996, he wrote a letter stating that he knew and worked with Keng for fifteen years and that, in his opinion, Keng is “a very competent, honest, professional prosecutor.... ” Explaining his current opinion of Keng, Penick stated that Keng was honest as a prosecutor but changed when he became a defense attorney, and his dealings with him as a defense attorney were not good.

Reed’s habeas counsel asked Penick why he used the present tense during the deposition in the civil case when recalling that he told Keng that he “has” all of the evidence he needed against Reed. In response, Penick explained that the question took him by surprise and that he failed to clear up the fact that Keng passed along the information involving Barnett four years after Reed’s trial. During that deposition, he realized that Keng made the disclosure during the Sykes case. Knowing this, Penick later pulled the District Attorney’s file on the Sykes case and determined that Keng would have been at the courthouse between January and October 2002.

Finally, Penick stated that the civil suit dealt with other allegations and that the alleged suppression of Barnett’s sighting did not provide a basis for the suit.

b. Blackwell

When testifying, Blackwell reiterated and added to the statements made in her affidavit. Blackwell stated that class was seated alphabetically in the academy and that she was seated near Angela Allred, Larry Franklin, and Fennell. When Blackwell was rewriting her notes in the classroom during a break, Fennell was standing up in the back of the room talking with the cadet who sat to his right. Blackwell overheard Fennell tell the cadet seated to his right that he would strangle his girlfriend if he discovered that she was cheating on him. Blackwell, who was seated at the table in front of Fennell, then looked over her shoulder and said, “Well, if you do that they’ll find your fingerprints all over her throat.” Fennell responded to Blackwell, telling her that he would use a belt. Blackwell found Fennell “to be extremely offensive when it came to his attitude towards wom[e]n in particular, not only women in police work but wom[e]n in general.” She also “found him to be conceited, arrogant, and that he regarded himself as a police officer having power over others in a way that police officers should not have power.” Recalling the incident in the parking lot in which Fennell directed her to stop talking to Stacey, Blackwell testified that when Fennell got into the truck with Stacey, she could tell from his facial expressions that he was yelling at her.

When Stacey was murdered, Blackwell was working as a Deputy Constable for Rocky Madrono in Travis County. The Bastrop County Sheriffs Department called Madrono’s office and requested help with the escort for Stacey’s funeral. This was the first time that Blackwell had learned about Stacey’s murder. Blackwell received permission to use one of Madro-no’s vehicles, and another deputy, who had been a cadet with Blackwell at CAPCO, accompanied Blackwell to the funeral. Blackwell attended the funeral and saw Fennell exiting the church. As Fennell followed Stacey’s casket, Fennell collapsed on one knee and needed assistance getting up. Blackwell returned to work and told Madrono that Fennell appeared to be putting on an act. She then informed him about the comments made by Fennell in class, even though, according to her testimony, she was unaware that the community and people attending the funeral were questioning what had happened to Stacey. During her testimony, Blackwell also recalled that she had previously told her best friend about what Fennell had said the night that Fennell made the statement.

According to Blackwell, at some point in 1998 when the weather was warm, she was introduced to Vasquez, the investigator appointed to assist Reed’s trial attorneys. From his introduction, Blackwell got the impression that Vasquez was actively involved in investigating Stacey’s murder and that he was working on behalf of the individual accused of murdering Stacey. When Vasquez told Blackwell that Stacey had been strangled with a belt, “bells and sirens went off” in Blackwell’s head, and Blackwell immediately told Vasquez about what Fennell had said in class. Presented with the CAPCO cadet-class roster and class photograph, Blackwell identified the cadet who sat on Fennell’s right as Christopher Dezarn. Vacillating, Blackwell later stated she could not remember who Fennell had made the statement to and that the proposed seating chart would not refresh her recollection.

On cross-examination, Blackwell conceded that she failed to notify authorities investigating Stacey’s murder of Fennell’s statement even though she believed that the statement was significant in light of her status as a peace officer, her professional training and experience, as well as her personal experience with Fennell.

Vasquez testified that he was visiting with Madrono a few weeks after Reed’s trial, following the conclusion of his official investigative duties. He was discussing Reed’s case with Madrono when Blackwell approached him and told him about Fen-nell’s statement. Vasquez then drove to CAPCO and confirmed that Blackwell and Fennell had been at the academy together. Vasquez documented his conversation with Blackwell in a memo. Contrary to the statement made in his affidavit about passing the information along to Penick, at the hearing, Vasquez stated that he gave the memo to Forrest Sanderson, a chief assistant district attorney in Bastrop County and a member of the trial team in Reed’s case, a week or so after he drafted the memo. Vasquez testified that he believed that he had given the information to Pen-ick, but when he saw Sanderson sitting in the courtroom, he remembered that he had in fact given the information to Sanderson. Vasquez was comfortable disclosing the information to the District Attorney’s Office because, he had known Penick and Sander-son for eight or nine years, believed that it was important to pass the information along to them, and was confident that they would do something with it. Vasquez did not call Garvie or Clay-Jackson. When asked if he gave the information to the previous investigator for the defense, Duane Olney, Vasquez stated that he believed that he had done so when he bumped into him on the street several months after he had given the information to Sanderson. Vasquez stated that he intended to give the information to Reed’s appellate attorneys. Vasquez, however, failed to reach out to Reed’s attorneys. Vasquez finally spoke to attorneys representing Reed after they initiated contact with him in 2003 or 2004.

Sanderson testified that Vasquez did not give him the information at issue and that, if Vasquez had given him such information, he certainly would remember it. Sander-son also expressed his opinion about how Penick would deal with exculpatory information relating to Reed’s case. Sanderson stated that Penick “would have been on it like white on rice.” Penick stated that Vasquez never handed him any document pertaining to the Reed case after Reed’s trial.

Contradicting Blackwell’s testimony, De-zarn testified that Fennell never told him that he would strangle his girlfriend with a belt if he caught her cheating on him.

Larry Franklin, another one of Blackwell’s and Fennell’s prior classmates at CAPCO, testified that he and Blackwell maintained a friendship after graduation and that the two would call one another. Sometime after Stacey’s murder, Blackwell called Franklin and asked him if he heard about the murder. Together, Blackwell and Franklin questioned whether Fennell had murdered Stacey. The two had other conversations on this topic, and during one of these conversations, Blackwell told Franklin about Fennell’s statement. Although Franklin had sat to the left of Fennell in class, Franklin did not hear Fennell make this statement and learned about it only through his conversation with Blackwell. Though Franklin stated that, as a peace officer, he felt that there was an ethical obligation to report such information, he admitted that he failed to do so.

Missy Wolfe was assigned to investigate the validity of Blackwell’s contentions for the State. She began by getting the class roster from CAPCO, which indicated that Fennell and Blackwell’s class had twenty-nine members. Wolfe and another investigator contacted all the individuals in the class, including Fennell and Blackwell. They obtained written or tape-recorded statements from everyone except Fennell and Blackwell. Wolfe testified that none of twenty-seven people in the class corroborated Blackwell’s contentions regarding Fennell.

Clay-Jackson stated that if she had found out about Fennell’s statement within time for filing a motion for a new trial, she would have moved for a new trial on that basis.

4. Trial Judge’s Findings of Fact, Conclusions of Law, and Recommendation

At the close of the evidentiary hearing, the trial judge, who succeeded the judge who presided over Reed’s trial, requested that the parties submit proposed findings of fact and conclusions of law. Adopting the State’s proposed findings of fact and conclusions of law, which we will explore in greater detail below, the trial judge recommended that we deny relief.

III. Issues for Resolution

When this case was returned to us, we noted, after conducting a careful review of the record, that a few of the trial judge’s factfindings were either unsupported by the record or appeared, in some fashion, to be misleading. Because of this and the sharply conflicting testimony offered at the evidentiary hearing, we filed and set this case for submission to decide whether Reed is entitled to relief under Brady. To facilitate our resolution of Reed’s claims on the record before us, we directed the parties to brief the following issues:

• Assuming, arguendo, that the court has entered a finding of fact or conclusion of law that has multiple sentences or phrases and that a portion of the finding or conclusion is supported by the record, while another portion is not, to what extent does this Court owe deference to the trial court on such a finding or conclusion? May the Court disregard the finding or conclusion in its entirety?

• Assuming, arguendo, that numerous findings and conclusions, or parts thereof, are not supported by the record, how should this affect the level of deference to the findings and conclusions as a whole?

We also ordered the parties to address whether Reed’s gateway-actual-innocence claim satisfied the requirements under Article 11.071, Section 5(a)(2).

IV. Analysis

A. Reed’s Brady Claims that Satisfied Article 11.071, Section 5(a)(1)

1. The Standard

To protect a criminal defendant’s right to a fair trial, the Due Process Clause of the Fourteenth Amendment to the United States Constitution requires the prosecution to disclose exculpatory and impeachment evidence to the defense that is material to either guilt or punishment.¹² This rule of law originated in 1963 in Brady v. Maryland and has been clarified and further refined in its progeny. Applying the rule in 1995, the Supreme Court, in Kyles v. Whitley, held that the rule encompasses evidence unknown to the prosecution but known to law-enforcement officials and others working on their behalf.¹³

Under its present incarnation, to succeed in showing a Brady violation, an individual must show that (1) the evidence is favorable to the accused because it is exculpatory or impeaching; (2) the evidence was suppressed by the government or persons acting on the government’s behalf, either inadvertently or -willfully; and (3) the suppression of the evidence resulted in prejudice (i.e., materiality).¹⁴ Evidence is material to guilt or punishment “only if there is a reasonable probability that, had the evidence been disclosed to the defense, the result of the proceeding would have been different.” ¹⁵ “A 'reasonable probability’ is a probability sufficient to undermine confidence in the outcome.” ¹⁶

For over forty years, our writ jurisprudence has consistently recognized that this Court is the ultimate factfinder in habeas corpus proceedings.¹⁷ The trial judge on habeas is the “ ‘original factfin-der.’ ” ¹⁸ Summarizing the role of the trial judge, we have explained that the judge

is the collector of the evidence, the organizer of the materials, the decisionmaker as to what live testimony may be necessary, the factfinder who resolves disputed fact issues, the judge who applies the law to the facts, enters specific findings of fact and conclusions of law, and may make a specific recommendation to grant or deny relief.¹⁹

Uniquely situated to observe the demeanor of witnesses first-hand, the trial judge is in the best position to assess the credibility of witnesses,²⁰ Therefore, in most circumstances, we will defer to and accept a trial judge’s findings of fact and conclusions of law when they are supported by the record.²¹ When our independent review of the record reveals that the trial judge’s findings and conclusions are not supported by the record, we may exercise our authority to make contrary or alternative findings and conclusions.²²

In answering the first two issues that we ordered the parties to brief, we conclude that it is appropriate to remain faithful to our precedent. Thus, we will afford no deference to findings and conclusions that are not supported by the record and will ordinarily defer to those that are. So where a finding or conclusion contains multiple sentences or phrases, we will pay deference to the sentences and phrases that are grounded in the record and reject or refuse to adopt those that are not. When our independent review of the record reveals findings and conclusions that are unsupported by the record, we will, understandably, become skeptical as to the reliability of the findings and conclusions as a whole. In such cases, we will proceed cautiously with a view toward exercising our own judgment. And when we deem it necessary, we will enter alternative or contrary findings and conclusions that the record supports. Furthermore, when we determine that the trial judge’s findings and conclusions that are supported by the record require clarification or supplementation, we may exercise our judgment and make findings and conclusions that the record supports and that are necessary to our independent review and ultimate disposition. However, where a given finding or conclusion is immaterial to the issue or is irrelevant to our disposition, we may decline to enter an alternative or contrary finding or conclusion.

