OPINION

BOGGS, Circuit Judge.

Nicholas T. Sutton is a Tennessee prisoner sentenced to death for murdering a fellow prisoner. He petitions for a writ of habeas corpus on the grounds that his trial counsel was constitutionally ineffective. We affirm the district court's denial of habeas relief.

I. Background

On January 15, 1985, Carl Estep, an inmate at Tennessee's Morgan County Regional Correctional Facility, was murdered in his cell. He was stabbed thirty-eight times in the chest and neck with two homemade knives, or "shanks," which were found near his body. Defensive wounds on his hands and arms, as well as blood on his body, the walls, and the bunk, indicated that there had been a struggle.

Sutton, Thomas Street, and Charles Freeman were charged with Estep's murder. At trial, the primary evidence against Sutton was the testimony of three other inmates, Harold Meadows, Estel Green, and Cary Scoggins. Meadows testified that, shortly before the body was discovered, he saw Sutton and Street enter Estep's cell and heard Estep scream. He also claimed that two days before, Sutton and Estep had a "physical" discussion, during which Sutton held a knife to Estep's throat. Green testified that he also saw Sutton and another inmate go into Estep's cell and that he heard Estep screaming while Sutton was inside. Scoggins explained that Sutton and Estep had been feuding over a drug deal, and that Estep had threatened to kill Sutton. Scoggins also testified that he saw Sutton, Street, and Freeman enter Estep's cell, and that he watched through the cell-door window as Sutton repeatedly stabbed Estep.

The jury convicted Sutton and Street but acquitted Freeman. Sutton was sentenced to death based on three statutory aggravating circumstances: (1) he had previously been convicted of a violent felony, first-degree murder; (2) he was incarcerated at the time of Estep's murder; and (3) Estep's murder was "heinous, atrocious, or cruel." See Tenn.Code Ann. § 39-2-203(i)(2), (5), (8) (1986). The Tennessee Supreme Court affirmed the conviction and sentence on direct appeal, State v. Sutton, 761 S.W.2d 763 (Tenn.1988), and the Tennessee Court of Criminal Appeals rejected Sutton's petition for postconviction relief, Sutton v. State, 1999 WL 423005 (Tenn.Crim.App. June 25, 1999).

Sutton appeals from the district court's denial of his petition for a writ of habeas corpus. He has received a certificate of appealability on four ineffective-assistance-of-counsel claims: (1) that his counsel failed to object to two aspects of courtroom security during the guilt phase; (2) that his counsel failed to object to three instances of prosecutorial misconduct during the guilt and penalty phases; (3) that his counsel failed to object to the penalty-phase jury instructions on the "heinous, atrocious, or cruel" aggravating circumstance; and (4) that his counsel failed to adequately investigate and present mitigating evidence of the amount of violence in Tennessee prisons and of his troubled background.

II. Standard of Review

Sutton's ineffective-assistance claims are governed by the familiar standard of Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). He "must show that counsel's representation fell below an objective standard of reasonableness" and "that there is a reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different." Id. at 688, 694, 104 S.Ct. 2052. Given the prejudice requirement, "counsel cannot be ineffective for a failure to raise an issue that lacks merit." Greer v. Mitchell, 264 F.3d 663, 676 (6th Cir.2001).

All of Sutton's claims were adjudicated on the merits by the Tennessee state courts on postconviction review. Therefore, we may not grant the writ unless the state court decision "was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court," or "was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding." 28 U.S.C. § 2254(d). An adjudication is contrary to clearly established law if, for example, the "state court applies a rule that contradicts the governing law set forth in [Supreme Court] cases." Williams v. Taylor, 529 U.S. 362, 405-06, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000). A state court decision is unreasonable if, for example, "the state court identifies the correct governing legal rule . . . but unreasonably applies it to the facts of the particular state prisoner's case." Id. at 407, 120 S.Ct. 1495. The application must be "objectively unreasonable," not merely incorrect. Id. at 409-10, 120 S.Ct. 1495.

III. Trial Security

Sutton's first claim is that his counsel was constitutionally ineffective for failing to raise two objections under Holbrook v. Flynn, which prohibits trial "practices" that "prejudice" the defendant without "sufficient cause." 475 U.S. 560, 568, 571, 106 S.Ct. 1340, 89 L.Ed.2d 525 (1986). He first argues that his counsel should have objected to "the conspicuous, or at least noticeable, deployment of" ten uniformed guards in the courtroom,1 seven of whom were armed, because they suggested "official concern or alarm" that he was "particularly dangerous or culpable," id. at 569, 106 S.Ct. 1340.

