Blankenship v. Hall

BLACK, Circuit Judge: Thirty years ago, Sarah Mims Bowen was found dead in her apartment. She had been raped and brutalized. Today, Roy Willard Blankenship, having been tried, convicted, and thrice sentenced to die for the killing, seeks from this Court a writ of habeas corpus. Arguing his counsel at his third and final sentencing trial was ineffective, he believes he is entitled to relief. We hold that he is not. I. A. 1978: The Crime At around 4:15 p.m. on March 2, 1978, officers from the Savannah Police Department responded to a call at 404 West 44th Street. They were directed to the second-floor apartment of Sarah Mims Bowen. Several members of Bowen’s family already had arrived, having been contacted by her downstairs neighbor. Inside the apartment, police found a blood-stained paper towel in the living room. In the bedroom, the body of 78-year-old Bowen lay dead and naked on her bed. She had bruises on her arms and hands, and her face was beaten and bloodied. A plastic bottle of hand lotion had been forced into her vagina. There were footprints found on the porch outside Bowen’s apartment. Police found similar prints inside the apartment. Outside the house, they traced the prints from the bannister supporting the porch southwest along the ground towards the street, in the general direction of the apartment of Roy Willard Blankenship. Dr. Rodrick Guerry performed an autopsy. He determined Bowen had been severely beaten, suffering repeated blows to her face. Bowen had preexisting chron *1257 ic pericarditis and arterioscleorosis, and the autopsist attributed Bowen’s death to heart failure precipitated by a severe assault. The autopsy also revealed she had been vaginally raped. Semen was found in her vagina, which tests demonstrated came from a blood type-0 individual. Both Blankenship and Bowen were type-O. In addition, Dr. Guerry stated the inside of Bowen’s mouth and throat were red and bloodied, injuries consistent with oral rape. However, tests did not reveal the presence of semen. Scrapings beneath the nails on Bowen’s right hand also tested positive for type-O. Based on the condition of the body, the coroner concluded Bowen had been raped while alive, was beaten, and suffered heart failure as a result. A fingerprint lifted from glass broken in from the balcony and found inside the apartment matched Blankenship. On March 11, an arrest warrant for Blankenship was prepared, as well as search warrants for his apartment. Inside the apartment, police found shoes belonging to Blankenship whose tracks matched those found in and around Bowen’s apartment. Police arrested Blankenship and he waived his right to remain silent. Blankenship spoke with police and described his presence in Bowen’s apartment in the early morning of March 2, 1978. His oral statement was transcribed, and he signed the transcription. In it, he confessed to the following: I went up on the iron rail on the side of the porch and climbed over the banister. I stood up there for a few minutes thinking, what the hell, I really didn’t know what to think. I had to be drunk. Stoned. And I kicked the window in and I waited. When I kicked the window in to see if anybody heard it, I could’ve got shot or something. I guess I should have. It would have been better. I went in through the window, I think. I scraped my arm on the window. I don’t think it cut it. I went into the next room, I saw no one. Just the bedroom. I looked around there and the door was opened into the next room. I went up to the door and started to go through when I saw a mirror straight ahead in the next room where the lady was. I seen her reflection through the mirror sitting in a chair so I stood beside the door for awhile watching her pray or something. Moaning. I don’t know. Then I grabbed her. I think her mouth so she did not scream, [sic] I covered her mouth and her nose and then she slid down in the chair. She fell on the floor and I fell on top of her. After I fell over on top of her I didn’t have to hold her mouth or anything. She was not screaming or kicking or anything. So this blood was coming out of her head, I think, on the right side. I think. I pushed this little stool back and I picked her up and I carried her and laid her on the bed. All right. I put her on the bed. She had some pajamas on, I think. I took them off. It’s crazy. When I put her on the bed and took her clothes off, I was drunk, I guess. I said I may as well go ahead and get some pleasure. That’s when I had the relationship with her. As far as I know, I thought I was in the right hole. After that I got up and was afraid that I might have hurt her. I thought I’d better get out of there. I left as soon as I did that shit. I left. I went the same way I came. I was wearing the same shoes that the police confiscated from my house today. I watched her about 10 minutes. After I grabbed her she fell to the floor and I put her on the bed. Right after that I shot off or got *1258 my pleasure or whatever you want to call it. I put back on my clothes and left. It probably was not long. I was in the house maybe 45 minutes or an hour all together. I don’t know why I did it. I was drunk. I know I had to be drunk. That time in the morning I had to be just coming back from the Orential [sic] Lounge. I came by myself. I had been at the bar with Joe and Alex. They left the bar about 1:30 or 2:00. I know I stayed until closing, 3:00. I walked from the bar to the house. The Orential [sic] Lounge on Abercorn Street. I shoot pool all the time. It takes me about five to seven minutes to get to my house walking. I never did make it home. I stopped at her house and went upstairs before I went home. I know the witnesses in the bar — waitresses, sorry. I know the waitresses in the bar. I don’t dance. I just shoot pool and get high and get drunk. I was drinking that night. I was drinking burbon and coke. I don’t remember anything about the plastic bottle. 1 Based on the confession and physical evidence, Blankenship was charged with burglary, rape, and felony murder. B. 1980: The First Trial and Sentencing On March 22, 1978, Bart Shea was appointed as Blankenship’s counsel. Shea passed away, however, and so his partner, John Hendrix, was appointed to represent Blankenship on July 17, 1978. Hendrix and his co-counsel, Penny Haas, would represent Blankenship through all three of his sentencing trials and direct appeals. Hendrix was an experienced attorney. He had been admitted to the bar in 1958 and had clerked on the federal court in Atlanta before moving to Savannah to practice law. When he was appointed counsel in Blankenship’s case, he previously had represented defendants in six to eight death penalty cases. Haas was a recent law school graduate, having earned her degree in 1978. She was working as an associate in Hendrix’s office when she was appointed to the case. Hendrix filed several motions as Blankenship’s attorney, including a request that an expert be appointed to examine the defendant’s mental condition, a request that a pathologist be appointed to assist the defense, and a petition to have an investigator to assist in the defense. The court ordered two investigators be appointed, William Friday and Richard Moesch, to assist Hendrix and Haas. Trial commenced on April 21,1980. The state of Georgia put forth as evidence of Blankenship’s guilt the shoes whose prints were found in and around Bowen’s apartment, the fingerprint found on the glass inside the apartment, the blood and semen samples taken from Bowen’s corpse, and Blankenship’s signed confession. Defense strategy focused around demonstrating the investigation into Bowen’s death was incomplete. Hendrix elicited testimony from the state of Georgia’s witnesses that a hair segment belonging (at least according to the state’s expert) to an African-American individual was found in the combings of Bowen’s pubic hairs. Bowen and Blankenship were both white. In addition, a state’s expert found evidence of a B-antigen in some of the fingernail scrapings taken from Bowen’s left hand, *1259 even though both Blankenship and Bowen were blood type-0. (The expert, however, could not repeat the results and so could not conclusively say there was type-B blood underneath Bowen’s fingernails.) Hendrix used the scientific evidence to suggest someone else was present in the apartment that evening and was responsible for Bowen’s rape and death. Blankenship testified in his own defense. During the time of his arrest, he worked at the Guerry Lumber Company. He said he was an alcoholic and also took Qualudes, a tranquilizer. Blankenship knew Bowen; in fact, he had been inside her apartment prior to the night of her death, performing odd jobs such as replacing blown light bulbs. He would frequently talk with her on weekends when he would pass by her apartment and she was sitting on her porch. Discussing the events of March 1 and 2, Blankenship noted he began drinking soon after he returned home from work. After some time, he went to the Oriental Bar. From 7:30 that evening until the bar closed at 3:00 a.m., Blankenship continued to shoot pool, drink, and ingest Qualudes. After the bar closed, Blankenship headed home alone. Instead of returning home, however, Blankenship scaled the balcony of Bowen’s apartment. Once on the balcony, he knocked on Bowen’s door. There was no answer. He kicked in a window and crawled into the apartment. As he made his way through the apartment, Blankenship saw Bowen in a mirror sitting in her chair. Bowen was speaking to someone near the area of her kitchen. Blankenship said he reached out to grab Bowen and she jumped, tripped over a foot stool, and fell to the floor. Bowen was now bleeding from her head and was unconscious. Blankenship picked her up and moved her into the bedroom. Once he placed her on the bed, he pulled her pajama bottoms partially off her body. He did not remove her pajama top, which Blankenship said already was unbuttoned. He tried to have sex with her, but could not achieve an erection. At this time, Bowen appeared to be regaining consciousness, so Blankenship left the apartment the same way he entered. Hendrix asked Blankenship whether the pictures of the crime scene matched his recollection of how he left it. He insisted they did not; he testified that, when he left her, her pajamas where still partially on, whereas in the photos she was completely naked. The plastic bottle was also not there when he said he left. In addition, he said her face was not in the