As recognized by our decisions, this standard of review accounts for the unparalleled position of the habeas judge to directly assess a witness’s demeanor. When listening to testimony, the habeas judge is tuned in to how something is being said as much as to what is being said. The judge is acutely aware of a witness’s tone of voice or inflection, facial expressions, mannerisms, and body language. There is no doubt that this type of assessment, the essence of which a cold record rarely captures, is a determinative factor in a trial judge’s credibility assessment and factfindings.

Next, we conclude that when numerous, but not all, findings and conclusions are not supported by the record, the determination of the level of deference to be accorded to the findings and conclusions as a whole is to be made on a case-by-case basis. It is impossible to establish any type of litmus test for determining when and under what circumstances the level of overall deference may be affected by numerous unsupported findings and conclusions. Because no two cases are alike, the level of deference accorded to the findings and conclusions as a whole where numerous findings are not supported by the record will depend on a thorough review and analysis of the specific facts and legal issues involved in a given case. The case may arise where the nature and number of unsupported findings and conclusions may render the findings and conclusions wholly unreliable and beyond repair. Under such circumstances, we may elect to take it upon ourselves to conduct all of the factfinding and to issue a ruling explaining our application of the law to the facts. However, we note that it will be under only the rarest and most extraordinary of circumstances that we will refuse to accord any deference whatsoever to the findings and conclusions as a whole.

In this case, in adopting the State’s proposed findings of fact and conclusions of the law, the trial judge concluded that Reed was not entitled to relief under Brady because he failed to establish that (1) the State suppressed evidence and (2) the evidence was material to guilt. We agree with the trial judge’s legal conclusion that Reed has not demonstrated that the State suppressed evidence and therefore find it unnecessary to render a decision regarding materiality. So in reviewing the trial judge’s factfindings, we will confine our discussion of the factfindings to those that are relevant to our determination that the State did not withhold any favorable information.

Our independent review of the trial judge’s remaining findings of fact (i.e., those irrelevant to our resolution) demonstrates that they are largely supported by the record. A select few of these findings, however, are inconsistent with the record or are somewhat misleading. For example, with respect to Barnett’s habeas testimony, the trial judge found that when Barnett was confronted with the fact that Fennell’s picture did not appear in any of the articles relating to Stacey’s murder from April 24, 1996, through May 30, 1998, “Barnett conceded that she knew Jimmy Fennell from ‘something completely independent of the Giddings newspaper,’ i.e., her DWI arrest.” This finding unfairly portrays Barnett’s testimony. Even though Barnett admitted that she knew Fennell from her DWI arrest, she was adamant that she recognized Fennell as the man she saw with Stacey at the Old Frontier store from a photograph in the newspaper. While the trial judge was entitled to find, based on her credibility determination, that Barnett recognized Fen-nell solely from her DWI arrest, especially given Barnett’s mother’s inability to specifically confirm Barnett’s identification of Fennell, the trial judge was not justified in finding as a matter of fact that Barnett conceded this point. We attribute this inaccuracy (and other like findings) to the fact that the State generated the proposed findings and they are therefore wholly representative of the State’s interpretation of the evidence. Mindful of the role of an advocate, the trial judge as a neutral arbiter should have more carefully scrutinized the State’s proposed findings to ensure that they accurately reflect the evidence in the record before adopting them verbatim. Regrettably, the trial judge’s decision to adopt the State’s proposed findings and conclusions verbatim has unnecessarily complicated our independent review of the record. Nevertheless, in this case, we conclude that the few instances in which the findings are inconsistent or misleading do not justify a decision to totally disregard the findings that are supported by the record and are germane to our resolution of Reed’s Brady claims.

2. Discussion

Relevant to Reed’s allegation that the State suppressed information concerning Barnett’s sighting of Stacey and Fen-nell, the trial judge found:

Stephen Keng testified that, at some point after speaking with Barnett, he told Bastrop County District Attorney, Charles Penick that he had a client who claimed to have seen [Stacey] with Fen-nell on the morning that she disappeared. According to Keng, this conversation took place sometime in February or March of 1998, on the second floor of the Bastrop County Courthouse, before Reed’s trial began. Keng testified that, in response, Penick laughed and told him ‘that he had all [t]he evidence he needed, and he just didn’t want to hear about it.’

Charles Penick testified that he recalled having a conversation with Keng, during which Keng told him that he had a client that knew something about the Reed case. Penick recalled that this conversation with Keng took place about four years after the trial in the law enforcement center during a docket call. Pen-ick stated that he thought Keng was joking and ‘didn’t take him seriously.’ Penick testified that he told Keng that he had enough evidence against Reed and ‘didn’t need to hear that....’ Pen-ick testified that Keng did not approach him with information regarding Martha Barnett at any point prior to, or during Reed’s trial.

In his April 16, 2002 affidavit, Keng stated that he believed the information Barnett had shared with him ‘was important because the newspaper reports indicated that Mr. Fennell had been excluded as a suspect because law enforcement could not explain how he committed the crime.’ Keng stated that information regarding Fennell’s elimination as a suspect came out in the newspaper before he spoke with Penick about Barnett. During cross-examination, Keng was confronted with the fact that articles reporting that Fennell had been eliminated as a suspect came out in May 1998, several months after he claimed to have spoken with Penick. Reed’s trial concluded on May 29, 1998. Stephen Keng testified that he did not participate in the making of a documentary movie in the instance [sic] case. Keng testified, ‘News to me.... I don’t know who interviewed me or where they had a camera.’

Ryan Polomski testified that he produced a documentary film about the instant case and Stephen Keng was interviewed on camera for that film and that parts of that interview were featured in the film.

Reed has not proven that the Bastrop County District Attorney’s Office was in possession of the information regarding Barnett prior to or during trial.

Clearly implied by the trial judge’s finding that the Bastrop District Attorney’s Office was not in possession of information regarding Barnett before or during trial is the determination that Keng failed to disclose that Barnett had information relating to Reed’s case until sometime after Reed was convicted and sentenced. And what necessarily flows from this is the trial judge’s implicit determination that Keng’s testimony about the timing of the disclosure to Penick’s memory is not credible. The trial judge’s credibility determination is supported by the record, and we therefore choose to adhere to her findings. However, we find it necessary to clarify and supplement the findings that pertain to Keng’s credibility.

First, the finding relating to Keng’s recognition of the importance of Barnett’s information suggests that it is unlikely that Keng truly realized the significance of Barnett’s sighting until sometime in May 1998, at the earliest, when the newspapers first began reporting that Fennell had been excluded as a suspect. This determination is reasonable regardless of Keng’s explanation that he wrote the affidavit years after the events took place, including when the newspapers began reporting Fennell’s exclusion as a suspect, and that the information therefore became part of his collective memory about the case. Because the impetus behind Keng’s disclosure to Penick is directly tied to the timing of his disclosure, anything that selves to undermine his credibility about the impetus also undermines his credibility about the timing. Further, when Keng was given the opportunity to offer a reason, independent of newspaper articles, for recognizing the significance of Barnett’s sighting when he claimed that he did, Keng failed to offer any alternative explanation. Accordingly, the trial judge was justified in finding Keng’s recollection about the timing of the disclosure to Pen-iek to be unconvincing.

Next, the trial judge’s findings about Keng’s appearance in the documentary film “State versus Reed” suggests that Keng testified untruthfully when he denied that he was interviewed for the documentary film. Our reading of the testimony indicates it is unlikely that Keng realized he was participating in a documentary film that would be released to the public. Po-lomski testified that he recalled telling Keng that he was a graduate student at the University of Texas and that he was working on his thesis project for a Master of Fine Arts in Film and Video Production. Polomski was unable to specifically recall whether he referred to the film as a documentary or a thesis project and stated that he did not believe that they talked about any showings, distributions, or screenings. However, the . record does support the broader, implicit finding that Keng’s testimony about being interviewed for the film was not credible. Even if Keng was unaware that Polomski’s film was a documentary, the trial judge, after observing Keng’s demeanor, was free to disbelieve Keng’s denial about being interviewed for a film devoted to Reed’s case. And even though this subject matter was unrelated to the disclosure of Barnett’s sighting, the trial judge was permitted to take Keng’s veracity on this issue into account when assessing his credibility on the timing of the disclosure.

The trial judge also expressly found Penick to be credible. Because the trial judge was positioned to witness Penick’s demeanor first-hand, we conclude that the trial judge’s credibility determination and resultant factfindings are supported by the record.

Reed argues that we should find that Keng is more credible than Penick. Keng, according to Reed, had no involvement in Reed’s case and has nothing to gain from the outcome of this case; he has placed his credibility and reputation on the line by stepping forward. Penick, on the other hand, should not be found credible because, when he testified at the habeas hearing, he had a financial interest in the case, his memory has proven to be selective, and he previously endorsed Keng’s character for honesty.

Pointing to Penick’s civil lawsuit, Reed argues that, if Penick was found to have suppressed evidence in this case, the impact on Penick’s case, which was unresolved at the time of the hearing, would be devastating. Reed claims that, if his case were reversed due to Penick’s failure to disclose Brady evidence, a jury in the civil case would be unlikely to award Penick damages, regardless of whether the particular evidence suppressed was the subject of Penick’s lawsuit.