The state appellate court reasonably rejected this claim because the underlying Flynn claim failed. The guards' presence likely caused little prejudice: as the trial judge testified during postconviction proceedings, they were "not overly conspicuous" because they were spaced out in the very full courtroom—four were behind the defense table, one was next to the jury, two were in the balcony, and one was posted at each of the courtroom's three doors. And we agree with the trial judge that the legitimate security concerns involved in trying three inmates for violently murdering a fourth inmate, where the defendants were not wearing upper-body restraints and six other inmates were testifying as witnesses, was "sufficient cause" for any prejudice. See Flynn, 475 U.S. at 571, 106 S.Ct. 1340 (holding that "the State's need to maintain custody over defendants who had been denied bail" as flight risks was "sufficient cause" for whatever prejudice resulted from the "spectacle of four [uniformed and armed] officers quietly sitting" behind the defendants); Bell v. Hurley, 97 Fed.Appx. 11, 16-17 (6th Cir.2004) (noting that visibly shackling the defendant—which, unlike the presence of guards, is "inherently prejudicial" under Flynn—was justified because he was accused of attacking a guard during a prison riot); see also Deck v. Missouri, 544 U.S. 622, 632, 125 S.Ct. 2007, 161 L.Ed.2d 953 (2005) (recognizing "the need to give trial courts latitude in making individualized security determinations"); United States v. Barger, 931 F.2d 359, 371 (6th Cir.1991) ("[T]he degree of security relating to a defendant is within the [trial] judge's discretion.").

Sutton also contends that his counsel should have raised a Flynn objection to what he calls the "shanks incident." Before introducing the murder weapons into evidence, the prosecutor placed them on the defense table, within reach of the defendants, for inspection by counsel. Sutton's counsel jerked away from him in fear and the guards reached for their weapons; there is conflicting testimony over whether any were actually drawn. Although the shanks were moved to the state's table for inspection without further incident, Sutton claims that his counsel's and the guards' reactions suggested that he was very dangerous and were therefore so "prejudicial as to pose an unacceptable threat to [his] right to a fair trial," Flynn, 475 U.S. at 572, 106 S.Ct. 1340.

The state court again rejected Sutton's ineffective-assistance claim because the underlying Flynn claim failed. This decision was not contrary to or an unreasonable application of clearly established Supreme Court law because "[n]o holdings of [the Supreme] Court required the [state court] to apply the test of . . . Flynn to" counsel's and the guards' reactions, Carey v. Musladin, 549 U.S. 70, 75-77, 127 S.Ct. 649, 166 L.Ed.2d 482 (2006). In Musladin, the Supreme Court held that Flynn's application to "private-actor[s'] courtroom conduct" was an open question under its case law, and therefore a state court's application of Flynn to spectators' conduct could be neither contrary to nor an unreasonable application of Flynn. Ibid. Sutton's counsel was a private actor, at least for these purposes, and therefore we cannot say the state court's decision that his reaction did not cause Flynn error was improper. See ibid.

Flynn's applicability to the guards' reaction is similarly uncertain. As Musladin noted, Flynn and its related case law focus exclusively on "state sponsored practices," id. at 75-77, 127 S.Ct. 649 (emphasis added); they do not address security's response to, e.g., courtroom outbursts or attacks. See Flynn, 475 U.S. at 567-71, 106 S.Ct. 1340 (discussing "practices" and "procedures" such as the presence of guards); Estelle v. Williams, 425 U.S. 501, 503-04, 96 S.Ct. 1691, 48 L.Ed.2d 126 (1976) (addressing the "practice" and "procedure" of requiring defendants to wear prison clothes); Illinois v. Allen, 397 U.S. 337, 340, 90 S.Ct. 1057, 25 L.Ed.2d 353 (1970) (addressing the binding and gagging of defendants). Furthermore, courts have long recognized that forceful reactions to courtroom outbursts or attacks may require a new trial even where there was "sufficient cause" for the response—and thus where Flynn would not imply a constitutional error. See, e.g., United States v. Serio, 440 F.2d 827, 830-31 (6th Cir. 1971) (collecting cases, on direct appeal, about requests for a mistrial based on courtroom outbursts or attacks); Braswell v. United States, 200 F.2d 597, 600-02 (5th Cir.1952) (holding on direct appeal that the trial judge improperly refused to grant a mistrial after two of the defendants violently attacked a United States Marshal and were forcefully subdued in front of the jury).2 The lack of a textual mandate, plus the existence of doctrine predating Flynn that more stringently evaluates events such as the shanks incident, leads us to conclude that, at the least, Flynn's applicability to the guards' reaction is an open question. Therefore, we cannot say that the state court improperly applied Flynn to the guards' reactions. See Musladin, 549 U.S. at 77, 127 S.Ct. 649.

IV. Prosecutorial Misconduct

Sutton's next claim is that his counsel failed to object to three statements in the prosecutor's closing arguments. First, Sutton argues that his counsel should have objected to the prosecutor's guilt-phase comment suggesting that Sutton was guilty because he was arrested for the murder:

The state appellate court rightly noted that this comment was improper, but it concluded that Sutton was not prejudiced by his counsel's failure to object.3 This was reasonable. Any prejudice from this comment affected only Sutton's factual guilt. However, the evidence of his factual guilt was strong: although the inmate-witnesses' testimony was contradictory and weak on some matters, it was consistent and persuasive on Sutton's involvement in and motive for Estep's murder. Therefore, we do not think there is a reasonable probability that an objection to this comment would have led to Sutton's acquittal.

