James T. Foster v. James M. Schomig, Cross-Appellee , 223 F.3d 626 ( 2000 )

MANION, Circuit Judge.

James Foster was convicted by an Illinois jury of murdering his girlfriend by beating her with a baseball bat, and was sentenced to death because he also sexually assaulted her by forcing pieces of the broken bat into her rectum. The Illinois Supreme Court affirmed and denied collateral relief. He brought this Section 2254 motion to argue that his counsel were ineffective for, among other things, failing to call a psychiatrist to testify during the sentencing phase of his trial. He asserts that had such an expert been called, he would have informed the court that Foster suffered from an extreme emotional disturbance, which is a mitigating factor under Illinois law. The district court accepted this argument and granted Foster’s petition. The State appeals, arguing that the district court erred in finding that Foster’s attorneys were ineffective. Foster cross-appeals, arguing that the district court erred in holding that the Antiterrorism and Effective Death Penalty Act (AEDPA) was applicable to this case and in finding that one of his attorneys was not ineffective for telling- the jury that Foster “killed the woman he loved.” Because the Illinois Supreme Court’s decision was not contrary to clearly established law, we reverse the district court in part and affirm in part.

I.

In 1985, James Foster was having an adulterous relationship with Jacqueline Simmons. Despite Foster’s own infidelity to his wife, he became angry at Simmons when he suspected she was unfaithful to him. So on the night of January 9, 1985, Foster went to confront her at an apartment she shared with her roommate — • Theresa Williams. He began yelling at Simmons and ordered her to disrobe in front of him so that he could look for signs of her involvement with other men. Screaming and crying, Simmons complied. Foster then threw her to the floor and proceeded to hit and kick her repeatedly, all the while ignoring her pleas to stop. He then picked her up by her hair with such force that tufts of it were torn out. After taking a short break, Foster proceeded to beat Simmons with a baseball bat while calling her a “bitch” and a “whore” and accusing her of messing around” on him. After the bat broke, Foster forced part of it into Simmons’s rectum. During the attack, Simmons’s young child and two other children were sleeping in another room of the apartment. Williams, Williams’s boyfriend, and a friend of Foster were also in an adjoining room during the beating. From there they could hear Simmons’s screams and they periodically witnessed portions of the two-hour attack. They were also the audience for Foster’s vaunts about how far he had inserted the bat into Simmons’s rectum. He showed them a piece of the bat and said “look at how much the bitch took.”

After hearing Foster’s boasts, Williams went to assist Simmons and found her breathing irregularly and suffering a seizure. Although Williams told Foster that Simmons desperately needed medical attention, Foster responded that Simmons was drunk and she just needed to sleep it bff. Foster then departed for a local tavern; when he returned about twenty minutes later, Simmons was dead. Not surprisingly, the coroner found that both of Simmons’s lungs were collapsed and that she suffered blunt trauma injuries on her head, chest, abdomen, and legs, as well as lacerations on her anus and liver. He determined that her death was caused by multiple blunt trauma, which resulted in swelling of the brain and internal hemorrhage of the liver. Foster eventually confessed to beating Simmons with a baseball bat, which made the jury’s task at the guilt phase of his trial easier. On June 7, 1985, after hearing witnesses describe the details of the crime and reading Foster’s written confession, the jury convicted him of murder. The State asked for the death penalty because the victim was killed in the course of an aggravated sexual assault. See 720 ILCS § 5/9-l(b)(6)(e). Because no judge in Kane County had ever sentenced a defendant to death, Foster and his attorneys decided to have the judge, rather than a jury, determine his penalty.

Foster was represented at trial and sentencing by attorneys Frank Giampoli, a part-time, Assistant Kane County Public Defender, and George Chabalewski, a full-time Public Defender. After the jury convicted Foster, his attorneys contacted a psychiatrist, Dr. Lyle H. Rossiter, Jr., to evaluate whether Foster suffered from any psychological condition which might prevent the imposition of the death penalty. Rossiter examined Foster on June 8 and June 19, and submitted a full report on June 21, 1985. The report states -that Foster had a history of serious head injuries and substance abuse and that he suffers from an antisocial personality disorder. While some aspects of Dr. Rossiter’s report indicated the presence of mitigating factors, it also candidly stated that Foster had “pronounced tendencies toward explosive anger and aggressive physical outbursts.”