Regarding Penick’s inconsistent testimony about the disclosure, Reed points to Penick’s evidentiary-hearing testimony addressing the first statements he made about the disclosure in 2003 during a deposition in the civil lawsuit. Referring to the transcript from the civil-suit deposition at the hearing, Reed’s habeas attorneys questioned Penick about his prior statements on cross-examination:

Q. Line 18: “ ‘Did you have any conversation with Mr. Keng before the Reed trial between the time Mr. Reed was indicted and the time it went to trial about a client of Mr. Keng’s who said that she had seen Stacey Stites and Jimmy Fennell together on the morning of Ms. Stites’ murder? And your answer was ‘no.’ ”

A. That’s correct.

Q. All right. Question: “ ‘You don’t remember it or it didn’t happen?’ ” And your answer is: “ ‘I don’t remember it happening, I don’t think it happened.’ ”

Question: “ ‘Okay. And so you didn’t tell him that you had all the evidence that you needed and you didn’t want to hear anything more about the case?’ ” Answer: “T do kind of remembering [sic] in passing making that statement to Steve.’ ”

Because Penick failed to correct this testimony and only mentioned, for the first time, during a subsequent deposition that the disclosure took place four years after the trial, Reed argues that we should credit his initial testimony, “not his carefully crafted answers presented months later.”

Finally, Reed argues that Penick could not reconcile accusing Keng of lying about the timing of the disclosure with the letter he had written in support of Keng’s reelection in 1996.

This information was before the trial judge, and she was free to resolve any tendency toward bias and any contradictory statements made by Penick in favor of finding Penick credible. After reviewing newspaper articles and the petition filed in the civil suit, Penick testified that the allegations involving Barnett had nothing to do with the civil suit. Reed’s contentions about the potential impact of this case on Penick’s civil suit is speculative, and the trial judge was permitted, in hearing Pen-ick’s testimony on the issue, to determine that his personal interests in the civil suit did not improperly influence his testimony.

The trial judge was also at liberty to believe Penick’s testimony explaining his failure to state that Keng’s disclosure took place after Reed’s trial during the civil-suit deposition and Penick’s reason for changing his personal opinion of Keng. Elaborating on his failure to state that Keng’s disclosure occurred four years after Reed’s trial when he was deposed in the civil suit, Penick admitted that he failed to clear up the matter during the deposition. He explained that the initial questions regarding Keng’s disclosure caught him off guard and he did not remember the conversation at first. The questions then prompted him to remember the conversation with Keng, and during the questioning, he thought about it and remembered that it occurred when Sykes was being prosecuted, four years after Reed’s trial.

I knew what case it was. I went back to that case and I found out what dates were that we have been over here because I remembered then, back in the first deposition, of the conversation I had with Steve Keng and it was over at the law enforcement center when he was representing Amanda Sykes on a murder case where she killed her husband.

Because Keng’s first affidavit relating to the disclosure was made in April 2002, Penick assumed that, based on his review of the docket dates for the Sykes case, Keng must have made the disclosure between January and April 2002.

Finally, regarding the letter endorsing Keng’s reelection, Penick explained that Keng was a friend of his and that, based on his opinion of Keng at the time he wrote the letter, he was being truthful when he said that Keng was an honest person. Penick stated: “[A]t the time he was in prosecution, I felt that way, but when he became a criminal defense lawyer he changed, he changed an awful lot.” Penick added that, had he known what he now knows about Keng, he would not have written the letter.

Next, although we question whether Fennell’s statement to Blackwell falls within Brady’s ambit because it was not alleged to have been disclosed until after Reed’s trial and therefore may be more properly characterized as newly discovered evidence,²³ we will nevertheless defer to the trial judge’s credibility determinations and factfindings because our independent review of the record establishes that they are supported by the record. Concerning Vasquez’s credibility, the trial judge found the following:

During cross-examination, Vasquez was confronted with the fact that he had sworn in his January 2, 2005, affidavit that he ‘spoke directly to Distinct Attorney Charles Penick.’ Vasquez testified that he thought at the time that he’d give[n] it to Penick but remembered when he saw Sanderson sitting in the courtroom that he had given the information to him, and not to Penick.

Charles Penick testified that John Vasquez never approached him with information pertaining to Reed’s case.

Forrest Sanderson testified that he did not recall Vasquez ever approaching him with information about Reed’s case. Sanderson testified that he would have rememberfed] if Vasquez had come to him with information pertaining to Reed’s case.

Given the inconsistencies in Vasquez’s testimony the Court finds him not to be credible.

This Court finds the testimony of Sand-erson, Penick, and Wolfe to be credible.

This Court finds that the Bastrop County District Attorney’s Office did not possess any evidence pertaining to Mary Best Blackwell prior to or during trial.

This Court finds that John Vasquez did not provide evidence pertaining to Mary Best Blackwell to the Bastrop County District Attorney’s Office until after Reed’s trial.

Based on the foregoing, we hold that the record supports the trial judge’s conclusion that the State did not suppress favorable evidence during trial in violation of Brady. Accordingly, Reed has not proven that he is entitled to relief.

B. Reed’s Schlup Claims Under Article 11.071, Section 5(a)(2)

Under our Legislature’s codification of the Supreme Court’s Schlup v. Delo24 standard, we may

not consider the merits of or grant relief on a subsequent application unless the application contains sufficient specific facts establishing that:

[[Image here]]

(2) by a preponderance of the evidence, but for a violation of the United States Constitution no rational juror could have found the applicant guilty beyond a reasonable doubt.²⁵

To obtain review of the merits of a procedurally barred claim, an applicant must make a threshold, prima facie showing of innocence by a preponderance of the evidence.²⁶ A Schlup claim of innocence is not an independent constitutional claim; it is “a gateway through which a habeas petitioner must pass to have his otherwise barred constitutional claims considered on the merits.”²⁷ Because Article 11.071, Section 5(a)(2) was enacted in response to the Supreme Court’s decision in Schlup,28 we conclude that standards set forth for evaluating a gateway-actual-innocence claim announced by the Supreme Court should guide our consideration of such claims under Section 5(a)(2). Therefore, to mount a credible claim of innocence, an applicant “must support his allegations of constitutional error with reliable evidence — whether it be exculpatory scientific evidence, trustworthy eyewitness accounts, or critical physical evidence — that was not presented at trial.”²⁹ The applicant bears the burden of establishing that, in light of the new evidence, “it is more likely than not that no reasonable juror would have” rendered a guilty verdict “beyond a reasonable doubt.”³⁰ To determine whether an applicant has satisfied the burden, we must make a holistic evaluation of “ ‘all the evidence,’ old and new, incriminating and exculpatory, without regard to whether it would necessarily be admitted under ‘rules of admissibility that would govern at trial.’ ”³¹ We must then decide how reasonable jurors, who were properly instructed, “would react to the overall, newly supplemented record.”³² In doing so, we may assess the credibility of the witnesses who testified at the applicant’s trial.³³

In this case, we must determine whether Reed has satisfied his gateway burden under subsection (a)(2) so as to permit us to review his proeedurally barred Brady claims, ineffective-assistance-of-counsel claims, and other constitutional claims. And in deciding whether Reed has met his burden, we will defer to the trial judge’s findings and conclusions when it is appropriate.

In support of his gateway-innocence claim, Reed relies on numerous items of evidence not presented at trial, some of which were offered in his prior applications. While we seriously doubt that some of the evidence Reed cites constitutes new evidence for purposes of our inquiry,³⁴ we will give Reed the benefit of all doubt and consider all of the evidence that was not presented at his trial, namely the evidence presented in all three of Reed’s applications. We will leave it for another day to decide exactly what new evidence, not presented at trial, may be considered in the purview of Section 5(a)(2)’s threshold showing of innocence.³⁵

1. Reed’s Initial Application

We start by examining the evidence presented in Reed’s initial application and the accompanying findings of fact that are pertinent to the particular items of evidence. We note that, reviewing this evidence in its entirety, the trial judge found that Reed failed to prove that he is actually innocent under the more stringent standard of Herrera and Ex parte Elizondo.36

a. Robbins’ Statements

First, Reed submits two statements from Robert and Wilma Robbins that were given to police in March 1998. The Rob-binses delivered the Austin American-Statesman Monday through Saturday in Bastrop. Their route included the Bas-trop High School. A month before Stacey’s murder, the Robbinses saw a gray/ blue Ford Tempo parked in the School parking lot a few days during the week between 4:30 and 5:00 a.m. The car was not there on the day of Stacey’s murder, and they never saw it parked at the School after Stacey was murdered. Robert also stated that he saw a Chevrolet full-size truck, which he believed was white, in the lot two or three days during the week. He did not see the truck the day of Stacey’s murder and did not see it in the lot again after Stacey’s murder.

The trial judge found that Carol Stites’s testimony about Stacey rarely using the Tempo to drive to work because of its unreliability was credible. The trial judge also determined that, because of Stacey’s work schedule, she could not have been at the Bastrop High School between 4:30 and 5:00 a.m. The trial judge found that, in April 1996, a 1988 Ford Tempo was registered to David Gonzalez, who worked for the High School in April 1996. As a result, the trial judge found no credible evidence that Stacey was driving Carol’s ear to the High School the month before her death.

Concerning the white truck, the trial judge found that when Robert was cross-examined at trial, he admitted that, when he was questioned two days after Stacey’s murder, he told investigators that the truck was silver. The Bastrop Police Chief investigated the information from Robert and concluded that the truck belonged to a school employee. Further, while Patty Timmons testified at trial that she saw three men in a white truck parked near Bluebonnet Road on April 23, 1996, between 6:30 and 7:00 a.m., she told investigators three weeks after Stacey’s murder that she saw the truck at 9:30 a.m. on April 22, 1996. As a result, the trial judge found no credible evidence that Stacey was murdered by three unknown men in a white truck.

b. Witnesses Affirming Relationship Between Stacey and Reed

Reed also presented several affidavits from witnesses claiming to know of an ongoing relationship between Stacey and Reed.

Kay Westmorland stated that Stacey and Reed came to her house three or four times between late January 1995 and April 1996. She knew Stacey from the H.E.B. and knew of Fennell because she saw him pick up Stacey at work. Westmorland knew Reed from the neighborhood. She heard that Fennell knew Reed was seeing Stacey and that Fennell was jealous. She claimed that she was “not surprised to see [Fennell] drive by [her] house on several occasions in the same truck that she had seen Stacey and [Reed] in.”