same condition it was when he left it, and that she had not been beaten up. Blankenship also testified as to his confession. He had been drinking before being arrested by the police in his apartment. He said he did speak with the police, and he did sign the statement. He said, however, that he had pointed out several errors in the typed statement to the interrogating officers, who told him to go ahead and sign it despite the errors. The errors he pointed out — including, for example, that instead of “sliding] down in the chair” she actually jumped up from the chair — essentially coincided with the story he had told the jury. In general, his testimony was that the story he told the detectives was the same as what he told the jury, and the typed confession was riddled with errors. He denied having intercourse with Bowen or ever hitting her. The jury found Blankenship guilty of burglary, rape and felony murder. The sentencing phase at the first trial was brief, and consisted of one witness, Victo *1260 ria Ray, who knew Bowen. The jury recommended the penalty of death after finding the aggravating circumstance that the offense of murder was outrageously or wantonly vile, horrible, or inhuman, in that it involved torture, depravity of mind, or an aggravated battery to the victim, or engaged in the commission of burglary. The judge sentenced Blankenship to death. On appeal, the Supreme Court of Georgia reversed the burglary conviction because the burglary charge was a lesser included offense of the felony murder charge. Blankenship v. State, 247 Ga. 590, 591, 277 S.E.2d 505, 507-08 (Ga.1981) (Blankenship I). In addition, the court reversed the sentence of death, finding the statements of one juror, who was excused due to conscientious opposition to the death penalty, were ambiguous. The dismissal of the juror was found to be a violation of Witherspoon v. Illinois, 391 U.S. 510, 88 S.Ct. 1770, 20 L.Ed.2d 776 (1968), and the case was remanded for resentencing. Blankenship I, 247 Ga. at 594, 277 S.E.2d at 509. C. 1982: The Second Sentencing The second sentencing trial took place in 1982. Prosecutors offered into evidence photographs of the crime scene and the autopsy photographs. The photos graphically depicted the apartment, including the blood on the floor and table where Bowen had been sitting. The photos also showed Bowen’s nude body lying on the bed with the plastic bottle inserted into her vagina. Blankenship’s confession also was introduced. The autopsist was again called, and he testified Bowen had been severely beaten and her vaginal and mouth areas were red and raw. He again testified the beating and trauma caused her heart failure. Several times Hendrix’s questions were curtailed by state objections and judicial admonishments not to raise issues of guilt during the sentencing phase. Hendrix questioned the officer who first arrived at Bowen’s apartment on the day of the murder. When the questions turned to the fingerprint lifted from the broken glass and the footprints found in and around the apartment, the state objected and the judge intervened: Mr. Kirkland [state attorney]: Your Honor, I would like to object to this line of questioning,— The Court: I’ve been trying to figure out what he was trying to show with all that. Mr. Kirkland: — I’m trying to limit this hearing — I will continue to limit this hearing to the sentencing phase of the trial— The Court: That’s all I’m concerned with. That’s all this jury is concerned with. The man has already been convicted of murder and rape. Now, let’s don’t retry that case again. Mr. Hendrix: Your Honor please, the defense does not care to retry the case insofar as guilt or innocence is concerned. But this jury is entitled to all of the evidence to select from it any matters that they may find to be of— The Court: I’ll disagree with you. Now, we are not going into the question of guilt or innocence of this defendant to the crime with which he has already been convicted. Mr. Hendrix: We understand that, Your Honor. The Court: Well, let’s confine our testimony to that. Mr. Hendrix: At the same time, Your Honor, we have to present any matter that can be of mitigation. *1261 Mr. Hendrix: Your Honor please, as I understand the word mitigation, any matter that this jury wishes to use as a mitigating fact. Not to lift the burden of guilt from him— The Court: I’ve already expressed myself on that subject. I don’t see any mitigating circumstances in what you are trying to do. I’ll be perfectly frank with you. Mr. Hendrix: Judge Harrison, I fully understand your reasoning in wanting to shorten this case. I have to present everything I can in mitigation for this boy’s life. (Sentencing Tr. 405-08, Sept. 20, 1982.) The state’s objection was sustained. Several more times Hendrix’s attempts to probe questions of guilt and innocence were short-circuited. He questioned one of the investigating officers regarding who removed evidence from the body and moved the body, in order to lay a foundation to discuss the fingernail scrapings and pubic hair combings. The Court: [W]hat are you trying to prove? Mr. Hendrix: I am trying to prove, sir, that there is physical evidence which shows that— Mr. Kirkland: Your Honor, I’m going to object to the statement that counsel is about to make. I think it’s intended to prejudice the jury. The Court: Is he trying to show that somebody else did it, I’m not going to listen to it. Mr. Hendrix: Your Honor please, for the record, the defense strenuously objects to being denied the opportunity to go into any matter— The Court: To show that somebody else might have done it? Mr. Hendrix: No, sir, to show that there could have been someone else there who could have been a party to it. The Court: Well, why would that exonerate him? Mr. Hendrix: If Your Honor please, it could mitigate in the minds of the jury the totality of the involvement of the defendant. (Id. at 426-27.) Six times Hendrix attempted to question witnesses or introduce evidence relevant to whether someone else was responsible for the crime, but the court continually blocked his efforts. In addition to attempting to elicit evidence relating to Blankenship’s responsibility for the crime and the possible role of others, Blankenship’s relatives testified in his defense. Nellie Fleming, his mother, said Blankenship was an “excellent teenager” and she “couldn’t ask for a better child.” She said he cared for his disabled grandparents for a year while they were sick. She said he never got into trouble and was churchgoing. Roy’s sister, Pearl Marie Blankenship, also testified. She said she was very close with her brother, and that he was a very good brother to her growing up. She also said Blankenship was not the sort of person to get in trouble. Blankenship again testified in his own defense. His testimony about the events of the evening before Bowen’s murder— including his consumption of alcohol and drugs — mirrored his testimony in the pre *1262 vious trial, 2 at least up until the point where he approached Bowen’s apartment. Unlike in the first trial, he said he did have a purpose for breaking into the apartment: he intended to steal her car and turn it over to some acquaintances for profit. Once he entered her apartment, he said he heard a conversation taking place. The voices grew louder, so Blankenship hid behind a dresser. After a few moments, he heard silence and crept out from beside the dresser. He saw Bowen lying unconscious on the floor. She was bleeding from her head, so he picked her up and put her on her bed. Other than the head wound, Blankenship said there were no other injuries. Blankenship said he used paper towels from a nearby dresser to stop the bleeding, but while he was applying them to Bowen’s head, she regained consciousness, began struggling, and started screaming for help. He informed her he intended to help her, but she continued struggling. He became scared, and while he thought about calling the police, he realized he would have to explain his presence in the room. He left immediately. Acknowledging his testimony was inconsistent with his statement to the police, Blankenship denied beating Bowen or having sex with her. He said he gave the police his statements because he was trying to protect a third party, although he refused to identify the person. He said he entered the apartment alone, but that someone else spoke with Bowen while he was there. On cross-examination, the state of Georgia pressed Blankenship on the confession he gave the police and his statements in the prior trial. As to some statements in the confession, he admitted to making them but said they were not accurate. As to others, he denied making them. Others he could not recall. The jury fixed Blankenship’s punishment as death. They found the murder was committed during the commission of another capital felony (rape), and that the murder was wantonly vile, horrible, and inhuman, in that it involved torture or depravity of mind. The Supreme Court of Georgia once again reversed the death sentence. Blankenship v. State, 251 Ga. 621, 308 S.E.2d 369 (Ga.1983) (Blankenship II). The Court held that when the sentencing phase of a death penalty case is retried by a jury other than the jury which determined guilt, evidence cannot be excluded simply because is speaks to the guilt or innocence of the crime, since the parties may offer any evidence relating to the circumstances of the crime. Id. at 624, 308 S.E.2d at 371. The judge overseeing the resentencing had excluded just such evidence. Blankenship’s death sentence was vacated and the case remanded for another sentencing proceeding. D. 1986: The Third Sentencing And so Blankenship, Hendrix, Haas and the state of Georgia once again found themselves in trial proceedings in Chat-ham County Superior Court to argue before a new jury whether or not Blankenship should receive the death penalty for Bowen’s death. The third and final sentencing trial took place in June 1986. *1263 The prosecution’s evidence was similar to the previous two trials: they introduced the photographs of the crime scene, testimony regarding the shoe prints in the room and on the balcony, testimony regarding the glass found in the apartment and the signs of a struggle, the autop-sist’s — Dr. Guerry’s — testimony regarding Bowen having been severely beaten and raped, Blankenship’s original confession to the police, and the blood scrapings and seminal fluid that were type-O. Hendrix’s opening statement laid out the defense strategy: he