PER CURIAM.

NOT RECOMMENDED FOR PUBLICATION
File Name: 20a0273n.06 ON APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE MIDDLE DISTRICT OF TENNESSEE BEFORE: COLE, Chief Judge; KETHLEDGE and WHITE, Circuit Judges.

PER CURIAM. Petitioner-Appellant Anthony Darrell Dugard Hines, a Tennessee death-row inmate, appeals from the district court's order denying his petition for a writ of habeas corpus filed under 28 U.S.C. § 2254. Because trial counsel were constitutionally ineffective for failing to investigate a crucial witness, and the state court's determination otherwise was an unreasonable application of the clearly established law of Strickland v. Washington, 466 U.S. 668 (1984), we REVERSE.

STATE COURT PROCEEDINGS

In 1986, a jury convicted Hines of first-degree murder and found three aggravating circumstances: (1) Hines was previously convicted of one or more felonies, other than the present charge, which involved the use or threat of violence to the person; (2) the murder was especially heinous, atrocious, or cruel in that it involved torture or depravity of mind; and (3) the murder was committed while Hines was engaged in committing, or was an accomplice in the commission of, or was attempting to commit, or was fleeing after committing or attempting to commit, any rape, robbery, or larceny. See Tenn. Code Ann. § 39-2-203(i)(2),(5),(7) (1982) (repealed). Hines was sentenced to death.

On direct appeal, the Supreme Court of Tennessee affirmed the conviction, but remanded the case for a new sentencing hearing. State v. Hines, 758 S.W.2d 515, 524 (Tenn. 1988). The Supreme Court set forth the following facts:

Between 1:00 and 1:30 p.m. on 3 March 1985 the body of Katherine Jean Jenkins was discovered wrapped in a sheet in Room 21 of the CeBon Motel off Interstate 40 at Kingston Springs. The victim was a maid at the motel and had been in the process of cleaning the room when she was killed. Her outer clothing had been pulled up to her breasts. Her panties had been cut or torn in two pieces and were found in another area of the room. A $20 bill had been placed under the wrist band of her watch.

The cause of death was multiple stab wounds to the chest. Four deep, penetrating wounds, ranging from 2.5 inches to 6.4 inches in depth, had been inflicted about the victim's chest with a knife similar to a butcher knife or a hunting knife. Other superficial cuts were found in the area of the neck and clavicle. There was also a knife wound which penetrated through the upper portion of the vagina into the mesentery in the lower part of the abdominal cavity. Dr. Charles Harlan who performed the autopsy on the victim's body testified that in view of the small amount of blood in the vaginal vault it was his opinion the wound occurred at or about the time of death. The victim also had what he described as "defensive wounds" on her hands and arms.

Jenkins had been left in charge of the motel at about 9:30 a.m. At that time the occupants of Rooms 9, 21 and 24 had not yet checked out. When the manager left her in charge she was given a Cheatham County State Bank bag containing $100 in small bills to make change for motel guests as they paid. The bank bag, bloody and empty, was discovered in the room with her body. It was her established habit to lock her automobile at all times and to keep her keys and billfold on her person when she worked. Her car keys, billfold and her 1980 silver-colored Volvo were missing.

On 1 March 1985 defendant had departed by bus from Raleigh, North Carolina. He had been given a non-refundable ticket to Bowling Green, Kentucky and $20 in spending money. The traveling time from Raleigh, North Carolina to Nashville, Tennessee was approximately 17 hours. Prior to his departure he was observed by a witness to be carrying a hunting knife in a sheath which was concealed beneath his shirt. The witness admonished him that he could not carry a knife like that on

the bus to which he responded "I never go anywhere naked." "I always have my blade." Sometime in the early morning hours of 3 March 1985 he checked in and was assigned to Room 9 at the CeBon Motel. He was wearing a green army-type fatigue jacket, fatigue pants and boots. He was next seen at approximately 9:30 a.m. walking in a direction from his room toward a drink machine. At that time he told the manager he was not yet ready to check out. He was also seen sometime prior to 9:30 purchasing a sandwich at a deli-restaurant across the street from the motel. The same witness who saw defendant also saw another stranger there somewhere between 1:30 and 2:30 who she described as taller than defendant with dark hair, kinky looking and wild-eyed. He departed the restaurant in the general direction of the CeBon Motel. The C[hea]tham County Sheriff testified that he responded to a call to the CeBon Motel at 2:37 p.m. When he arrived on the scene blood spots in the room were beginning to dry and the body was beginning to stiffen. Defendant was seen between 11:00 and 11:30 a.m. walking from the direction of the Interstate toward the CeBon Motel. At 12:40 p.m. a witness saw the victim's Volvo automobile pulling out from the CeBon Motel driveway. It was being operated by a person who appeared to be a man with very short, light colored hair. The vehicle crossed over the Interstate and turned east on Interstate 40. She followed behind and endeavored to catch up but it sped off toward Nashville at a high rate of speed. Defendant was next identified in possession of the car a few miles past Gallatin on Interstate 65, heading in the direction of Bowling Green, Kentucky. A group of young people first endeavored to help him start the stalled automobile and then gave him a ride to Bowling Green.

During the trip to Bowling Green one of these witnesses observed some dried blood on the right shoulder of his shirt. He carried a jacket which he kept folded. After he arrived at his sister's home in Bowling Green defendant told her he had endeavored to pay another day's rent at a motel when he was attacked by the motel operator. He demonstrated to her how he had stabbed the man. He also related to her he had a sum of money. She could not remember whether he said $35,000 or $3,500. Defendant also told his sister's husband he had earned approximately $7,000 working as a mechanic in North Carolina. He displayed a set of keys to a Volvo automobile and explained that a man who had given him a ride attempted to rob him. Defendant purportedly grabbed the steering wheel and when the car ran off the road he grabbed the keys and ran. According to the witness he was wearing an army fatigue jacket which had something large, heavy and bulky in the pocket. The witness had previously seen defendant with a survival knife with a 6 ½ to 7 inch blade hanging from his belt.

When defendant was taken into custody he volunteered the statement that he had taken the woman's car but had not killed her. According to the arresting officer he had not advised the defendant that a woman had been killed prior to the volunteered statement. There was evidence however that defendant was aware he had been charged in Tennessee on a murder warrant. The victim's wallet was found wrapped in a thermal underwear shirt a short distance from where her car was found abandoned. The key to Room 9 of the CeBon Motel was found at the site where

defendant had been camping out near Cave City, Kentucky. When asked by a [Tennessee Bureau of Investigation ("TBI")] agent to tell the truth about the death of Katherine Jenkins defendant stated that if the officer could guarantee him the death penalty he would confess and tell him all about the murder and that he could tell him everything he wanted to know if he was of a mind to. There were marks on the wall of Room 9 at the CeBon Motel apparently made by someone stabbing a knife into the wall. When shown photographs of the marks on the wall defendant responded that they were knife marks. These marks were obviously made by a knife larger than [the] two taken from defendant at the time of his arrest.

In March 1997, Hines sought post-conviction relief. The trial court held evidentiary hearings and denied relief. The Tennessee Court of Criminal Appeals affirmed the decision. Hines v. State, No. M2002-01352-CCA-R3-PD, 2004 WL 112876, at *39 (Tenn. Crim. App. Jan. 23, 2004). In June 2004, the Tennessee Supreme Court granted Hines's application to appeal and remanded the case to the court of criminal appeals to reconsider its determination that the trial court submitted an incorrect version of the aggravating circumstance in Tenn. Code Ann. § 39-2-203(i)(5) to the jury. On remand, the court of criminal appeals held that the (i)(5) aggravating circumstance instruction had been proper and again denied relief. Hines v. State, No. M2004-01610-CCA-RM-PD, 2004 WL 1567120, at *1 (Tenn. Crim. App. July 14, 2004). The Tennessee Supreme Court denied leave to appeal.

Hines's second post-conviction petition seeking funds and authorization to conduct DNA testing was unsuccessful. Hines v. State, No. M2006-02447-CCA-R3-PC, 2008 WL 271941, at *1 (Tenn. Crim. App. Jan. 29, 2008). The Tennessee Supreme Court denied leave to appeal.

FEDERAL COURT PROCEEDINGS

In January 2005, Hines filed a pro se petition for a writ of habeas corpus in the district court. After Hines received appointed counsel, he filed an amended petition in June 2005. One month later, Hines filed another amended petition in which he asserted thirty-one claims of constitutional error. The warden filed a response. In September 2005, Hines filed a motion to conduct discovery. In November 2005, the district court held the case in abeyance to allow Hines to pursue state-court remedies under Tennessee's Post-Conviction DNA Analysis Act of 2001. Hines's petition for DNA testing was ultimately denied, and the denial was affirmed on appeal. Hines v. State, M2006-02447-CCA-R3-PC, 2008 WL 271941, at *8 (Tenn. Crim. App. Jan. 29, 2008), perm. app. denied (Tenn. Dec. 8, 2008).

The federal case resumed in February 2009. In October 2010, the district court granted Hines permission to conduct DNA testing. In February 2013, the district court held the case in abeyance pending the issuance of Trevino v. Thaler, 569 U.S. 413 (2013), in light of the Supreme Court's decision in Martinez v. Ryan, 566 U.S. 1 (2012). In May 2014, the warden filed a motion for summary judgment. After holding an evidentiary hearing, the district court granted the warden's motion and denied Hines's petition. The district court certified for appeal all claims related to the death sentence. Following a remand for reconsideration of its certified claims under Slack v. McDaniel, 529 U.S. 473 (2000), the district court narrowed the scope of the claims certified for appeal. We expanded the certification.

STANDARD OF REVIEW

The district court's denial of a habeas petition is reviewed de novo. Adams v. Bradshaw, 826 F.3d 306, 309 (6th Cir. 2016), cert. denied, 137 S. Ct. 814 (2017). The district court's findings of fact are reviewed for clear error, and its legal conclusions on mixed questions of law and fact are reviewed de novo. Id. at 309-10. Hines's petition was filed in January 2005 and is subject to the requirements of the Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA). Id. at 310. Under AEDPA, a writ shall not be granted unless the state court's adjudication of the claim:

(1) resulted in a decision that was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States; 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.

DISCUSSION

The following claims were certified for appeal: (1) whether Hines was entitled to an evidentiary hearing concerning (a) DNA and fingerprint evidence that would have supported an actual innocence claim to overcome a procedural bar, and (b) declarations by Hines's trial and post-conviction counsel concerning their omissions; (2) whether trial counsel were ineffective for failing to: (a) challenge the jury panel as to the underrepresentation of women, (b) make a closing argument at the resentencing hearing, (c) challenge the underrepresentation of women on the petit and grand juries, present evidence of Hines's personal history as well as his alcohol and drug abuse, and object to the prosecution's failure to provide notice of aggravating circumstances, (d) interview and conduct an effective cross-examination of Ken Jones, (e) investigate and present evidence of residual doubt, (f) challenge Dr. Charles Harlan's testimony, and (g) challenge the imposition of the death penalty as arbitrary and unconstitutional because the trial judge rejected a plea agreement which would have resulted in a life sentence; and (3) the prosecution withheld exculpatory evidence and elicited false testimony in violation of Brady v. Maryland, 373 U.S. 83 (1963), and Giglio v. United States, 405 U.S. 150 (1972). On appeal, Hines does not address his claims of trial counsel ineffectiveness for failing to make a closing argument at the resentencing hearing and to object to the prosecution's failure to provide notice of aggravating factors. Hines has thus waived those claims. Elzy v. United States, 205 F.3d 882, 886 (6th Cir. 2000).

For the reasons described below, we hold that trial counsel were ineffective for failure to interview and conduct an effective cross-examination of Ken Jones, and for the related failure to investigate and present evidence of residual doubt in relation to Ken Jones at the penalty phase of the trial. The state court's contrary ruling was an "unreasonable application of . . . clearly established Federal law, as determined by the Supreme Court of the United States" in Strickland v. Washington, 466 U.S. 668 (1984), and we thus reverse the district court's denial of Hines's petition for a writ of habeas corpus. 28 U.S.C. § 2254(d)(1). Before discussing this meritorious claim, we explain below why we reject Hines's other claims of error. I. Whether Hines's claim that trial counsel were ineffective for failing to investigate and conduct forensic testing warranted an evidentiary hearing.

Hines's habeas petition asserted the ineffective assistance of trial counsel (IATC) arising from trial counsel's failure to investigate and conduct forensic testing of various pieces of evidence. Hines now argues that he was entitled to an evidentiary hearing on this IATC claim because the evidence demonstrated his actual innocence. Acknowledging that this IATC claim is procedurally defaulted, Hines argues that a showing of either actual innocence or the ineffective assistance of post-conviction counsel can overcome this default.

The district court's decision not to hold an evidentiary hearing is reviewed for an abuse of discretion. Hodges v. Colson, 727 F.3d 517, 541 (6th Cir. 2013) (citations omitted). No evidentiary hearing is required "if the record refutes the applicant's factual allegations or otherwise precludes habeas relief." Schriro v. Landrigan, 550 U.S. 465, 474 (2007).

The district court held the habeas proceedings in abeyance and permitted Hines to exhaust state-court remedies concerning the DNA testing of certain evidence. In state court, Hines filed a successive post-conviction petition seeking to subject the following evidence to DNA testing: Jenkins's underwear; Jenkins's dress; Jenkins's slip; a bloody bank bag; a cigarette butt from Room 21; a twenty-dollar bill that was found on Jenkins; and a plastic spray bottle found in Room 21. Hines, 2008 WL 271941, at *3. The trial court denied the petition and that decision was affirmed on appeal. Id. at *6-8.

When the habeas proceedings resumed, the district court granted Hines discovery and permitted DNA testing of the same evidence. Test results revealed that a section of Jenkins's underwear contained "a mixture of DNA from at least three individuals." (R. 124-1, PID 1349.) The results further clarified that the sample contained genetic material from "at least two male individuals," and that Hines was excluded as a contributor. (Id.) The district court did not hold an evidentiary hearing on this matter because Jenkins's murder was not committed in the context of sexual assault. (R. 145, PID 2310 ("If this murder involved sexual intercourse, the Court would be inclined to agree with Petitioner about this DNA evidence warranting an evidentiary hearing. Yet, the victim's death was caused by multiple and deep knife wounds to her chest area including her heart, lungs[,] and diaphragm.").) The district court concluded that the IATC claim was procedurally defaulted because it was not raised in state court and, alternatively, lacked merit.