Meller Marie Aldridge stated that, when she was at a friend’s house, Stacey came and picked up Reed. Her friend identified Stacey as Reed’s girlfriend. Meller Marie Aldridge knew Stacey from the H.E.B.

On June 13, 2000, Meller Marie Aldridge gave a second affidavit to the State elaborating on her first sworn statement. The young woman, whom she saw pick up Reed, had driven a full-size truck, worked at the customer-service center at H.E.B., and was “best friends” 'with a Hispanic woman named Rose, who worked at H.E.B. and lived in the “projects” in Bas-trop. The trial judge found that the only truck Stacey drove belonged to Fennell and that it was not a full-size truck. Further, according to the general manager of the Bastrop H.E.B., Stacey did not work at the customer-service center. That position required special training, which Stacey never received. The general manager also maintained that Stacey did not regularly hang out with a Hispanic woman named Rose and that she was not drawn toward any particular coworker. The trial judge found that there was no Hispanic woman named Rose who worked at the H.E.B. or was Stacey’s best friend. The trial judge found that the evidence presented by Meller Marie Aldridge concerning a relationship between Stacey and Reed was unpersuasive.

Shonta Reed stated that Stacey had come by her house looking for Reed when he was not at home and that Stacey returned to pick him up when he got back home.

Elizabeth Keehner stated that she saw Reed, whom she “knew quite well,” walking out of the H.E.B. “holding hands with a very pretty white girl” a few months before Stacey’s death. When she saw Stacey’s picture in the paper, she thought that Stacey might have been the girl with Reed at the H.E.B.: “The familiarity was there.”

The trial judge found that Keehner was a bondswoman and a close friend of Reed’s family. She often bonded Reed out of jail. Before trial, Keehner gave a more detailed statement to police. In addition to mentioning her sighting of Stacey and Reed at H.E.B., Keehner detailed a conversation she had with Chris Hill’s grandmother-in-law, Betty Wallace. Wallace, who occasionally worked for Keehner, told Keeh-ner, in Chris Hill’s presence, that everyone at H.E.B. knew that Stacey and Reed were dating. Hill also worked at H.E.B., and Keehner stated that Hill responded in the affirmative when Wallace asked whether it was common knowledge at H.E.B. about Stacey and Reed dating. The State obtained a statement from Hill in 2000. He denied any knowledge of the conversation and stated that he did not have any personal knowledge of a relationship between Stacey and Reed. The trial judge found that the State could have subjected Keeh-ner to significant impeachment if she had testified at Reed’s trial.

Walter Reed, Reed’s father, stated that Kelly Bonguli, who had worked at the H.E.B., told him that he knew where Stacey was the night she was killed. Bonguli also told Walter that he and his family had been “tailed” during Reed’s trial. He then said that he wanted to talk with someone before he said anything about the case.

Considering these statements, the trial judge found that they were not credible or persuasive. Reed failed to submit an affidavit from Kelly Bonguli. The State obtained an affidavit from Bonguli that discredited Walter Reed’s statements. Bonguli stated that he “never told Waiter-Reed that I knew where Stacey Stites was on the night she was killed. All I ever-told Walter was that Rodney Reed was a erackhead who raped girls on the R.R. tracks. I have no idea where Stacey Stites was when she died.”

In an affidavit submitted by Reed, Ron Moore states that he had a conversation with Debra Pace and Jane Campos about Stacey’s murder in January 1999. According to Moore, Campos told him that Reed did not kill Stacey and that she had overheard a conversation between Fennell and his coworker, Curtis Davis. Davis told Fennell “not to worry that ‘it was all taken care of ” in response to Fennell’s complaint about Stacey’s affair with Reed. Pace told Reed’s trial investigator, Olney, about the conversation. Olney submitted an affidavit attesting to his conversation with Pace.

The trial judge found that Moore’s and Olne^s statements were not persuasive or credible. Reed failed to provide the trial judge with affidavits from Campos and Pace. In an affidavit obtained by the State, Campos stated that she never told anyone that Reed did not kill Stacey or that she overheard a conversation between Fennell and Davis in which the two discussed an affair between Stacey and Reed. Pace also executed an affidavit at the State’s request. In it, she asserted that she never told Moore or Olney about anything Campos said; when Moore and Ol-ney came to her house, she refused to talk to them. Pace read Moore’s and Olney’s affidavits and stated that the two are “bald face liars.” Campos said that Reed did not do it when Pace was talking with Campos and Moore about them personal opinions about the case. Pace stated that Campos’s tone was serious, but that for all she knew it was only Campos’s opinion. Finally, Curtis Davis submitted an affidavit denying that he ever had such a conversation with Fennell.

Jon Chris Aldridge submitted an affidavit stating that he saw Stacey and Reed together during the three months before she was murdered. Around April 1st, when Jon and Reed were walking, Fennell stopped them and told Reed he knew about him and Stacey. Fennell then told Reed that he was going to “pay.”

Jon Aldridge gave the State a more detailed affidavit on June 14, 2000. He stated that he was at Shonta Reed’s house when a large full-size pick-up truck pulled up. When Reed introduced Jon to the driver, he told Jon that her name was Stacey and that they were dating. The three then rode around and purchased crack cocaine. After Stacey and Reed smoked the crack cocaine, Stacey dropped them off at a local bar. Jon asserted that Fennell and another law-enforcement official whom he did not know stopped them in Bastrop. Fennell was wearing plain clothes, and the other officer was wearing a uniform. Jon stated that they were in a Bastrop County Sheriffs Department vehicle with a star embossed on the side. He stated that he knew Fennell because Fen-nell had booked him into the Bastrop County Jail.

The trial judge determined that Jon’s statements were neither persuasive nor credible. Fennell, the trial judge found, was a Gidding’s police officer at the time— not a Bastrop Sheriffs Deputy. Additionally, after reviewing Jon’s booking sheets, which the State submitted, the judge found that Fennell never booked Jon into the jail. Regarding the allegation of Stacey’s crack cocaine use with Reed, the trial judge found that the toxicology report from Stacey’s autopsy was negative for drugs and alcohol. The drug screen conducted by H.E.B. before Stacey was hired was also negative. Further, the trial judge found:

Prior to trial, the State sent samples of Stacey Stites’ hair to National Medical Services, Inc. in Pennsylvania. That laboratory analyzed 32 centimeters of her hair in order to determine whether cocaine or its metabolites were present. As that laboratory’s report indicates, two different analyses were negative for cocaine. Since hair grows at an approximate rate of one centimeter per month, the State was prepared, through the use of these analyses, to prove that Stacey Stites was not a cocaine user for the last 32 months of her life.

Finally, the trial judge found that many of these affidavits were from Reed’s family members. Jon Aldridge, Shonta Reed, Meller Marie Aldridge, and Ron Moore are Reed’s cousins, and Walter Reed is Reed’s father. The trial judge also determined that, at the time of the habeas proceedings, many of these family members had criminal records. Jon Aldridge, who is Meller Marie Aldridge’s son, had a lengthy arrest record. He had been convicted several times of theft by check and had been convicted for failure to identify himself as a fugitive from justice. Shonta Reed had been convicted of theft four times and convicted of assault once. Linda West-morland had been convicted of felony theft, and the State’s motion to revoke her probation on that cause was pending at the time of the habeas proceedings. She also had forgery charges pending in Dallas.

In light of his earlier findings, the trial judge found that the evidence of a secret affair between Stacey and Reed was unpersuasive. Moreover, the trial judge determined that the evidence of Fennell’s awareness of a “secret affair” and vow to get revenge was unpersuasive.

c. Statements from Allison and Hawkins

Reed also submitted written statements taken from Jason Allison and Neal Hawkins while they were in custody during the investigation into Arldt’s murder. Both Allison and Hawkins recounted the murder. Hawkins stated that Lawhon confessed to killing Stacey immediately after he killed Arldt. Lawhon told Hawkins that he “did the girl in Bastrop.”

Without judging Allison’s and Hawkins’s credibility, the trial judge found no credible evidence that Lawhon is guilty of murdering Stacey. The judge, who had presided over Reed’s trial, made this determination after recalling the evidence at trial and reviewing the habeas evidence. The habeas evidence specifically included a written statement from Macy, Lawhon’s ex-girlfriend, who would meet Lawhon in the Walmart parking lot, written statements from Lawhon’s parents asserting that Lawhon was at home on the night Stacey was murdered, and information showing that Lawhon had been excluded as a contributor to the DNA on the beer can (item number 24).

d. Fennell’s Deceptive Polygraph Results

Finally, Reed pointed to Fennell’s two polygraph results. The polygraphs were conducted during the investigation into Stacey’s death, and both results indicated that Fennell was deceptive when he was asked if he strangled, struck, or hit Stacey. At trial, the results were offered by Reed’s attorneys, and the trial judge ruled that they were inadmissible. On direct appeal, we affirmed the trial judge’s ruling.³⁷

After exhaustively considering all of the trial and habeas evidence, the trial judge determined that there was no credible evidence that Fennell is guilty of murdering Stacey.

2. Reed’s First Subsequent Application

We now turn to the beer-ean-DNA evidence presented in Reed’s second application. As previously discussed at length above, Reed submitted Young’s DNA-test results on the beer can (item number 24) found on the road near Stacey’s body in his first subsequent state habeas application. Young could not exclude Stacey, Officer Hall, or Investigator Selmala as DNA contributors. But Reed’s trial expert, Dr. Johnson, did exclude all three through Po-lymarker testing.

At the end of 2000, when Reed’s first and second applications were before the trial judge, the State ordered additional, more discriminating DNA testing on the beer can. With intervening advances in DNA testing, Young conducted Short Tandem Repeat (STR) testing on the can and compared the results with the genetic profiles for Stacey, Officer Hall, and Investigator Selmala. STR testing is more discriminating than the previous testing conducted by both parties’ trial experts. Young examined thirteen STR loci. Based on his evaluation of the results, Young was unable to exclude Hall from ten loci. He was also unable to exclude Stacey and Investigator Selmala from five loci. Young documented these findings in a report on January 22, 2001.