expected to introduce evidence (including Blankenship’s own testimony) showing that someone other than Blankenship was involved in the crime. He said the scientific evidence would show the existence of hair and blood found on Bowen’s corpse that did not belong to Blankenship. The defense also planned to show the police department’s investigation was inadequate, in that a similar rape and murder had occurred near Bowen’s apartment in the two weeks prior to Bowen’s death with a different culprit. In addition, the defense planned to present evidence suggesting the injuries suffered by Bowen were not as severe as the state suggested. To these ends, Hendrix questioned prosecution witnesses on the physical evidence and the rape and murder. He asked Detective William McGuire, one of the investigating officers, if he had ever investigated whether the culprit in the rape and murder of Valerie Armstrong, which took place two weeks before Bowen’s death, was responsible for Bowen’s murder. Pointing out that Dr. Guerry noted the similarities between the crimes when he conducted the autopsy, Detective McGuire conceded he did not investigate the connection further. Hendrix also cross-examined Dr. Guer-ry. He elicited testimony that Bowen did not suffer internal injuries. He also questioned Dr. Guerry on the Valerie Armstrong case, and the Doctor admitted there were similarities between the cases — both victims had been raped and murdered. The state’s redirect, however, revealed differences between the crimes: Armstrong was a six-year-old black child, whereas Bowen was a 78-year-old white woman. In addition, Armstrong had been stabbed, whereas Bowen had died of heart failure brought on by severe trauma. The forensic serologist, Linda Tillman, also testified. She conceded that the type-0 seminal fluid found in Bowen could have come from a non-type-O, non-secreting individual; since Bowen was a type-0 se-cretor, her secretions could have infected the sample. In addition, she testified that Gary Nelson, the man responsible for the Armstrong murder, was a type-0 secretor, like Blankenship and Bowen. Her findings would have been as consistent with Nelson being the culprit as Blankenship. Tillman also said she found evidence of a B-antigen in one set of fingernail scrapings, but there was not sufficient material to repeat the results and say with any confidence there was type-B blood. She also said the evidence would be consistent with the culprit being a type-B non-se-cretor. The defense also questioned Roger Par-ien, the director of the Savannah branch of the state crime lab who performed analysis on the hair samples. He confirmed one hair segment taken from Bowen’s pubic combings appeared to come from an African-American. He also said he was never asked to compare the hair with any of the hairs from the Nelson case (Nelson was an African-American). On cross-examination, Parten admitted to the state prosecutor the hair segment could have come from anywhere. *1264 Hendrix called Dr. Joseph Burton, chief medical examiner for Dekalb and Cobb Counties, as an expert witness. He reviewed the reports from the state crime lab and Dr. Guerry’s autopsy report. Like Tillman, he said that the seminal fluid extracted from Bowen’s vagina could hypothetically come from a person who was a non-secretor, since Bowen herself was a type-0 secretor and could account for the result of the test. He also said that the existence of B-antigen material beneath the fingernails, if it did exist, would not be consistent with their Bowen’s or Blankenship’s blood type. Burton also provided extensive testimony regarding Bowen’s bruises and scratches. Specifically, Burton said some of the bruising on her arm and elbow could have a number of explanations that were not violent in nature. Although Burton said some of the facial wounds could result from a beating, there was no evidence below the neck of a beating. Burton also testified as to the effect of heavy alcohol and drug use on a person’s ability to perform sexually. He could not say whether a person could have an erection under such circumstances, although it was hypothetically possible they could not. On cross-examination, Burton explained the wounds on the arms and legs could have come from a struggle. Blankenship again testified in his own defense. Haas examined him, and he provided some information on his family background. He explained his family had a significant problem with “nerves,” and that his biological father, along with his aunt and uncle, were killed in a motel room by a carbon monoxide leak when he was eight years old. After his death, his mother went into a nerve coma and he stayed with his grandparents. His mother recovered, and he returned to her care; she had married an alcoholic with whom she “had a lot of trouble.” He would “com[e] home and see[] the house torn up” and his mother and stepfather fighting. Blankenship then described his time in the armed forces. He was in the service for less than two years. After being AWOL for visiting his sick grandfather, he was dropped in rank and the service ultimately agreed to discharge him. He soon moved to Savannah and began work at the lumber company. Blankenship recounted his drinking and drug habits, and then discussed the events of the evening and early morning of Bowen’s death. His testimony was similar to that in his 1982 resentencing trial. He recounted that he entered her apartment to steal her car and heard her speaking to someone. Although he heard the voice, he did not see who the person was. There was a commotion, and he found her lying on the floor, bleeding. He picked her up and placed her on the bed. He also said his statement to the police was what they wanted to know and what they asked him to say; he said he was drunk during the police’s questioning and simply wanted to be left alone. Haas also asked Blankenship whether his statement in the first trial, that he attempted to rape Bowen but was unable to do so, was true. He first responded that if he did attempt to rape her he was not able to do it, but then said he did not think he attempted to rape Bowen. Haas asked why he testified as he did in his first trial and he responded “[m]aybe for protection,” but refused to elaborate. On cross-examination, the state asked Blankenship specific details about his signed confession. Blankenship was also *1265 asked questions about the alleged third person present in the apartment. When asked about his contradicting testimony in the first trial, Blankenship said he had lied for a reason, but he refused to tell the reason in court. He said he had made a vow to God and he could not say because it involved someone else. During closing argument, Hendrix asked the jury to spare Blankenship’s life due to residual doubt over his guilt: I would like to take a moment and review with you items that the defense thinks you can consider that create doubt, that create sufficient doubt to determine in favor of life imprisonment as opposed to killing Roy Blankenship. As opposed to taking his life. There’s doubt. (Sentencing Trans. 498, June 1986.) After recounting reasons for doubt, including the blood and hair evidence, and criticizing the lack of a thorough investigation into Bowen’s murder, Hendrix did reference Blankenship’s upbringing: Roy Blankenship is different in one regard. He is the product of mountains. Mountain people are different. He was born up in one of those valleys up in West Virginia. Now, I don’t know exactly where it is but I will tell you how far back up in there it is. When I get letters from his mother and from his sister the post office still uses a hand stampled [sic] canceled stamp. They don’t have a fancy machine up there yet. So I know that he’s different because I was born up in one of those valleys up in the mountains up in North Georgia and I know just how far back up in there folks live. He came down here to work. And he got himself a job. It was the only job he had. Apparantly [sic] he worked pretty hard. He says so. (Id. at 502-03.) After again emphasizing the existence of doubt and faulting the police for their investigation, the defense finished its summation, stating “If you impose the life sentence then you create the possibility of this matter being further investigated.” (Id. at 503.) The jury recommended the death sentence be imposed, finding the offense was horrible and inhumane in that it involved aggravated battery and depravity of mind. This time, the Supreme Court of Georgia affirmed Blankenship’s sentence on appeal. Blankenship v. State, 258 Ga. 43, 365 S.E.2d 265 (Ga.1988) (Blankenship III). A petition for certiorari was denied by the Supreme Court, and Blankenship’s direct appeals came to an end. E. 1990: The State Habeas Petition & Evidentiary Hearing Blankenship next began pursuing his state habeas relief. His state petition raised a litany of issues, among them the assertion that his trial counsel were ineffective. A hearing was held in February 1990 where Haas and Hendrix testified as to their representation of Blankenship. 1. Haas’ testimony. Haas testified first. Blankenship’s state habeas counsel began his questioning by asking Haas why she and Hendrix made several motions during the 1986 resentenc-ing. In response to a question regarding their motion for an investigator, Haas responded: [T]he third trial was more geared toward the idea of there possibly having been someone else there the night that this happened and the fact of this unexplained Negroid hair and the unexplained B antigen blood that was found *1266 and that sort of thing, so we were looking for an investigator to help us especially in that line. (State Habeas Trans. 