Hines concedes that this IATC claim is procedurally defaulted. Review of this claim is thus barred "unless the prisoner can demonstrate cause for the default and actual prejudice as a result of the alleged violation of federal law, or demonstrate that failure to consider the claims will result in a fundamental miscarriage of justice." Coleman v. Thompson, 501 U.S. 722, 750 (1991). "[I]n an extraordinary case, where a constitutional violation has probably resulted in the conviction of one who is actually innocent, a federal habeas court may grant the writ even in the absence of a showing of cause for the procedural default." Murray v. Carrier, 477 U.S. 478, 496 (1986). To successfully assert actual innocence, a petitioner "must show by clear and convincing evidence that, but for a constitutional error, no reasonable juror would have found the petitioner eligible for the death penalty under the applicable state law." Sawyer v. Whitley, 505 U.S. 333, 336 (1992). Because Hines's IATC claim is procedurally defaulted, the actual innocence claim "is thus 'not itself a constitutional claim, but instead a gateway through which a habeas petitioner must pass to have his otherwise barred constitutional claim considered on the merits.'" Schlup v. Delo, 513 U.S. 298, 315 (1995) (quoting Herrera v. Collins, 506 U.S. 390, 404 (1993)). Hines must show that "it is more likely than not" that "in light of the new evidence, no juror, acting reasonably, would have voted to find him guilty beyond a reasonable doubt." Id. at 327-29.

To support his claim of actual innocence, Hines characterizes his trial as one involving a sexual assault and murder, and relies on House v. Bell, 547 U.S. 518 (2006), to underscore the importance of the DNA test results.

In House, the petitioner Paul House was convicted of the first-degree murder of Carolyn Muncey and sentenced to death. Id. at 521. At trial, the prosecution presented evidence suggesting that DNA material found on Muncey's nightgown and underwear belonged to House, and that the blood found on House's jeans belonged to Muncey. Id. at 528-30. In closing arguments, the prosecution argued that House's desire to have sex with Muncey was a possible motive for the crime. Id. at 531-32. House unsuccessfully filed two state post-conviction petitions, the second of which contained several IATC claims that were found to be waived. Id. at 533-34. On habeas review, the district court and this court held that House had not shown actual innocence to overcome the default of his IATC claims. Id. at 534-36 (citing House v. Bell, 386 F.3d 668 (6th Cir. 2004) (en banc)).

The Supreme Court disagreed. The Court noted that subsequent testing revealed that Muncey's nightgown and underwear contained DNA from her husband rather than from House, eliminating "the only forensic evidence at the scene that would link House to the murder." Id. at 540-41. The Court observed that this new information altered the prosecution's theory of the case: "When the only direct evidence of sexual assault drops out of the case, so, too, does a central theme in the State's narrative linking House to the crime." Id. at 541. Additional testing prompted one expert to opine that the blood on House's jeans "came from the autopsy samples, not from Mrs. Muncey's live (or recently killed) body." Id. at 543. The Court explained that the prosecution's efforts to discredit the expert's opinion were undermined by a police officer's statement that he saw "reddish brown stains" on House's blue jeans and that "[t]he pants were in fact extensively soiled with mud and reddish stains, only small portions of which are blood." Id. at 547. Other evidence showed that Muncey's marriage had involved physical abuse and that Muncey's husband had purportedly confessed to killing her. Id. at 549-50. The Court granted relief because "the central forensic proof connecting House to the crime—the blood and the semen—has been called into question, and House has put forward substantial evidence pointing to a different suspect." Id. at 554.

Hines also relies on Mills v. Barnard, 869 F.3d 473 (6th Cir. 2017). The plaintiff in Mills had been convicted of rape of a child, aggravated sexual battery, and casual exchange of a controlled substance. Id. at 478. The primary evidence in support of his conviction was the victim's statement that she had engaged in sexual intercourse with Mills and corroborating DNA evidence suggesting that Mills was a possible source of a DNA sample. Id. On habeas review, Mills presented new DNA analysis that directly contradicted the state's inconclusive evidence and excluded Mills as a contributor of the DNA. Id. at 478-79. The Tennessee Court of Criminal Appeals overturned all of Mills' convictions, finding that the DNA evidence called into question the victim's statement. Id.

Here, unlike in House and Mills, DNA evidence was not used to establish Hines's guilt. At trial, Sheriff Weakley testified that Room 21 did not contain Hines's blood or fingerprints and that Jenkins's blood was not found on Hines's clothing. At the 1989 resentencing hearing, the medical examiner, Dr. Harlan, testified that "[a] visual inspection [of Jenkins] was performed. Since there was no material that was indicative of semen, no scientific or laboratory study was performed, since there was no such material to evaluate." (R. 173-9, PID 4737.) He confirmed that he did not see any visual indication that a sexual assault had occurred. When asked whether that precluded a "penile sexual assault," Dr. Harlan answered: "Not necessarily. It means that there is no semen present, so there was no ejaculation." (Id. at PID 4740.) In closing argument at Hines's resentencing, the prosecutor mentioned the injury to Jenkins's vagina, but did not describe the injury as a sexually motivated crime; rather, the wound was evidence of a "reprehensible," "vile" act by "a depraved mind." (R. 173-11, PID 5029, 5032, 5037.) Hines's reliance on House and Mills is thus misplaced because the new evidence does not undermine any evidence central to his conviction.

Hines also submits that new fingerprint evidence supports his actual-innocence claim. Max Jarrell, a former fingerprint examiner for the Federal Bureau of Investigation (FBI), declared that Hines's fingerprints did not match fingerprints found at the scene and on the relevant evidence. As with the DNA evidence, however, this does not demonstrate Hines's actual innocence. The prosecution did not rely on fingerprint evidence to convict Hines, and the existence of other people's fingerprints at the scene does not undermine any of the evidence that the prosecution presented at trial.

The prosecution offered the following evidence at trial. Sheriff Weakley testified that Hines confessed to stealing Jenkins's car, which had the keys in it, and left the motel at either 8:30 or 9:00 a.m. on Sunday. Jenkins's husband testified that Jenkins locked her car "religiously" and used a key ring emblazoned with the words "I love my Volvo," with a heart symbol in the place of "love." (R. 173-1, PID 3826-28.) Gay Doyle, who managed the CeBon Motel, testified that Jenkins "always locked her car" and "always kept [her keys] in her pocket with her." (Id. at PID 3850-52.) Doyle testified that she saw Hines walking from his room to a vending machine when she was leaving the motel "[b]etween 9:25 and 9:30." (Id. at PID 3849.) Penny Rust, who worked with Jenkins on a part-time basis, testified that she saw Jenkins's car leave the motel at 12:40 p.m. on Sunday; though she could not determine the driver's gender, she knew that Jenkins was not driving because she "would never drive [that] fast." (R. 173-2, PID 3883-84.)

Daniel Blair testified that he and his friends were driving to Bowling Green, Kentucky when they saw Hines stranded on the side of the road in a silver car that had overheated. After looking for water for thirty to forty-five minutes, Blair and his friends gave Hines a ride to Bowling Green, where Hines's sister lived. This journey lasted "around an hour." (Id. at PID 3908.) Hines told Blair that he had purchased the car from an "old lady" for "three or four hundred dollars." (Id. at PID 3910-11.) Blair noticed that Hines had a key attached to a "black thing" that had a "9" on it. (Id. at PID 3913-14.) Blair and his friends dropped Hines off at "3:00 or 4:00 o'clock." (Id.)

Victoria Daniel, Hines's sister, testified that Hines was in her home when she arrived there between two o'clock and five o'clock on Sunday afternoon. Daniel testified that Hines told her that he was attacked when he attempted to pay rent for another night at a motel but "got [the attacker] in the side, you know, and in the chest" with a knife. (R. 173-2, PID 3966-68.) Daniel noticed that Hines had "something reddish" on his t-shirt, which she described "as blood at first." (Id. at PID 3974.) She also saw Hines with a key ring that had the words "I love Volvo" on it. (Id. at PID 3975.)

Robert Daniel, Hines's brother-in-law, testified that he noticed that Hines had an "I love Volvo" key chain. Hines explained that he grabbed the key chain from the car's ignition after a man who had picked him up tried to rob him.

Hines has not established that it is more likely than not that no reasonable juror would have convicted him in light of the new forensic evidence. See Schlup, 513 U.S. at 327. Although a "petitioner's showing of innocence is not insufficient solely because the trial record contained sufficient evidence to support the jury's verdict," id. at 331, the lack of Hines's DNA on Jenkins's underwear and the failure to find his fingerprints on various items in the hotel does not contradict the evidence presented by the prosecution at trial. As such, Hines cannot satisfy his burden. Cf. Souter v. Jones, 395 F.3d 577, 596-97 (6th Cir. 2005) (finding that petitioner established actual innocence by presenting new evidence that diminished impact of trial evidence linking petitioner to commission of crime).

Alternatively, Hines argues that the ineffectiveness of counsel for failing to raise the IATC claim on post-conviction review can overcome the procedural bar. In Martinez, the Supreme Court held that a successful claim asserting the ineffective assistance of post-conviction counsel could excuse the procedural default of an IATC claim if the IATC claim could only be raised on collateral review:

Where, under state law, claims of ineffective assistance of trial counsel must be raised in an initial-review collateral proceeding, a procedural default will not bar a federal habeas court from hearing a substantial claim of ineffective assistance at trial if, in the initial-review collateral proceeding, there was no counsel or counsel in that proceeding was ineffective.

A successful claim of counsel ineffectiveness requires a showing that counsel's performance was deficient and that the deficient performance prejudiced the defense. Strickland v. Washington, 466 U.S. 668, 687 (1984). Deficient performance occurs when counsel makes "errors so serious that counsel was not functioning as the 'counsel' guaranteed the defendant by the Sixth Amendment." Id. Prejudice occurs when counsel's errors are "so serious as to deprive the defendant of a fair trial, a trial whose result is reliable." Id.

In support, Hines relies on a declaration by his post-conviction counsel, Donald E. Dawson, who stated in relevant part:

Though the post-conviction petition raised a general claim that counsel was ineffective for failing to properly obtain and examine the physical evidence at the original trial, our office never presented any evidence in support of such a claim at the evidentiary hearing. We did not raise a similar claim related to the resentencing hearing. We never secured the physical evidence or otherwise employed forensic experts (such as fingerprint or serology experts) to assist us in the investigation of the physical evidence and claims relating to counsel's ineffectiveness for failing to obtain, examine, and/or test the physical evidence. We had no tactical reason for not doing so.

Hines next contends that trial counsel were ineffective because they did not challenge the systematic underrepresentation of women on the venires in Cheatham County, Tennessee, or the systematic exclusion of women from serving as a grand jury foreperson in Cheatham County. The warden argues that the state court's resolution of the former claim was not an unreasonable application of Supreme Court precedent. As to the latter claim, the warden argues that the claim is procedurally defaulted.

Hines must show that trial counsel performed deficiently, that is, "made errors so serious that counsel was not functioning as the 'counsel' guaranteed the defendant by the Sixth Amendment." Strickland, 466 U.S. at 687. In so doing, Hines "must overcome the presumption that, under the circumstances, the challenged action 'might be considered sound trial strategy.'" Id. at 689 (quoting Michel v. Louisiana, 350 U.S. 91, 101 (1955)). Second, Hines must show that the deficient performance resulted in prejudice, meaning, "counsel's errors were so serious as to deprive the defendant of a fair trial, a trial whose result is reliable." Id. at 687.

Hines argues that trial counsel should have challenged the underrepresentation of women on the venires. Under the Sixth Amendment, a criminal defendant is guaranteed the right to be tried by a jury selected from a fair cross-section of the community. Taylor v. Louisiana, 419 U.S. 522, 537 (1975). A prima facie showing of the denial of this right requires a defendant to show:

(1) that the group alleged to be excluded is a 'distinctive' group in the community; (2) that the representation of this group in venires from which juries are selected is not fair and reasonable in relation to the number of such persons in the community; and (3) that this underrepresentation is due to systematic exclusion of the group in the jury-selection process.

Hines correctly identifies women as a distinctive group within a community. See id. at 364 (citing Taylor, 419 U.S. at 531); Ford v. Seabold, 841 F.2d 677, 681 (6th Cir. 1988). Thus, Hines has satisfied the first prong.

To satisfy the second prong, Hines refers to the Tennessee Court of Criminal Appeals' opinion concerning the disparity between the percentage of women living in Cheatham County and the percentage of women who have comprised the venires:

Based upon the report of Dr. James O'Reilly which provided that the percentage of women in Cheatham County between 1979 and 1990 was 50.6 to 50.7% of the population, but the percentage of women in the Cheatham County jury venire for that same time period was between 10 and 22%, the State conceded that the first two prongs of the Duren test had been satisfied.

The issue then is whether Hines has established a systematic exclusion of women from the jury selection process. Unlike in Taylor and Duren, Hines does not argue that statutory provisions form the basis of the exclusion. Rather, he relies on testimony offered during the initial post-conviction proceedings to support this claim.

Jennie Delores Harris Moulton, who worked in the clerk's office of the Cheatham County circuit court during Hines's trials, testified about the procedure for selecting a venire. A judge appointed three individuals as jury commissioners who would meet with the court clerk "each month to draw out names to get a panel of jurors for the next upcoming court." (R. 174-2, PID 5354.) This was called the "sheriff's voir dire." (Id. at PID 5355.) The sheriff served summons for those individuals, who, when they appeared, were placed on "a jury list." (Id.) "The judge, at that time, drew out a panel for the grand jury" and "the petit jury—the trial jury." (Id.)

The first step of the process involved "charging the jury box," which, as Moulton explained, meant that "they gather new names and all and put [the names] back into the box," a procedure that occurred every two years "unless the box is getting low and you need to do it more than that." (Id. at PID 5357.) Moulton explained that the jury commissioners obtained the names for the box from "the voter registration list, because we had more access to it. And they would go randomly, maybe, every sixteenth one or twentieth one down and write the name and address on a little jury ticket." (R. 174-2, PID 5357-58.) When charging the box, each of the three jury commissioners worked from a separate section of the County's voter registration list and would independently select names. Moulton testified that to assemble the list that would become the sheriff's voir dire, either a child or a blindfolded person would draw names from the box. Two men and one woman served as the jury commissioners; Ms. Adkisson, one of the commissioners, wrote down the names as the other two commissioners called them out.

Moulton testified that the jury commissioners tried "to get good solvent jurors" and would remove a name from consideration "if they knew at that time if there was someone that was deceased or someone that was real sick or in the hospital or if it was a student that they knew was off to college somewhere, [or] someone that maybe was in jail." (Id. at PID 5368.) When the selected person was a school teacher, the jury commissioners "would pitch it back—or lay him over to the side to put back into the box where they could—he could serve maybe during the summer months . . . when he's not in school." (Id. at PID 5368-69.) When asked whether women with children were also removed, Moulton replied: "Well, no, not all women with children. But if they had just had a baby or something and they knew it, yeah, you know, they—they did." (Id. at PID 5369.)

Moulton testified that once the jury commissioners compiled a list of individuals, the list was given to the sheriff, who prepared the jury summonses; the person was either served or instructed by telephone to pick up the summons, which contained the date for the individual to appear for jury service. When those summoned appeared in court, their information was collected and they were assigned a number. A judge then selected numbers from a box; the first twelve individuals who had a corresponding number served on the grand jury, and the remainder on the petit juries.