Dr. Ranajit Chakraborty, whom Young testified was “one of the country’s most definitive experts in the field of population genetics,” submitted an affidavit concurring with Young’s determination. “Review of the electropherograms indicates that the conclusions reached by the DPS laboratory are accurate and they are scientifically valid.” Dr. Ranajit Chakraborty noted, however, that Young’s results raised questions. He stated that Officer Hall is excluded based on three loci and Stacey and Investigator Selmala are each excluded based on eight loci. “[T]he exclusion of each of the three persons (based on multiple loci) are consistent with the inference that they are NOT part contributors of DNA in the mixture sample (of item # 24).”

In a deposition, Reed’s habeas expert, Dr. Arthur Eisenberg, disagreed with two of Young’s conclusions. In his opinion, the data from Young’s testing did not support the finding that Stacey and Investigator Selmala are included as contributors. Officer Hall, however, cannot be excluded. Dr. Eisenberg opined that DPS’s protocol was correct but stated that his results were obtained using an alternative interpretation method. Explaining DPS’s method, Dr. Eisenberg maintained that DPS protocols mandate a peak-height-minimum of 150 RFU units when making an allele or loci call designation. Another value, a stochastic cutoff level, is 50 RFU units. Dr. Eisenberg asserted:

to make an allele designation, it needs to be a minimum of 150 RFU units to be used for what we refer to as inclusionary purposes. However, there is an area between 50 and 150 RFU where there are peaks that are clearly visually detectable but are typically only used for purposes of exclusion....

Using the lower threshold of 50 RFU units, so that the three loci not previously identified by Young were now visible, Dr. Eisenberg could not exclude Officer Hall. Ninety-nine percent of the Caucasian, African-American, and Hispanic populations would have been excluded. But Dr. Eisen-berg made clear that he could not say whether Officer Hall put his saliva on the beer can. He also stated that the absence of an exclusion would have to be looked at in conjunction with other evidence relating to Officer Hall. As for Stacey and Investigator Selmala, Dr. Eisenberg found no reason to include them as DNA contributors. Because Dr. Eisenberg regarded Dr. Chakraborty as a friend, he spoke with Dr. Chakraborty and showed him the elee-tropherograms where the loci were called at 50 RFU. Dr. Chakraborty, according to Dr. Eisenberg, changed his opinion and agreed that only Officer Hall could not be excluded. Dr. Eisenberg had the impression that Dr. Chakraborty was not given the electropherograms when he reviewed Young’s conclusions. Referring to Dr. Johnson’s previous exclusion of Officer Hall, Dr. Eisenberg stated that if the Poly-marker test was properly conducted, he would have no problem relying on the results.

Considering the results of Young’s 2001 DNA analysis on the merits, the trial judge concluded that Reed’s free-standing-innocence claim did not entitle him to relief. The trial judge found that the jury’s guilty verdict would not have differed if the report had been admitted into evidence at trial. The trial judge also determined that Reed failed to establish by clear and convincing evidence that no reasonable juror would have convicted him in light of the report.³⁸

3. Reed’s Second Subsequent Application

a. Beer-Can-DNA Evidence

Reed again points to the beer-can-DNA evidence in this application. He theorizes that Officer Hall could have assisted Fen-nell with either committing the murder or returning to Giddings from Bastrop on the morning of the murder. Because Officer Hall is six feet, one inch tall, Reed contends that he cannot be excluded as the driver of Fennell’s truck.

In a follow-up affidavit to his deposition testimony, Dr. Eisenberg asserts that his subsequent review of Dr. Johnson’s Poly-marker testing does not change his opinion about the inability to exclude Officer Hall based on the STR results. “The STR systems are in general several times more sensitive at detecting minute amounts of DNA, and the visualization of the STR profiles on an electropherogram is better at discerning mixtures as compared with DQ-Alpha and Polymarker systems.”

b. Officer Davis

Reed further contends that Officer Curtis Davis could have assisted Fennell with either committing the murder or returning home. Reed asserts that Officer Davis reported for the night shift on the night before Stacey was murdered, but he signed out an hour later, taking sick leave. He was then absent from work for a few days after the murder to comfort Fennell.

c. Barnett and Blackwell

Reed also relies on the information that originated with Barnett and Blackwell, the particulars of which we have fully discussed above.

In relation to Reed’s Brady claims, the trial judge made findings on Barnett and Blackwell’s credibility. The trial judge found that both were not credible.

Regarding Barnett, the trial judge found that she was not credible or persuasive for the following reasons:

• Barnett failed to give a satisfactory explanation about why she failed to report her sighting of Stacey and Fen-nell on the morning of April 23rd to police and why she did not report it to anyone until she spoke with Keng, over a year and a half after the murder. Barnett: (1) knew Stacey’s murder was huge news where she worked; (2) agreed that it was common knowledge that H.E.B. offered a $50,000 reward; and (3) was aware that her sighting would have been important to law-enforcement officers investigating Stacey’s murder.

• The timing of Barnett’s disclosure, because of her DWI arrest by Fennell shortly before her disclosure to Keng, suggests an apparent bias and motive underlying her testimony.

• Given Attorney General Investigator Miranda’s testimony and collection of local newspaper articles about Stacey’s murder, Barnett’s credibility is undermined by the fact that she could not have identified Fennell from a photograph in the newspaper because his photo was never in the newspaper.

• Barnett’s credibility is damaged because her testimony about the time she saw Stacey and Fennell was inconsistent with her sworn statements made in her 2002 affidavit. Viewed in conjunction with the State’s argument that it would have been impossible for Barnett to have seen Stacey and Fen-nell at the store between 5:00 and 5:30 a.m., the time discrepancy of Barnett’s sighting at the hearing is significant.

• Barnett’s testimony is incredible when considered with the following facts developed at trial: (1) Stacey was scheduled to be at the Bastrop H.E.B. for work at 3:30 a.m.; (2) Stacey was a prompt employee who was never late to work; (3) Stacey was partially dressed in her H.E.B. uniform when her body was discovered; (4) Fennell’s red truck, which Stacey drove to work, was found at the Bastrop High School at 5:23 a.m. on the 23rd; (5) Stacey had already been killed when Fennell’s truck was found; (6) Carol Stites woke Fennell up at 6:45 a.m. to tell him that Stacey failed to arrive at work; and (7) Carol had to give Fennell a set of keys to her Tempo so he could go look for Stacey.

The trial judge then concluded that Barnett did not see Stacey and Fennell at the Old Frontier store on the morning of April 23rd.

Finding Blackwell’s testimony neither credible nor persuasive, the trial judge entered the following findings:

• Blackwell’s testimony about Fennell feigning grief at Stacey’s funeral is undermined by the testimony of Gid-dings Police Chief Nathan Lapham. Lapham testified that Fennell “appeared to be very upset, emotionally upset, he was crying, I believe before and after the funeral.... He was very distraught.”

• Blackwell’s testimony is severely undermined by Wolfe’s testimony that none of the other cadets in the CAPCO class could corroborate the conversation that Blackwell said that she had had with Fennell in which Fennell threatened to strangle his girlfriend with a belt if he ever caught her cheating on him. Further, none of the cadets could corroborate Blackwell’s claim that Fennell acted abusively toward Stacey.

• Blackwell’s testimony is undermined by the testimony of Dezarn, who was assigned to sit next to Fennell during the class. Dezarn never heard or participated in the conversation with Fen-nell that Blackwell described.

• Larry Franklin’s testimony also undermines Blackwell’s testimony. Franklin did not hear the conversation described by Blackwell and never heard Fennell say anything disparaging about Stacey.

• Blackwell’s testimony that she did not make a connection between Stacey’s murder and Fennell’s statement until she spoke to Vasquez is undermined by her testimony that she recalled Fennell’s statement when she returned to work after the funeral and Franklin’s testimony that he and Blackwell discussed whether Fennell could have killed Stacey. And during that conversation, Blackwell told Franklin about Fennell’s statement.

• Blackwell’s testimony is undermined because, as a peace officer, she failed to report information relevant to a homicide investigation.

• Blackwell’s credibility is undermined by the fact that she originally told Vasquez that Fennell was joking when he made the statement but testified at the hearing that Fennell was “absolutely” serious.

• “Given the fact that Blackwell lived in Austin, attended [Stacey’s] funeral, and knew Fennell from the academy, it is implausible that Blackwell was entirely unaware, as she claimed, of the circumstances surrounding [Stacey’s] death.”

d. Statements of Jennifer and Brenda Prater

Reed has submitted affidavits from Jennifer and Brenda Prater. Both women claim that they saw Stacey with a man, who was not Reed, in the early morning hours of April 23rd. Jennifer maintained that her husband, Paul, woke her up that morning because there was a suspicious car behind their house. The car was light in color, and Jennifer did not recognize the driver or the passenger. The man in the driver’s seat had a dark complexion but was not an African-American. The woman in the passenger’s seat was pale and had “big” ham. Jennifer and Paul went outside to get a better look at the occupants. Jennifer recalled that the two people in the car saw them and drove off. Jennifer stated that she got a good look at the two because the interior light in the car was on. Later she saw a picture of Reed and was sure that he was not the driver of the car. The man in the driver’s seat had a lighter skin tone and different facial features. When Jennifer’s mother-in-law showed her Stacey’s picture in the paper on the 25th, Jennifer knew that Stacey was the woman who had been in the passenger’s seat.

On the 25th, Jennifer’s mother-in-law told her that the police had been to Jennifer’s house when she was not home. The police walked into the house when Jennifer’s kids did not answer the door. Jennifer’s mother-in-law, who lived across the street from Jennifer and Paul, saw the police arrive. She went to Jennifer and Paul’s house and confronted the police about their entry into the house. Jennifer’s mother-in-law told Jennifer that the police threatened to call Child Protective Services because the kids were home alone. After Jennifer’s mother-in-law explained that she was watching the children until Jennifer returned home, the police left, stating that they would come back later.

When the police returned later that day and spoke to Jennifer, they asked her about the car she saw on the 23rd. Aware that she was lying, Jennifer told the officers that she did not know anything. Jennifer did not want to be involved in a criminal investigation, did not trust the police, and was angry at the police for enteiing her house.

Brenda Prater lived in a house a block away from her brother, Paul, and her sister-in-law, Jennifer. During the early morning hours on April 23rd, she was writing in her journal. She was awake because her husband, whom she was in the process of divorcing, called and harassed her. She called Paul and asked him to keep an eye out for her husband. Between 1:00 and 3:00 a.m., while Jennifer was sitting outside in her front yard, she saw a light-colored car pass by with three occupants.