Believing that Dr. Rossiter’s testimony would do more harm than good, and after seeing the prosecutors effectively cross-examine some of the other defense witnesses, Foster’s attorneys decided not to present Dr. Rossiter’s testimony. Instead, they called various family members and friends who characterized Foster as a generous, hard-working, religious, family man. The mother (not his wife) of two of his six children testified that his generosity occasionally entailed the provision of financial support for the two children. The defense also tried to show that one shooting incident in which Foster was purportedly involved was simply a case of mistaken identity. Foster’s young son also testified that his father played basketball with him, exhorted him to lead a good life, and all in all was an excellent father. Other family members and friends testified that he was generous with his time and would help sick relatives and repair automobiles for free.

Of course, the State also presented its own witnesses to show the presence of aggravating factors. They described Foster’s previous convictions for battery, armed robbery, theft, and illegally possessing firearms. One witness testified that Foster and his comrades gang-raped him while they were in jail. Another witness described how Foster beat and raped her at gunpoint when she was fifteen years old and how when he was finished with her Foster ordered her to urinate on herself. Other witnesses described an incident similar to the beating of Simmons. Foster and his friends tied up a woman and repeatedly beat her with a stick, after which Foster picked her up and dropped her on her face. The judge also heard that Foster once stabbed a police officer with a concealed knife while he was in police custody. On another occasion, Foster forcibly entered the home of one family by blowing apart their door with a shotgun, the blast from which wounded one of the occupants. The prosecution closed its case by describing the crime and asking the court to remember Foster’s lack of remorse and the pain suffered by Jacqueline Simmons..

Judge, I ask you when you go back into chambers to consider this, to look, again, at the photograph of this woman. This is an attractive young girl who was turned into this mess of mush with a broken jaw and a torn liver.

And what does he do when he is finished with her? As she’s lying there dying, he brags about it. This is a good time for him. He revels. He revels in the torture that he inflicted on this woman. He takes the bat, holds it up to the people who are in the next room, and with excrement all over it, he says “Look how much the bitch took up her ass.” And then, he goes to a local bar for a drink because to him this is a reason to party. This is a good time.

Sent. Tr. at 2540-41.

After considering all of the evidence, the judge determined that Foster merited the death penalty and set his execution date for September 18, 1985. The Illinois Supreme Court affirmed his conviction and sentence on direct appeal. People v. Foster, 119 Ill.2d 69, 115 Ill.Dec. 557, 518 N.E.2d 82 (1987). Later, it also denied the relief which Foster sought under the Illinois PosNConviction Hearing Act. People v. Foster, 168 Ill.2d 465, 214 Ill.Dec. 244, 660 N.E.2d 951 (1995).

Foster subsequently filed the present petition which alleges that his counsel were ineffective for failing to call Dr. Rossiter as a witness and for admitting during closing argument of the guilt phase that Foster killed Simmons. The district court succinctly dispatched Foster’s complaint about the closing argument because Foster showed neither deficient performance nor prejudice. But the district court elected to hold an evidentiary hearing on the ineffective assistance at sentencing claim. The court heard testimony from attorneys Giampoli, Chabalewski, and Dr. Rossiter. Although Dr. Rossiter’s 1985 report makes no mention of the term “extreme emotional disturbance,¹” he testified at the evidentiary hearing that had he been called as a witness at the sentencing hearing, he would have told the court that Foster suffered from such a disorder. But his testimony also indicated that, when effectively cross-examined, he would have revealed that Foster would exploit others for his own personal gain, would rationalize brutality, and was capable of inflicting great harm without feeling any remorse.

Chabalewski and Giampoli testified that they were both experienced litigators at the time of Foster’s trial and had both used Dr. Rossiter’s expert testimony in other cases. They testified that they made a tactical decision not to present Dr. Rossiter as a witness because they feared that his assessment of Foster’s violent proclivities and inability to feel remorse might harm Foster’s chances of escaping the death penalty. Despite this testimony, the district court found that Foster’s counsel were constitutionally ineffective at sentencing. It therefore decided to grant the writ unless Foster was afforded a new sentencing hearing. By its own terms, the district court’s order was stayed pending appeal. United States ex rel. Foster v. Gilmore, 35 F.Supp.2d 626, 633 (N.D.Ill.1999).

II.