Second, Sutton challenges his counsel's failure to object to the prosecutor's guilt-phase discussion of inmate Carl Crafton's testimony. Crafton testified as a defense expert on prison life, describing the violence then endemic in the Tennessee prison system because of overcrowding and guards' inability to keep order, as well as the culture and incentives this created for prisoners. In particular, he explained that once an inmate's life had been threatened, "his only defense[] is to make the offensive move." Sutton relied on this testimony to argue that, if he did kill Estep, he did so in self defense because Estep had threatened to kill him.

Sutton claims that the prosecutor "deliberately misled the jury by claiming that Crafton advocated violent prisons where inmates lived by their own rules":

The state court concluded that this comment was also improper but not prejudicial. This was reasonable because Sutton's preemptive-strike self-defense theory failed as a matter of state law. See State v. Leaphart, 673 S.W.2d 870, 873 (Tenn. Crim.App.1983) (holding that the lack of an imminent threat precludes preemptive killing from qualifying as self-defense).

Third, Sutton contends that his counsel should have objected to the prosecutor's penalty-phase future-dangerousness argument:

These statements were not improper as a matter of federal law: they were not inflammatory, and future-dangerousness arguments are permissible under federal law, Jurek v. Texas, 428 U.S. 262, 275, 96 S.Ct. 2950, 49 L.Ed.2d 929 (1976). The state court concluded that they were improper under state law because they suggested that the jury consider deterrence, which is not relevant to any statutory aggravating factor. Nonetheless, the court decided that Sutton was not prejudiced,4 which was reasonable given the overwhelming aggravating circumstances. Furthermore, any impropriety under state law was harmless because we conclude that the prior-violent-felony and prison-murder statutory aggravators allowed the jury to consider the prosecutor's argument that only death will prevent Sutton from killing again, and that death is the proper punishment for someone who killed while in prison for murder. Cf. Brown v. Sanders, 546 U.S. 212, 220, 126 S.Ct. 884, 163 L.Ed.2d 723 (2006) (explaining that, in capital cases, the consideration of sentencing factors that are improper under state law is constitutional error, but holding that such error is harmless if the federal court determines that other, proper factors allow consideration of "the same facts and circumstances").5

V. Penalty Phase Jury Instructions

Sutton's third ineffectiveness claim is that his counsel should have argued that the penalty-phase jury instructions on the "heinous, atrocious, and cruel" aggravating circumstance were unconstitutionally vague. This claim fails for lack of prejudice. The Tennessee Supreme Court reviewed and affirmed the jury's finding of the aggravator on direct appeal. Because there is no "affirmative indication to the contrary, we must presume that it" applied its well-established, and permissible, narrowing construction of the aggravator, thereby "cur[ing] any error in the jury instruction." Bell v. Cone, 543 U.S. 447, 453-56, 125 S.Ct. 847, 160 L.Ed.2d 881 (2005) (per curiam) (rejecting an identical challenge to Tennessee's "heinous, atrocious, and cruel" aggravator for this reason); see also Payne v. Bell, 418 F.3d 644, 653-60 (6th Cir.2005) (same).

VI. Mitigating Evidence

Sutton's final claim is that his counsel failed to adequately investigate and present mitigation evidence. At the sentencing hearing, counsel presented additional testimony from Crafton to support the statutory mitigating circumstance that Sutton was acting under the belief that his actions were morally justified. Counsel also presented the testimony of the Burchetts, a family who had known Sutton since he was in high school and who visited him regularly in prison. This testimony was intended to humanize Sutton and to show that he is capable of having normal relationships with law-abiding citizens.

Sutton first argues that, in addition to Crafton's testimony about the violent conditions in Tennessee prisons, counsel should have presented the testimony of a particular TDOC corrections officer and documents from Grubbs v. Bradley, a federal lawsuit that, from 1980 to 1993, evaluated the violent conditions in Tennessee prisons, see, e.g., 552 F.Supp. 1052 (M.D.Tenn.1982). The state court reasonably rejected this claim for lack of prejudice. The additional evidence would have been cumulative and added little or no extra mitigating value, particularly since we think it is unlikely that the jury did not believe Crafton that prison was violent given his status as a court-sanctioned expert and the nature of the crime at issue. See Hill v. Mitchell, 400 F.3d 308, 319 (6th Cir.2005) ("[I]n order to establish prejudice, the new evidence that a habeas petitioner presents must differ in a substantial way—in strength and subject matter—from the evidence actually presented at sentencing.").