The interior light was on. The driver was a man who had a darker eomplection [sic], but was not black. I thought that he was Mexican. There was a woman in the passenger seat. She was light complected [sic] with big dark hair. I remember that, as the car drove by the first time, the woman in the passenger seat turned her head toward the driver. I got a very good look at her face as they went by. There was a white male in the back scat [sic]. At first I was afraid that the man in the back seat was my husband. I got a better look at him when the car went through the second time and realized that he was not my husband. I later spoke with my brother Paul and Jennifer and they told me they saw the same car in back of their house.

Jennifer went to work the next day, and a coworker asked if she heard about Stacey’s murder and showed her Stacey’s picture in the newspaper. Realizing that she had seen Stacey on the night she disappeared, she began to 'yell, “When, when, when[?]”

e. Fennell’s Deceptive Polygraph Results

Again, Reed directs our attention to Fennell's polygraph exams, which led both examiners to conclude that he was deceptive when asked about Stacey’s murder.

f. Faulty Forensic Analysis and Collection of Forensic Evidence

Reed asserts that the various aspects of the forensic testimony offered by the State and admitted into evidence at trial lack a foundation in science. To support his theory, Reed refers to an affidavit from Dr. Leroy Riddick, an affidavit from Ronald Singer, and to medical literature. Reed included each of these items in his habeas application.

Dr. Riddick is a medical examiner for the State of Alabama. His affidavit, for the most part, is devoted to criticizing Dr. Bayardo’s conclusions. First, Dr. Riddick contends that Dr. Bayardo’s time of death estimate is unreliable.

In order for the time of [Stacey’s] death to have been reliably determined, rigor mortis, post mortem lividity, and body temperature should have been recorded at the scene where her body was found. These measurements are the most common means for calculating time of death, but none of this information was recorded at the scene in this case. By the time Dr. Bayardo saw [Stacey’s] body for the autopsy at 1:50 p.m. on April 24, the body had been handled by multiple people, turned and had been refrigerated for nearly a full day. Consequently, it was too late for accurate assessments of rigor mortis, lividity, and body temperature to produce a reliable determination of time of death.

Next, Dr. Riddick asserts that the evidence of anal intercourse is inconclusive. Dr. Riddick faults Dr. Bayardo for failing to preserve the slides that he used to determine the presence of intact sperm in Stacey’s rectum. “[W]ithout the slides on which Dr. Bayardo claimed to have seen the presence of sperm heads from swabs from [Stacey’s] rectum, this conclusion cannot be verified.” Dr. Riddick states that Dr. Bayardo’s conclusion about the presence of intact sperm would be more reliable if the rectal swabs he used had been taken at the scene. Noting the pubic-hair-tape lifts between Stacey’s labia and rectum conducted by Blakley and the manner in which Stacey’s body was moved, Dr. Riddick contends that “there were several opportunities for leakage by the time that Dr. Bayardo took the rectal swabs.” Dr. Riddick asserts that “there is no evidence of anal dilation at the time that [Stacey’s] body was recovered.” When Dr. Bayardo examined Stacey, it was twenty-four hours after she died. “Rigor mortis begins to pass 24 hours after death and makes dilation of the anus easier, whether by finger, swab or another object.” Dr. Riddick also maintains that “it cannot be concluded with any degree of scientific certainty that [Stacey’s] anus was lacerated and that those lacerations occurred around the time of death.” The photographs taken at the autopsy do not show redness, according Dr. Riddick. Redness is associated with laceration and “would have accompanied a laceration incurred at or around the time of death.” Regarding Bayardo’s testimony, Dr. Riddick states that Dr. Bayardo mentioned lacerations on direct-examination but described them as scrapes on cross-examination. Further, Dr. Bayardo’s report mentioned only abrasions. Dr. Riddick contends that lacerations and abrasions are not the same.

Lacerations of the anus could be consistent with anal intercourse. In contrast, abrasions of the anus are not as accurate an indicator of anal intercourse, much less an anal assault. Abrasions of the anus can and do occur naturally, for example, due to constipation or hemorrhoids .... ‘Scrapes’ would be consistent with an abrasion, not a laceration.

As for the cause of death, Dr. Riddick attacks Dr. Bayardo’s conclusion that Stacey died of asphyxiation resulting from strangulation associated with sexual assault. Dr. Riddick claims that it is unknown whether the sexual contact was consensual. “The best indicator of non-consensual sexual contact is the existence of other injuries, such as being held down. There are no injuries of this type.” Dr. Riddick contends that there is no evidence that Stacey died from ligature strangulation because, with the exception of the exterior injuries to Stacey’s neck, other common injuries associated with strangulation were not present. Dr. Riddick suggests “some other modality, such as smothering....”

Finally, regarding the collection of evidence at the scene where Stacey’s body was found, Dr. Riddick claims that no effort was made to collect evidence from under Stacey’s fingernails. Even though Stacey’s nails were short, Dr. Riddick claims that the investigators could have collected evidence with a toothpick. Dr. Riddick contends that the video of the crime scene does not show the authorities, who collected evidence, changing gloves between tasks. In making this critique, Dr. Riddick admits that he is an expert in crime-scene evidence collection only in so far as it relates to establishing time and cause of death.

Dr. Riddick viewed the following items of evidence in rendering his opinion: (1) Dr. Bayardo’s autopsy report; (2) photographs of Stacey’s body at the scene of discovery, Stacey’s clothing, and Stacey’s body at the autopsy; (3) the video of the scene where Stacey’s body was discovered; (4) the trial testimony of Dr. Bayardo; (5) the trial and habeas testimony of Blakley; (5) DPS Crime Lab reports; (6) crime scene reports; and (7) police reports of witness interviews.

In his affidavit, Singer, a consulting forensic scientist who works at Tarrant County Medical Examiner’s Office Crimi-nalistics Laboratory, focuses on problems with the investigation of the scene where Stacey’s body was discovered and with Blakley’s testimony. Singer offers some of the following personal observations and conclusions:

• Law enforcement officials exercised poor security and control at the scene where Stacey’s body was discovered; “a perimeter should have been established around the scene with only one entrance and exit. Entrance and egress should have been limited until that area was thoroughly searched for tire prints, shoe prints, and other evidence.”

• “The law enforcement authorities depicted on the tape demonstrated poor technique in dealing with, and taking evidentiary samples from [Stacey’s] body.” The origin of the blanket used to cover her body is unknown and the video does not show whether it was inspected for trace evidence. “It was not good technique for one of the crime scene analysts to put his gloved hand into his pocket — as the video shows — and then later handle trace evidence without having changed to fresh gloves.” Ungloved personnel touched the body and “one of the individuals who moved [Stacey’s] body to a stretcher does not appear to be gloved.”

• It is probable that Blakley contaminated Stacey’s bra and breasts with trace evidence because there is no evidence that the criminalist changed gloves between taking evidentiary samples. “This contamination likely occurred when Ms. Blakley handled [Stacey’s] brassiere and breasts after taking swabs and tape lifts from [Stacey’s] pubic area.”

• It is troubling that the criminalist failed to swab the pieces of belt for DNA evidence. “If, as the prosecution later theorized, this belt was used to stoangle [Stacey], the pieces likely would have had trace DNA evidence from [Stacey] and her attacker.”

• The videotape was poorly done; there is no time marker and it begins after the crime scene team arrived. As a result, it does not completely depict the activities of law enforcement. “A better practice is to record continuously from the moment that the police get to the scene after the first responding officer.” “[A] valuable record of evidence collection — which could support or impeach the integrity of the prosecution’s evidence — has been lost.”

• Blakley exhibited poor forensic practices and repeatedly went beyond her area of expertise. Blakley testified beyond her expertise when stating how long Stacey was dead, identifying marks on Stacey’s body and dating the marks, and opining that it was a crime of passion. “[T]raining as a criminalist does not give one the ability to estimate how long someone has been dead. This determination is the province of a pathologist.” And “[o]nly pathologists can determine that a mark is, in fact, a bruise, cigarette burn, scratch or bite, or how old the mark is and how it was' incurred.”

Finally, Reed cites to a book written by Dr. William Green in 1998, entitled: “Rape: The Evidential Examination and Management of the Adult Female Victim.” In the book, Dr. Green surveys studies conducted on the presence of nonmotile intact sperm in the cervix and vagina. Dr. Green notes that one study found intact sperm ten days after intercourse.³⁹ Other studies found the presence of intact sperm in the cervix or vagina anywhere from two days to nine days after intercourse.⁴⁰ Reed contends that Blakley’s testimony estimating the length of time that sperm can remain intact in the “cervix” is patently false.

g. Fennell and the Giddings Police Department’s Reputation for Violence

Reed maintains that both Fennell and the Giddings Police Department have a reputation for violence. Concerning Fen-nell, Reed points to a state-eivil-rights lawsuit filed against the City of Giddings, Giddings Police Chief Dennis Oltmann, Giddings Officer Nathan Lapham and Fen-nell for using excessive force against suspects a year before Reed’s trial.

Reed also asserts that Fennell was violent toward women he dated. Reed directs us to an affidavit from Pamela Duncan, Fennell’s girlfriend from August 1996 to September 1997. Duncan describes Fennell as abusive, possessive, controlling, and extremely prejudiced toward African-Americans. When Duncan broke up with Fennell, he stalked her until he left Gid-dings; she was afraid for her safety and that of her children.

He would drive by my house, night after night, and shine a spotlight into the house. It got so bad that I finally put tin foil up in my windows, to reflect the light. He would stand outside my house at night, screaming at me, calling me a ‘bitch’ and other obscenities. He would come by my job at the Circle K, and just sit parked out front, with the headlights shining into the store. He would stay there, sitting in his car and watching me, for anywhere from two minutes to two hours ... Once he came into the store and wouldn’t let me out of the office— we had to call the police to get someone to escort him out, so I could leave. He would hassle any guy I tried to date until it scared them away. For instance, I dated one guy who delivered beer in town. After we started dating, Jimmy sta[r]ted pulling him over and giving him tickets. He got so many tickets he couldn’t keep his job anymore.

Summarizing the end of her relationship, Duncan states that it was the worst time in her life.