To prevail on his Section 2264 claims, Foster must show that a decision by the Illinois state courts “was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States,” or “was based on an unreasonable determination of the facts in the light of the evidence presented in the State court proceeding.” 28 U.S.C. § 2254(d)(1) & (2);² see Williams v. Taylor, 529 U.S. 362, 120 S.Ct. 1495, 1521, 146 L.Ed.2d 389 (2000). We review these questions de novo. Schaff v. Snyder, 190 F.3d 513, 522 (7th Cir.1999). Findings of fact made by the state courts are presumed to be correct and this presumption may be rebutted only by clear and convincing evidence. 28 U.S.C. § 2254(e)(1). In ineffective assistance of counsel claims, the habeas petitioner must show that his attorneys’ performance was deficient and that then-errors caused him prejudice. Strickland v. Washington, 466 U.S. 668, 687-88, 692, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984).

Foster argues here, as he did before the Illinois Supreme Court and the district court, that his attorneys acted incompetently by declining to call Dr. Rossiter to testify that Foster suffered from an extreme emotional disturbance at the time he killed Simmons.

In assessing the performance of counsel, “a court must indulge a strong presumption that counsel’s conduct falls within the wide range of reasonable professional assistance; that is, the defendant must overcome the presumption that, under the circumstances, the challenged action ‘might be considered sound trial strategy.’” Strickland, 466 U.S. at 689, 104 S.Ct. 2052. Keeping this presumption in mind, our task is to reconstruct the circumstances surrounding the challenged conduct and determine whether they were reasonable. This determination must not be distorted with hindsight. Rather we must evaluate the conduct from counsel’s perspective at the time the relevant decision was made. Id. A defense attorney’s performance is acceptable when he chooses a professionally competent strategy that secures for the accused the benefit of an adversarial process. Kokoraleis v. Gilmore, 131 F.3d 692, 696 (7th Cir.1997). If the attorneys’ strategic decisions were sound when they were made, these decisions cannot support a claim of ineffective assistance. Certainly the decision not to call a witness to testify can be a strategic decision. United States v. Kozinski, 16 F.3d 795, 813 (7th Cir.1994). Such a decision is sound “if it is based on the attorney’s determination that the testimony the witnesses would give might on balance harm rather than help the defendant.” Hall v. Washington, 106 F.3d 742, 749 (7th Cir.1997).

The Illinois Supreme Court found that Giampoli and Chabalewski’s decision not to call Dr. Rossiter was “a matter of trial strategy” which they concluded made sense at the time the decision was made. 660 N.E.2d at 963. The district court, on the other hand, concluded that Dr. Rossi-ter’s testimony would have demonstrated the statutory mitigating factor of extreme mental or emotional disturbance, that Dr. Rossiter was highly credible, and that his testimony was effective and persuasive. 35 F.Supp.2d at 631. It believed that the failure to present Dr. Rossiter as a witness deprived the sentencing judge of valuable testimony which would have necessarily altered the judge’s findings. The court concluded that the failure to present Dr. Rossiter’s testimony “rises to the level of constitutionally ineffective assistance of counsel.” Id. at 633.

In arriving at his conclusion, the court referred to the sentencing transcript where the trial judge stated that he could not find any evidence that indicated or showed him that the murder was committed while the defendant was under the influence of extreme mental or emotional disturbance. “There just isn’t any evidence of that factor.” The district court twice repeated the “there just isn’t any evidence” statement to emphasize trial counsel’s deficiency in presenting mitigating evidence. But a review of the sentencing transcript shows that this statement was made in the process of the court methodically going through the five mitigating factors set out in the Illinois statute.³ The first factor to be considered was prior criminal activity. The record was replete with evidence of Foster’s prior criminal activity, so there was obviously no mitigation there. The second factor concerned whether the defendant was “under the influence of extreme mental or emotional disturbance.... ” The sentencing judge said that he could not find any evidence which indicated or showed him that the murder was committed while the defendant was under such an influence. This is where he said “There just isn’t any evidence of that factor.” Contrary to the district court’s characterization, the sentencing judge wasn’t distressed by the fact that none was presented, nor did he question whether there was any possibility of presenting such evidence. As we discuss below, there is a clear reason for this. He then went on with the next three mitigating factors making the same cryptic statement that there was no evidence indicating that the defendant qualified for any of them.

Unfortunately, the hearing before the district court, some thirteen years after the trial, left something to be desired. Although both defense attorneys testified, neither of them had taken the opportunity to review the sentencing record, which included extensive discussion among the attorneys and the sentencing judge. Dr. Rossiter did have an opportunity to at least review his report. But all three witnesses' — defense attorneys Giampoli and Chabalewski, and Dr. Rossiter — had only vague recollections of what took place pri- or to and during the sentencing hearings. But even with the witnesses’ diminished recollection, the record of the hearing before the district court supports the conclusion of the Illinois Supreme Court that the decision not to call Dr. Rossiter was a matter of reasonable trial strategy. Both Chabalewski and Giampoli testified that they had worked with Dr. Rossiter before and thus knew how he would testify on cross-examination. They both carefully considered presenting Dr. Rossiter’s testimony at the sentencing hearing but thought better of it when they learned that he would describe Foster as remorseless and prone to violence.