Moulton testified that the number of individuals who appeared pursuant to the summonses varied. Sheriff Weakley, who was sheriff during both of Hines's trials, would be given a list of 150 individuals to summon, but only a third or less would appear. By contrast, under Sheriff Weakley's successor, approximately 75 to 80 percent would appear. There was no discipline for individuals who failed to appear, and no effort was made to understand why compliance was low during Sheriff Weakley's tenure. The individuals who did appear were divided into petit jurors and grand jurors.

When asked whether a conscious effort was made to exclude women from jury service, Moulton's response was characterized as inaudible in the transcript. Moulton denied that Black individuals were excluded from jury service. Moulton was again asked whether there was a conscious effort to exclude a particular group of people, and she responded: "I don't think it was an intentional thing. But Ms. Martha Adkisson, she didn't like too many women on the jury . . . She would say, [']Getting too many women, getting too many women.[']" (R. 174-2, PID 5383.) Moulton expanded her answer:

[S]he would be writing [the names] down, she would tell the guys, say, ["]We're getting too many women, getting too many women.["] I think they wanted to equal it out, but she had a thing about putting too many women on the jury. So when I wound up at one time, whenever I was clerk—or even when Mr. Harris was—was clerk—we wound up with a big box of women.

Lloyd Harris testified that he had "never—never seen anyone" on the jury commission do something that would prevent a group of people from serving on the jury. (R. 174-3, PID 5417.) Harris denied hearing Ms. Adkisson say that the jury list contained too many women. Harris acknowledged, however, that some judgments were made:

It was two men and Ms. Adkisson. She was a school teacher and she done the writing down. We'd have somebody draw them out of the box, a small child or somebody blindfolded. They would draw the names out and she would write them down. And she'd come across one, maybe, was a school teacher that she knew in the county. And she'd say, ["]It's going to be hard for her to serve because she's a school teacher.["] And back then, you couldn't get nobody, you know, to—fill in.

When asked about women being excused from jury service, Harris testified that "[i]t was easy for them to get off" because "most of them had children at home and had to take care of them. They didn't have no babysitter. That's the number one thing." (R. 174-3, PID 5419.) Harris was asked whether he "observe[d] whether the court seemed to be more inclined not to let folks just be off jury duty just because they wanted to be off jury duty?"; he responded: "I think so, yes, sir." (Id. at PID 5420.)

On re-direct examination, Harris was asked whether Adkisson chose jurors, and he responded: "Only times she'd ask them not to [be] put on there if it was a school teacher or if it was a woman she knowed (sic) that had a bunch of children, [and had] nobody to stay with them[.]" (Id. at PID 5434-35.) Harris testified that, because the commission met once or twice a year during the school year, multiple possible jurors may have been excluded per year. Harris testified that, notwithstanding Adkisson's possible opinion regarding there being "too many women," he did not observe that opinion having any effect on the way that names were chosen. (Id. at PID 5435.)

Hines relies on Duren v. Missouri to support this claim. 439 U.S. 357 (1979). In Duren, there were two opportunities for the systematic exclusion of women from venires to occur: (1) when the questionnaires were sent to those randomly selected from the voter registration list, they contained language stating that a woman could elect not to serve by indicating that desire on the form and returning the questionnaire to the jury commissioner; and (2) a woman's failure to respond to a summons was treated as a claimed exemption, whereas other individuals had to do more to benefit from an exemption. Id. at 361-62, 366-67. Women comprised 54% of the Jackson County, Missouri, population at the relevant time. Id. at 362. Noting that "the percentage of the women at the final, venire, stage (14.5%) was much lower than the percentage of women who were summoned for service (26.7%)," the Court concluded that the petitioner "demonstrated that the underrepresentation of women in the final pool of prospective jurors was due to the operation of Missouri's exemption criteria." Id. at 367.

Hines fails to show an entitlement to habeas relief on this ineffective-assistance-of-counsel claim with regard to either his 1986 or 1989 trials. As to Hines's 1986 trial, the Tennessee Court of Criminal Appeals concluded that "[t]he record supports the post-conviction court's finding that Hines was not prejudiced" by counsel's failure to "challenge the 1986 venire." Hines, 2004 WL 1567120, at *36. We cannot say that the Tennessee court's conclusion was an unreasonable application of clearly established Supreme Court precedent. See 28 U.S.C. § 2254(d). Hines must show prejudice to succeed on his IATC claim, even though the alleged underlying error was structural in nature. See Ambrose v. Booker, 684 F.3d 638, 652 (6th Cir. 2012). To succeed on a fair-cross-section theory in the context of his ineffective-assistance claim, Hines must show that, given the underrepresentation of women in the jury venire, there was a reasonable probability that "a properly selected jury [would] have been less likely to convict." Id. at 652 (quoting Hollis v. Davis, 941 F.2d 1471, 1482 (11th Cir. 1991)); see also Garcia-Dorantes v. Warren, 801 F.3d 584, 596-98 (6th Cir. 2015). Hines has not shown a reasonable probability that the result of his trial would have been different if the jury venire more adequately represented the community, particularly in light of the fact that three women served on the petit jury.

Regarding Hines's challenge to his 1989 resentencing jury, AEDPA also bars Hines's relief because the Tennessee Court of Criminal Appeals reasonably determined that Hines "failed to show that he was prejudiced" by 1989 counsel's decision not to challenge the venire. Hines, 2004 WL 1567120, at *36. The Tennessee court's decision was not an unreasonable application of Strickland.

The same is true of Hines's assertion that women were systematically excluded from serving as the foreperson of the grand jury. "[A] criminal defendant's right to equal protection of the laws has been denied when he is indicted by a grand jury from which members of a . . . group purposefully have been excluded." Rose v. Mitchell, 443 U.S. 545, 556 (1979). To prevail, Hines must (1) identify a distinctive and recognizable group, (2) determine the disparity between the identified group and the proportion of the group called to serve over a period of time, and (3) establish that the selection procedure is not gender-neutral. Id. at 565 (citing Castaneda v. Partida, 430 U.S. 482, 494 (1977)).

Here, Moulton testified that the grand jury foreman was selected and sworn in by the judge. The grand jury foreman was appointed for two-year terms; Mouton recalled that two men—Buddy Frazier and Billy Ellis—were appointed repeatedly. Hines also provided an affidavit from Gaye Nease, an investigator with the Office of the Federal Public Defender, stating that no woman had been chosen as a grand-jury foreperson between 1919 and 1985, when Hines was indicted.

Again, Hines fails to show prejudice. We also note that the grand jury foreperson's duties and powers in Cheatham County during this time appear to be merely "ministerial," and therefore do not present a risk of prejudice. Hines, 2004 WL 1567120, at *36 n.3. Contrary to the Supreme Court's description of the role of the Tennessee grand jury foreperson in Hobby v. United States, 468 U.S. 339, 348 (1984), the relevant state statute establishes that the foreperson has the same voting power as any other grand juror, Tenn. Code Ann. § 40-1506, and therefore does not have "virtual veto power over the indictment process." Hobby, 468 U.S. at 348. Thus, Hines is not entitled to relief on this claim. III. Whether the prosecution withheld exculpatory evidence and elicited false testimony.

Next, Hines contends that he was denied due process because the prosecutor (1) did not disclose records indicating the existence of exculpatory DNA evidence and (2) presented testimony that falsely denied the existence of that evidence.

Before trial, defense counsel was given a report from the Tennessee Bureau of Investigation (TBI) that stated that the medical examiner's swabs "failed to reveal the presence of spermatozoa." (R. 175-6, PID 5790.) Post-conviction counsel later discovered some handwritten "raw notes" from the TBI stating that the swabs that had been sent to the TBI for testing had molded. (Hines Br. at 20; R. 175-4, PID 5787.) The final report did not disclose this fact. Hines now argues that the TBI did not find an absence of spermatozoa on the swabs. Rather, "the raw notes" reveal "no sperm was seen microscopically (likely because of mold)" and "no testing for semen was done on the swabs because of their molded condition." (Hines Br. at 20.) Thus, Hines argues, the final report was misleading. Further, because defense counsel was unaware that there might have been spermatozoa present, they were unable to explore the theory that the crime was committed by an unknown assailant by conducting their own testing for semen or DNA.

Hines also asserts that, had the raw notes been disclosed at trial, defense counsel would have been afforded an opportunity to impeach Dr. Harlan's testimony that there was "no material that was indicative of semen" present. (R. 173-9, PID 4737.)

The warden responds that Hines cannot overcome the procedural default of this claim.

Hines first raised this claim on habeas review, asserting:

The prosecution also knowingly presented false testimony from [Dr.] Harlan that there was no evidence of semen and that there was no study performed on any such evidence, and the prosecution withheld evidence which demonstrated the falsity of that testimony and which was otherwise material to the jury's guilt and death verdicts, including proof of the results of any such scientific or laboratory study concerning the existence and nature of any semen.

First, there is not any scientific evidence that the mold was caused by the timing of the TBI laboratory testing. The possibility of the mold impacting any semen is speculative. There is not any scientific proof that if Petitioner's trial counsel had seen the laboratory working papers [the raw notes] about the molded swab that any testing could have been conducted. As discussed infra, Dr. Harlan's trial testimony

was based upon his visual examination of the victim, not a laboratory test. At Petitioner's trial, a TBI laboratory technician testified about the testing of the victim's swabs. To date, the proof remains that the several swabs taken from the victim did not contain semen. Petitioner's tying of the inferences about another suspect was found by the State courts to be "farfetched" and this Court agrees. The cited suspect was not seen in the area at the time of the murder and the witness did not testify that this suspect, described as a wild person, was going to the motel.

"[T]he suppression by the prosecution of evidence favorable to an accused upon request violates due process where the evidence is material either to guilt or to punishment, irrespective of the good faith or bad faith of the prosecution." Brady, 373 U.S. at 87. "[E]vidence is material only if there is a reasonable probability that, had the evidence been disclosed to the defense, the result of the proceeding would have been different." United States v. Bagley, 473 U.S. 667, 682 (1985). "A 'reasonable probability' is a probability sufficient to undermine confidence in the outcome." Id. The Brady rule applies to both exculpatory and impeachment evidence. Giglio, 405 U.S. at 154. The omission of such evidence "must be evaluated in the context of the entire record." United States v. Agurs, 427 U.S. 97, 112 (1976).

At the guilt phase, Dr. Harlan testified that he performed Jenkins's autopsy on March 4, 1985. Jenkins suffered a stab wound to her vaginal vault. It was the final wound inflicted on Jenkins.

At the resentencing hearing, defense trial counsel asked Dr. Harlan on cross-examination whether he conducted any tests to determine whether sperm was present or a sexual assault had occurred, and he responded: "A visual inspection was performed. Since there was no material that was indicative of semen, no scientific or laboratory study was performed, since there was no such material to evaluate." (R. 173-9, PID 4737.) Trial counsel then asked, "So you didn't even observe visually anything indicating any type of sexual assault[,] is that right?"; Dr. Harlan responded, "That is correct." (Id.) The prosecution pursued this topic on re-direct examination:

Q: Dr. Harlan, when you said there was no evidence of sexual assault, you meant there was no—what type of evidence did you mean?
A: I meant that there was no evidence of ejaculation; that, there was no semen present.
Q: There would be no penile sexual assault, then? Would that define it better?
A: Not necessarily. It means that there is no semen present, so there was no ejaculation.

In a deposition taken during the initial post-conviction proceedings, Dr. Harlan explained that an autopsy can determine whether sexual contact had occurred but not sexual assault because "sexual assault . . . has a legal connotation" related to whether there was consent "that goes beyond whether there's semen present or not." (R. 141-1, PID 2045-46.) He stated that he would consider a stab wound to the vagina a suspicious circumstance that would prompt him to examine for sexual contact during an autopsy. He affirmed that he would use swabs if he saw no evidence of sexual contact with the naked eye but suspected that sexual contact had occurred. In this case, he sent swabs to the TBI, but could not recall whether he examined them. Dr. Harlan stated that he had not seen the TBI report "until this date." (Id. at PID 2060.)

When cross-examined, Dr. Harlan stated that sending anal and vaginal swabs to the TBI was standard practice. He added that he did not personally test the swabs and did not have the facility to do so. He denied that a visual inspection of Jenkins's body led him to believe that semen was present. He affirmed that he prepared the swabs out of "an abundance of caution." (Id. at PID 2066.)

In the evidentiary hearing held by the district court, Mike Turbeville, a forensic scientist supervisor at the Forensic Biology Unit at the TBI Crime Lab in Nashville, testified that Dr. Harlan's office submitted a form dated March 4, 1985, requesting toxicology testing and a review of the vaginal and anal swabs taken from Jenkins. Turbeville testified that a TBI report dated July 5, 1985, stated that the swabs—Exhibits No. 41A and 41B—"failed to reveal the presence of spermatozoa." (R. 142, PID 2080-81.)

At the hearing, Turbeville read aloud the raw notes' description of Exhibit No. 41A, the vaginal swabs : "Two swabs were molded when received. Numerous RBS, that stands for reddish brown stains, on tubes and—that second I can't make out. I don't know if it's slits or sheets. Micro, which is a microscopic exam, July 1, 1985. Scattered epithelial cells, bacteria, yeast. No SP. No sperm." (Id. at PID 2087 (quoting R. 175-4, PID 5787).) Concerning the description of Exhibit No. 41B, the anal swabs, Turbeville quoted the notes as saying: "Rectal swabs, quotes, Jenkins Catherine, rectum, end quotes. Two swabs with RBS, reddish brown stains, and fecal material. Micro, July 1, 1985. Scattered debris, some epithelial cells, no sperm." (Id. at PID 2087 (quoting R. 175-4, PID 5787).) Turbeville testified that the notes meant that testing to determine the presence of semen had not occurred, but that the final reports do not reflect this fact. Turbeville acknowledged that it may have been "appropriate" for the TBI report to indicate that no testing for semen had occurred because the sample "was somewhat compromised by mold." (Id. at PID 2089-90.)

While this background demonstrates that laboratory notes were not turned over to trial counsel, Hines is not entitled to relief on his Brady claim because he has not demonstrated that the undisclosed notes are material. That is, Hines has not shown a reasonable probability that, had the raw notes been disclosed to the defense, the result of the proceeding would have been different. See Bagley, 473 U.S. at 682. Hines argues that had defense counsel been given the notes, they "would have used [them] . . . to establish that the prosecution could not exclude the reasonable hypothesis that someone other than Darrell Hines left semen on the victim," such as the "suspicious and wild-eyed individual - not Hines" seen by "the clerk of the store across from the motel . . . near the time of the killing." (Hines Br. at 23-24.) However, there is no indication that there was in fact semen on the swabs or the victim's body, so Hines's argument would have been mere unconvincing speculation. Dr. Harlan testified that he found no evidence of semen upon a visual inspection and that he only ordered testing out of "an abundance of caution" and not because he saw evidence of semen, as Hines speculates. (R. 141, PID 2065-66.)