Claiming that the Giddings Police Department had a long-standing reputation for brutalizing suspects and targeting non-whites at the time of Reed’s trial, Reed relies on a federal-civil-rights action initiated against the Giddings Police Department and another Giddings officer. Attached to the plaintiffs petition in that case is an affidavit from Keng. Keng recalled several instances of alleged misconduct involving officers with the Gid-dings Police Department using excessive force and recalled some specific instances of alleged misconduct. He also recalled requesting that the Texas Rangers investigate abuse allegations when Chief Olt-mann failed to give him a satisfactory explanation about the alleged abuse. A Ranger told Keng that there was not much he could do because Chief Oltmann was supporting his officer. In closing, Keng stated: “For the past ten years, the Giddings Police Department has had a reputation in Lee County of roughing up suspects during then- arrest.”

h. Statement of James Robinson

James Robinson contends that he had a separate relationship with Stacey and Reed. Robinson knew Reed from the nursing home where they worked and knew Stacey from “school.” Robinson saw Stacey and Reed together on numerous occasions, kissing and calling each other “baby.” He went to parties where Stacey and Reed would meet. Lawhon would often be at the same parties, and Stacey would say hello to him. After Stacey was murdered, Reed seemed sad and angry. Robinson was in the Bastrop County Jail while Reed was being held there on this case. At the jail, Reed told Robinson that he did not kill Stacey. Robinson was told that he would be transferred to another county and that he could not stay in Bas-trop to testify at Reed’s trial. Robinson also declares that he was with Chris Al-dridge and Reed when Fennell stopped them, telling Reed that he knew about Reed’s relationship with Stacey and that he would pay.

4. Discussion

We hold that all of the reliable evidence, both old and new, presented by Reed does not compel the conclusion that it is more likely than not that no reasonable juror would have voted to convict Reed.⁴¹ Initially, we note that what separates this case from the majority of gateway-innocence cases is the complete lack of a cohesive theory of innocence. Reed’s claim of innocence is seriously disjointed and fragmented' — -he presents numerous alternative but critically incomplete theories. By focusing on a romantic relationship between himself and Stacey as well as pointing to several alternative suspects— Fennell, Lawhon, and some unknown dark-skinned man — the new evidence before us fails to tell a complete, rational exculpatory narrative that exonerates Reed. None of Reed’s theories meets the gateway standard of innocence.

As Chief Justice Roberts recognized in his concurring opinion in House v. Bell, “Implicit in the requirement that a habeas petitioner present reliable evidence is the expectation that a factfinder will assess credibility.”⁴² Here, consistent with our writ jurisprudence, we follow the credibility determinations and factfindings made by the two judges who presided over Reed’s habeas proceedings. Both judges had the opportunity to assess the demean- or of the witnesses who appeared before them. Further, the trial judge who presided over Reed’s first and second habeas proceedings also presided over Reed’s trial. Based on our review of the record, the findings entered by the trial judges and discussed above are supported by the record; thus, in several instances Reed has failed to provide us with reliable evidence of innocence. The evidence that we reject as unreliable includes: the Robbinses’ statements; the witnesses who affirmed a relationship between Reed and Stacey; Allison’s and Hawkins’s statements, even if regarded as credible; and the information from Barnett and Blackwell. Further, regarding Barnett’s sighting, given the evidence developed during the habeas proceedings about Officer Hall’s alibi, which has not been undermined, and the lack of any reliable evidence suggesting that Fen-nell had an accomplice, we conclude that Barnett’s information is not credible or reliable.

Additionally, we find that Robinson’s statement is not credible for several reasons. First, his statement is not sworn. Second, he contends that he knew Stacey from school and that, as of 2000, he has known Reed for eight or nine years. The evidence at trial, however, establishes that Stacey moved to Bastrop after graduating from Smithville High School; therefore, Robinson’s statement is suspect. Third, Robinson’s statements about seeing Stacey and Reed together are general; Robinson offers no specific facts that have been or could be corroborated. Fourth, this statement lacks credibility because Jon never mentioned that Robinson was present when Reed was threatened by Fennell, even though he gave two statements.

Based on the above, we refuse to credit the foregoing evidence in assessing whether Reed has made a prima facie showing that, in light of all of the evidence before us, no reasonable juror would have convicted him. We now consider the remaining new evidence as it relates to the various alternative theories of innocence offered by Reed.

a. Fennell

Excluding the items of evidence that we have rejected, we consider the following evidence that, according to Reed, suggests Fennell’s involvement in Stacey’s murder: (1) Fennell’s deceptive polygraph results, regardless of their admissibility,⁴³ even though we question them reliability;⁴⁴ (2) the DNA-beer-can-test results that cannot exclude Officer Hall; (3) evidence that Fennell’s coworker, Officer Davis, took sick leave shortly after beginning his shift on the night of April 22nd; and (4) evidence that Fennell and the Giddings Police Department had a reputation for violence.

Although this new evidence may indeed arouse a healthy suspicion that Fennell had some involvement in Stacey’s death, we are not convinced that Reed has shown by a preponderance of the evidence that no reasonable juror, confronted with this evidence, would have found him guilty beyond a reasonable doubt. The evidence of vaginal assault, which we will discuss more fully below, and the circumstantial evidence admitted against Reed at trial have not been undermined and still support a guilty verdict.

b. Consensual Sexual Relationship

Reed contends that the evidence from Dr. Riddick, Singer, and Dr. Green’s book establishes only that he and Stacey had sexual relations at some point before her death and that there is no credible evidence that Stacey was raped. We disagree. When considered in conjunction with the trial evidence, Reed’s new evidence does not verge on establishing that it is more likely than not that no reasonable juror would have convicted Reed.

Reed contends that the evidence of anal intercourse is inconclusive. From our reading of Reed’s briefing, it is apparent that Reed theorizes that, if Stacey was not anally sodomized, then the uncontested forensic evidence of vaginal intercourse was from a consensual encounter and Reed is therefore not her killer. This theory is illogical. Any deficiency in the evidence suggesting anal intercourse does not necessarily support Reed’s theory that he and Stacey engaged in consensual vaginal intercourse. Likewise, evidence of anal intercourse does not conclusively establish that the encounter was forced. Nevertheless, the competing evidence that semen leaked from the vagina to the anus was before the jury. Blakley stated that she did not see a significant amount of leakage in Stacey’s underwear and therefore could not conclude that semen from her vagina was transferred into her rectal cavity. Blakley observed only “[f]our small, maybe less than dime-sized spots” of semen in Stacey’s underwear, which was atypical for a significant amount of leakage. According to Blakley, this indicated that Stacey did not move much after intercourse. Game cross-examined Blakley about the movement of Stacey’s body after it was discovered. Blakley stated that she rolled Stacey’s body onto the stomach so they could look at the back side. Stacey’s body was then rolled back, transferred onto the gurney, and transported to Dr. Bayardo’s office in Austin.

Dr. Johnson claimed that leakage from the vagina is common and stated that semen could be detected in areas surrounding the vagina, including the anal area. Movement of the body, according to Dr. Johnson, makes leakage more likely. She added, “A very small number of sperm that would be collected in an area would [be] much more likely to come from a contamination of the swab touching one area as it’s inserted into another, or drainage from around that area.” Semen in low numbers is not indicative of an ejaculate and is more likely to be discovered due to leakage.

Dr. Riddick’s contentions that moving Stacey’s body created several opportunities for leakage, which in Reed’s view supports his theory that there was no anal intercourse, was presented to the jury and is therefore cumulative. Because of this, we cannot say that Reed’s new evidence regarding leakage would have had any appreciable impact on the jury’s verdict.

Additionally, Dr. Riddick’s opinions that there is no evidence that Stacey’s anus was dilated and that it cannot be concluded with any degree of scientific certainty that Stacey’s anus was lacerated merely presents differing opinions that a jury could reject.

In any event, when the conflicting evidence about anal penetration is viewed in conjunction with the evidence at trial, Reed has not shown by a preponderance of the evidence that no reasonable juror would convict him.

Compelling, independent circumstantial evidence showed that Reed forced Stacey to have vaginal intercourse. When discovered, Stacey’s body was partially disrobed, her pants were unzipped, the top of her pants were parted, the zipper was broken and was “jammed down onto the metal that holds — the piece of metal that clamps the zipper together, a tooth from the zipper was pulled off and missing, and her underwear was bunched down around her hips.” Contrary to Dr. Riddick’s opinion that Stacey had no other injuries consistent with an assault, Blakley noted a darkened area on the inside of the elbow on Stacey’s left and right arms. The bruise was there before Stacey died because bruising does not occur or increase after the heart stops beating. On Stacey’s right arm, Blakley also noted a mark that was “very consistent with a fingernail being dug into the flesh.” Blakley believed these marks suggested physical violence. Blakley opined that the bruises resulted from “a small area of pressure being applied to the skin, either from a fingertip or instrument, something sharp but localized.” Regarding her ability to differentiate between old and recent bruises, Blak-ley noted that Stacey had older yellow and green bruises on her upper thighs, which were consistent with Stacey carrying boxes at H.E.B. Dr. Bayardo documented pre-mortem injuries to Stacey’s head that suggested that she had been hit with a closed fist.

Furthermore, Stacey’s life circumstances leading up to her death strongly support a finding that she did not willingly participate in vaginal intercourse with Reed. When Stacey was murdered, the wedding she carefully planned and helped pay for by working the early-morning shift at the H.E.B. was only eighteen days away; she and Fennell were in the final stages of preparing for the wedding and their future as a married couple. Stacey devoted every free moment of her time to planning the details of the wedding. Her mother, Carol, suffered from a nervous condition that caused her to get depressed. When Carol’s exhaustion and stress from helping Stacey with the planning came to a head the day before Stacey was murdered, Carol asked Stacey if she was certain that she wanted to marry Fennell. Stacey reassured her mother, stating, “I love Jimmy[,] and I’m going to marry him.” Stacey also told her mother that her mother needed to get over her anxiety about the wedding.

The evidence at trial also establishes that Stacey consistently arrived to work on time. Further, Stacey’s body was partially dressed in the H.E.B. uniform when it was discovered. This shows that she was en route to work and fully intended to be there at 3:30 a.m., as scheduled. Stacey was murdered at some point before 5:23 a.m., when Officer Alexander first noticed Fennell’s truck at the High School with a piece of Stacey’s belt lying on the ground outside the door. Finally, despite Reed’s efforts, he presents no credible evidence showing that he had a romantic relationship with Stacey.