Q. Whose decision was it not to call Dr. Rossiter?

Giampoli: I think we both thought long and hard about it and felt that his testimony would hurt more than help.

Evid. Hear. Tr. at 44. Chabalewski’s recollection was consistent with Giampoli’s.

Q. Why did you fail to use Dr. Rossi-ter’s report?

Chabalewski: My recollection is that on balance, it hurt us as much as it helped us and gave, I think, as much reason for imposition of the death penalty as not.

Q. And precisely what portions of the report did you believe were more harmful to the case than helpful?

Chabalewski: The one thing that I recall is I think the doctor’s assessment that Mr. Foster was a sociopath.

* * *

Q. And in your opinion, after going through Dr. Rossiter’s report, you felt that it would be more harmful' — harmful to the defendant than beneficial at the sentencing stage, am I correct?

Chabalewski: Correct.

Q. And that’s why you didn’t call Dr. Rossiter?

Chabalewski: Correct.

Evid. Hear. Tr. at 92,106.

At the sentencing hearing, the state put on extensive evidence of Foster’s prior criminal activity. In mitigation, the defense presented witnesses who disputed some of the facts regarding the criminal activity referred to by the state. But the bulk of the mitigating evidence involved testimony of family and friends regarding Foster’s good works, his helpfulness to family members and friends in need, and his role as a father. The hearing was then continued because there was some question whether Dr. Rossiter would testify. At a point in the hearing the court stated to the defense attorney,

As I understand your comments this morning, you are saying that you still, as you stand here now, intend to go forward with the medical — or, with the psychological evaluation, to present that as evidence in mitigation.

Accordingly, I will then permit the state to move forward with their examination.

Sent. Tr. at 2518-19.

At the hearing on June 25, 1985, the court learned that the defense would not be calling Dr. Rossiter to testify. After the attorneys discussed several documents that would or would not come into evidence, the prosecutor stated to the court:

I have one other matter that I would like to put on record in regard to the expert testimony that the defense was seeking to procure in this case.

Our office talked personally to Dr. Rossiter, and also, received a report from him indicating that he had examined the defendant.

It’s also come to our attention that various medical tests were performed on the defendant at Community Hospital as part of the evaluating process by Dr. Rossiter.

We lined up a witness to come to court today in rebuttal.

Since the defendant indicated they would not call Dr. Rossiter, obviously, we have called off our witness.

We were ready to proceed if, in fact, the defense wanted to call Dr. Rossiter.

What I’m indicating in summary, Judge, is that all of the tests that the defense sought to have performed on the defendant were, in fact, carried out. And it was a strategy decision on the part of the defense not to call Dr. Rossi-ter. He was available, and he did com-píete all of the testing that they were seeking.

Sent. Tr. at 2534-35.

In response, attorney Giampoli stated:

Your Honor, basically, the only comment attacking that comment is that this Court cannot infer that those reports would have a negative or positive effect because, obviously, we just decided we didn’t want to call the expert.

The Court: I understand. I was just about to stop the defendant if any comment was made to what the tests may or may not indicate because, as far as I am concerned, that is the choice you people made.

I think the record does reflect that you did have the opportunity to have the defendant evaluated and tests performed and things of that nature, and that’s' — and that’s as far as it goes.

Sent. Tr. at 2535-36.

Given this exchange, it is readily apparent that the decision not to call Dr. Rossiter was a thoughtful and strategic one and that the prosecution was not only going to bring in its own expert psychiatric witness in the event Dr. Rossiter testified, but there were also certain documents that would reveal the results of a number of medical tests. Without saying why, the defense attorneys emphasized to the court that it should not infer that those reports would have a negative or positive effect. This further demonstrates that defense counsel had every opportunity to review those reports, and in view of the opposing expert that would have been called, made a strategic decision not to call Dr. Rossiter. Given the context of the actual hearing, and not based on what the witnesses could or could not recall thirteen years later, not calling Dr. Rossiter was entirely reasonable.⁴