Nor can Hines show that Dr. Harlan testified falsely. "[A] prosecutor violates a criminal defendant's due process rights when she knowingly allows perjured testimony to be introduced without correction." Thomas v. Westbrooks, 849 F.3d 659, 666 (6th Cir. 2017), cert. denied, 138 S. Ct. 390 (2017) (citing Agurs, 427 U.S. at 103). To prevail, Hines must show that: "(1) the statement was actually false; (2) the statement was material; and (3) the prosecution knew it was false." Rosencrantz v. Lafler, 568 F.3d 577, 583-84 (6th Cir. 2009) (citing Coe v. Bell, 161 F.3d 320, 343 (6th Cir. 1998)).

Hines's argument focuses on Dr. Harlan's statements that based on a "visual inspection . . . there was no material that was indicative of semen," that there was "no such material to evaluate," and "there were no semen present." (Hines Br. at 83 (quoting R. 173-9, PID 4737, 4740).) Because Hines cannot show that the vaginal swabs contained any evidence of semen, he cannot show that Dr. Harlan's testimony was actually false. Hines argues that Dr. Harlan's office's request for the TBI to test "seminal type" indicates that it "concluded there were materials 'indicative of semen,'" contrary to Dr. Harlan's testimony. (Hines Br. at 25 (emphasis removed).) However, Dr. Harlan has explained that he requested this test out of "an abundance of caution," (R. 141, PID 2065-66), and this testing request does not establish that Dr. Harlan's testimony was actually false, Rosencrantz, 568 F.3d at 583-84. In fact, post-conviction forensic testing of Jenkins's underwear indicated that while there were bloodstains from which DNA testing could be conducted, the test for semen produced "negative results." (R. 124-1, PID 1347.) Thus, we reject Hines's claim that the prosecution withheld exculpatory evidence or knowingly elicited false testimony. IV. Whether trial counsel were ineffective during the penalty phase of trial.

Hines contends that trial counsel were ineffective during the resentencing phase because they did not: (1) present available mitigating evidence; (2) challenge Dr. Harlan's testimony about the amount of time Jenkins survived after being wounded; or (3) object to the death penalty as arbitrary and unconstitutional under the circumstances of this case. We address, and reject, each of these sub-claims in turn.

A. Failure to present mitigation evidence

Hines contends that trial counsel were ineffective because they did not present available mitigation evidence at the penalty phase. Specifically, Hines asserts that trial counsel should have presented evidence that Hines endured physical and sexual abuse by his stepfather as well as sexual abuse by an uncle, suffered head injuries as a child, endured physical and psychological abuse at Green River Boys Ranch, sniffed gasoline and glue and consumed alcohol and drugs as an adolescent, suffered from paranoia and chronic post-traumatic stress disorder, which affected his brain function, and has a deficit of serotonin in his brain. The warden responds that the state court's decision was not an unreasonable application of Supreme Court precedent.

As an initial matter, Hines asserts that the state post-conviction appellate court applied the incorrect standard in reviewing his mitigating-evidence claim, and therefore we should review this claim de novo. Strickland requires that a petitioner show that there was a "reasonable probability" that, but for the counsel's failure to introduce the mitigating evidence, "the result of the proceeding would have been different." Strickland, 466 U.S. at 694. Hines argues the state court apparently applied a stricter standard, as evidenced by the use of the phrase "would not have affected" in its opinion:

In Goad v. State, 938 S.W.2d 363 (Tenn. 1996), our supreme court set out the relevant factors to consider when determining if prejudice had resulted from a trial attorney's failure to present mitigating evidence during the penalty phase of a capital trial. . . . "[C]ourts have considered whether there was such strong evidence of aggravating factors that the mitigating evidence would not have affected the jury's determination."
In the present appeal, the post-conviction court found that counsel were not deficient in their representation of the petitioner, saying that "[i]n view of the overwhelming strength of the aggravating factors in Petitioner's case . . . , the mitigating factors would not have affected the jury's determination[.]" Accordingly, under the principles enunciated in Goad, the post-conviction court found that the petitioner was not prejudiced. . . . We conclude that the record supports this determination.

It is unclear what standard the post-conviction court actually applied. The opinion identifies the proper Strickland standard in its "Standard of Review" section but does not use the "reasonable probability" language anywhere else in the opinion—other than in footnote 2, where the court discusses a case, Wiggins v. Smith, 539 U.S. 510, 516 (2003), which the Tennessee court found inapplicable. See Hines, 2004 WL 1567120, at *22-23, 31-32, 32 n.2. Although habeas review includes a "presumption that state courts know and follow the law," Woodford v. Visciotti, 537 U.S. 19, 24 (2002), we have previously found that a state post-conviction court's failure to apply the Strickland test warranted de novo review in circumstances similar to those here. See Vazquez v. Bradshaw, 345 F. App'x 104, 112 (6th Cir. 2009) (finding the de novo standard appropriate where the state post-conviction court reviewed whether the trial's outcome "would have been different" due to new evidence, rather than whether new evidence presented a "reasonable probability" of a different outcome). Here, although the court identified the correct legal standard early in the opinion, it used different language in explaining its decision. When a state court applies a decisional rule contrary to clearly established federal law, "'a federal court [is] unconstrained by § 2254(d)(1),' and de novo review is appropriate." Fulcher v. Motley, 444 F.3d 791, 799 (6th Cir. 2006) (quoting Williams v. Taylor, 529 U.S. 362, 406 (2000)). Thus, we will proceed with a de novo analysis. Hines's claim fails even under this standard.

Trial counsel presented extensive mitigation evidence at the 1989 penalty phase. First, counsel presented testimony from Therman Page, a counselor at the Tennessee State Prison. Page worked with Hines "on several occasions relating to visits and various other problems that he's had." (R. 173-9, PID 4745.) Page testified that Hines had been disciplined by "receiv[ing] two or three different writeups[,]" but that none related to violence against another individual. (R. 173-10, PID 4759-60.) Page described Hines as follows: "[He is] somewhat [of] a loner. He does not have a lot of close friends as far as the other people that he's incarcerated with. The times that I have talked to him and been with him, he has talked to me freely, but he does not have the friends that a lot of other people have in prison." (Id. at PID 4760.) According to Page, Hines "does not have very many visits at all. I don't recall, right off, any family members at all coming to see him, the time that I've known him," which was "close to three years." (Id. at PID 4760-61.)

Trial counsel next introduced the transcript of John Croft's testimony from Hines's first trial. Croft and his wife, Nancy, lived in Cave City, Kentucky and were Hines's grandparents. Croft testified that Hines's mother, Barbara, first married a man named Dugard, with whom she had three children, including Hines. Dugard abandoned the family when Hines was "maybe ten or eleven." (Id. at PID 4767-68.) Barbara then married Bill Hines. Croft and his wife kept the children because Barbara worked; he described Hines as "a well-mannered boy, a well minding boy. And the keeping of the children was kind of left to the older girl." (Id. at PID 4768.) When Hines was "about thirteen years old," Barbara and her family moved, and "it was no longer convenient for [Croft and his wife] to have the children." (Id. at PID 4769.) According to Croft, Hines was "a little high tempered" but not troublesome. (Id.)

Croft testified that Hines stayed with him briefly when he was a teenager, that Hines called him "Big John," "never g[a]ve [him] a minute's trouble," and "was always well behaved, and he minded good." (Id. at PID 4770.) When asked about discipline in the Hines's home, Croft replied that he "never figured it was so much a discipline problem as it was the separation. You know, there was a loneliness." (Id. at PID 4770-71.) Croft was asked whether Hines minded him, and he responded: "Always. I had the best respect from him; you know, sometimes even more respect from him than I did one of my boys. But mine were in and out, gone a lot, school and all. But Darrell always did mind, and he minded his grandmother, as well." (Id. at PID 4771.)

During this stay, Croft noticed a change in Hines's behavior, searched his room, and saw evidence that Hines had been sniffing glue. Croft viewed Hines as "a changed personality from that day on" and harbored "doubts" about having him live there permanently. (Id. at PID 4772-73.) Croft knew that Hines had been incarcerated and suspected that it changed Hines's behavior: "He had a chip on his shoulder—a chip on his shoulder all the time. He just wasn't the same. He wasn't himself. You'd have to see the change in him to know it." (Id. at PID 4774.)

Croft testified that Hines had a three-year-old son. Hines and his son's mother suffered "an emotional conflict" because she had had a relationship with another man. (Id. at PID 4775.) Croft also offered insight into the crime that led to an earlier incarceration for Hines:

When he was in that first assault, or whatever it was that he was sent up for, I begged the county judge at that time—I explained to the county judge that the boy needed help, that he didn't need confinement at that time, that he needed help. And I was laughed at. He needs help now. He's needed it all these many years, help that he didn't get. And since I'm sworn to tell the truth, this is the truth. I begged Basil Griffith, the county judge, to get him help, you know, on that.

On cross-examination, Croft testified that Barbara, Hines's mother, had a drinking problem until 1979, when she stopped drinking. Croft confirmed that he did not visit Hines during his incarceration for assault and only saw him one or two times before Jenkins's murder. On re-direct examination, Croft testified that he believed that Hines could be helped and noted that Hines offered no resistance to his arrest.

Trial counsel next presented the testimony of Pamela Mary Auble, Ph.D., an expert in clinical psychology. Dr. Auble conducted a psychological examination of Hines, relying on three primary sources of information: test data; a clinical interview; and various records, such as school records and "interviews by various investigators." (Id. at PID 4796, 4800.) Dr. Auble administered several psychological tests, measuring Hines's I.Q., Hines's ability to learn and recall new information, Hines's adaptability, and other metrics of Hines's mental, creative, and motor abilities.

Dr. Auble summarized her analysis of Hines's neurological history. She testified that as an eight-year-old, Hines fell from a hay wagon, resulting in a loss of consciousness and a dent to his skull. She testified that Hines had "a history of a lot of alcohol and drug abuse, including glue sniffing, amphetamines, cocaine, heroin[], barbiturates, and a lot of those kinds of things." (Id. at PID 4803.) Dr. Auble testified that tests were inconclusive as to the existence of brain damage, however.

Dr. Auble also examined Hines's family history. She testified that Hines was "ignored a lot when he was growing up" and "didn't have a lot of interaction with his parents" because both "worked full time, and [Hines] reported that when they were not working, they tended to drink a lot." (Id.) Hines's mother "took Valium for a lot of years" and "was in a psychiatric hospital once when [Hines] was nineteen, for nerve problems." (Id. at PID 4803-04.) Dr. Auble saw "some evidence of physical abuse"; Hines "reported that he was beat with a tobacco stick by his stepfather" and "reported whippings when his parents just didn't know when to stop." (Id. at PID 4804.) She recalled that Hines "hid in the woods for several days because he was afraid his stepfather would kill him" after he broke a tractor. (Id.)

Dr. Auble testified that "[t]here was also indications of some kind of sexual issues within [Hines's] family" as Hines reported "growing up early, sexually, somewhere around six years old" and has one sister who is bisexual and another sibling who is transgender. (Id. at PID 4804, 4806.) Dr. Auble testified that "the fact that he has [siblings] with such unusual sexual orientations suggests that in the family, there's something a little weird that they both turned out that way." (Id. at PID 4807.) Dr. Auble testified that Hines "has some issues about masculinity, that sexuality is a sensitive and troubling area for him, that he has had a lot of difficulty with, which is also consistent with [his] . . . history." (Id.) On cross-examination, Dr. Auble stated that she did not know whether Hines's early sexual contact at the age of six occurred with a relative; she did not ask him about the details, as Hines was "reluctant to talk about these things, which, I mean, is understandable, I think." (Id. at PID 4843.)

Dr. Auble observed that Hines's family did not help him with his problems. When Hines started abusing alcohol and drugs, the family "move[d] him into his own apartment when he was fifteen; in a sense, just to get him out of the family and put him out on his own, rather than trying to help him solve any of the problems." (Id. at PID 4807.) She added that the family "ha[d] a history of repeatedly turning him in to the authorities," resulting in his incarceration. (Id. at PID 4807-08.) Further, Hines had "trouble" in school partly because he "had to work on his [step]father's farm a lot—his [stepfather was a farmer—and he wasn't able to attend school as often as he should have, so he wasn't there a lot." (Id. at PID 4808.) She noted that her testing suggested a learning disability that would make learning more difficult.

Dr. Auble summarized the test results. She testified that Hines "is emotionally pretty immature, that he's never really grown up," adding that his "emotional level of maturity is that of about a teenager." (Id.) She testified that Hines has "very poor" self-esteem, "has a lot of trouble with criticism," "has a lot of trouble trusting other people easily," and "expects other people to harm him and is reluctant to trust people enough to confide anything, to tell people his troubles." (Id. at PID 4808-09.) She testified that Hines is "real insecure about his masculinity" and has "underlying depression and anger," which he managed by either avoiding the underlying problem or engaging in self-destructive behavior such as drug and alcohol abuse. (Id. at PID 4809-10.) She noted that "testing did not indicate that he possesses an alcoholic-type personality" and characterized the alcohol and drug use as "more an escape from the negative feelings that he has than something he's just got a natural weakness for." (Id. at PID 4810.) When stressed, Hines will become "angry and destructive,"; Dr. Auble recalled "reports of a lot of fights, whenever he's provoked in some way, that he just boils up and then stuff comes out of him." (Id.)

Dr. Auble diagnosed Hines as having paranoid-personality disorder and dysthymia. She explained that the former diagnosis requires the individual to exhibit four of the following symptoms: "expect without basis, to be exploited or conned by other people"; "question the trustworthiness of friends and associates without justification"; "tend to read threatening messages or insults into remarks or events that aren't really all that threatening or insulting"; "tend to bear grudges" and "don't forgive insults"; reluctance "to confide in people because they have this fear that information will somehow be, either, used against them or that they'll be betrayed"; "react [quickly] with anger"; and "question the faithfulness of their spouse or sexual partner without justification." (Id. at PID 4810-12.) Dr. Auble explained that Hines had many of these characteristics.

Dr. Auble testified that dysthymia is similar to depression, and Hines suffered the following symptoms of dysthymia: a depressed mood for at least two years; insomnia; low self-esteem; poor concentration, and a sense of hopelessness. She testified that the paranoid personality disorder, in particular, would affect Hines's ability to handle stress: "The worst thing in the world to happen to those people is if they finally do trust somebody and then they get betrayed. That's the worst. And, in particular, that kind of stress would be the hardest for him to handle." (Id. at PID 4815.)