Sutton also argues that his trial counsel should have discovered and presented evidence of his troubled background. This evidence was presented at postconviction proceedings by Dr. Gillian Blair, a licensed clinical psychologist who evaluated Sutton. The state appellate court summarized her testimony:

The probability of a different sentence had this evidence been presented is a function of the strength of the aggravating circumstances and the net mitigating value of Sutton's troubled background. See Wiggins v. Smith, 539 U.S. 510, 516-18, 534, 536-38, 123 S.Ct. 2527, 156 L.Ed.2d 471 (2003) ("In assessing prejudice, we reweigh the evidence in aggravation against the totality of available mitigating evidence."); Williams, 529 U.S. at 396-98, 120 S.Ct. 1495 (considering prejudice de novo and balancing the proposed evidence's mitigating value, its potential harm, and the aggravating circumstances); Carter v. Mitchell, 443 F.3d 517, 530-33 (6th Cir.2006) (holding that the state court's decision was reasonable because the proposed troubled-background evidence likely "would have done more harm than good"). The state appellate court weighed these factors and concluded that Sutton's troubled-background evidence would not have created a reasonable probability of a life sentence. We cannot say that the state court unreasonably considered, or gave unreasonable weight to, any factor, or that its ultimate balancing of the variables was unreasonable.6

The aggravating circumstances were, as the state court suggested, overwhelming: while in prison for first-degree murder, Sutton planned and carried out the killing and mutilation of another inmate because of a drug deal. The prior-violent-felony and prison-murder statutory aggravators are undisputable and weigh strongly against Sutton, and the jury's finding of a third statutory aggravator, that the murder was "heinous, atrocious, or cruel," reflects the brutality of the murder and suggests that the jury believed that the aggravating circumstances were extraordinary.

Given these aggravating circumstances, only evidence with an overwhelming net mitigating value could produce a reasonable probability of a life sentence. The state court reasonably noted two points suggesting that, while the mitigating value of Sutton's troubled background evidence might be substantial, it was not overwhelming. First, the court observed that Sutton offered "little positive or redeeming evidence." This is true, as the mitigating value of Sutton's background comes primarily from its troubled nature, and what little was positive—primarily that he once saved a guard during a prison fight—was not overwhelming. Thus that avenue of mitigation would have been of little or no benefit to Sutton.

Second, the court noted that Sutton's troubled past did not reflect the sort of extreme deprivation or mental and emotional problems that might be thought to reduce his culpability to a critical degree. Sutton v. State, 1999 WL 423005, at * 18. Sutton's background is undeniably chaotic and unfortunate, but he did have one constant and positive influence: his grandmother, who raised him and adequately provided for him. And, as Dr. Blair conceded, Sutton had no mental disease or severe emotional disturbances; he merely had a personality disorder reflecting unexceptional maladjustment. See Wickline v. Mitchell, 319 F.3d 813, 821 (6th Cir.2003) (noting that the defendant "did not suffer from any mental condition" and suggesting that his depression was a "weak mitigating factor" (internal quotation marks omitted)).

This limited mitigating value must be weighed against the potential harm its introduction might have done to Sutton's mitigation case. It is well established that facts such as Sutton's extensive involvement with drugs and his discharge from the Navy are often viewed by juries as harmful, and this must be counted against the proposed evidence's mitigating value. See, e.g., Williams, 529 U.S. at 396, 120 S.Ct. 1495 (balancing the unfavorableness of the defendant's juvenile convictions against other, favorable evidence); Burger v. Kemp, 483 U.S. 776, 793, 107 S.Ct. 3114, 97 L.Ed.2d 638 (1987) (noting that evidence of the defendant's "involve[ment] with drugs" "could have affected the jury adversely"); Carter, 443 F.3d at 530 (noting that evidence of, inter alia, a "history of drug use and alcohol abuse" would be "double edge[d]").

The state court also reasonably considered the possibility that presentation of Sutton's troubled-background evidence would have "potentially opened the door" to devastating rebuttal evidence of his prior violent acts, in particular the fact that he brutally beat his grandmother to death after she found out that he had murdered two other people. See Mason v. Mitchell, 543 F.3d 766, 780-83 (6th Cir.2008) (weighing the possibility that damaging rebuttal evidence would be introduced); Scott v. Mitchell, 209 F.3d 854, 880-81 (6th Cir. 2000) (explaining that the proposed background evidence's mitigating value would have been "largely, even overwhelmingly, negated" because it would have allowed the prosecutor to elicit "evidence that his background includes commission of robbery, assault, kidnaping, [sic] and other violent acts upon innocent citizens"). Even discounted by a relatively low probability of admission, the expected value of such devastating rebuttal evidence is quite large, and it significantly reduces the net mitigating value of Sutton's background. Cf. Rickman v. Bell, 131 F.3d 1150, 1157-60 (6th Cir.1997) (excoriating counsel for introducing evidence of his client's violent and troubled background that "creat[ed] a loathsome image . . . that would make a juror feel compelled to rid the world of him").