The preclusion of whatever aggravating evidence the prosecution was marshaling was not the only thing that made the decision not to call Dr. Rossiter reasonable. As we have noted before in cases like this one, there is a strong possibility that the defendant’s mitigation evidence might turn out to be aggravating. See Emerson v. Gramley, 91 F.3d 898, 906 (7th Cir.1996). Dr. Rossiter’s preliminary report foretold enough aggravating factors that could offset whatever evidence of extreme emotional disturbance might have been mitigating. For instance, the report stated that Foster has “an antisocial personality disorder” and has “pronounced tendencies toward explosive anger and aggressive physical outbursts.” Because the report describes Foster in an unflattering light, to say the least, it would have caused any competent attorney to pause and consider how devastating this testimony could be to Foster’s case. From Dr. Rossiter’s testimony at the evidentiary hearing thirteen years later we further see how he would have described Foster if he had been called to testify at the sentencing phase.⁵ Dr. Rossiter stated:

I saw him [Foster] as a particularly self-centered individual, probably capable of exploiting others for his own personal gain without regard to their rights, let alone their sensibilities.

I think that he is capable of rationalizing a good deal of the brutality in this particular case. I think he was capable of denying to me or representing to me certain events in somewhat less than entirely candid detail.

[C]ertainly antisocial disorder, with it by definition lack of the usual constraints of conscience and the usual ideal that provide the standards to which one conforms conduct, that — that characteristic of the antisocial personality disorder would be a strong aggravating factor in one sense.

The irony of the situation, if I may, your Honor, is that irom a psychiatric point of view, an antisocial personality disorder is a very extreme developmental disorder. It’s a very defective development of the human personality. The absence of the constraints of conscience and the ideals that guide and constrain one’s behavior, that is an abnormality of human nature. So in itself, it constitutes, from a clinical point of view, a very severe aberration.

It is not typically regarded that way from a forensic legal point of view, of course, but the fact is that persons who have such disorders are very severely crippled in their ability to lead a normal life, to cooperate, to relate to other people in a normal manner.

Evid. Hear. Tr. at 180-81 (emphasis added). When cross-examined, Dr. Rossiter was forced to expand upon his description of Foster as a emorseless criminal.

Q. Okay. Dr. Rossiter, it is my understanding that an individual who suffers from an antisocial personality disorder, that it’s typical of that type of individual to have a criminal history, is that correct?

A. That’s correct.

Q. And you had testified that this starts with burglaries and purse snatch-ings and crimes of that nature, am I correct?

A. Correct.

Q. And that it’s not uncommon for an individual who suffers from an antisocial personality disorder to graduate into more violent crimes, am I correct?

A. That’s true.

Q. And in fact, Mr. Foster has a history of violent crimes?

A. That’s true.

Q. And he has the antisocial personality disorder, is that right?

A. That’s correct.

* * *

Q. It’s my understanding as well that people with an antisocial personality disorder are generally considered to be remorseless, is that fair?

A. Correct?

Q. Devious?

A. Yes.

Q. Dishonest?

A. Yes.

Q. Self-centered?

A. True.

Evid. Hear. Tr. at 172-73. When asked whether Foster might become increasingly violent, Dr. Rossiter conceded that this might happen, and in his deposition testimony he went so far as to say that Foster was predisposed to violence. Although Dr. Rossiter opined in his deposition that his report would have been more helpful to Foster’s case than harmful, he conceded that there were aggravating factors in his report, and that based on his twenty-five years of experience in criminal matters, “the diagnosis of antisocial personality disorder is usually taken in legal circles as a damning kind of finding because the individual is characterized as being predatory, malicious, unconscionable, and calculating in his criminal activities.”

As Dr. Rossiter emphasized several times, extreme emotional disturbance is a legal term, not a medical one. When the district court pressed him at the hearing, Dr. Rossiter acknowledged that he would discuss the pros (mitigating) and cons (aggravating) with defense counsel. Dr. Rossi-ter stated, “I call them as I see them. I try to inform the consulting — the attorney for whom I’m consulting of the potential two-edged sword of my testimony if there is one, because what I have to say may be mitigating or damning, depending on which dimension of the human nature under question I’m talking about.” Evid. Hear. Tr. at 182.⁶

This kind of testimony could have substantially damaged Foster’s case. Foster obviously would not have wanted the sentencing judge to hear an expert opinion that Foster’s psychological aberrations cause him to disregard the rights and feelings of others to such an extent that he would feel no remorse even after killing a young mother while her child was sleeping in the next room. Dr. Rossiter’s testimony risked designating Foster’s treatment of Jacqueline Simmons as something he would be inclined to do again. It would have suggested that Foster’s violent behavior was not the exception but the rule, making him a constant threat to prison guards or fellow inmates. Given the context in which Dr. Rossiter’s testimony would have been presented, it was a reasonable strategic decision for the defense attorneys not to expose Dr. Rossiter to the prosecution’s cross-examination, especially because it would have opened the door to the testimony of the government’s psychiatric expert.