Dr. Auble offered insight into Hines's only significant relationship with a woman, Melanie, whom Hines dated for about a year starting in 1981 and with whom he had a son. The relationship ended because Hines perceived Melanie's mother and brother to be "freeloading on him," as "they weren't willing to pull their own weight in the household." (Id.) Hines's mother, with whom Hines's son lived, did not allow Hines to "keep him for any length of time." (Id. at PID 4815-16.) In 1985, following his release from prison, Melanie invited Hines to visit her in North Carolina; he did and they discussed reconciliation. According to Hines, Melanie had changed and was engaged in drug use and prostitution. Hines told Dr. Auble that Melanie's mother wanted him to fight Melanie's ex-husband, which was problematic because Hines was trying to adjust to life outside of prison. Dr. Auble described the situation as "particularly hard on [Hines] because, first, he is very insecure about his masculinity." (Id. at PID 4817.) "He has a general difficulty trusting people; and once he does trust somebody, the worst thing in the world for him would be to be betrayed, and that's sort of what he felt like happened." (Id.) Further, Hines "tends to avoid negative feelings," "doesn't deal with them[,]" and "just puts them inside [while]. . . they get worse and worse." (Id.) Hines decided to leave for Kentucky on a bus, intending to get his son and move to Montana. Hines's parents in Kentucky did not give him his son, called the police, and dropped him along a Tennessee highway. He later checked into the CeBon Motel.

Dr. Auble described Hines's state of mind during his time at the motel. After checking into his room, he drank and watched television throughout the night, getting little sleep. His mental state was "very fragile," he "fe[lt] worthless" and insecure," and he had "a whole lot of anger and disappointment and sadness towards Melanie and his parents for rejecting him"; she added that Hines's "masculine image is just extremely vulnerable," meaning that he would "just be very sensitive to anything that might be seen as criticism." (Id. at PID 4818.) Dr. Auble surmised that "any provocation at all would probably result in an explosion of all these feelings." (Id. at PID 4819.) Dr. Auble found the "stab marks on the wall of the motel room where he was staying" telling because "most people don't go around with a knife and stab the walls of motels for fun or profit," but do it because they are "suffering from some kind of mental disorder or [are] under a lot of stress at the time they're doing stuff like that." (Id.)

Dr. Auble testified that she reviewed Hines's prison records and concluded that he had "done pretty well in prison" and "hasn't been any trouble to anybody, basically" as he "kind of keeps apart from other prisoners." (Id. at PID 4821.) She explained that Hines is "able to make a pretty good adjustment when he's not in a situation which brings up all these feelings" associated with Melanie and his family. (Id.) She testified that Hines suffered from a mental disease or defect on the day of the murder but did not pose a threat in a prison environment.

Trial counsel presented the testimony of Floyd Eugene Collins, Hines's childhood friend who first met Hines in their neighborhood at fourteen years old. Collins testified that he saw Hines sniff gasoline and glue, drink beer and whiskey, and consume marijuana and pills. He testified that Hines sniffed glue "all the time" and that "[e]very time you'd see him late in the evenings or something like that, he'd be—have him a glue bag or something like that, just sniffing away." (Id. at PID 4861.) Collins testified that, compared to the other neighborhood children, Hines "was always more hyper, always going somewhere, always talking," but "was crazy" and "needed help." (Id. at PID 4861-62.) Collins testified that there was once a "gasoline party" in Hines's backyard where everyone was "just sniffing gas," and Hines displayed that he "wasn't right" by repeatedly riding a bicycle into a chain-link fence. (Id. at PID 4864.)

Trial counsel also presented the testimony of Charles Preston Smith, who knew Hines around 1975-76 and testified that Hines "used to sniff model-car glue, gasoline, [and] smoke some dope[.]" (Id. at PID 4873-74.) Smith testified that Hines "seemed like he would get pretty comatose. Go back there and talk to him, and he wouldn't even know you was there." (Id. at PID 4875.) Smith recalled that Hines had his own apartment at about fifteen years old and that he never saw any of Hines's family there.

Finally, trial counsel presented the testimony of sociologist Ann Marie Charvat, Ph.D. Dr. Charvat testified that she interviewed Hines's family, individuals from his neighborhood, and individuals who worked with him as a child. She also reviewed a variety of records and obtained "as many different kinds of reports as were available." (R. 173-11, PID 4925.) Dr. Charvat used the information to "set up a life-cycle study where I was able to get substantiation on various elements of his report." (Id.) To be included in her report, Dr. Charvat "had to hear [the information] not only from [Hines], but also from another person, or read it in one of his documents, or it had to be consistent with what we already know about these different elements." (Id.)

Dr. Charvat explained how she collected data:

Originally, what I started with was a social history. That's what came from him. Then I developed four criteria, four possible substantiations. If I could get a substantiation from conversation or interview with a primary relationship or with a secondary relationship or with an historical document or with my scholarly research, the literature in my field, if I got two of those, then I included it in the life-history section.

Dr. Charvat had access to a third category of information, historical documents, including: educational records from elementary school, junior high, and high school; medical records; and prison records. She reviewed additional records, such as "social histories, psychological evaluations, physical exam[s], medical history, FBI record of charges and dispositions, [a] psychological evaluation, a mitigation evaluation prepared by Capital Case Resource Center, correspondence from [defense counsel], and the testimony of [Hines's grandfather] John Croft." (Id. at PID 4929.) Dr. Charvat testified that Green River Boys Ranch sent no records, but she talked to Mr. Courtney, who was Hines's counselor there. Dr. Charvat also reviewed the research and literature published in her field.

Dr. Charvat testified that Hines lived outside of Bowling Green for the first twelve years of his life—the first two years with his mother and her husband, Billy Frank Dugard, who had an "unstable" and "violent" marriage, and the next ten years with his mother and her second husband, Bill. (Id. at PID 4931-32.) Barbara and Bill, a farmer, lived on a rented farm with Hines's two sisters and brother. The family was "socially isolated" and did not participate in any community activities such as attending church. (Id. at PID 4932.) The parents were "very hardworking people," but "there was not very much supervision on the kids" and "an absence of rules within the family[.]" (Id. at PID 4933, 4935.) Dr. Charvat also testified that she found "evidence of violence" that she would categorize as "very serious abuse," and that "there were situations in this family that I found to be beyond and into the criminal violent category on Darrell and his older" sibling. (Id. at PID 4933.) Dr. Charvat testified that she suspected sexual "irregularities":

Another issue about the family was I did not find sexual abuse—I did not—although, I did find that there were irregularities from sexual norms. And because the family was very self contained, there was no evidence whatsoever that [Hines] was sexually molested; however, there is—It's a very difficult topic to get information about, even in anonymous situations on the telephone.

Addressing his education, Dr. Charvat described Hines as "slow" and explained that he was passed to the next grade "because he was not too much of a problem and because it was, simply, time to move him." (Id. at PID 4938.) Dr. Charvat testified that Hines's "formal education ended in the sixth grade," but added that he attended one school for the first six grades, and attended several different schools until ninth grade, which he repeated once, without evidence "of any kind of successful completion" of any of these grades. (Id. at PID 4938-39.) For a two-year period beginning when Hines was twelve years old, the family moved between Bowling Green and the farm, ultimately settling in a dangerous neighborhood in Bowling Green, which required Hines to learn social skills for the new environment.

Dr. Charvat testified that Hines started "getting into trouble" at fifteen years old, taking "a variety of drugs at this point in time, some of them glue, some of them gasoline," and "dropp[ing] out of school, basically without any objections from anybody, as far as I could determine." (Id. at PID 4941.) Around that time, Hines "moved out of the family unit" and "at various points, he would go back; but, essentially, he either had an apartment, sometimes stayed with his family, or stayed with other people in the neighborhood." (Id. at PID 4941-42.) Dr. Charvat explained that the separation constituted "a significant, serious breach of a [familial] bond at that point in time." (Id. at PID 4942.) Hines had several experiences with juvenile court, which often resulted in warnings and probation; on one occasion, he was ordered to see a psychiatrist, and did so a couple times, but declined additional visits and no one made him return. Various reports noted that Hines possessed "an inability or an unwillingness to cooperate, and [that] no successful treatment has ever been noted." (Id. at PID 4948.)

Dr. Charvat testified that Hines was sent to Green River Boys Ranch at seventeen years old. Hines was subjected to intensive group therapy there, referred to as "grouping," in which "the bad behavior of one person in that group will lose the privileges for everybody in it." (Id. at PID 4946.) Dr. Charvat testified about what she had heard regarding practices at Green River:

[T]hey would get boys on the ground and shout at them in their faces; that at various points, it would be physical, with all the guys in this group losing their privilege because of the bad behavior of this one guy participating in the grouping of another guy.

Now, [Hines] described a situation to me in which he and another guy were getting grouped. And one of the things that occurred in this grouping was that somehow this thing moved out to where the sewage w[as], and there was pushing into the sewage as a result of this grouping to try to get about this behavior.

Finally, Dr. Charvat identified several factors in Hines's past that literature has found can lead to criminal behavior:

In [Hines]'s specific case, these included: the level of physical abuse was too great, there was neglect of the children, there was social isolation of the family, there was evidence that there was uninformed parenting, irregular sexual norms, excessive adult responsibility, poor performance, achievement testing was significantly below grade level, excessive truancy, early onset of delinquency, early onset of drug use, self-abusive tendencies, lack of adult supervision, ineffective involvement with the juvenile justice system—it didn't work—terminated education, he had violent police models, his incarceration, his treatment at Green River. Basically, if I were working on predicting delinquency, each and every one of these would be found to be important contributors.

Hines argues that the evidence presented on state post-conviction review demonstrates that trial counsel's presentation of mitigation evidence at the resentencing hearing constituted ineffective assistance. He also argues that we should consider mitigation evidence presented for the first time in his habeas proceeding in district court, citing Martinez. Because this claim was adjudicated on the merits in state court, we cannot consider any evidence offered for the first time in federal district court. See Cullen v. Pinholster, 563 U.S. 170, 181 (2011) ("We now hold that review under § 2254(d)(1) is limited to the record that was before the state court that adjudicated the claim on the merits."). And Hines's reliance on Martinez is unavailing because the IATC claim was adjudicated on the merits in state court, rather than found to be defaulted. Martinez, 566 U.S. at 9 ("The precise question here is whether ineffective assistance in an initial-review collateral proceeding on a claim of ineffective assistance at trial may provide cause for a procedural default in a federal habeas proceeding.").

Hines has not demonstrated trial-counsel ineffectiveness related to the presentation of evidence concerning his use of alcohol and drugs because Croft, Collins, and Smith each offered testimony on that subject. Additionally, Dr. Auble testified about Hines's head injuries and that, after conducting neurological testing for brain damage, she found the tests inconclusive. Dr. Charvat also talked about Hines's substance abuse, as well as his difficult experience in the Green River Boys Home. The additional evidence presented during the state post-conviction proceedings concerning these topics was cumulative, and therefore Hines cannot show any prejudice resulted from trial counsel's failure to present this evidence at his resentencing. See Hill v. Mitchell, 842 F.3d 910, 943-44 (6th Cir. 2016), cert. denied, 138 S. Ct. 82 (2017) ("A petitioner does not establish prejudice if he shows only that his counsel failed to present 'cumulative' mitigation evidence"); Fautenberry v. Mitchell, 515 F.3d 614, 626-27 (6th Cir. 2008).

As to trial counsel's alleged ineffectiveness concerning the presentation of evidence about sexual and physical abuse, William D. Kenner, M.D., a psychiatrist, testified during the initial post-conviction proceedings that he learned from Hines's older sibling, Lee, that Hines was sexually abused by his stepfather and uncle. Dr. Kenner acknowledged that he did not have a first-person account of the events, as all of his information came from Lee, while Hines did not discuss the events—nor did Dr. Kenner know if Hines even remembered them. Dr. Kenner stated that "it's not uncommon for even adults to, you know, have a blank spot in their memory when something traumatic like that happens." (R. 174-2, PID 5291.)

Hines's difficulty with sexuality, as well as the possibility that he was sexually abused, was discussed at the resentencing hearing. But, unlike Dr. Kenner, Dr. Charvat was unable to contact Lee for unknown reasons and instead only spoke with Hines's other siblings. Because Lee was the only source of that information and did not testify, there is a question whether the court would have found the evidence admissible, as the trial court only permitted Dr. Charvat to base her opinion on "reliable hearsay," to the extent she relied on hearsay. (R. 173-10, PID 4920.) At any rate, although additional evidence of sexual and physical abuse might have made the mitigation case stronger, the new evidence does not differ from the information presented at the 1989 penalty phase in such a substantial manner to entitle Hines to relief. See Tibbetts v. Bradshaw, 633 F.3d 436, 444-45 (6th Cir. 2011) (explaining that new mitigation evidence that covers the same subject as evidence presented during the penalty phase of trial, but in greater detail, is insufficient to demonstrate prejudice).

Hines also takes exception to Dr. Auble's testimony, as she did not tell the sentencing jury that Hines suffered from post-traumatic stress disorder, nor that he had brain damage. But Hines was not prejudiced simply because trial counsel failed to retain "some other hypothetical expert." Smith v. Mitchell, 348 F.3d 177, 208-09 (6th Cir. 2003). Furthermore, "[a]bsent a showing that trial counsel reasonably believed that [Dr. Auble] was somehow incompetent or that additional testing should have occurred, simply introducing the contrary opinion of another mental health expert during habeas review is not sufficient to demonstrate the ineffectiveness of trial counsel." McGuire v. Warden, 738 F.3d 741, 758 (6th Cir. 2013) (citing Black v. Bell, 664 F.3d 81, 104-05 (6th Cir. 2011)). Hines makes no such showing.

Finally, Hines argues that trial counsel were ineffective for failing to tell the jury that Hines lacks a sufficient quantity of serotonin, "which makes it difficult for him to modulate and control his behavior." (Hines Br. at 34.) The Tennessee Court of Criminal Appeals denied this claim, explaining:

[Paul] Rossby[, Ph.D., a molecular biologist,] acknowledged that he did not work on developing this issue in a criminal case until approximately 1992, three years after the petitioner's resentencing trial. Further, he said that he did not actually testify on the issue of serotonin until 1999, ten years after the petitioner's resentencing trial, and knew of no one who had testified on the issue prior to that. As the post-conviction court stated: "Petitioner's counsel at re-sentencing could not reasonably have been expected to search for experts on a subject which they did not know existed." The record supports this conclusion.

At the initial post-conviction hearing, Dr. Rossby testified that molecular neurobiology involves "the study of the brain and the central nervous system at the level of molecules and systems." (R. 176-5, PID 6478.) He explained serotonin's effect on the brain:

Serotonin is a naturally occurring neuromodulator in the brain. It comes under the broad heading of neurotransmitters but it is a neuromodulator. Serotonin is essentially produced in one very small region of the brain and then projected to every part of the brain. Projected meaning that it is, it is synthesized in one place and then it [is] sent to all parts of the brain. Serotonin essentially has an inhibitory effect on the neuronal firing that I was describing before. Serotonin blocks pain for example. Serotonin is released in tons [sic] of great stress and it opposes the stressful reaction or the fight[ or] flight reaction. Serotonin appears and there has been a tremendous amount of research on the function of serotonin. Serotonin appears to orchestrate various systems of inhibition within the brain. And there is a tremendous amount of data that indicate that serotonin orchestrates these systems of inhibition within the brain.