Perhaps Sutton is correct that competent counsel likely could have presented evidence of his troubled background without making such rebuttal evidence relevant, and thus admissible. See Carter v. Bell, 218 F.3d 581, 597-600 (6th Cir.2000) (concluding that, under Tennessee law, competent counsel likely could have introduced troubled-background evidence without allowing the introduction of rebuttal evidence of his prior violent acts (citing Cozzolino v. State, 584 S.W.2d 765, 767-68 (Tenn.1979))). However, the state court evaluated Sutton's troubled-background evidence under state law and concluded that there was at least a colorable chance that the trial judge would have ruled that its presentation would allow the introduction of such rebuttal evidence, perhaps because the background evidence touched on Sutton's juvenile crimes and violent behavior. We presume that the state court properly applied its own law, see Cone, 543 U.S. at 453-56, 125 S.Ct. 847, and we defer to its estimation that there was some recognizable "potential[]" that the rebuttal evidence would be admitted since it is better suited to predict state law than we are. See Carter, 218 F.3d at 599-600 (noting that this court "g[ives] deference" to state court determinations of the possible admissibility of rebuttal evidence in these situations); Scott, 209 F.3d at 880-81 (deferring to such a state court determination).

Nor can we say that the state court's balancing of these variables was unreasonable. The mitigating value of Sutton's background was not overwhelming, and it is countered by the double-edged nature of significant portions of his background and the small, though potentially devastating, danger of rebuttal evidence. As a result, the evidence's net mitigating value was dwarfed by the extreme aggravating circumstances. And while the prejudice inquiry is unavoidably case-specific and fact-intensive, we are satisfied that our decision is in the mainstream of failure-to-introduce-mitigating-evidence case law decided under AEDPA's strictures. See, e.g., Wickline, 319 F.3d at 821-22 (holding that the state court reasonably concluded that the mitigating value of evidence about the defendant's mental health, good behavior, and troubled background was insufficient to establish prejudice); Williams v. Cain, 125 F.3d 269, 277, 279-80 (5th Cir.1997) (holding that the state court reasonably concluded that the defendant was not prejudiced by counsel's failure to present evidence of "`chaotic, violence-filled childhood'" because the evidence "likely would have had little mitigating effect against the aggravating evidence concerning the brutal, premeditated murder . . ., [his] prior criminal history, and the fact that [he] hid [evidence]"); Campbell v. Kincheloe, 829 F.2d 1453, 1464 (9th Cir.1987) (concluding that the defendant was not prejudiced by counsel's failure to present evidence that his "father was an alcoholic; [he] was the victim of child abuse; he suffered from various medical problems as a young child; he had a history of drug and alcohol abuse; [and] he had reportedly attempted suicide on one occasion" because the mitigating value of this evidence did not outweigh the aggravating circumstances that he sought out, beat, strangled, and killed a woman who had testified against him, and then cut the throats of her daughter and another witness).

We also note that in every case on which Sutton relies, the federal court weighed the mitigating and aggravating circumstances de novo, rather than evaluated the reasonableness of the state court's weighing; thus they are inappropriate comparisons. See Wiggins, 539 U.S. at 516-18, 534, 536-38, 123 S.Ct. 2527 (weighing de novo because no state court addressed prejudice); Williams, 529 U.S. at 398, 120 S.Ct. 1495 (weighing de novo because the state court did not apply the correct legal rule and did not consider all of the mitigating evidence, see Magana v. Hofbauer, 263 F.3d 542, 551 (6th Cir.2001)); Harries v. Bell, 417 F.3d 631, 634, 640 (6th Cir.2005) (weighing de novo because the habeas petition was filed before AEDPA's effective date); Coleman v. Mitchell, 268 F.3d 417, 427, 452 (6th Cir.2001) (same); Carter, 218 F.3d at 591, 593, 600 (same).

Nor can we say that it was unreasonable to conclude that the totality of the proffered mitigating evidence—the Burchetts' testimony, all of the evidence about the violence in Tennessee's prisons, and the evidence of Sutton's troubled background—did not create a reasonable probability of a life sentence. The net mitigating value of all of this evidence is too low, and the aggravating circumstances are too strong.

VII. Conclusion

We AFFIRM the district court's denial of habeas relief.

MARTHA CRAIG DAUGHTREY, Circuit Judge, concurring.

I concur in the reasoning and the result in Judge Boggs's opinion but write separately to address the admissibility of Sutton's prior criminal offenses. This issue is key, in my judgment, as it relates to the prejudice prong of the Strickland analysis necessitated by Sutton's claim of ineffective assistance of counsel. The dissenting opinion reads Judge Boggs's opinion as "conced[ing]" that the elicitation of "testimony about Sutton's past crimes . . . is an unlikely possibility." This interpretation is wrong in two respects, at different levels of significance.