Thus, we agree with the Illinois Supreme Court that the decision not to call Dr. Rossiter as a mitigation witness was a matter of sound strategy. Foster has failed to rebut the presumption that his counsel’s decision was based on a legitimate strategy or that his counsel’s performance fell below an objective standard of reasonableness.⁷

Even if we assume for the sake of the argument that his attorneys were incompetent in failing to present Rossiter’s testimony, Foster has failed to show that this caused him prejudice. To demonstrate prejudice, a defendant must show that “there is a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different. A reasonable probability is a probability sufficient to undermine confidence in the outcome.” Strickland, 466 U.S. at 694, 104 S.Ct. 2052. The prejudice inquiry “focuses on the question whether counsel’s deficient performance renders the trial unreliable or the proceeding fundamentally unfair.” Williams, 120 S.Ct. at 1513 n. 17. In the penalty phase of a capital case, to show prejudice the mov-ant must demonstrate that “a reasonable probability exists that, but for counsel’s substandard performance, the sentencer ‘would have concluded that the balance of aggravating and mitigating factors did not warrant death.’ ” See Hall, 106 F.3d at 751-52 (quoting Strickland, 466 U.S. at 695, 104 S.Ct. 2052).

The Illinois Supreme Court concluded that “[tjhere is no reasonable probability that the introduction of Dr. Rossiter’s testimony would have prompted the sentencing judge to find that there were mitigating circumstances sufficient to preclude imposition of the death penalty.” Foster, 214 Ill.Dec. 244, 660 N.E.2d at 963. Our review of the record leads us to agree with the Illinois Supreme Court. At best, Dr. Rossiter’s testimony would have helped Foster by suggesting that he was suffering from schizophrenia, was depressed at the time he killed Simmons, and that due to psychological factors beyond his control, he was -prone to violent outbursts. The presence of these factors caused Dr. Rossi-ter to believe (at least at the time of the hearing before the district court) that when he Wiled Simmons, Foster was under the influence of extreme mental or emotional disturbance. Assuming that the sentencing judge would have believed Dr. Rossiter, and would have disbelieved any psychiatric evidence the state presented to the contrary, Dr. Rossiter might have put some weight on the mitigation arm of the scale. We say “might have” because we have noted before that the presence of an emotional or mental disturbance can be seen by some not as a mitigating factor, but as an aggravating one. “Mitigation ... after all, may be in the eye of the beholder.” Burger v. Kemp, 483 U.S. 776, 794, 107 S.Ct. 3114, 97 L.Ed.2d 638 (1987). Sentencing judges “may not be impressed with the idea that to know the cause of viciousness is to excuse it; they may conclude instead that when violent behavior appears to be outside the defendant’s power of control, capital punishment is appropriate to incapacitate. Burris v. Parke, 130 F.3d 782, 784-85 (7th Cir.1997).

But even if Dr. Rossiter’s testimony definitely would have established the existence of a mitigating factor, the presence of one (or more than one) mitigating factor does not preclude the imposition of the death penalty. Rather, Illinois law requires the sentencing judge to consider all mitigating and aggravating factors, including those aggravating factors intentionally or unintentionally set forth by á “mitigation” witness. 720 ILCS § 5/9-l(c). Thus, the sentencing judge was obliged to consider those aspects of Dr. Rossiter’s testimony which spoke to Foster’s lack of rehabilitative potential, his propensity to commit violent acts, his inability to assimilate himself into society, his failure to accept responsibility for his crimes, and his lack of remorse. See People v. Ward, 187 Ill.2d 249, 240 Ill.Dec. 636, 718 N.E.2d 117, 127 (1999); People v. Shatner, 174 Ill.2d 133, 220 Ill.Dec. 346, 673 N.E.2d 258, 268 (1996); People v. Anderson, 284 Ill.App.3d 708, 220 Ill.Dec. 302, 672 N.E.2d 1314, 1319-20 (1996); People v. Kerkering, 283 Ill.App.3d 867, 219 Ill.Dec. 454, 671 N.E.2d 368, 372 (1996); People v. Moore, 250 Ill.App.3d 906, 189 Ill.Dec. 615, 620 N.E.2d 583, 589 (1993). The district court did not take into account these facets of Dr. Rossi-ter’s testimony when it concluded that the defense attorneys rendered ineffective assistance by not calling Dr. Rossiter to testify.