Dr. Rossby assessed Hines's serotonin level and concluded that it "is at the extreme[ly] low level in our society" and the effects of that low serotonin are "exacerbated by his Type II alcoholism[.]" (Id. at PID 6490.) Dr. Rossby offered the following insight:

Essentially all of the studies that . . . have accumulated over the past twenty to twenty-five years, the low serotonin has been the central feature that distinguishes the impulsively violent offender from the non-impulsively violent offender. But there has also been a very strong correlation with Type II alcoholism and so, based on everything that I know and everything that I have read about the case and based on this analysis of his serotonin levels in his cerebral spinal fluid I would say that he is virtually, for biological reasons, he is virtually incapable of opposing his, his behavior, his spontaneous behavior. He is organically impaired.

Dr. Rossby testified that Hines's alcoholism is significant because, in a research study, "Type II alcoholism was detected in almost all of the violent offenders who had also low serotonin levels." (Id. at PID 6493.) Dr. Rossby described an individual who is a Type II alcoholic as having "low harm avoidance, gets bored easily and needs a lot of stimulation and is always out seeking alcohol." (Id. at PID 6494-95.) Dr. Rossby explained that serotonin, in conjunction with Type II alcoholism, affects impulse control:

We are talking about an organic capacity to limit, to regulate or to control impulse and it doesn't determine what the impulse may be it just, we are talking about a failure of inhibitory systems and the systems are really designed to inhibit any kind of impulsive behavior, instinctual compulsive behavior.

On cross-examination, Dr. Rossby testified that he first contributed serotonin research to a criminal case in 1992 and first testified as an expert on the subject in a criminal case in 1999. Dr. Rossby was not personally aware of anyone testifying about this topic before 1992.

Hines has not shown that trial counsel performed deficiently by failing to present evidence that he has low serotonin. In anticipation of the 1989 sentencing hearing, trial counsel retained a mental-health expert, Dr. Auble, who conducted a psychological examination that included testing for brain damage. She testified about the difficulty Hines had controlling his behavior when provoked. As noted previously, trial counsel is not ineffective for failing to retain a specific mental-health expert to obtain a specific outcome. McGuire, 738 F.3d at 758. Further, Dr. Rossby stated he was not aware of the use of the low-serotonin argument in a criminal case at the time of the resentencing hearing. Hines's counsel at resentencing were therefore not ineffective for failing to put forward mitigation theories that it was reasonable for them to be unaware of.

B. Failure to cross-examine Dr. Harlan's testimony

Hines next contends that trial counsel were ineffective because they did not effectively challenge Dr. Harlan's testimony about how long Jenkins would have remained conscious and would have survived following the infliction of her injuries. The warden responds that the state court's resolution of this claim was not unreasonable.

At resentencing, the prosecution relied on the aggravating factor that the offense was "heinous, atrocious, or cruel in that it involved torture or depravity of mind," Tenn. Code Ann. §39-2-203(i)(5)(1982), relying on Dr. Harlan's testimony that the victim was conscious for three to four minutes after being stabbed, during which she tried to fight off her attacker.

Dr. Harlan testified about the time elapsed between infliction of the wounds and the resulting death:

The fact that there's hemorrhage from all three of these wounds indicate[s] that they occurred at approximately the same time. The amount of hemorrhage or bleeding from these wounds wou[l]d indicate that death occurred within a short period of time after the time of the infliction of these wounds, probably within a few minutes; most likely, within a span of probably four to five minutes, maybe six minutes.

Q: And I believe you stated that after receiving those wounds, the victim would have died within about four to five minutes?
A: Yes, sir, that is correct.
Q: And you indicated that consciousness would have lasted somewhere between, I think you said, three and four minutes?
A: Yes, sir, that is correct.
Q: All right. Now that's something that you can't absolutely find; that's your opinion based upon the appearance of the wounds, right?
A: That is my opinion based upon the type of wounds that were present, upon the lack of damage to the central nervous system, and on the fact that the wounds all occurred at approximately the same time.

On state post-conviction review, Dr. Sperry opined that the period between the infliction of Jenkins's injuries and her death was shorter than Dr. Harlan had stated:

In my opinion, all of the injuries except for the vaginal stab wound were sustained very very rapidly. That, from the time of the attack had [en]sued to when she . . . collapsed and was receding into unconsciousness because of internal bleeding was [happening] very rapidly. Again, less than a minute and probably less than thirty second[s] realistically.

Not with two stab wounds involving the heart like this. That is not possible. People collapse very rapidly. And, in fact, . . . there is no evidence that there is any blood elsewhere other than again right beneath this air conditioner thing, on the inside of the door and then over where her body ultimately was found which is an indicator from the scene alone that she collapsed very rapidly and lost consciousness very rapidly. But irrespective of that, it is just not physically possibl[e] for someone to sustain wounds like this and stay conscious for four minutes. They will be dead by that time and have lost consciousness long before that.

Dr. Harlan testified during the post-conviction proceedings, expressly disagreeing with Dr. Sperry's assessment and reiterating his prior opinion. Dr. Harlan specifically addressed the conflict between the two opinions:

The—the problem that you have with the statements that he made is that he made—by he, I'm talking about Dr. Sperry—is that his statements are to the effect that the heart just stops beating immediately. Well, if the heart stops beating immediately, then you don't get the blood out there. If the blood, as we know is out there, because we can see it, then the heart has to have continued to beat. And the heart beating will allow blood to continue to flow to the brain in gradually reducing quantities.

On state post-conviction review, the trial court found that trial counsel were deficient for failing to challenge Dr. Harlan's testimony. The court concluded, however, that there was no prejudice because the jury would have found that the offense was "depraved," even if forensic proof established that the offense was not torturous, as the prosecution maintained. Hines, 2004 WL 112876, at *33 (discussing district court decision). The state trial court relied on State v. Williams, 690 S.W.2d 517, 529-30 (Tenn. 1985), which held that "depravity of mind" "may be inferred from acts committed at or shortly after the time of death." Hines, 2004 WL 112876, at *33 (citing Williams). The Tennessee Court of Criminal Appeals referred to portions of the trial court's opinion and affirmed the decision. Id. at *32-33. On habeas review, the district court determined that the Tennessee Court of Criminal Appeals' decision was not an unreasonable application of Supreme Court precedent.

Hines is required to show prejudice—a "reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different." Strickland, 466 U.S. at 694. Both Dr. Sperry and Dr. Harlan agreed that the wound to the victim's vaginal cavity occurred at, or shortly after, the time of death. Regardless whether the victim was unconscious, dying, or had just died, the infliction of such a wound may evince a "depravity of mind." See State v. Zagorski, 701 S.W.2d 808, 814 (Tenn. 1985) ("infliction of gratuitous violence" on a victim who was "already helpless from fatal wounds" indicates "a depraved state of mind at the time of the killing" and is sufficient to support a finding by the jury that the murder was especially heinous, atrocious and cruel); Van Tran v. Colson, 764 F.3d 594, 622 (6th Cir. 2014) (under Tennessee law, "'depravity of mind' can be found even where there is no gratuitous infliction of severe pain, physical or mental, that amounts to torture" (citation omitted)). In Williams, the Tennessee Supreme Court explained:

If acts occurring after the death of the victim are relied upon to show depravity of mind of the murderer, such acts must be shown to have occurred so close to the time of the victim's death, and must have been of such a nature, that the inference can be fairly drawn that the depraved state of mind of the murderer existed at the time the fatal blows were inflicted upon the victim. This is true because it is "the

murderer's state of mind at the time of the killing" which must be shown to have been depraved.

Finally, Hines argues that the state appellate court's holding that Dr. Harlan's testimony could have shown depravity to support the heinous, atrocious, and cruel aggravating circumstance was contrary to Supreme Court precedent because "a finding of 'depravity' [is] unconstitutional[ly vague]." (Hines Br. at 94-95.) We have previously held otherwise. See, e.g., Van Tran, 764 F.3d at 622-23 (holding that Tennessee's "depravity of the mind" aggravator "avoids a constitutional vagueness problem"); Strouth v. Colson, 680 F.3d 596, 606 (6th Cir. 2012) (same). Hines's argument thus fails.

C. Failure to object to death sentence in light of prosecutor's agreement to life sentence

Hines contends that trial counsel were ineffective for failing to object to his death sentence because the prosecutor had agreed to a sentence of life imprisonment. The warden argues that this claim is procedurally defaulted because it was not presented to the state courts and that Hines cannot show cause and prejudice to excuse the default.

On direct appeal after his resentencing, Hines argued that the trial court erred in rejecting the plea agreement between the parties, in which Hines would have pleaded guilty to a new offense and received a life sentence. Relying on Tennessee Rule of Criminal Procedure 11, the Tennessee Supreme Court denied relief:

In this case, the trial judge felt that the facts of the case, even when mitigating circumstances were considered, should be decided by a jury. He expressed the view that the interest of justice did not allow a plea bargain and he rejected it. We find that the trial judge acted within his authority under Rule 11 in rejecting the plea bargain.

Hines is not entitled to relief. "Counsel is not ineffective merely for failing to obtain a desired ruling from the court." Hodge v. Haeberlin, 579 F.3d 627, 645 (6th Cir. 2009). Here, the trial court rejected the plea agreement, which the Tennessee Supreme Court determined to be proper. Hines has not shown that an objection by trial counsel had any probability of producing a different result. See Mapes v. Coyle, 171 F.3d 408, 427 (6th Cir. 1999) ("Counsel could not be constitutionally ineffective for failing to raise . . . meritless arguments."). Regarding his Eighth Amendment claim, Hines relies on Adamson v. Ricketts, 865 F.2d 1011 (9th Cir. 1988), but that case is distinguishable. In Adamson, the trial court accepted a plea after considering "the presentence report, the matters in the file, the preliminary hearing transcript, the plea agreement, and the proceedings at the previous hearing." 865 F.2d at 1021. Thus, as the Ninth Circuit reasoned, the trial court's acceptance of the plea agreement constituted a judicial determination that the plea was the appropriate punishment and reflected the trial court's belief that the defendant would be appropriately punished by a prison sentence rather than death. Id. at 1021-22. But when the trial court later imposed a death sentence on the same information and "for the same conduct for which he had previously found a prison term 'appropriate,'" the Ninth Circuit found the imposition of the death penalty arbitrary. Id. at 1022-23. Here, the trial court did not make a judicial determination as to the appropriate punishment; rather, as the Tennessee Supreme Court explained, "[t]he trial judge rejected the plea bargain agreement because he felt that the case should be decided by a jury." Hines, 919 S.W.2d at 577-78. Trial counsel was thus not ineffective for failing to object to the trial court's rejection of the plea agreement. Hines has not cited any directly applicable cases in support of his Fourteenth Amendment claim, and we therefore find it is also without merit. V. Whether trial counsel were ineffective for failing to investigate and present evidence regarding Ken Jones.

Having rejected each of Hines's claims thus far, we now turn to the claim on which we reverse the district court's denial of Hines's petition for a writ of habeas corpus. During the guilt phase, trial counsel failed to investigate or effectively examine Ken Jones, who was at the motel at the time of Jenkins's death and had an apparent motive for the murder. During the penalty phase, defense counsel also failed to present evidence of residual doubt in relation to Ken Jones. Because the state court's decision that this did not constitute ineffective assistance of counsel was an unreasonable application of Strickland, we reverse the district court's denial of Hines's petition for a writ of habeas corpus.

A. Failure to investigate and present evidence regarding Jones at the guilt phase

Ken Jones was the witness at Hines's 1986 trial who testified to having first discovered Jenkins's body. Hines argues that trial counsel were ineffective because they did not investigate and present evidence explaining Ken Jones's presence at the CeBon Motel at the time of the murder. The Tennessee Court of Criminal Appeals affirmed the state trial court's rejection of this argument on post-conviction review, finding that Hines could not satisfy Strickland's prejudice requirement:

Missing in the petitioner's theory, which the post-conviction court described as "farfetched," is any motive or reason why Jones would want to kill the victim, except the petitioner's suggestion, recounted in the post-conviction findings, that the victim was killed because she had "thwarted" the sexual liaison between Jones and [Vernedith] White. In effect, the petitioner argues that fifty-one-year-old Ken Jones, accompanied by his twenty-one-year-old girlfriend, Vernedith White, following their normal Sunday morning routine and checking into the same motel where they had been together approximately 100 times before and were known by the staff, including the victim, stabbed the victim to death, with Jones driving White to another location, cleaning blood from himself and his vehicle, and then returning to the scene to report the crime and wait for law enforcement officers to arrive. We agree with the post-conviction court that, given the strength of proof against the petitioner, making the argument that Ken Jones was the actual killer would have been "farfetched" and could have resulted in a loss of credibility for the defense.

The district court rejected this claim on habeas review:

Petitioner has not presented any evidence to suggest that Jones could have been the murderer. Jones's motivation for being at the motel was undisputed. Given the State's proof and Petitioner's statement to the officers, the Court concludes that there is not any basis to suggest any other identifiable person as the perpetrator of this horrendous crime. The Court also concludes that Petitioner has not demonstrated any prejudice for this claim. Given the state courts' finding of the absence of prejudice required by Strickland, the Court concludes that this claim was reasonably decided by the state courts applying clearly established federal law.

The following evidence was presented regarding Ken Jones at Hines's 1986 guilt trial. Mary Sizemore of the Cheatham County Ambulance Service testified that she received a call at 2:36 p.m. from a woman named Maxine to go to the CeBon Motel because a man reported that a woman had been stabbed. According to Sizemore, Maxine worked at the CeBon Restaurant, which was across the street from the motel. Sizemore testified that Maxine indicated the man who discovered the body "was a man coming to rent a room there and was looking for the maid to make arrangements," but Sizemore did not know who the man was. (R. 173-2, PID 3899.) Sizemore arrived at the motel four minutes after the call, and Maxine then walked to the motel and met Sizemore. Sizemore found Jenkins on the floor of Room 21 "laying on her back wrapped in a bedspread." (Id. at PID 3890.) Sizemore "unwrapped the top of the bedspread to look at her chest to see where the blood was coming from and noticed the stab wounds." (Id. at PID 3891.) Sizemore testified that she saw Jenkins's underwear—torn into two pieces—in the room.