First, on a superficial level, the term "concedes" overstates Judge Boggs's rather tentative1 observation that Sutton's criminal record should, or perhaps could, be "discounted by a relatively low probability of admission." Second, and of much greater importance, is the fact that in the peculiar circumstances of the trial in this case, the admissibility of Sutton's prior history would almost certainly have been admissible to rebut much of the proposed mitigation evidence introduced at the evidentiary hearing in district court.

True, there is a line of cases in the Tennessee law of sentencing that has a somewhat limiting effect on the scope of negative evidence admissible in the penalty phase of a capital case. It traces back to the Tennessee Supreme Court's opinion in Cozzolino v. State, 584 S.W.2d 765 (Tenn. 1979), one of the first death-penalty cases reviewed on appeal after the reinstatement of the death penalty in 1977. The Tennessee Supreme Court held that the trial judge erred during the sentencing hearing in permitting the prosecutor to introduce evidence during its case-in-chief that Cozzolino had committed several armed robberies subsequent to the murder for which he was on trial. The state had succeeded in convincing the trial judge that the evidence in question was relevant under (then) Tenn.Code Ann. § 39-13-204(c):

The Cozzolino court rejected the state's broad reading of the statute, instead holding more narrowly that:

Cozzolino, 584 S.W.2d at 768. Hence, the court concluded, "evidence is relevant to the punishment, and thus admissible, only if it is relevant to an aggravating circumstance, or to a mitigating factor raised by the defendant." Id. As a result, the court held, it was error to permit the introduction of the evidence to rebut mitigating circumstances that had not yet been offered by the defendant, for "[o]ne cannot rebut a proposition that has not been advanced." Id. Noting that this error "might have been made harmless by the later introduction of evidence to which the State's proof of subsequent crimes was relevant in rebuttal," the court offered this observation:

Id.

In this case, the state post-conviction courts at both the trial and appeal levels applied state law as summarized above and concluded that Sutton did not establish that he was prejudiced by his counsel's failure to delve into and present evidence of Sutton's troubled childhood. They did so in a principled and thorough fashion. See Nicholas Todd Sutton v. State of Tennessee, No. 03C01-9702-CR-00067, 1999 WL 423005 (Tenn.Crim.App. June 25, 1999). Yet, the dissent characterizes the state courts' application of Strickland to the facts of this case as unreasonable without taking into account their analysis of the issue. This despite the United States Supreme Court's admonition in Bell v. Cone, 543 U.S. 447, 455, 125 S.Ct. 847, 160 L.Ed.2d 881 (2005), that a federal court should not "presume . . . lightly that a state court failed to apply its own law [in a reasonable fashion]." This despite the fact that 28 U.S.C. § 2254(d) dictates a "highly deferential standard for evaluating state-court rulings" and "demands that state-court decisions be given the benefit of the doubt." Id. (internal citations and quotation marks omitted). And, this despite cautionary advice in Williams v. Taylor, 529 U.S. 362, 411, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000), that "a federal court may not issue the writ simply because the court concludes in its independent judgment that the relevant state court opinion applied clearly established federal law erroneously or incorrectly."

In this case, too, it behooves us to recall Strickland's mandate that we "must indulge a `strong presumption' that counsel's conduct falls within the wide range of reasonable professional assistance because it is all too easy to conclude that a particular act or omission of counsel was unreasonable in the harsh light of hindsight." Strickland v. Washington, 466 U.S. 668, 689, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). If defense counsel had more fully investigated Sutton's background and presented the defense that Sutton now says should have been put before the jury, it is difficult to see how he could have prevented the state from introducing devastating rebuttal evidence. It appears, for example, that Sutton's trial attorney managed to keep from the jury the underlying details of Sutton's prior conviction for a crime of violence, one of the three aggravating circumstances established at the sentencing hearing. But if counsel had sought to introduce testimony about Sutton's "troubled childhood," the state undoubtedly would have been successful in drawing attention to the fact that this aggravating circumstance involved Sutton's murder of his grandmother, the same grandmother who had raised him almost from birth and was apparently the only mother figure Sutton ever had. The state might also have been successful in presenting testimony establishing the motive for that murder, which was Sutton's response to his grandmother's negative reaction to learning that he had admitted killing two other people in North Carolina.

The dissent is critical of Judge Boggs's analysis of the probable result of the jury's deliberations had the jury been able to consider additional evidence of Sutton's troubled childhood. But weighed against information that Sutton's conviction for the Estep murder was, in fact, the fourth murder for which he had been found guilty, it seems almost preposterous to think that even one member of the jury would have held out against the death penalty. The same is true of the mitigation evidence offered in the district court to demonstrate Sutton's need to resort to violence for self-protection, based on the prison conditions that existed at the time of Estep's death. In rebuttal, the state would surely have argued that prison conditions could not similarly excuse the commission of Sutton's three prior murders, none of which occurred while he was incarcerated. All in all, whatever one thinks of the level of the trial attorney's professional competence, it is remarkable that he was able, for the most part, to avoid having the underlying details of his client's criminal record put before the jury. It is this absence of prejudice that convinces me that Sutton cannot succeed on his claim of ineffective assistance of counsel.