23, Feb. 28, 1990.) Counsel also asked Haas why they requested a psychiatrist at the third trial, and Haas responded that the motion had been denied. In addition, Haas explained that a motion for a forensic serologist and lab technician was filed during the third sentencing trial because they hoped to discuss the existence of the B-antigen on Bowen’s body and the suggestion that another person was present. Counsel next asked Haas if she recalled speaking with Blankenship’s sisters or mother. Haas said Hendrix was the person who communicated with the family, and her own contact had been only intermittent. On cross-examination, Haas explained in greater detail her duties during the first trial. Her role in the 1980 trial, as a recent law school graduate, was minor. She in large part took notes during the trial and prepared the appeal. But she became more experienced as time wore on. By the 1986 trial, she was a more seasoned attorney (having been practicing for six years) and took a more active role: although the main function was to prepare an appeal, by then she had been examining witnesses in court. In fact, she had examined Blankenship himself when he testified at the 1986 resentencing. She also handled the drafting of most pretrial motions. The state’s counsel questioned Haas about the psychiatric examination performed on Blankenship prior to the first trial. Two examinations were performed — one by the state, and one by an expert, Dr. Wolfe, selected by the defense. Haas said the state examination was normal, and Dr. Wolfe’s examination did not “give us anything that we felt could help us and so we sort of let that die on the vine.” (Id. at 51.) As for examinations after the first trial, Haas and Hendrix asked for funds, but were denied. Haas also discussed conversations she had with Blankenship prior to the first trial. She said he gave them the names of certain family members, who provided information on Blankenship’s background: I really don’t recall the specifics of it now, like just that he came from sort of a hard background, and we hoped to be able to use some of that. And certainly his mother we hoped to use as someone who could give us some insight into him, and to what had happened here, and to what kind of a person he was. (Id. at 52.) When asked whether she recalled if Blankenship’s mother testified at the second sentencing trial, she responded that she remembered the mother at one of the trials, but at the third one she did not come. “As I recall, Mr. Blankenship didn’t want us to contact her, I think.” (Id. at 54.) Later, state’s counsel returned to the question about Blankenship’s background: Q. Did you ever discuss Petitioner’s background with him? A. With him, and again, Mr. Hendrix had discussed it with his family. Q. Did he ever tell you about any psychological disorders that some of the family members may have had? A. I don’t recall it. Q. And again, Dr. Wolfe gave you no information from his psychological examination that you thought would be helpful? A. That’s correct. (Id. at 72-73.) State counsel asked Haas if any additional investigation took place prior to the *1267 third sentencing trial, and Haas responded: A. Again, we were centering in — especially in the third trial, we were centering in on that hair and the blood type, so that would have been the nature of whatever other investigation we did. We were severely limited because we weren’t able to get the funds for any other assistance, but that was where we were headed. Q. So would it be a fair statement to say that your strategy at this third sentencing trial was to show that someone else was present and committed the murders and plant some sort of reasonable doubt to avoid a death sentence? A. Yes. That there was someone else there, yes. (Id. at 54-55.) This was reiterated when counsel asked if the strategy changed between the first, second and third trials: Q. [D]id you alter your strategy at the second sentencing trial based upon the outcome of the first sentencing trial or was your strategy essentially the same? A. ... [I]n the second trial, as I say, our main strategy had been to try to really hone in on this information and really get the information tied down about this blood and this hair and so forth, and we weren’t able to do that, and of course, that made it very difficult to try the case and it was reversed. Q. And what about the third trial? A. We were trying to come at it from the same direction again. (Id. at 58-59.) Haas also mentioned that Blankenship relied on them to guide the proper strategy. Finally, counsel for Georgia asked Haas about the allegation in Blankenship’s state habeas petition that she and Hendrix were ineffective for presenting no witnesses at sentencing other than those going to guilt and innocence: Q. Again, what was the theory at the third sentencing trial? A. I think at that point we were, as I said before, really honing in on the blood and the hair in that part of the case. Again, I believe there was some family members that couldn’t make it .... Our theory mainly had to do with those hairs and that blood type and. that part of it. That’s really where we were coming from. (Id. at 67.) The point was reiterated on redirect by Blankenship’s habeas counsel: Q. Is it fair to say that throughout the course of these proceedings the theory of defense focused on this Negroid hair to some extent? A. The hair, the blood sample and the fact that there was a possibility that there might have been someone else there that night, yes. Q. So the hair would have been a crucial piece of evidence? A. Correct. Q. And the small amount of blood sample which indicated B antigen would have been crucial? A. Correct. Q. And the possibility that a third person may in fact have been present would have been crucial? A. Yes. (Id. at 74.) 2. Hendrix’s testimony. Next Hendrix testified. Blankenship’s habeas counsel asked if he had used Dr. *1268 Wolfe in the third sentencing trial. Hendrix replied they did not use Dr. Wolfe, because the judge denied their motion. Hendrix also testified as to the theory of defense at the trials: Q. Would it ... be fair to say that one of the themes of the defense at trial and actually at the various trials focused on the presence of [the hair evidence]? A. Yes, indeed. Q. Would it be fair to say that this hair would be a crucial piece of evidence? A. Not as much as Roy’s testimony. Q. But it was the subject at least of one of the theories of defense; is that fair? A. Yes. (Id. at 92.) Later, the questioning returned to this subject: Q. Now, is it fair to say that your defense theory, or a large part of your defense theory throughout these trials was the involvement of another unidentified person in this particular event? A. Yes, sir. (Id. at 153.) Hendrix further explained that the hair segment and the B-antigen were vital to defense strategy. Hendrix was asked whether he spoke with Blankenship’s mother and sisters pri- or to the third sentencing trial. He did not speak with the mother, and recalled speaking with his sisters “[u]p to a point.” (Id. at 108.) He said Blankenship’s sister Pearl’s inquiries involved when her brother was set for trial and whether Hendrix thought he would prevail. Later, Hendrix mentioned Blankenship’s resistance to involving his family: “[W]hat you have to understand is that most all of these people [other people in Blankenship’s background] Roy did not want involved in his case just as he directed us to not communicate with his family. He wanted to protect his family from his case.” (Id. at 117.) He further elaborated about the introduction of mitigating evidence: Q. In terms of potential mitigating evidence, did you have discussions with Mr. Blankenship about what mitigating evidence was? A. Yes. Q. And did you ever discuss his background with him? A. I think so. I think that you could say that we knew a great deal about Roy and his childhood in West Virginia, both through him and from talking with members of his family, and then from just reading their letters. Q. Do you recall Mr. Blankenship ever telling you that he did not want to involve his family? A. Oh, absolutely. Absolutely. Q. At what point in your representation did this occur? A. Well, I think it happened in more than one fashion and on more than one occasion. I think that the words “involve my family” probably came about at about the second trial and then at a later point, he admonished me not to write or correspond with or talk on the telephone with his family. I think he wanted them protected. (Id. at 135-37.) Hendrix was also asked about Blankenship’s state habeas petition complaining they only used witnesses addressing guilt and innocence during the sentencing phase: *1269 Q. Did you have any [information] from Petitioner regarding his background that you thought would have been beneficial to present in his behalf? A. No, I’m not aware of any specific information that he had given us concerning his background that we felt mandatory to introduce. Q. What about that that you did have about Mr. Blankenship, would you characterize it as lukewarm, or having potential for more harm than good ultimately, or could you characterize it at all? A. Well, I felt certain today that my feeling was, we won’t be helped by any of this. And there again, that’s a decision you have to make at that time and at that place. (Id. at 143-44.) At no time did Hendrix elaborate on the extent of his knowledge of Blankenship’s background and upbringing. 3. The affidavits. Blankenship’s state habeas counsel obtained affidavits from Blankenship’s sisters, Debbie Blankenship and Pearl Dalton, his mother, Nellie Fleming, and a clinical psychologist, Harry Krop. Hendrix and Haas were not aware of the affidavits when they testified, nor were they cross-examined on their contents. In fact, the affidavits were not even filed in the state habeas proceedings until May 1990, two months after the hearing in which Hendrix and Haas testified. These affidavits provided specific details on Blankenship’s background. They describe his difficult upbringing, alcohol abuse, and family history of mental illness. His family’s affidavits describe a disturbing childhood, where he was subjected to a series of alcoholic, abusive and sadistic father figures (including his biological father, who died in a motel of carbon monoxide poisoning), and an unstable mother. He was also once raped by a neighbor when he was a child. The details included incidents where his mother was abused and severely beaten by jealous husbands, who also abused (psychologically and physically) and threaten to kill the children. The affidavits also describe how Blankenship was a sickly child and was repeatedly hospitalized for fevers and other illnesses. When he grew up and joined the service, his family said he would experience blackouts and tell them he thought people were after him. They worried about his mental health. Krop’s affidavit attributed Blankenship’s drug and alcohol abuse to his traumatic upbringing. Speaking specifically about the night in question, Krop said Blankenship’s thought processes would have been highly disordered due to his alcohol and drug intake and, as a result, his capacity to form intent was diminished. Krop’s evaluation revealed that Blankenship was not sociopathic and has good insight