Ken Jones testified at trial that he stopped at the CeBon Motel on March 3, 1985, and was "acquainted" with the "older couple" who had run the motel. (Id. at PID 3941-42.) Jones "first pulled [in] around about 12:30 or possibly a few minutes past" and "stopped [for] a few minutes." (Id. at PID 3942.) Jones parked his car and walked to the motel's office. "[T]here was no one in the office," but Jones noticed a "key laying in [a] little box." (Id. at 3949.) Jones testified that he then went back to his car, at which time he saw a woman in a maroon car, accompanied by a child, drive up to Room 21 around 12:40. The woman got out of her car, knocked on the door of Room 21, and left when no one answered. Jones added that before leaving, the woman backed around and asked him if he "knew where the people were that run it." (Id. at PID 3957.) Jones said that he then left, going "up to [a roadside convenience store] for a few minutes" before coming back to the motel. (Id. at PID 3942.)

He came back "just past 1:00." (Id. at PID 3943.) He explained that when he returned to the motel, he "just sat there a few minutes" in his car and then realized that he had to use the bathroom and remembered the key that he had seen in the office. (Id. at PID 3953.) Jones explained that he "went to the office and there was no one there . . . so when no one showed up in a few minutes I took the key and left a note that I had the key that I was going to the restroom in that particular room." (Id. at PID 3943.) Jones testified that he went to the room at "probably about twenty after 1:00, in that neighborhood." (Id.)

Jones opened the door to the locked motel room, Room 21, using the key he had taken. He testified that he "saw a vacuum cleaner on the left, and I proceeded to go toward the bathroom, and I saw hair, a head of hair sticking out from behind the bed." (Id. at PID 3944.) When asked whether he recognized the individual as female or male, Jones responded: "No sir. Not then." (Id. at PID 3945.) When asked how close he was to the body, Jones testified that he was "heading for the restroom and was three feet, two-and-a-half to three feet I guess[.]" (Id.) When asked to describe his reaction, Jones answered: "Lord, I don't know. All I knowed [sic] to do is get out and call somebody." (Id.) Jones said, "I wasn't in that room but a second." (Id. at PID 3947.) Jones testified that he left the motel room, got into his car, and went straight to the restaurant across the street to call the sheriff. He observed a dark blue car parked near the motel's Room 8. He testified that a woman at the restaurant placed the call for him to the sheriff. When Jones was asked if he "stay[ed] there and talk[ed] to the sheriff," Jones replied, "Yes sir." (Id. at PID 3948.) Jones testified that he knew Sheriff Weakley, and on that afternoon he told the sheriff what he had seen.

At closing argument of the guilt phase, Hines's trial counsel emphasized the significance of Jones's testimony, notwithstanding his failure to meaningfully cross-examine Jones:

Now, this gets me. This confuses me. This causes me considerable reasonable doubt right here. We've got this Mr. Jones, Kenneth Jones. We already had one girl that said Mrs. Jenkins['s] car pulled out at 12:40. I don't know what time Mr. Jones was fooling around at that motel that Sunday afternoon or that Sunday morning. Or what he was really up to. But you can kind of gather from his testimony, kind of reading between the lines, he wasn't a traveling salesman just coming through; he had a usual spot where he always went to; he was meeting somebody. He said he got there around—what did he say—12:00 o'clock? Something like that? He saw a maroon car pull right up to Room 21 and a woman get out and bang on the door, a baby crying. A blue car parked right in front of #9

at that time. Was there anything about a silver car being there? I wonder if whoever he was meeting had a husband? I wonder if whoever he was meeting might have thought Mrs. Jenkins was this man's girlfriend in Room 21. Maybe somebody hired somebody to go down there and do something. I don't know. It causes me some concern though I'll tell you that; it causes me a lot of concern.

And I'll tell you something else that causes me some concern. Here we are, there's a murder that's been commit[t]ed, and you got this man, you saw how nervous Mr. Jones was, boy he was quivering, he was wanting to get in here and out. You saw that. Why didn't he tell Sheriff Weakley—and I feel sorry for Sheriff Weakley on this—why didn't he tell Sheriff Weakley, look, Sheriff Weakley, I saw a blue car right beside #9 and I saw a maroon car and I saw a woman get out and knock on Room 21. And wasn't it a casual relationship just going up and taking Room 21 key out and going up there and him just barging into Room 21. There was a lot of something going on up there that day.

On post-conviction review, Hines presented the testimony of one of his trial counsel, William G. Wilkinson, who stated that he did not attempt to interview Jones before trial. Wilkinson testified about a conversation he had with Sheriff Weakley concerning Jones:

Sheriff Weakl[e]y told me that [Jones] was the person who opened the door and discovered the body. He told me that he was married and that he was meeting there for the purpose of having an affair and had been there just a very short period of time and that he didn't want his wife to find out that he was carrying on the affair and that all he knew was that he was assigned that room, opened the door and saw the body and that is all that he knew about it.

Hines's other trial counsel, Steve Stack, also admitted that Jones was not interviewed, and explained:

We never interviewed Mr. Jones. We were told early on in the case by, I was told by Sheriff Weakley that Mr. Jones had been over to the motel that day to have a meeting with a lady friend and that he didn't, all that he did was go in and discover the room and that he was there for a brief period of time and he had no further information and he didn't want him to be embarrassed by having it brought out that he had been over there to meet with a lady friend.

I remember being told by [Assistant Attorney] General Kirby who just gave a brief synopsis saying that Ken Jones will be testifying because he was the one, the person that found the body. Sheriff Weakley, like I said, talked to me more in detail explaining there that he was just there a few minutes. [H]e went up, opened the door, made the discovery and then left. I don't recall that he even described him going into the room or anything. Quite frankly, Sheriff Weakley's main concern was just that we didn't make an issue of him being there to protect him from his wife.

And the Sheriff had asked me not to bring out what [Jones] was [at the motel] for. The Sheriff made it clear to me that Ken Jones had nothing to do with this case. If Do[r]ris Weakley had told me right now that it was going to rain so hard this afternoon that I will need a boat to get home I would be buying a boat right now. I mean, I would take that man's word for anything in the world. He say's [sic] this hadn't got a dog in the hunt, don't embarrass the man. I wasn't going to embarrass the man.

Stack admitted that the failure to interview Jones presented difficulties with the defense offered at trial, because defense counsel were unable to resolve factual discrepancies between Jones's testimony and that of other witnesses. For example, Jones testified that he called for the ambulance from the CeBon Restaurant around 1:20, but Sizemore testified that the ambulance service received the call at 2:36. Further, Sizemore testified that the caller identified the victim as a woman who suffered from stab wounds, but Jones testified that he was unable to determine whether the victim was male or female when he found Jenkins's body in Room 21. And given the way Jenkins's body was wrapped in a bedspread, Jones would have been unlikely to determine the source of her wounds. In short, "[k]nowing now, going back and looking at things[,] definitely we should have interviewed him." (Id. at PID 6415.)

Trial counsel's performance was clearly deficient because they abandoned any effort to interview Jones based on nothing more than an assurance by the sheriff that Jones was not involved in Jenkins's murder. In Strickland, the Court explained that a reviewing "court should keep in mind that counsel's function, as elaborated in prevailing professional norms, is to make the adversarial testing process work in the particular case." 466 U.S. at 690. In Towns v. Smith, we interpreted that language to assign to trial counsel an "obligation to investigate all witnesses who may have information concerning his or her client's guilt or innocence." 395 F.3d 251, 258 (6th Cir. 2005) (citation omitted). In determining whether a particular decision not to investigate constituted ineffective assistance, "the relevant question is not whether counsel's choices were strategic, but whether they were reasonable." Id. (citations omitted). Hines's trial counsel made no effort to interview or investigate Jones, even though Jones was a "known and potentially important witness" who clearly had information relevant to Hines's defense. See id. at 259 (citation omitted). This decision was unreasonable—as defense counsel Stack openly admitted—and constituted deficient performance under Strickland.

Hines has also shown prejudice, and the Tennessee court's conclusion to the contrary, see Hines, 2004 WL 1567120, at *27, was an unreasonable application of Strickland. See 28 U.S.C. § 2254(d)(1). Hines was prejudiced by his counsel's failure to investigate Jones because it hindered Hines's ability to effectively challenge the prosecution's theory and timeline of events, as well as undermined Hines's ability to build an affirmative argument pointing to Jones as an alternative suspect. See Ramonez v. Berghuis, 490 F.3d 482, 489-91 (6th Cir. 2007) (concluding a state court's finding of no prejudice under Strickland was objectively unreasonable—and thus the habeas petitioner was entitled to relief—where trial counsel was ineffective for failing to interview witnesses who could corroborate petitioner's version of events while undermining a key prosecution witness's testimony).

The failure to interview or investigate Jones left defense counsel without key information regarding Jones's relationship with Jenkins and his activities at the motel prior to and on the day of the murder—much of which could have helped the defense to credibly cast Jones as an alternative suspect, or at the very least seriously undermine his testimony. At a deposition taken during the initial state post-conviction proceedings, Jones offered information that would have significantly aided Hines's defense at the trial. Jones stated that he and Vernedith White were at the CeBon Motel to use a room on the day of the murder because they were having an affair, which had been on-going for "[a] couple of years." (R. 174-5, PID 5674-75.) They went to the motel almost every Sunday, arriving "between 10:00 and 11:00." (Id. at PID 5675.) Jones explained that he usually contacted Jenkins about obtaining a room: "The girl that was dead, usually she took care of me when I was there." (Id. at PID 5676.) Most Sundays, Jones would get a motel room key from Jenkins instead of the motel owners. Jones stated that he "usually paid $20" for the room rather than the full rate. (Id. at PID 5692.)

On the day of Jenkins's murder, Jones was specifically looking for Jenkins when he arrived at the motel. Jones and White were at the motel for about an hour before finding the body. During that hour, they briefly left to go to the parking lot of a restaurant on the top of the hill before quickly returning—but Jones did not enter the restaurant because he wanted to watch the motel parking lot from the top of the hill to see if anyone would return who would be able to get him a room.

Jones also stated that, contrary to his trial testimony, he did not stick around after the call to the sheriff was made, and instead drove his female companion, Vernedith White, home. When he later returned to the motel, Jones spoke with Sheriff Weakley, whom Jones knew because "one of them boys of mine was always in trouble." (Id. at PID 5694.)

Jones explained that Sheriff Weakley had tried to put him at ease about the problem of why he was at the hotel when he discovered the body, and Jones "understood" that he would not be asked the reason for his presence at the motel by either party at the trial. (Id. at PID 5688-89.) Jones confirmed that he never spoke with an investigator or an attorney for Hines before trial. Jones said that he knew "[n]othing" about the murder other than finding Jenkins's body. (Id. at PID 5695.) Notably, at this deposition Jones gave a different timeline for when he found the body, explaining that he arrived at the motel at "[a]bout 10:30 or so," and found Jenkins's body "[a]round 11:00." (Id. at PID 5695-96.)

Much of this information provided by Jones in his post-conviction deposition could have provided ample fodder for defense counsel to focus on Jones as a reasonable alternative suspect. Jones, who was married, was at the motel due to an affair with a younger woman—an affair which he had a clear motive to hide. Jones had known the victim, Jenkins, through weekly interactions for approximately two years. Jenkins knew of Jones's secret affair with White, and she helped to facilitate the affair by getting Jones room keys and giving him a discounted rate for use of rooms. On the day of Jenkins's murder, Jones arrived at the motel specifically looking for her and monitored the parking lot of the motel closely for an hour. Jones typically paid for the rooms he rented with $20, and a $20 bill was found under the wrist band of Jenkins's watch when her body was found. Furthermore, Mary Sizemore of the Cheatham County Ambulance Service testified that she learned from dispatch that Jones had reported "there was a woman stabbed," but when Sizemore entered Room 21, she was unable to determine that Jenkins had been stabbed until she unwrapped the bedspread around Jenkins, which only then revealed Jenkins's stab wounds. (R. 173-2, PID 3890-91.) If defense counsel had investigated and presented evidence of these suspicious circumstances regarding Jones, there is a reasonable probability that Hines would have been able to convincingly argue at trial that reasonable doubt existed due to Jones's role as a viable alternative suspect for Jenkins' murder. See Poindexter v. Booker, 301 F. App'x 522, 531 (6th Cir. 2008) (affirming a district court's grant of habeas relief where the petitioner was prejudiced by trial counsel's failure to investigate witnesses who could have implicated a third party as the shooter).

Contrary to the state court's determination, Jones's desire to keep his affair a secret from his wife could serve as motive, and the time he spent away from the motel could have been used to dispose of important evidence. Jones's motive and opportunity to commit the crime are at least as compelling as that offered by the prosecution for Hines, if not more compelling. There was no clear motive for Hines to have committed a murder so gruesome of a woman he had never met before, in which her body was brutally stabbed in the vagina even after she was incapacitated and possibly already dead.

Pointing to Jones as an alternative suspect may have been a viable path for the defense, as the evidence of Hines's guilt was not overwhelming. See English v. Romanowski, 602 F.3d 714, 730 (6th Cir. 2010) (finding prejudice from counsel's failure to investigate where the "evidence of [the habeas petitioner's] guilt. . . [wa]s not overwhelming" and "[t]he government presented no physical evidence"). In so concluding, we are cognizant of the difference between overwhelming evidence and sufficient evidence. We do not question that there was sufficient evidence to sustain Hines's conviction for first-degree murder. The dissent recounts this evidence, which we have also carefully considered. But Strickland's prejudice inquiry differs from a sufficiency-of-the-evidence analysis. See Ferensic v. Birkett, 501 F.3d 469, 474 (6th Cir. 2007). Under Strickland, we ask whether there is a "reasonable probability" that one juror would have voted differently but-for counsel's deficient performance. See English, 602 F.3d at 730. Here, there was ample room for defense counsel to point to Jones as an alternative murder suspect. There was no DNA or fingerprint evidence connecting Hines to Jenkins's murder—not on Jenkins's body, not in the room where the murder took place (Room 21), and not on Hines's clothing. In addition, no witness testified to seeing Hines near Room 21. In contrast, Jones was clearly in Room 21 on the day of the murder, had a plausible motive to kill Jenkins, and knew information about the circumstances of Jenkins's injuries that would not have been available to someone who just happened upon her wrapped body.

Defense counsel's closing argument at the guilt phase alluded to the idea that Jones's presence at the motel had been suspicious. Yet without any evidence collected from an investigation into Jones to support this argument, defense counsel likely undermined the defense's credibility with the jury by making this implication. Where defense counsel fails to corroborate statements to the jury, "the jury may well have counted this . . . against [Hines] and his attorney." English, 602 F.3d at 729. "[T]he jury would naturally assume" that defense counsel's uncorroborated Jones theory "lacked reliability," without knowing that the lack of corroboration was instead a function of defense counsel's negligence in failing to investigate. See Stewart v. Wolfenbarger, 468 F.3d 338, 360 (6th Cir. 2006). Had defense counsel collected evidence to properly show the jury why Jones's behavior that morning and relationship with Jenkins were highly suspicious, the strength of Hines's defense likely would have looked much different. "The difference between the case that was and the case that should have been is undeniable." Id. at 361.