BOYCE F. MARTIN, JR., Circuit Judge, dissenting.

Nicholas Sutton's childhood was horrific. The undisputed facts elicited at his habeas hearing in the district court from a licensed clinical psychologist who had evaluated Sutton, Dr. Gillian Blair, showed "an unstable, often violent and threatening home life where the supervision and structure were inadequate." His brutal, mentally-ill father held Sutton and his mother at gun point during a stand-off with the police. When Sutton's father later died of hypothermia and exposure while Sutton was a child, the death was never explained to him. Sutton was also abandoned by his mother before the age of one and by his maternal grandparents at the age of two. His paternal grandfather died when Sutton was only seven or eight and he was raised by his paternal grandmother alone. He was shot in the eye at the age of nine, suffered several head injuries during his teenage years and was shot in the knee at sixteen. By the time he was an adolescent, he used a "wide variety of drugs" and sold drugs to earn money. He was sent to live with his aunt and uncle in Knoxville for high school because of his juvenile problems and drug abuse, but his lack of an education was not addressed, and he dropped out of high school during the eleventh grade. Though he joined the Navy at the age of seventeen, he was unable to adjust to military life because he was overwhelmed by the training and could not cope with the emotional pressure. Shortly after enlisting, Sutton received an honorable discharge. Dr. Blair diagnosed Sutton with Axis II personality disorder.

Sutton's trial counsel did not present any of this evidence at the penalty phase of Sutton's trial—not because he made a tactical trial strategy decision that the evidence would be unhelpful or would, as the state courts mused, potentially open the door to introduction of other damaging evidence, but because trial counsel simply did not deign to ask his client. A thorough inquiry into a client's childhood and background is high on an attorney's list of things to do when defending a capital case, along with "show up," "wear a suit," and "stay awake." Sutton's counsel's failure to make this basic inquiry constitutes ineffective assistance of counsel per se. See Wiggins v. Smith, 539 U.S. 510, 123 S.Ct. 2527, 156 L.Ed.2d 471 (2003) (due to minimal investigation, counsel presented no evidence of defendant's family history, which included severe childhood abuse); Hamblin v. Mitchell, 354 F.3d 482 (6th Cir. 2003) (counsel failed to seek mitigating evidence and thus did not learn of defendant's unpleasant childhood); Frazier v. Huffman, 343 F.3d 780 (6th Cir.2003) (counsel presented no mitigating evidence except defendant's one-sentence statement).

Thus, I turn to the second Strickland prong, whether counsel's deficient performance prejudiced Sutton. Stated differently, had counsel introduced evidence of Sutton's troubled upbringing and argued the evidence in mitigation of imposition of the death penalty, is it reasonably possible that the jury would have imposed life in prison instead of lethal injection? I believe so. As the Supreme Court observed in Williams v. Taylor, 529 U.S. 362, 398, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000): "Mitigating evidence unrelated to dangerousness may alter the jury's selection of penalty, even if it does not undermine. . . the prosecution's death-eligibility case." In Williams, the Court recognized that "the reality that [the defendant] was `borderline mentally retarded[ ]' might well have influenced the jury's appraisal of his moral culpability." Id.

The same is true here. Had the jurors been confronted with the mitigating evidence of Sutton's extremely troubled childhood, the probability that at least one juror would not have decided that the aggravating circumstances of the case outweighed the mitigating circumstances beyond a reasonable doubt "is a probability sufficient to undermine confidence in the outcome." Strickland, 466 U.S. at 694, 104 S.Ct. 2052; see also Wiggins, 539 U.S. at 535, 123 S.Ct. 2527 ("Had the jury been able to place petitioner's excruciating life history on the mitigating side of the scale, there is a reasonable probability that at least one juror would have struck a different balance."); Frazier, 343 F.3d at 798-99; cf. Tenn.Code Ann. § 39-13-204(i) (providing that "[n]o death penalty or sentence of imprisonment for life without possibility of parole shall be imposed but upon a unanimous finding that the state has proven beyond a reasonable doubt the existence of one (1) or more of the statutory aggravating circumstances").1 And, because I believe that it is reasonably possible that the jurors would not have unanimously agreed to impose the death penalty upon Sutton had they heard all of the evidence, I would find that the violation of Sutton's right to effective assistance of counsel at the penalty phase of his trial was prejudicial, and that the state court's conclusion to the contrary is an unreasonable application of clearly established Supreme Court precedent. I would, therefore, remand this case for resentencing so Sutton may present evidence of his upbringing to a jury, or at least receive competent advice of counsel as to whether to present the evidence. Cf., e.g., Wiggins, 539 U.S. at 534, 123 S.Ct. 2527; Hamblin, 354 F.3d at 494; Frazier, 343 F.3d at 798.