The State also presented relevant circumstantial evidence implicating Reed at trial. For example, authorities knew Reed routinely walked around Bastrop late at night and in the early morning. Authorities frequently saw Reed walking near the roads that Stacey traveled on her way to the H.E.B. Further, it was convenient for Reed to leave Fennell’s truck parked at the Bastrop High School. The School was near Reed’s house, where Reed walked at odd hours. With Reed’s height at six feet, two inches, the position of the driver’s seat and the rearview mirror also supports the State’s theory that Reed was the last person who drove the truck. Importantly, Reed denied knowing Stacey when he was first questioned by authorities. This made Reed’s claim of a consensual sexual relationship, offered for the first time at trial, look like a manufactured and implausible explanation to account for the presence of his semen.

Reed also takes issue with Blakley’s testimony about the viability of sperm, and in doing so, Reed points to Dr. Green’s survey of studies on nonmotile-intact sperm. To Reed, the small amount of leakage of semen from Stacey’s vagina is consistent with Stacey having sex with Reed at least a day before her death. The studies cited by Dr. Green do not fully support Reed’s contentions. For example, the study that reported finding intact sperm after ten days was based on an analysis of cervico-vaginal scrapings. In this case, Blakley used vaginal swabs. Next, Blakley’s testimony that the outside length of time for finding the presence of intact sperm is twenty-six hours was not the only testimony on the issue. When Dr. Bayardo conducted the autopsy at 1:50 p.m. on the 24th and obtained his own vaginal swabs, he documented the presence of intact sperm and testified at trial that this meant the semen was introduced into the vagina a day or two before the exam. Thus, Dr. Bayardo’s estimate about the length of time that sperm can remain intact in the vagina exceeded the length of time that Blakley testified to and is consistent with Reed’s theory that he and Stacey had consensual vaginal intercourse at least a day before her death. Furthermore, even if we assume that Blakley and Dr. Bayardo underestimated the length of time that sperm will remain intact, we conclude that, given the other evidence in this case, Reed has failed to meet his burden.

Finally, citing cross-contamination, Reed contends that testimony at trial that the breast swabs taken by Blakley contain saliva is unreliable. Reed claims that the State simply found epithelial cells, which are present in semen along with sperm. In support of this, Reed relies on Dr. Riddick’s statement that it is likely that Blakley contaminated Stacey’s breasts with trace evidence. At trial, Dr. Johnson testified that the swabs taken from Stacey’s breasts contained saliva samples. Dr. Johnson identified the substance as saliva based on an amylase test. Amylase is a primary component of saliva, according to Dr. Johnson. Dr. Johnson stated that it was likely that the saliva got there after Stacey’s last shower, which was the night before she was murdered.

To refute Dr. Johnson’s testimony, Reed points to Singer’s affidavit. Singer maintains:

Amylase testing is a procedure that is helpful as a screening device. It is, however, a general test and cannot be relied upon to identify a specific body fluid such as saliva with accuracy. We have discovered, for example, that amylase testing routinely indicates a presumptive positive in reaction to certain plant matter as well as vaginal fluid and non-human body fluid.

The possibility that the substance on the breast swabs was not saliva was before the jury. Dr. Johnson stated that amylase is found in other fluids. Furthermore, Singer offers only an alternative theory that the jury could have chosen to disregard. However, even if we assume that the type of substance is unreliable because of cross-contamination, considering the evidence at trial, it is highly unlikely that any reasonable juror would view the presence of Reed’s semen in Stacey’s vagina as the byproduct of a intimate, consensual interlude between the two.

c. The Unidentified Male or Men

The statements from Jennifer and Brenda Prater also fail to make a threshold showing of innocence. First, we question their reliability because they did not come forward with this information until September 2002, even though the investigation into Stacey’s death was well known in Bas-trop.⁴⁵ Further, we find that Jennifer’s credibility is also suspect because her husband, Paul, failed to corroborate his wife’s account in an affidavit. However, we need not linger on this point. This evidence has no continuity with any of the other new evidence offered by Reed and does not fit within the chronicle of events that the trial evidence supports. Thus, when the information about Stacey from Jennifer and Brenda is viewed alongside the evidence at trial, we cannot say that Reed has established that it is more likely than not that no reasonable juror would have convicted him.

Because, after reviewing the cumulative force of all the foregoing evidence, Reed has failed to satisfy the gateway standard under Article 11.071, Section 5(a)(2), we refuse to reach the merits of Reed’s Brady and ineffective assistance of counsel claims.

V. Conclusion

In reviewing Reed’s Brady claims that satisfied Article 11.071, Section 5(a)(1), we hold that Reed has failed to show that the State did not disclose favorable evidence. We also hold that Reed has not made a threshold, prima facie showing of innocence by a preponderance of the evidence under Article 11.071, Section 5(a)(2). Therefore, we refuse to consider the merits of Reed’s other constitutional claims. We deny relief.

KELLER, P.J., filed a concurring opinion.

PRICE, J., filed a concurring opinion.

WOMACK, J., concurred.

1 . Reed v. State, No. AP-73, 135 (Tex.Crim.App. Dec. 6, 2000) (not designated for publication), cert. denied, Reed v. Texas, 534 U.S. 955, 122 S.Ct. 356, 151 L.Ed.2d 270 (2001).

2 . Id. at *9.

3 . Id. at *22.

4 . Ex parte Reed, No. WR-50,961-01 (Tex. Crim.App. Feb. 13, 2002) (not designated for publication).

5 . 373 U.S. 83, 83 S.Ct. 1194, 10 L.Ed.2d 215 (1963).

6 . Ex parte Reed, No. WR-50,961-02 (Tex.Crim.App. Feb. 13, 2002) (not designated for publication).

7 . Id.

8 . Reed v. Dretke, No. A-02-CA-142-LY (W.D.Tex., Mar. 22, 2004).

⁹

10 . 506 U.S. 390, 113 S.Ct. 853, 122 L.Ed.2d 203 (1993).

11 . See Cox Tex. Newspapers, L.P. v. Penick, 219 S.W.3d 425, 432-33 (Tex.App.-Austin 2007, pet. denied).

12 . Strickler v. Greene, 527 U.S. 263, 280, 119 S.Ct. 1936, 144 L.Ed.2d 286 (1999).

13 . 514 U.S. 419, 438, 115 S.Ct. 1555, 131 L.Ed.2d 490 (1995); see also Thomas v. State, 841 S.W.2d 399, 402-04 (1992) (tracing history and developments of Brady rule).

14 . Strickler, 527 U.S. at 281-82, 119 S.Ct. 1936.

15 . United States v. Bagley, 473 U.S. 667, 682, 105 S.Ct. 3375, 87 L.Ed.2d 481 (1985).

¹⁶

17 . Ex parte Young, 418 S.W.2d 824, 829 (Tex.Crim.App.1967); Ex parte Adams, 768 S.W.2d 281, 288 (Tex.Crim.App.1989); Ex parte Van Alstyne, 239 S.W.3d 815, 817 (Tex.Crim.App.2007) (per curiam).

18 . Ex parte Van Alstyne, 239 S.W.3d at 817 (quoting Ex parte Simpson, 136 S.W.3d 660, 669 (Tex.Crim.App.2004)).

19 . Ex parte Simpson, 136 S.W.3d at 668.

20 . Ex parte Van Alstyne, 239 S.W.3d at 817.

²¹

22 . Ex parte Adams, 768 S.W.2d at 288 (citing Ex parte Davila, 530 S.W.2d 543, 544 (Tex.Crim.App.1975); Ex parte Bagley, 509 S.W.2d 332, 335 (Tex.Crim.App.1974); Ex parte Williams, 486 S.W.2d 566, 568 (Tex.Crim.App.1972)). See e.g., Ex parte White, 160 S.W.3d 46, 51-55 (Tex.Crim.App.2004).

23 . See generally Petition for Writ of Certiorari, District Attorney's Office for the Third Judicial District, et al. v. Osborne, - U.S. -, 129 S.Ct. 488, 172 L.Ed.2d 355 (No. 08-6), granted Nov. 3, 2008 (arguing that the Ninth Circuit Court of Appeals in 42 U.S.C. § 1983 civil rights action improperly "created a post-conviction right of access to evidence under the Due Process Clause by extending the doctrine of Brady ... and its progeny.”).

24 . 513 U.S. 298, 115 S.Ct. 851, 130 L.Ed.2d 808 (1995).

25 . Tex.Code Crim. Proc. Ann. art. 11.071 § 5(a)(2) (Vernon Supp.2008).

26 . Ex parte Brooks, 219 S.W.3d 396, 400 (Tex.Crim.App.2007).

27 . Schlup, 513 U.S. at 315, 115 S.Ct. 851.

28 . Ex parte Brooks, 219 S.W.3d at 399.

29 . Id. at 324.

30 .Id. at 327.

31 . House v. Bell, 547 U.S. 518, 537-38, 126 S.Ct. 2064, 165 L.Ed.2d 1 (2006) (quoting Schlup, 513 U.S. at 327-28, 115 S.Ct. 851).

32 . Id. at 538, 126 S.Ct. 2064.

33 . Id. at 539, 126 S.Ct. 2064.

34 . See Jay Nelson, Note, Facing up to Wrongful Convictions: Broadly Defining “New” Evidence at the Actual Innocence Gateway, 59 Hastings L.J. 711, 718-20 (2008) (surveying approaches adopted by federal circuit courts in defining new evidence under Schlup standard).

35 . Compare with Ex parte Brown, 205 S.W.3d 538, 545-46 (Tex.Crim.App.2006) (discussing what constitutes new evidence for purposes of a substantive claim of innocence under Ex parte Elizondo).

36 . 947 S.W.2d 202 (Tex.Crim.App.1996).

37 . Reed, No. AP-73,135. at *12-14.

38 . See Ex parte Elizondo, 947 S.W.2d at 208.

39 . William M. Green. M.D., Rape: The Evidential Examination and Management of the Adult Female Victim 107 (Lexington Books 1988).

40 . Id. at 107-08.

41 . Tex.Code Crim Ann. art. 11.071 § 5(a)(2).

42 . 547 U.S. at 556, 126 S.Ct. 2064; see also Schlup, 513 U.S. at 332, 115 S.Ct. 851.

43 . See Schlup, 513 U.S. at 327-28, 115 S.Ct. 851.

44 . See United States v. Scheffer, 523 U.S. 303, 309-12, 118 S.Ct. 1261, 140 L.Ed.2d 413 (1998).

45 . Goldblum v. Klem, 510 F.3d 204, 230 (3rd Cir.2007) (considering the timing of a disclosure and credibility of a witness in assessing the probable reliability of a statement).