into his behavior. According to Dr. Krop, the lack of sociopathy and Blankenship’s near normal intelligence makes the chance of rehabilitation more likely. Overall, Krop found no specific signs of neuropsychological disease. Both Pearl and Debbie stated in their affidavits that, if Blankenship’s counsel had asked them about his background they would have told them and would have been willing to testify regarding the background to a jury. His mother’s affidavit stated the same. In addition, Nellie said she phoned Hendrix prior to the 1986 sentencing trial and explained Blankenship’s sisters’ history of schizophrenia and the mental illness of her father’s brother. An affidavit from Hendrix confirms he knew about the history of mental illness in the *1270 family and informed Dr. Wolfe of the history — including the two sisters’ history of schizophrenia — in September 1982. Amidst this backdrop, the state habeas court rejected Blankenship’s ineffective assistance claim in a summary fashion. In an order dated March 6, 1991, the state habeas court stated: Petitioner has also raised ineffective assistance of counsel as a ground for relief in his petition. He alleges numberous [sic] areas in which he contends his attorneys were ineffective. Petitioner’s claim is without merit. Petitioner was represented by two competent attorneys who hotly and ably contested the state’s case at every phase of Petitioner’s trials. In fact, the Petitioner seeks to have two attorneys declared ineffective notwithstanding the fact that they secured reversals of two prior sentences of death entered against Petitioner. The Court finds on review of the record and consideration of the evidence presented in this Habeas proceeding that Petitioner was afforded effective assistance of counsel. Petitioner seeks to hold his attorneys to a standard of perfection which is impossible to attain for any man or woman. He was entitled to and received effective assistance of counsel as mandated by the Constitution. After the Georgia Supreme Court rejected his application for a certificate to appeal, this avenue of state habeas relief was exhausted. F. 2005: The Federal Habeas Petition In October 2005, Blankenship filed his federal habeas petition, the resolution of which is the subject of this appeal. 3 Among the claims argued was that Blankenship’s state trial counsel provided constitutionally deficient performance by failing to investigate and present evidence of Blankenship’s background at his 1986 re-sentencing. The district court did not defer to the state habeas court’s resolution of the claim but still denied the petition, finding this Circuit’s case law defeated Blankenship’s ineffective assistance claim. The district court also denied all other grounds for relief. Blankenship appeals only his ineffective assistance claim; he argues counsel were ineffective for failing to investigate and introduce evidence of his traumatic childhood during the third sentencing trial. II. “An ineffective assistance of counsel claim is a mixed question of law and fact that the court reviews de novo.” Williams v. Allen, 458 F.3d 1233, 1238 (11th Cir.2006). It is the petitioner’s burden to establish his right to habeas relief and he must prove all facts necessary to show a constitutional violation. Jones v. Walker, No. 04-13562, slip op. at 34-35, 2008 WL 3853313, at *13-14, 540 F.3d 1277, 1291-93 (11th Cir. August 20, 2008) *1271 (en banc); Romine v. Head, 253 F.3d 1349, 1357 (11th Cir.2001). In other words, Blankenship has the burden of demonstrating counsel’s performance was ineffective and must elicit the facts necessary to prove the claim. A question presented by the circumstances of this appeal is whether the strictures of the Anti-terrorism and Effective Death Penalty Act of 1996 (AEDPA), Pub.L. No 104-132, 110 Stat. 1214 (1996), 28 U.S.C. § 2241, et seq. apply to federal review of the state court’s rejection of the ineffectiveness claim. The deference afforded to state courts by federal courts sitting in review prohibits the federal court from granting a petition unless the state court’s adjudication on the merits of the claim (1) resulted in a decision that was contrary to, or involved an unreasonable application of, clearly established federal law, or (2) resulted in a decision that was based on an unreasonable determination of the facts in light of the evidence presented in the state court proceeding. 28 U.S.C. § 2254(d). Key to this case is whether the state court “adjudicated on the merits” the ineffective assistance claim when it summarily rejected Blankenship’s arguments in its 1991 order. While both parties and the district court seem to think no deference is owed to such summary adjudications, our case law is clear: We have repeatedly held “a state court’s summary rejection of a claim qualifies as an adjudication on the merits under § 2254(d) so as to warrant deference.” Ferguson v. Culliver, 527 F.3d 1144, 1146 (11th Cir.2008); Herring v. Sec’y, Dep’t of Corr., 397 F.3d 1338, 1347 (11th Cir.2005); see also Wright v. Moore, 278 F.3d 1245, 1255 (11th Cir.2002) (“The [AEDPA] language focuses on the result, not on the reasoning that led to the result, and nothing in that language requires the state court adjudication that has resulted in a decision to be accompanied by an opinion that explains the state court’s rationale.”). Therefore, the Georgia state habeas court’s summary rejection of Blankenship’s claims is entitled to AEDPA deference. As to what constitutes clearly established Supreme Court law at the time of the state habeas decision, the question of AEDPA deference makes little difference: Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984), was clearly established by 1991, and it remains the standard for evaluating ineffective assistance claims. Williams, 458 F.3d at 1238. Therefore, Blankenship must show the state court’s application of Strickland was unreasonable. 4 This is no easy task. Blankenship “must do more than satisfy the Strickland standard. He must also show that in rejecting his ineffective assistance of counsel claim the state court applied Strickland to the facts of his case in an objectively unreasonable manner.” Rutherford v. Crosby, 385 F.3d 1300, 1309 (11th Cir.2004) (emphasis added) (quotation omitted). As the second aspect of § 2254(d) makes clear, we also defer to the *1272 state court’s reasonable factual determinations. Summary adjudications, as a rule, do not have explicit factual findings to which a court can easily defer. Our body of case law discussing implied findings of fact are instructive as to the appropriate deference afforded to factual determinations in summary rulings. We have previously recognized a state court’s “disposi-tive ruling may contain implicit findings, which, though unstated, are necessary to that ruling.” Hightower v. Terry, 459 F.3d 1067, 1072 n. 9 (11th Cir.2006) (post-AEDPA habeas case) (citing United States v. $242,484,00, 389 F.3d 1149 (11th Cir.2004) (en banc)). State court findings of fact can be inferred from its opinion and the record. Freund v. Butterworth, 165 F.3d 839, 859 n. 30 (11th Cir.1999) (citing Cave v. Singletary, 971 F.2d 1513, 1516 (11th Cir.1992)). 5 Moreover, implicit findings of fact are entitled to deference under § 2254(d) to the same extent as explicit findings of fact. See Mathis v. Zant, 975 F.2d 1493, 1495 (11th Cir.1992); Cunningham v. Zant, 928 F.2d 1006, 1011 (11th Cir.1991). Thus, we can “make the common sense judgment that material factual issues were resolved by the trial court in favor of the judgment when it was reasonable for that court to have done so in light of the evidence.” Hightower, 459 F.3d at 1072 n. 9 (quotations and alterations omitted). In other words, since we apply AEDPA deference to summary adjudications, we may uphold the state court’s decision that counsel was not constitutionally deficient if our review of the record reveals that a reasonable view of the facts before the state court supports such a conclusion. In this case, the district court erroneously afforded no AEDPA deference to the state court’s adjudication of Blankenship’s ineffective assistance of counsel claim. With the proper principles of deference in mind, we turn to the merits of Blankenship’s habeas challenge. III. A The Law of Ineffective Assistance of Counsel As stated above, Strickland is the touchstone for all ineffective assistance of counsel claims. The Sixth Amendment right to counsel includes the right to effective assistance of counsel, since the purpose of the right to counsel more generally is to ensure a fair trial. Strickland, 466 U.S. at 686, 104 S.Ct. at 2063-64. In order to prevail on an ineffective assistance of counsel claim, a petitioner must satisfy two prongs: first, the petitioner must show that counsel’s performance was deficient in that he “made errors so serious that counsel was not functioning as the ‘counsel’ guaranteed the defendant by the Sixth Amendment.” Id. at 687, 104 S.Ct. at 2064. Second, the petitioner must show the deficient performance prejudiced the defense. Id. All that a defendant at trial is entitled to is reasonably effective assistance; therefore, the petitioner must show *1273 his counsel’s representation fell below some objective standard of reasonableness. Id. at 688, 104 S.Ct. at 2064. Stated generally, the standard is “reasonableness under prevailing professional norms.” Chandler v. United States, 218 F.3d 1305, 1313 (11th Cir.2000) (en banc). The range of reasonable behavior permitted by this standard is broad. Id. Petitioner must show the course of action taken by counsel would not have been taken by any competent counsel. Id. at 1315; Alderman v. Terry, 468 F.3d 775, 792 (11th Cir.2006). Thus, “[¡judicial scrutiny of counsel’s performance must be highly deferential.” Strickland, 466 U.S. at 689, 104 S.Ct. at 2065; Chandler, 218 F.3d at 1313. “Courts must indulge the strong presumption that counsel’s performance was reasonable and that counsel made all significant decisions in the exercise of reasonable professional judgment.” Chandler, 218 F.3d at 1314 (quotations and alterations omitted) (citing Strickland, 466 U.S. at 689-90, 104 S.Ct. at 2065-66). We are to avoid the “distorting effects of hindsight” and judge the reasonableness of counsel’s action from the reference point of the time of counsel’s conduct. Strickland, 466 U.S. at 689-90, 104 S.Ct. at 2065-66. Among the duties owed by minimally competent counsel is the duty to make reasonable investigations or to make a reasonable decision that makes said investigations unnecessary. Id. at 690-91, 104 S.Ct. at 2066. “[Strategic choices made after less than complete investigation are reasonable precisely to the extent that reasonable professional judgments support the limitations on investigation.” Id. “[I]n evaluating the reasonableness of the investigation, ‘a court must consider not only the quantum of evidence already known to counsel, but also whether the known evidence would lead a reasonable attorney to investigate further.’ ” Alderman, 468 F.3d at 792 (quoting Wiggins v. Smith, 539 U.S. 510, 527, 123 S.Ct. 2527, 2538, 156 L.Ed.2d 471 (2003)). In addition to the duty to reasonably investigate avenues of defense (or make a reasonable decision to not do so), counsel’s choice of strategy is subject to review, but “strategic choices made after thorough investigation of law and facts relevant to plausible options are virtually unchallengeable.... ” Strickland, 466 U.S. at 690-91, 104 S.Ct. at 2066 (emphasis added). In order to assess Blankenship’s claim of ineffective assistance, then, we are faced with two questions. First, was Hendrix and Haas’ investigation of Blankenship’s background reasonable? Second, was the strategy selected reasonable? Each question has a factual and legal dimension to it. In the former, we must determine the actual scope of Hendrix and Haas’ investigation before we can determine whether the investigation was reasonable. In the latter, we must determine the strategy actually pursued by counsel at the 1986 resentencing before we can assess its reasonableness. We will address each question in turn. B. Was Counsel’s Investigation Reasonable? 1. What did counsel know? At the heart of Blankenship’s challenge lies the question of Hendrix and Haas’ investigation into his background. Blankenship argues the pair failed to investigate his background, and that said investigation — e.g., interviews with members of the family — would have revealed his troubling background, alcohol use, and *1274 the existence of a family history of schizophrenia. Our task is to determine what counsel knew and when they knew it. We reiterate that in habeas proceedings, unlike direct appeals, the petitioner bears the burden of establishing his right to relief; Blankenship must prove the facts necessary to demonstrate his counsel’s performance was constitutionally defective. See Jones v. Walker, No. 04-13562, slip op. at 34-35, 2008 WL 3853313, at **13-14, 540 F.3d 1277, 1291-93; Romine, 253 F.3d at 1357. Because of this burden, when the evidence is unclear or counsel cannot recall specifics about his actions due to the passage of time and faded memory, we presume counsel performed reasonably and exercised reasonable professional judgment. Romine, 253 F.3d at 1357-58; Williams v. Head, 185 F.3d 1223, 1227 (11th Cir.1999). In this case, our task is complicated in two ways. First, Blankenship’s state ha-beas counsel failed to elicit from either Hendrix or Haas details as to the extent of their knowledge of Blankenship’s background. Instead, both attorneys made general statements about their investigation and they were not asked to explain in detail what parts of Blankenship’s background they had learned about. Although both Haas and Hendrix testified in the state habeas proceedings and Blankenship’s counsel had ample opportunity to question them, the record remains incomplete. In fact, the key affidavits Blankenship points to in this appeal — his family’s affidavits and Dr. Krop’s affidavit — were not filed until months after the state habeas hearing in which Haas and Hendrix testified. Thus, the state of Georgia was never afforded the opportunity to cross-examine Hendrix and Haas on the contents of the family’s affidavits, and counsel were never able to explain what they knew about Blankenship’s personal background. Nor was the state able to cross-examine the family members themselves on the veracity of their claims. 6 Second, Blankenship’s trial counsel testified they had difficulty remembering aspects of the case. The state habeas hearing in which they testified took place twelve years after their representation of Blankenship began, and four years had passed between the 1986 resentencing trial and the hearing. Memories were understandably clouded. For example, Haas testified she could not “recall the specifics” of what she knew of Blankenship’s background, and she could only say she knew he had come from a hard background. Blankenship’s failure to exact detailed testimony from trial counsel and gaps created by faded memory make his attempts to *1275 satisfy his already heavy burden even more difficult. Although we are without the benefit of clear statements from Blankenship’s trial counsel, we can discern from the record that they knew generally about the three key elements of Blankenship’s background discussed in the affidavits of his family members and in Dr. Krop’s analysis: his alcohol and drug abuse, the family history of psychological problems, and his difficult childhood. The record reveals Hendrix and Haas knew Blankenship struggled with drugs and alcohol; indeed, it was a substantial aspect of their defense throughout trial. Blankenship himself discussed his use of drugs and alcohol during his several trials. In his confession to police, he admitted to being drunk the night of the death. His testimony at his first trial revealed he had been drinking and ingesting tranquilizers much of the evening and early morning of Bowen’s death. In the second trial, he testified that his lumber yard supervisor told him his drinking was excessive. In the third trial, he again recounted his systemic drinking and drug use. Moreover, Hendrix sought to elicit from Dr. Burton— both at the first trial and the third trial— testimony about the effects of such alcohol and drug use on an individual’s ability to commit the acts for which Blankenship was on trial. Therefore, there can be no doubt Hendrix and Haas knew about Blankenship’s issues with drugs and alcohol, given the repeated references to it throughout all of the proceedings. Equally plain in the record is counsel’s knowledge about the family history of schizophrenia. Nellie’s affidavit states she discussed the family history of mental illness with Hendrix. Hendrix’s supplemental affidavit and documentary evidence provided in the state habeas proceedings confirms he knew about the history of mental illness. A memorandum demonstrates in 1982 Hendrix informed Dr. Wolfe, the defendant’s expert who examined Blankenship prior to the first trial, of the history of family illness — including the ongoing paranoid schizophrenia of Blankenship’s sisters and the institutionalization of Nellie’s father’s brother. Blankenship himself told the jury in the 1986 resentenc-ing his family had a problem with “nerves.” Moreover, Hendrix testified Dr. Wolfe’s report — which is not in the record- — did not give them anything they felt was helpful, and both Haas and Hendrix testified the court denied them the funds necessary to have an expert examine Blankenship for the third trial. (Indeed, Dr. Krop’s affidavit says he found no evidence of neuropsychological disease and that Blankenship lacked sociopathy and had near normal intelligence.) State trial counsel had ample knowledge of the family’s history of mental illness but did not have evidence Blankenship himself suffered from mental illness. The record also demonstrates Blankenship’s trial counsel knew he came from a difficult background. At the state habeas hearing, Haas testified she knew Blankenship came from a hard background, and Hendrix stated he knew a “great deal about [Blankenship] and his childhood in West Virginia, both through him and from talking with members of his family, and then from just reading their letters.” Blankenship’s mother and one sister testified at his second sentencing in 1982, although other than expressing positive opinions about Blankenship they did not discuss his background. Blankenship himself did discuss some of his familial background in his testimony in the 1986 sentencing. He said his biological father and *1276 his aunt and uncle died of carbon monoxide poisoning in a motel. He also said after the death, his mother was incapacitated with “nerves” and he briefly lived with his grandparents. Blankenship also told the jury that when he returned to live with his mother, she married an alcoholic with whom she consistently fought. Blankenship also provided details about his brief stint in the armed forces. Thus, it is clear state trial counsel did know about Blankenship’s alcohol abuse, the history of mental illness, and the existence of a difficult background. What the family’s affidavits provide are details of these areas of Blankenship’s life. There is no doubt these details are horrifying and disturbing, describing a severely fractured home life with incidents of physical abuse, psychological abuse, and rape. But they are details of a hard background about which Haas and Hendrix testified they were aware. Haas said she knew Blankenship had a “hard background” and Hendrix said he knew a “great deal” about Blankenship’s background, but no one ever asked either counsel the extent of their knowledge about the background. We are left to speculate as to the exact scope of their knowledge about the details of Blankenship’s upbringing. It is true the family’s affidavits state that if Hendrix or Haas had asked about Blankenship’s background, they would have been willing to recount the detailed history of abuse and neglect. However, this is not particularly helpful in determining the extent of trial counsel’s actual knowledge of his background or the full nature of their investigation. If believed, the affidavit testimony of Blankenship’s mother and sisters would only tell us that the family members themselves never discussed Blankenship’s background with Hendrix or Haas. In his testimony, Hendrix stated his knowledge of Blankenship’s background came from the family and from Blankenship himself. Indeed, the petitioner is often in the best position to inform his counsel of salient facts relevant to his defense, such as his background. Newland v. Hall, 527 F.3d 1162, 1202 (11th Cir.2008). Blankenship was fully capable of appraising Hendrix and Haas of his background and we do not know what, if anything, Blankenship himself did or did not tell counsel. Blankenship’s voice is noticeably absent from the state habeas record, and at no time has he provided testimony or an affidavit suggesting he did not tell his counsel about his personal background. In order to show his trial counsel unreasonably failed to investigate his background, Blankenship would have to show that they did not in fact know about the facts he claims they failed to discover. In this case, the evidence simply does not exist to show Hendrix and Haas did not know about the details of Blankenship’s background. The evidence demonstrates they knew about his alcohol difficulties, his family history of mental illness, and his hard background. We cannot simply assume Blankenship failed to inform them of the details. In light of what the record does (and does not) contain, a reasonable view of the record could find trial counsel were fully aware of petitioner’s life history. 