The majority, of course, disagrees. It attempts to reconstruct the sentencing phase and jury deliberations, assuming the introduction of Sutton's appalling childhood and background and speculatively admissible evidence of Sutton's prior crimes, to show that Sutton was not prejudiced by his counsel's deficient performance. To accomplish this feat, the majority employs an impressively sterile, faux-mathematical dissection that, as best I can follow, involves subtracting an ethereal "net mitigating value" of this evidence from the aggravating circumstances and dividing by the square root of maybe. After performing this devil's arithmetic,2 the majority concludes that Sutton was not prejudiced.

Now, I fully believe that judges are ordinarily good at looking back to determine when an error in a trial court was prejudicial to the outcome, at least with a greater likelihood of accuracy than a layman. And I fully concede that for our appellate and collateral review systems to work, our better-than-average accuracy rate is sufficient for most cases. The fact remains, however, that an appointment to the federal bench affords judges a black robe, not a crystal ball. Capital cases, and especially the sentencing phase of capital cases, are not "most cases" and better than average just will not do to accomplish Strickland's overarching goal of ensuring confidence in the outcome when the penalty for getting it wrong is the ultimate price.3

The majority's analysis on this point reveals just how inadequate our look-back abilities are in death penalty sentencing cases. Its attempt to recreate what would have happened during the penalty phase and then what would have happened during jury deliberations is certainly plausible, but not utterly convincing. One could easily imagine things unfolding much differently had the evidence of Sutton's upbringing been introduced: Defense counsel introduces evidence of Sutton's terrible upbringing—his constantly changing home life, a violent, mentally unstable father, having been shot at least twice before the age of eighteen, several head injuries, extensive drug use during his teens, and any other evidence a diligent investigation uncovers—and brings in a psychiatrist to testify as to how these childhood experiences contributed to Sutton's violent tendencies as an adult. Even if this opened the door to allow the prosecutor to elicit testimony about Sutton's past crimes (which the majority concedes is an unlikely possibility), including beating his grandmother to death after she found out that Sutton had killed two other people, defense counsel could plausibly have argued that Sutton's violent nature derives entirely from his horrific childhood. In other words, both his prior violence and the instant murder stem from the same rotted roots. Moreover, the evidence suggests, and counsel could have argued, that Sutton has tried to better himself. He tried to join the Navy, but could not adjust to the emotional pressure of the training, no doubt due to his early years, and successfully obtained his GED while in prison.

Would this line of argument have changed the minds of all twelve jurors? I cannot say for certain, but in this case Sutton only needed one: If one juror had been affected by the story of Sutton's past, Sutton would not have been sentenced to death.4

Furthermore, because this background information appeals to the emotions of the jurors and at least tries to paint a picture of Sutton as a human being, it strikes me as a more persuasive prediction of jury deliberations than the majority's effort to analogize capital sentencing jury deliberations to impassive algebraic gymnastics. As an attempt to persuade the reader that Sutton suffered no prejudice, the majority's macabre mathematics do not recognize how emotional the decision between life and death is for a juror, and for that reason fails. The decision whether to sentence someone to death is highly emotional. This is a matter of common sense, and it is well supported by the literature. See, e.g., Scott E. Sundby, A Life and Death Decision: A Jury Weighs the Death Penalty 177 (2005) ("However one feels about the death penalty, the jurors' stories lead to one indisputable conclusion: At bottom, a jury's effort to decide between life and death is a distinctly human endeavor infused with emotion and moral judgment. Despite the efforts of legislatures and courts to make the death penalty decision a legal judgment that is reached by following a series of rules, in the end the determination of whether the defendant lives or dies results unavoidably from the intersection of twelve jurors' individual beliefs and views.");5 Terry Maroney, Emotional Common Sense as Common Law, 62 Van. L.Rev. 851, 913 (2009) (noting that the Supreme "Court can openly defend its judgments about emotion as normative ones . . . where there is general consensus on the nature and prevalence of the emotional phenomenon but disagreement on the weight it should be accorded in the jurisprudential calculus"). It stands to reason that at least one juror would have had a much harder time making the emotional leap to condemn a man to die had he or she known who Sutton was and how he became the man that he or she had just convicted of murder.

As I said, though I believe it to be the case, I cannot say for certain that my attempt to reconstruct events is more accurate than the majority's. Nor can the majority honestly say that its reconstruction is more likely accurate than mine. However, one thing is certain—counsel's failure to obtain and present to the jury evidence regarding Sutton's awful early life robbed Sutton of his clearly established right to show himself as a human being in the jury's eyes and made easier what should be the most difficult decision a jury can make. This is precisely the potential to undermine confidence in the outcome that Strickland stands to protect defendants and the courts against. I would therefore reverse and remand for resentencing with competent defense counsel.