The district court believed that because the sentencing judge was required to consider whether Foster was emotionally disturbed, “it is axiomatic that the presentation of such evidence would have influenced the sentencing judge.” 35 F.Supp.2d at 632. No doubt it would have had some influence, but that influence could well have been negative.

But even if Dr. Rossiter’s testimony could be construed as having no aggravating elements, and even if it would not have opened the door to psychological evidence from the prosecution, in light of the other evidence it seems that Dr. Rossiter could hardly have established that Foster was under an extreme emotional disturbance at the time he murdered Simmons. As the Illinois Supreme Court pointed out, “An extreme emotional disturbance occurs when defendant’s emotional state at the time of the murder is ‘at such a fragile point as to leave him with little or no emotional control.’ ” 214 Ill.Dec. 244, 660 N.E.2d at 963 (quoting People v. Phillips, 127 Ill.2d 499, 131 Ill.Dec. 125, 538 N.E.2d 500, 514 (1989)). Apparently Foster had been looking for Jacqueline Simmons throughout the evening of January 9, 1985. After several unsuccessful attempts, he finally found her at the apartment later that night. The brutal beating with a baseball bat, witnessed intermittently by three adults, lasted nearly two hours. After showing off the bat handle that he had inserted in Simmons’s rectum, Foster left with a friend to visit a bar. On the way out he dismissed Theresa Williams’s warning that Simmons needed to go to the hospital and simply observed that “Simmons was drunk and they should let her sleep it off.” 518 N.E.2d at 85. When he returned about 20 minutes later, Simmons was dead. As the Illinois Supreme Court concluded:

Defendant exhibited rational and logical conduct on the night of the murder. Defendant spoke to the victim and the others in the apartment that night in a coherent and rational manner. He tried to resuscitate Simmons and removed evidence of the beating from the apartment. Defendant apparently expected the police to investigate and he concocted a story to tell them in order to protect himself. In light of all the evidence which the trial court considered at sentencing, including defendant’s history of criminal activity, defendant failed to demonstrate a reasonable probability that the sentencing judge .would not have imposed the death penalty if trial counsel had presented Dr. Rossiter’s testimony in an effort to show that defendant was under an extreme emotional disturbance at the time of the murder.

660 N.E.2d at 963-64.

We agree with the Illinois Supreme Court. Even if we were to conclude that counsel were ineffective, which we do not, Foster could not demonstrate prejudice. Because the totality of Dr. Rossiter’s testimony would not have provided a net benefit to Foster’s case, and instead would likely have harmed it, Foster has failed to show that there is a reasonable probability that but for Dr. Rossiter not testifying the sentencer would have concluded that the balance of aggravating and mitigating factors did not warrant death. See Hall, 106 F.3d at 751-52. Thus, he has failed to satisfy either the deficient performance or prejudice prong of Strickland.

Foster’s final argument is that Giampoli doomed his case by stating during closing argument of the guilt phase of the trial that Foster “killed the woman he loved.” The Illinois Supreme Court and the district court correctly determined that Foster failed to show either that the statement constituted deficient performance or that prejudice resulted from this remark. Obviously, Giampoli was attempting to direct the jury’s attention to the intent element, the element on which Foster hoped to prevail. Giampoli testified as much. It is reasonable to focus a jury’s attention on the element which provides the greatest chance of success and we previously held that this is a sound trial strategy. See Underwood v. Clark, 939 F.2d 473, 474 (7th Cir.1991); see also United States v. Wilks, 46 F.3d 640, 644 (7th Cir.1995) (conceding guilt on one count was a reasonable tactic because it lent credibility to counsel). So contrary to Foster’s assertions, merely conceding the presence of one element was not an admission of guilt and certainly was not unreasonable in this context. Also, because the jury had already heard testimony that Foster confessed to beating Simmons, had seen his written confession, heard uncontradicted testimony from witnesses to the beating, and heard the coroner testify as to the cause of Simmons’s death, Giampoli didn’t tell the jury anything it didn’t already know. This statement hardly could have caused prejudice. Therefore, we agree with the district court that the Illinois courts did not err in denying relief on this claim.

III.