2. Was the investigation reasonable? In light of the above holding, we cannot say the investigation was unreasonable because we cannot say they did not know about the details of Blankenship’s life history. Even if the record demonstrated Hendrix and Haas did not know about Blankenship’s background, however, another fact could make their failure to in *1277 vestigate reasonable: Blankenship himself instructed them not to contact his family. The record certainly does demonstrate Hendrix and Haas spoke to Blankenship’s family — Hendrix would speak with the family to update them on the trial, and Nellie’s affidavit says she spoke with Hendrix. Hendrix, however, testified Blankenship asked them not to involve his family. In fact, Hendrix said “at about the second trial and then at a later point, he admonished me not to write or correspond with or talk on the telephone with his family. I think he wanted them protected.” Hendrix also said Blankenship “directed” them not to speak with family members. “We have ... emphasized the importance of a mentally competent client’s instructions in our analysis of defense counsel’s investigative performance under the Sixth Amendment.” Newland, 527 F.3d at 1202. Significant deference is owed to failures to investigate made under a client’s specific instructions not to involve his family. Id. at 1203. Assuming Hendrix did not know the details of his client’s background, Blankenship’s admonishment to him not to contact his family cannot be ignored. We cannot say it would be unreasonable for Hendrix not to discuss Blankenship’s background with his sisters and mother if Blankenship indeed instructed him not to contact them out of a sense of protection. 7 Putting aside Blankenship’s instruction to his counsel, however, we hold Blankenship failed to satisfy his burden of proving his state trial counsel did not uncover the evidence of his background. As such, state counsel’s investigation of his life history was neither unreasonable nor inadequate. C. Was Counsel’s Choice of Strategy Reasonable? The federal district court found Hendrix and Haas opted for a “mixed” *1278 strategy at the 1986 resentencing, presenting both residual doubt evidence and mitigation evidence. At trial, however, counsel elicited only a few statements from Blankenship about his background and only briefly mentioned his upbringing in closing argument. Since they decided to pursue life history mitigation as part of their trial strategy, Blankenship argues they were ineffective for presenting life history mitigation evidence in such a cursory manner. We disagree. 1. What strategy did counsel select? The district court’s and Blankenship’s belief that trial counsel sought a “mixed” defense of residual doubt and mitigation is belied by ample record evidence. On several occasions, Hendrix and Haas were asked about their strategy at the 1986 resentencing. Haas explained that “especially in the third trial, we were centering in on that hair and blood type .... ” When asked whether it was fair to describe the strategy at the third sentencing trial as demonstrating someone else was present at the scene, she responded affirmatively. She again confirmed that by the third resentencing trial, their “theory mainly had to do with those hairs and that blood type and that part of it. That’s really where we were coming from.” She also said the hair and blood type were crucial pieces of evidence. Hendrix’s testimony dovetails with Haas’ account. He was asked twice if his strategy at trial was focused on the hair and blood evidence and demonstrating the existence of a third person; he responded both times affirmatively. His testimony was that the evidence suggesting another person was present was vital to their strategy. These statements make plain counsel’s strategy: the physical evidence, including hair and possible blood evidence not belonging to Blankenship, could suggest someone else was present at Bowen’s death. Hendrix had won reversal on the 1982 resentencing based on his inability to pursue the strategy that counsel in 1986 ultimately employed. Hendrix doggedly pursued a residual doubt strategy during the 1982 resentencing, only to be thwarted at every turn. The trial court repeatedly blocked his attempts to pursue lines of questioning and introduce witnesses and evidence which would raise questions as to the involvement of another individual in the crime. For example, Hendrix’s questioning of one of the officers at the scene of Bowen’s death was prematurely halted by the prosecutor and the trial court judge because he was seeking information regarding the presence of Blankenship’s fingerprint found on broken glass in the apartment. As Hendrix then saw it, attacking the police investigation and suggesting another person was involved “could mitigate in the minds of the jury the totality of the involvement of the defendant.” As Hendrix himself told the court in the 1982 hearing, he “ha[d] to present everything [he could] in mitigation for [Blankenship]^ life.” Having had his attempts to raise questions of guilt during the hearing undercut, state trial counsel prevailed on appeal on this very point, securing a reversal based on the inability to discuss Blankenship’s guilt during the sentencing phase. In the 1986 sentencing trial, counsel was free to pursue the strategy they thought best appropriate. They did so. At no point did either attorney testify they intentionally sought out mitigation evidence based on Blankenship’s life history. There is no evidence in the record that Blankenship’s counsel anticipated presenting to the 1986 resentencing jury a strategy of mitigation based on compelling life *1279 history testimony. In fact, Hendrix testified Blankenship’s background was not helpful to them. Specifically, he said he “was not aware of any specific information that [Blankenship] had given us concerning his background that we felt mandatory to introduce.” Again he said: “I feel certain today that my feeling was, we won’t be helped by any of this. And there again, that’s a decision you have to make at that time and at that place.” 8 State trial counsel’s testimony at the hearing makes clear only one strategy was pursued at the 1986 resentencing: residual doubt. Blankenship’s brief discussion of his background in the 1986 resentencing and Hendrix’s minor reference to his home town during closing arguments are not sufficient to suggest counsel were pursuing a “mixed” strategy and purposely introduced life history evidence to persuade the jury to spare Blankenship’s life out of mercy. This was not a situation where counsel sought a particular strategy and only introduced cursory evidence in support. Counsel’s opening and closing statements, in addition to the bulk of the evidence introduced in the 1986 resentencing and the testimony of Hendrix and Haas, make clear they pursued only a residual doubt strategy. It would be unreasonable to describe the strategy as “mixed.” 9 2. Was the strategy reasonable? Because we are not dealing with an instance where counsel sought to present a *1280 “mixed” strategy at trial, we are left with a situation where we presume counsel was informed of Blankenship’s background but opted to pursue a residual doubt strategy in lieu of discussion of petitioner’s life history. This strategy was eminently reasonable. “Creating lingering or residual doubt over a defendant’s guilt is not only a reasonable strategy, but is perhaps the most effective strategy to employ at sentencing.” Parker, 331 F.3d at 787-88. In this case, counsel was faced with a brutal rape and murder of an elderly woman. In light of the gruesome facts, including the foreign object left in the victim’s body, reasonably competent counsel could have decided the best chance for sparing Blankenship’s life was to convince the jury some residual doubt existed. Counsel could have concluded the inclusion of extensive mitigating evidence addressing Blankenship’s life history might cloud the jury from focusing on the question of residual doubt, or would simply have been unpersuasive in light of the gruesome nature of the crime. 10 Moreover, Hendrix and Haas’ strategy was far from baseless. There was evidence to suggest the presence of a third person at Bowen’s death. The B-antigen sample, although inconclusive as to whether type-B blood actually existed, arguably suggested the presence of someone other than Blankenship, who was type-O. Similarly suggesting residual doubt was the hair segment found in Bowen’s pubic combings, which did not appear to belong to either her or Blankenship. Finally, counsel presented evidence that another man was responsible for a murder-rape which took place two weeks prior to Bowen’s death and within a few miles of her apartment. Counsel made a reasonable strategic choice to pursue a lingering doubt strategy, and we do not second-guess that decision. Strickland, 466 U.S. at 690-91, 104 S.Ct. at 2066. Indeed, Hendrix and Haas had good reason to believe a residual doubt strategy was the best option: since the 1986 trial was a resentencing trial, they were facing a jury who had not passed judgment on Blankenship’s guilt. When counsel has an opportunity to convince a new jury to spare a defendant’s life, the selection of residual doubt as a trial strategy is particularly sensible. In contrast to asking a jury who just convicted a defendant to spare his life due to residual doubt, a new jury might be more willing to entertain *1281 arguments on guilt and innocence. Blankenship’s counsel fervently argued for the opportunity to introduce residual doubt evidence during appeal from the second sentencing, and they followed through by pursuing the strategy at the final resentencing trial. There was