Robert Miller was accused of breaking into a woman's home on September 2, 1986, and raping and murdering her. He was also accused of trying to break into another home early on Thanksgiving Day 1986, and of breaking into a third victim's home and raping and murdering her.

From the evidence, Oklahoma police officers knew that the suspect was a black male and an A secretor. They canvassed neighborhoods and asked black men to provide blood samples. Miller was one of twenty-three men who obliged and was found to be an A secretor. Police then took him, willingly, to police headquarters in order to give more samples. While at the station, police asked Miller if he could help them solve these crimes and Miller responded by saying he could see through the killer's eyes. The police, who had a hidden camera set up, taped the twelve-hour interview that followed. In this interview, police detectives took advantage of Miller's fragile mental state, playing along with Miller's admission that he had special powers and was deeply religious, even joining him in group prayer. Police eventually led him to what the State called an admission of guilt. Later, an examination of the taped interviews revealed many inconsistencies between Miller's statements and the actual circumstances of each crime. There were several key differences; for example, Miller claimed that one of the victim's was only a little older than himself. She was eighty-three at the time of her death and Miller was twenty-eight.

Miller was charged with two counts of rape in the first degree, two counts of murder in the first degree, two counts of burglary in the first degree and one count of attempted burglary in the first degree. His trial was conducted between May 9 and May 19, 1988. At trial, the jury found the state's allegations that the murders were especially heinous, atrocious or cruel, that they were committed for the purpose of avoiding or preventing a lawful arrest or prosecution, and that a probability existed that the defendant would commit criminal acts of violence that would constitute a continuing threat to society to be true, sentencing Miller to death for the two counts of murder. For the first count of rape, Miller was sentenced to 287 years in prison, and for the second count of rape, he was sentenced to 392 years. For the two counts of burglary, he was sentenced to twenty years each, and for the count of attempted burglary, he was sentenced to ten years. This sentence was formally imposed on May 27, 1988.

On April 6, 1988, 131 pieces of evidence were submitted to the Oklahoma Public Defender's Office, whereupon a serological examination was requested. Many of the samples contained semen but, at that time, the examination could only conclude that the semen donor was an A secretor, meaning the donor's blood was Type A and the type could be determined from bodily fluids. In 1988, Allo-Type Genetic Testing Inc., performed immunoglobulin allotyping from evidence submitted by Joyce Gilchrist from the state laboratory. The findings revealed that foreign allotypes were detected in the nightgown stain, the stained sheet, and two of the mixed blood and body fluid stains found on the sheet of one victim were consistent with Robert Miller and not with Ronald Lott, another suspect. Further, these foreign allotypes are found 33.33% of the time in the black population and only .56% of the time in the Caucasian population. This information translated to the statistic that there was a 98.33% probability that the assailant was black and only 1.67% chance that the assailant was white.

At trial, Janice Davis Lyhane testified that the donor of the stain found on the sheet of the other murder victim was a blood type A secretor. This finding matched Miller's blood profile. The samples of saliva and blood were obtained by Ms. Lyhane from Joyce Gilchrist. Lyhane testified that, upon examining the hair samples obtained at the crime scene, all that could be concluded was that the hair had "Negroid characteristics," thereby excluding the victim as the possible donor. However, Lyhane could not come to any other conclusions about the origin of the hair sample. In cross examination, Lyhane conceded that two scalp hairs found on a pink blanket belonging to the victim could not have originated from Miller. In cross examination, the defense tried to raise the issue of DNA, but because the witness was not knowledgeable of DNA testing procedures, and DNA evidence had not yet been accepted by Oklahoma courts, the court only allowed the defense to ask questions pertaining to the extent of Lyhane's knowledge of DNA testing in general, and not referring specifically to Miller's case.

Joyce Gilchrist also testified for the state. In her testimony, she described the PGM subtyping she did on a stain from a sheet obtained from one of the victims. Gilchrist obtained a PGM subtype (1+) which is consistent with the victim's PGM blood marker. Although Gilchrist testified that Mr. Miller's blood type had a subtype PGM (1-), she did not exclude him as the possible donor of the semen found in the stain on the sheet. Gilchrist's reasoning was that she had also identified the ABO blood group secretor substance A in the stain that was foreign to the victim and did not eliminate Miller as a possible donor of the semen. Gilchrist also stated that she was unsure if any of the stains were pure semen stains, but she believed there was one and after testing that stain she found Peptidase A, no ESD and PGM markers that were consistent with Miller's blood type and blood markers. Tests were also performed on semen/blood stains from the victim's nightgown that did not exclude Miller as the possible donor. Gilchrist had performed tests on blood and saliva samples from Miller and identified Peptidase A (1), Esterase D (2-1), and Phosphoglucomutase (1). She subtyped the PGM marker and determined the PGM subtype to be (1-). Tests performed on the vaginal posterior fornix swab of the victim revealed PEPA (1), ESD (1), and PGM (1) and the PGM subtype (1+). Although these findings are inconsistent with Miller's types, Gilchrist claimed that he could not be excluded as the potential donor because the swabs were bloody and all the markers that she found were consistent with the genetic markers of the victim, and therefore, Gilchrist determined that the PGM subtyping of (1+) was the result of the victim's blood mixing with the seminal fluid. At the time, the serological forensic examination had not yet been concluded and the seminal stains were being sent to a Dr. Schanfield for further testing.