Because Foster has not shown that his conviction or sentence was contrary to clearly established law or was an unreasonable application of law as set out in Strickland, he cannot prevail. Accordingly, we Affirm the district court’s denial of his Section 2254 motion on his ineffective assistance claim regarding his counsel’s closing argument during the guilt phase. We REVERSE the district court’s granting of his Section 2254 motion on his ineffective assistance claim with respect to the sentencing phase of his case.

1 . "Extreme emotional disturbance” is a legal term rather than a medical one. In assessing a defendant's fitness for the death penalty, Illinois law requires a sentencing judge to consider all aggravating and mitigating factors, including "whether the murder was committed while the defendant was under the influence of extreme mental or emotional dis-turbance_" 720 ILCS § 5/9-1 (c)(2). Under Illinois law, a "defendant is under the influence of an extreme emotional disturbance when the defendant’s emotional state at the time of the murder is at such a fragile point as to leave him or her with little to no emotional control.” People v. Evans, 186 Ill.2d 83, 237 Ill.Dec. 118, 708 N.E.2d 1158, 1166 (1999).

2 . Because Foster filed his petition on January 16, 1997, the AEDPA applies. Williams v. Taylor, 529 U.S. 420, 120 S.Ct. 1479, 1486, 146 L.Ed.2d 435 (2000). He argues that the date he sought appointment of counsel (September 1, 1995) should constitute the initiation date of his petition, but we have rejected this argument before and now do so again. Gosier v. Welborn, 175 F.3d 504, 506 (7th Cir.1999).

3 . 720 ILCS § 5/9— 1 (c) provides:

Mitigating factors may include but need not be limited to the following:

(1) the defendant has no significant history of prior criminal activity;

(2) the murder was committed while the defendant was under the influence of extreme mental or emotional disturbance, although not such as to constitute a defense to prosecution;

(3) the murdered individual was a participant in the defendant’s homicidal conduct or consented to the homicidal act;

(4) the defendant acted under the compulsion of threat or menace of the imminent infliction of death or great bodily harm;

(5) the defendant was not personally present during the commission of the act or acts causing death.

4 . Our review here includes not only the record of the hearing before the district court, but also the state court opinions and record, most significantly the record of the guilt and sentencing phases of the trial including the sentencing court's findings and conclusions. We certainly owe deference to the district court in its determination of witness credibility and the findings from the hearing on the habeas petition. This court has noted that in an appeal from a ruling on a petition for habeas relief, we review a district court's findings of fact for clear error and its rulings on issues of law de novo. Warren v. Richland County Circuit Court, 223 F.3d 454, 456-57 (7th Cir.2000); Gardner v. Barnett, 199 F.3d 915, 918 (7th Cir.1999). Claims of ineffective assistance involve mixed questions of law and fact which we review de novo. United States v. Shukri, 207 F.3d 412, 418 (7th Cir.2000); Hall, 106 F.3d at 748. Here, where the court based its ruling on findings from its own hearing combined with findings it gleaned from the cold record of the original sentencing transcript, we are inclined to apply the clearly erroneous standard. That said, we disagree with the finding that Foster’s counsel did not make a strategic decision regarding whether to call Dr. Rossiter as a witness. Because we conclude the court’s determination was unreasonable and contrary to the record, it can appropriately be called a clear error. Concrete Pipe and Products of Cal., Inc. v. Construction Laborers Pension Trust for S. Cal., 508 U.S. 602, 622, 113 S.Ct. 2264, 124 L.Ed.2d 539 (1993) ("A finding is 'clearly erroneous' when although there is evidence to support it, the reviewing [body] on the entire evidence is left with the definite and firm conviction that a mistake has been committed.").

5 . We consider Dr. Rossiter’s testimony only to the extent it indicates what he told Foster's attorneys concerning what his testimony would have been had he testified at the sentencing hearing. Considering anything else would be an impermissible use of hindsight. See Strickland, 466 U.S. at 689, 104 S.Ct. 2052.

6 . At the hearing he had no recollection one way or the other whether thirteen years earlier he talked to the defense attorneys about these pros and cons.

7 . One -might argue that defense counsel should have gambled on presenting Dr. Rossi-ter’s testimony, as Foster already had a dismal case and had little to lose. While it's true that the sentencing judge may have inferred from the depravity of the crime itself that Foster possessed some of the attributes which Dr. Rossiter ascribed to him, Dr. Rossiter would have provided an expert opinion that Foster was a fundamentally flawed individual whose characteristics included a predisposition to violence and an inability to feel remorse. Our review of the sentencing hearing transcript indicates that no other witness explicitly made these assertions.