In any event, Hines does not need to show that Jones was the actual killer to succeed on his claim before this court. "Even though the jury could have discredited" the theory that Jones was the true murderer, "there certainly remained a reasonable probability that the jury would not have," and that is sufficient to show prejudice under Strickland. Ramonez, 490 F.3d at 491.

Furthermore, even absent attempting to affirmatively argue that Jones was an alternative suspect, pre-trial investigation into Jones could have allowed defense counsel to effectively challenge the prosecution's case by, at the very least, seriously undermining Jones's testimony and calling the prosecution's timeline of events into question. As trial counsel explained during post-conviction proceedings, there were numerous inconsistencies between Jones's testimony and the testimony of others—such as an hour-long gap between when Jones allegedly found the body at 1:20 and when dispatch was called at 2:36, as well as the inexplicability of Jones's first report claiming that a woman had been stabbed given that Jenkins's body was wrapped in a bedspread and the cause of her injury would not be apparent unless someone attempted to unwrap her. Defense counsel were unaware of these inconsistencies before trial—and thus did not investigate them further—due to their failure to interview or investigate Jones.

Given that Jones gave an entirely different timeline for his presence at the motel and discovery of the body during his post-conviction deposition—estimating he found the body around 11:00, more than three hours before dispatch was called—there is ample reason to think a pre-trial interview of him would have provided defense counsel further evidence to argue that the prosecution's timeline was flawed and that Jones was an unreliable witness. Yet defense counsel was unprepared to challenge the government's case in this manner, as counsel made no attempt to investigate or interview Jones before trial—thereby prejudicing Hines. See Stewart, 468 F.3d at 360-61 (finding Strickland prejudice, and in turn the habeas petitioner entitled to relief, where trial counsel's failure to investigate allowed the prosecution's evidence "to go unchallenged").

The state court, in concluding there was no prejudice from counsel's failure to investigate Jones, unreasonably ignored the key evidence learned at Jones's post-conviction deposition. See Hines, 2004 WL 1567120, at *27. The state court ignored the serious inconsistencies and gaps in Jones's story, and it ignored the new evidence of the extent of Jones's established relationship with the victim. See id. The court mentioned that Jones was "known by the staff" at the motel, "including the victim." Id. But Jones's deposition testimony clearly evidences more than that. It shows that Jenkins was the specific person at the motel who, on a weekly basis for two years, would provide Jones with a motel room to facilitate his affair while allowing him to circumvent the motel's established daily rate. And on the morning of the murder, Jones was lurking in the parking lot for about an hour, specifically looking for Jenkins—not just any staff member at the motel. Only by ignoring this evidence did the state court conclude that pointing to Jones as an alternative suspect would have been "farfetched." See id. For the state court's analysis to have ignored this evidence was objectively unreasonable, as "weighing the prosecution's case against the proposed witness testimony" that was not elicited due to counsel's ineffectiveness "is at the heart of the ultimate question of the Strickland prejudice prong." Ramonez, 490 F.3d at 491. As the Supreme Court explained in Williams, a state court's "prejudice determination" is "unreasonable insofar as it fail[s] to evaluate the totality of the available . . . evidence." 529 U.S. at 397-98.

Given that presenting Jones as an alternative suspect would not have been "farfetched" in light of Jones's deposition testimony, the state court was similarly unreasonable in concluding that presentation of this theory "could have resulted in a loss of credibility for the defense." Hines, 2004 WL 1567120, at *27. Armed with evidence to emphasize the suspiciousness of Jones's activities at the motel—which is now apparent due to Jones's post-conviction testimony—Hines could have made a convincing argument that Jones was a viable alternative suspect. Hines's counsel alluded to this argument at trial anyway, without any evidence from an investigation in support—and that, in turn, was what undermined the defense's credibility with the jury. See English, 602 F.3d at 729. The state court's decision ignored the fact the defense counsel in closing had already pointed at Jones, and ignoring the trial record in its prejudice determination was objectively unreasonable. See Williams, 529 U.S. at 398 (explaining a state court's "prejudice determination was unreasonable" where it "failed to even mention the . . . argument . . . that trial counsel did advance").

Hines needs only to show "a reasonable probability"—not a certainty—"that, but for counsel's unprofessional errors, the result of the proceeding would have been different." Strickland, 466 U.S. at 694. Hines has carried his burden. For the result of Hines's guilt trial to have been different, he only would have needed to sow reasonable doubt in at least one juror's mind based on evidence related to Jones. "[T]he negative consequences of defense counsel's failure to conduct a sufficient pre-trial investigation" into Ken Jones "sufficiently creates a reasonable probability that at least one juror would have struck a different balance had defense counsel not performed deficiently." English, 602 F.3d at 730. The state court's contrary ruling was an unreasonable application of Strickland.

B. Failure to present evidence of residual doubt at the sentencing phase

Hines also contends that trial counsel were ineffective because they did not present evidence regarding Jones in support of residual doubt at the penalty phase. The warden argues that this claim is procedurally defaulted because it was not raised in state court and because Hines cannot establish the requisite cause and prejudice to excuse the default under Martinez.

The warden is correct that Hines did not raise this claim in the state trial court. Rather, he raised it at oral argument on post-conviction appeal at the Tennessee Court of Criminal Appeals, in conjunction with the IATC claim arising from counsel's failure to interview Jones. Hines, 2004 WL 1567120, at *26 ("We will review this argument along with the related claim, made at oral argument, that trial counsel could have created residual doubt by properly dealing with Ken Jones."). The Tennessee Court of Criminal Appeals subsequently denied the claim for the same reasons it denied the IATC claim. Id. at *28.

The Tennessee Court of Criminal Appeals thus adjudicated this claim—at least as to Jones's involvement—on the merits without imposing a state procedural bar. The claim is thus not defaulted and we address its merits. See Ylst v. Nunnemaker, 501 U.S. 797, 801 (1991) ("If the last state court to be presented with a particular federal claim reaches the merits, it removes any bar to federal-court review that might otherwise have been available.").

There is no Eighth Amendment right to a jury instruction concerning residual doubt in the penalty phase. See Franklin v. Lynaugh, 487 U.S. 164, 173 (1988). Under Tennessee law, however, a capital defendant may present evidence of residual doubt at the penalty phase. State v. Hartman, 42 S.W.3d 44, 55-56 (Tenn. 2001) (citing State v. Teague, 897 S.W.2d 248, 256 (Tenn. 1995)). The Tennessee Supreme Court explained that "residual doubt is established by proof that casts doubt on the defendant's guilt. It is not limited to proof that mitigates the defendant's culpability for the crime." Id. at 57.

As discussed, trial counsel were ineffective for failing to investigate Jones's conduct at the motel when Jenkins was killed, and were similarly ineffective for failing to present this evidence in the penalty phase of trial. If presented with this evidence regarding Jones, there is a "reasonable probability" that the sentencing jury would have reached a different verdict, see Strickland, 466 U.S. at 694, and the Tennessee court's contrary conclusion was an unreasonable application of Strickland.

CONCLUSION

Because Hines's trial counsel were ineffective for failing to investigate Ken Jones, and the state court's determination otherwise was an unreasonable application of Strickland, we REVERSE the district court's order denying relief and REMAND for proceedings consistent with this opinion.

KETHLEDGE, Circuit Judge, dissenting. Respectfully, the majority opinion makes precisely the same mistake for which our court was summarily reversed in Etherton v. Rivard, 800 F.3d 737 (2015), rev'd sub nom. Woods v. Etherton, 136 S. Ct. 1149 (2016) (per curiam). Specifically, the opinion "nowhere gives deference to the state courts, nowhere explains why their application of Strickland was unreasonable rather than merely (in the majority's view) incorrect, and nowhere explains why fairminded jurists could view [the petitioner's] claim only the same way the majority does. The opinion, in other words, does exactly what the Supreme Court has repeatedly told us not to do." Etherton, 800 F.3d at 756-57 (dissenting opinion).

Here, neither Hines nor the majority has remotely shown that Hines was prejudiced by his trial counsel's failure to investigate Ken Jones. To begin, the evidence that Hines killed Katherine Jenkins was overwhelming. Two days before the murder, Hines boarded a bus in North Carolina with a one-way ticket to Kentucky. He had a large hunting knife sheathed beneath his shirt. His girlfriend's mother—who had bought the ticket because Hines could not afford it himself—admonished him for taking the knife on the bus, but Hines responded, "I never go anywhere naked. I always have my blade." R. 173-4, Pg. ID 4201.

Shortly after midnight on March 3, Hines checked into Room 9 of the CeBon Motel in Kingston Springs, Tennessee. Later that morning, around 9:30 a.m., the motel's manager put maid Katherine Jenkins in charge of the motel's operations and gave her a bank bag containing $100 in small bills. Three hours later, around 12:40 p.m., another maid saw a man driving Jenkins's Volvo away from the motel. The maid got into her own car and gave chase, but the Volvo sped off, heading east toward Nashville.

Around the same time, Ken Jones arrived at the CeBon Motel. Nobody was at the front desk, so Jones eventually took the key to Room 21 and left a note saying that he was using the restroom there. (Testimony at a state post-conviction hearing revealed that Jones was there that day with Vernedith White, with whom he had been having an affair for 11 years.) When Jones walked inside Room 21, however, he found Jenkins's body wrapped in a bedspread, on the floor on the far side of the room's two beds. He ran out of the room and across the street to a restaurant, where he asked someone to call the county sheriff.

Sheriff's deputies arrived soon thereafter. They searched Room 21 and, in addition to Jenkins's body, found the bank bag—bloody and empty—along with an unfiltered cigarette burned down to a nub. Then they examined the body. Someone had pulled Jenkins's clothing up to her breasts; her underwear was cut in two pieces and scattered across the room. Her neck had superficial wounds, consistent with "some firm sharp object [held] to [her] neck," and her hands showed defensive wounds as if she had tried to "ward off injury." R. 173-5, Pg. ID 4304. But the fatal wounds were to her chest—five "deep, penetrating wounds, ranging from 2.5 inches to 6.4 inches in depth." Hines v. State, No. M2004-01610-CCA-RM-PD, 2004 WL 1567120, at *2 (Tenn. Crim. App. July 24, 2004); see also R. 173-4, Pg. ID 4168; R. 173-5, Pg. ID 4283. A final knife wound, likely inflicted after Jenkins had died, went through her vagina and penetrated her abdominal cavity. The deputies also discovered stab holes with similar widths and depths in the walls of Room 9—the room that Anthony Hines stayed in the prior night. Missing altogether from the scene was Jenkins's wallet, keys (which were attached to an "I love Volvo" keychain), and her Volvo.

Meanwhile, a group of young adults spotted the Volvo—along with Anthony Hines—on the side of the road near Gallatin, Tennessee. The car's engine had overheated—perhaps from being driven at high speeds—and the youths tried to help Hines cool it off. When that failed, Hines offered them $10 for a ride to his sister's house in Bowling Green, Kentucky. They accepted. On the way, the youths said, Hines "seemed real nervous," his eyes wide and bright; and he "talked a lot"—saying, for example, that he had bought the Volvo from "an old lady for $300 or $400." R. 173-2, Pg. ID 3910, 3932-33; R. 173-3, Pg. ID 4022. One of the youths noticed dried blood on Hines's shoulder. During the drive, Hines carried a jacket that he kept folded.

Hines arrived in Bowling Green sometime between 3:00 and 4:00 p.m. His sister too noticed blood on his shirt. Hines explained that someone had attacked him at the CeBon Motel, and that he had stabbed the attacker "in the side . . . and in the chest[.]" R. 173-2, Pg. ID 3967. But he told his brother-in-law a different story: that he had hitchhiked a ride with a stranger driving a Volvo, that the stranger had tried to rob him, and that during the ensuing struggle the stranger's Volvo had run off the road and flipped over. Afterward, Hines said, he had grabbed the Volvo's keys and escaped. He showed his brother-in-law the keychain, which said something like, "I love Volvo." The brother-in-law gave Hines a ride to Cave City, Kentucky, where Hines's grandparents lived. When Hines arrived in Cave City, he bought a grill as a gift for his sister and brother-in-law.

The police found the Volvo around 4:45 p.m., precisely where Hines had abandoned it. They also found Jenkins's wallet about 20 feet in front of the car, wrapped in a shirt. Any cash that had been in the wallet was gone.

For the next eight days, Hines hid out in the hills around Cave City. On March 11 he turned himself in to a Kentucky sheriff. Before the sheriff said anything about the murder, Hines volunteered that he had stolen the Volvo but said that he had not killed Jenkins. Later that day, Hines told deputies that he would confess to the murder if they would guarantee that he would be sentenced to death. Deputies eventually investigated Hines's campsite and found, among other items, unfiltered cigarettes—much like the one discovered in Room 21.

The jury heard all this evidence at trial. They heard that Hines always carried a large hunting knife; that Jenkins's neck had wounds suggesting that someone had held her at knifepoint; that her chest and vagina had knife wounds consistent with holes in the wall in Hines's motel room; that on March 1 Hines could not afford a $20 bus ticket, but that on March 3—hours after Jenkins's murder—he was flush with cash and bought a grill for his sister; that Hines had stolen Jenkins's wallet, keys, and car; that Hines had blood on his shirt that afternoon; that he told his sister that he had stabbed an "attacker" at the motel; and that he volunteered to tell sheriff's deputies "all about the murder" if they guaranteed him the death penalty.

The question here is whether every "fairminded jurist" would agree that, if only Hines's counsel had investigated Ken Jones, there would have been a "reasonable probability" that the result at Hines's trial would have been different. See Harrington v. Richter, 562 U.S. 86, 101 (2011); Strickland v. Washington, 466 U.S. 668, 694 (1984). But neither Hines nor the majority has even attempted to make that showing. Nor could they. At trial, Jones offered no testimony regarding Hines's guilt, instead testifying about his discovery of the body. Any post-investigation to impeach him on that score would have been a waste of time—which makes this case easily distinguishable from the cases cited by the majority. See Ramonez v. Berghuis, 490 F.3d 482, 489-91 (6th Cir. 2007) (investigation could have led to impeachment of the prosecution's key witness); Stewart v. Wolfenbarger, 468 F.3d 338, 361 (6th Cir. 2006) (same). And there is zero reason to think that, after investigation, counsel could have presented Jones as the "real killer" at trial. Quite the contrary: in post-conviction proceedings, Jones and White testified that they were regulars at the CeBon Motel, and that they came to the motel on March 3 to do what they had done at least "100 times"—namely, to carry on their affair, as part of their "normal Sunday routine." Hines, 2004 WL 1567120, at *27. And White testified that Jones was in Room 21 that morning for "less than a minute"—with her watching him the whole time—before he came running out, scared and—unlike Hines—without any blood on his clothes.

In sum, the Tennessee Court of Criminal Appeals had every reason to reject Hines's Strickland claim on the ground that it was "farfetched." See id. And we have no reason whatever to grant habeas relief on that same claim here. I respectfully dissent.