nothing unreasonable about counsel’s actions. IV. Blankenship has failed to overcome the “strong presumption” that his counsel’s performance at the 1986 resentencing was reasonable. See Conklin v. Schofield, 366 F.3d 1191, 1204 (11th Cir.2004). For the reasons stated above, a reasonable view of the record demonstrates Blankenship has not proved counsel was unaware of his life history and did not make a reasonable, strategic choice to pursue residual doubt. Therefore, the state court did not unreasonably apply Strickland in finding Blankenship’s counsel were not ineffective at the final resentencing trial. The district court’s denial of his habeas petition is AFFIRMED. Footnotes [1] Blankenship did not simply give a narrative account of the evening. His statement was a mixture of his narrative and responses to questions and comments from the interrogating officers. Only his statements were recorded, typed and signed as a confession. [2] Blankenship did go into more detail about his drinking. He said a supervisor at the lumber yard where he worked advised him his drinking was excessive and that his friends would get him in trouble. [3] The gap between the conclusion of the state habeas proceedings and the commencement of this federal habeas petition is explained by a procedural history not relevant to this appeal. Suffice it to say in 1993 Blankenship filed a federal habeas petition. Due to a change in Georgia case law, the petition was dismissed without prejudice under certain exhaustion rules. His subsequent Georgia state habeas petition languished in Georgia state courts through 2004, where he was ultimately unsuccessful. None of those proceedings involved the ineffective assistance claim, and the state of Georgia does not raise an argument implicating the procedural timeline. [4] That the state court did not specifically mention Strickland is of no moment. "The state court's failure to cite the relevant Supreme Court precedents does not mean that AEDPA deference does not apply." Parker v. Sec’y for Dep’t of Corr., 331 F.3d 764, 776 (11th Cir.2003). A decision "that does not rest on procedural grounds alone is an adjudication on the merits, regardless of the form in which it is expressed.” Wright, 278 F.3d at 1254-56. It is clear from the state court's resolution of the issue that it recognized Blankenship had raised a constitutional claim of ineffective assistance of counsel, and rejected the claim on non-procedural grounds. [5] We recognize this Circuit’s caution in Cave that, while state court findings of fact can be implied, "they cannot be imagined from thin air.” Cave, 971 F.2d at 1516. We do not disagree. In this case, however, we can easily discern the evidence available before the district court on the question whether Blankenship's counsel were ineffective at the 1986 resentencing. Based on this evidence, we can comfortably decide whether the state court’s application of Strickland was reasonable. [6] Indeed, there is at least some vagueness in Fleming’s affidavit as it relates to Hendrix's account. Her affidavit states she realized the importance of the family history of mental illness before the 1982 resentencing, but she spoke with Hendrix about the family history of mental illness prior to the 1986 third sentencing trial. The evidence supplied by Hendrix in his supplemental affidavit demonstrates he informed Dr. Wolfe of this exact history in September 1982, prior to the second sentencing hearing. This is the type of vagueness which could have been clarified and explained by Hendrix and Haas had Blankenship’s state habeas counsel provided the affidavits before the hearing in which counsel testified. Cf. Waters v. Thomas, 46 F.3d 1506, 1513-14 (11th Cir.1995) ("It is common practice for petitioners attacking their death sentences to submit affidavits from witnesses who say they could have supplied additional mitigating circumstance evidence .... But the existence of such affidavits, artfully drafted though they may be, usually proves little of significance."). [7] Blankenship argues the failure to investigate would be similar to that found to be deficient in Rompilla v. Beard, 545 U.S. 374, 125 S.Ct. 2456, 162 L.Ed.2d 360 (2005). In Rompilla, the attorneys for the petitioner found their client to be unhelpful and uninterested in providing them with mitigation evidence. Id. at 381, 125 S.Ct. at 2462. Counsel also interviewed several family members in a detailed manner, and they told petitioner’s attorney they did not know him well and could not provide much insight into his background. Id. at 381-82, 125 S.Ct. at 2463. Finally, counsel marshaled the opinions of three mental health experts who did not reveal anything useful about petitioner's background. Id. Since the Supreme Court found counsel’s performance deficient in Rompilla, Blankenship reasons Hendrix and Haas' performance should similarly be found deficient, since their efforts did not rise to the level of the counsel in Rompilla. But this argument is easily dismissed because the Court in Rompilla did not pass judgment on whether counsel’s interviews with family, the petitioner himself, and mental health experts were sufficient. Rather, Rompilla found counsel constitutionally deficient because they failed to examine the record in petitioner’s prior felony cases, which would have revealed life history mitigation evidence. Id. at 383, 125 S.Ct. at 2463. Their obligation to review the prior felony case stemmed from their knowledge the prosecutor would use petitioner’s history of felony convictions and violence to seek the death penalty. Id. Because the file would be used to establish aggravating circumstances, counsel had a duty to investigate it. Thus, Rompil-la stands for the proposition that a reasonably competent counsel will investigate a prior felony conviction it knows the prosecution will rely upon in seeking the death penalty. This holding has no bearing on this case, since Blankenship cannot point to any evidence the prosecution relied on in prosecuting Blankenship that his counsel ignored but would have led to the discovery of the details of his life history. [8] To reiterate: we do not know what is meant by "this” and “background” because Hendrix was never asked to explain the extent of his knowledge. [9] In his reply brief, Blankenship argues his trial counsel's conduct is similar to conduct the Supreme Court found to be constitutionally deficient in Wiggins v. Smith, 539 U.S. 510, 123 S.Ct. 2527, 156 L.Ed.2d 471 (2003). We think Wiggins is distinguishable. In Wiggins, counsel sought a bifurcated sentencing proceeding to allow them to first present evidence of residual doubt and then, if necessary, present mitigation. Id. at 515, 123 S.Ct. at 2532. This was denied, and counsel presented a residual doubt case. Counsel mentioned Wiggins' hard life in the opening statement, but did not present any evidence of his life history. Id. In his state post-conviction proceedings, counsel presented evidence of a life history of childhood abuse and neglect, including severe sexual and physical abuse, his trial counsel, a public defender, failed to uncover. Id. at 516-17, 123 S.Ct. at 2532-33. He argued counsel’s failure to uncover this evidence in his sentencing proceedings amounted to constitutionally deficient performance. In agreeing counsel was ineffective, the Supreme Court first noted counsel failed to compile a social history report, even though it was standard practice in Maryland at the time and one for which the public defender’s office provided funds. Id. at 523-24, 123 S.Ct. at 2536-37. Second, the Court said petitioner's Department of Social Service records — which were in the possession of counsel — indicated he had been shuttled from foster home to foster home, had a mother who was an alcoholic, and was on one occasion left home with his siblings without food. Id. at 525, 123 S.Ct. at 2537. In fact, Wiggins himself had described his childhood as "disgusting.” Id. at 523, 123 S.Ct. at 2536. A reasonably competent attorney would have further investigated these hints at a troubling history. Id. Third, counsel’s decision to pursue residual doubt over mitigation was not the result of reasonable judgment because counsel — up to the day of sentencing — sought leave to bifurcate the proceedings to first introduce residual doubt and then introduce mitigating evidence. Counsel intended to present mitigating evidence if the bifurcation motion was successful, which suggested the failure to investigate was not the result of a reasoned, strategic choice. Finally, counsel put forth a 'halfhearted” mitigation case at the sentencing trial. Id. Blankenship's situation is factually distinguishable. We have found Blankenship failed to show Hendrix and Haas were not aware of his background. Unlike counsel in Wiggins, Blankenship's counsel did not fail to investigate in light of records or social history reports suggesting a troubling history. More *1280 over, unlike Wiggins, they did not pursue a mixed strategy at the 1986 resentencing hearing, and therefore did not engage in a “halfhearted” mitigation case. In addition, absent from Wiggins is any suggestion the petitioner instructed counsel not to investigate his background. In this case, Blankenship "admonished” his counsel not to involve his family in his case, so any failure on their part not to investigate is reasonable in light of Blankenship's own actions. [10] We are struck by the similarities between this case and Stewart v. Dugger, 877 F.2d 851 (11th Cir.1989). In that case, the defendant was invited into the home of an elderly, slight woman. Once inside, he attacked the woman and brutally raped her. He then killed her by strangling her with the cord extending from an iron. Id. at 853. In his habeas appeal, the defendant argued his counsel was ineffective for focusing on residual doubt to the exclusion of other possible mitigating evidence. Id. at 856. "Trial counsel made a strategic decision that in light of the atrocious nature of the offense, [the defendant]'s only chance of avoiding the death penalty was if some seed of doubt, even if insufficient to constitute reasonable doubt, could be placed in the minds of the jury .... Trial counsel cannot be faulted for attempting to make the best of a bad situation.” Id. Blankenship's counsel found themselves in a similar situation: they were faced with the heinous facts of a brutal death, and opted to seek a life sentence by planting a seed of doubt with the jury. This was reasonable.