Because DNA testing methods had advanced since the time of trial, new tests were conducted in June 1992. These tests were conducted on the original biological samples that had been preserved since the first tests in 1988. In this second round of tests, analyses were conducted on a blood sample from the first murder victim, a vaginal and anal swab, a white sheet from her home, a blood sample from the second victim, as well as a sheet, pillow, bed sheet, and nightgown from her home, a vaginal swab, and a blood sample from Miller. Using PCR based DNA testing, DNA was extracted and subjected to DQ Alpha typing. The chief forensic serologist at the Serological Research Institute, Brian Wraxall, concluded that, assuming that there was one semen donor on the anal swabs and white sheet from the first victim and the nightgown and pillow from the second, the semen donor was DQ Alpha type 3, 4. The semen on these items could not have come from Miller since he is a DQ Alpha type 3, 3. On July 18, 1995, Marcia Eisenberg at LabCorp came to different conclusions. According to Dr. Eisenberg, the DNA profile obtained from the non-sperm fraction of the vaginal swab from the frist victim was consistent with the victim's DNA profile and another individual with a n A,B profile at the GC locus. Robert Miller, along with suspects Lott and Wilson could all be potential donors. The DNA profile obtained from the sperm fraction of the cervical swab of the second victim was consistent with a mixture of DNA profiles from he victimr and any individual with a 3 allele at the DQ Alpha locus, thereby not excluding Miller or suspect Lott. A similar conclusion was obtained from the an analysis of the DNA profile from the non-sperm fraction of the rectal swab obtained from the second victim, as well as the non-sperm fraction from her pillow and sheet. Yet another test was conducted in 1994, this time at the Roche Biomedical Laboratories by Richard Guerrieri. According to Guerrieri, Miller was excluded as the source of the genetic material detected in the sperm fraction found on the sheet and pillow, as were the victims.

In 1994, Skip Palenik at Microtrace examined hair samples after they had already been mounted and previously analyzed by Joyce Gilchrist. According to Palenik, the comparison of fragments to the Miller's scalp hair was completely unjustified because scalp hair has a large degree of variation.

New DNA tests began in 1995 at LabCorp. This first set of tests did not exclude Miller as the potential donor of the genetic material. However, further tests in 1996 came to a different conclusion. In these tests, the laboratory used PCR based DNA testing on numerous pieces of evidence that contained sperm and nonsperm samples and compared them with samples from the two murder victims as well as from suspects Ronald Lott, Roderick Wilson, and Robert Miller. Dr. Eisenberg concluded that Miller and Wilson were excluded as possible contributors of the genetic material in all of the samples tested.

With these results, the defense appealed in 1996. The court ruled that the videotapes used in the capital case against Miller would have to be reviewed in order to determine whether or not Miller should be retried. The court ruled that there was insufficient evidence to keep Miller in jail. The state prosecutors appealed. Although prosecutors had dropped the murder charges against Ronald Lott, who had provided a sample as well, they proceeded with the murder charges against Robert Miller because of his supposed confession. The defense claimed that Miller never admitted to the crime, but instead adamantly denied, over seventy times throughout the course of the interview, any involvement. The defense also pointed out that, during the original trial, Judge Charles Owens characterized Miller's taped statements as being neither a confession nor an admission to anything. The statements made by Miller were either general public knowledge, multiple guesses resulting in a correct answer, the product of leading questions posed by the police, outright denials of any involvement, or completely incorrect factual statements. The state's primary source of evidence was highly flawed and could not be used as the primary source for Miller's involvement in the crimes. The case came to rest primarily on biological evidence.

The 1996 DNA tests ultimately led to Miller's exoneration and release in 1998. The tests also identified the true perpetrator, Ronald Lott, a man whose samples had been included in all rounds of testing. The prosecution had known that DNA tests had identified Ronald Lott but continued to maintain that Miller had been present at both crime scenes. Robert Miller had been on death row for nine years. For six of those years, the state had exculpatory DNA test results in their possession. In 1997, Robert Miller was granted a new trial and acquitted, based on DNA evidence, of murder and rape. In December 1997, while Ronald Lott was in jail serving a forty year sentence for the rape of two elderly victims, the prosecutor tried to negotiate with Lott's lawyers. Lott was told that he would face the death penalty because his semen had been found at both murder scenes from the Miller case, unless he could implicate Miller. Lott rejected the deal, leaving the prosecution with no case. On Thursday, January 22, 1998, Robert Miller was finally exonerated.