In this death penalty case, we must decide whether the district court properly granted petitioner's request for relief under 28 U.S.C. Sec. 2254 on the ground that petitioner received ineffective assistance of counsel. For the reasons stated below, we conclude that the district court should have granted no relief.BACKGROUND

Rogers has been convicted of murdering his neighbor, Grace Perry, in Floyd County, Georgia, in 1980. Perry died when she was impaled with a rake handle. Rogers' first conviction for this crime was overturned by Georgia's Supreme Court. Rogers v. State, 250 Ga. 652, 300 S.E.2d 490 (1983). His second trial in June 1985 resulted in a conviction for murder of Perry and aggravated assault on another woman, Perry's housemate. At sentencing for the murder, the jury found the second and seventh statutory aggravating circumstances,1 O.C.G.A. Sec. 17-10-30(b)(2) & (7); and Rogers was sentenced to death.

There was evidence Rogers had been drinking beer on the day of the murder. On that day, Rogers, before Perry was killed, might have smoked a marijuana cigarette laced in angel dust (also known as PCP). Before trial, Rogers alternately denied and admitted the alcohol and drug use; and witnesses offered varying accounts of his sobriety on the evening of the murder. One police interview with Rogers evidenced that Rogers smoked the cigarette knowing it had been dipped in PCP.

Rogers' trial counsel were David Smith, who was appointed more than two years before the trial, and Kenneth Fuller, who was appointed four months before trial. Fuller had practiced in Floyd County since 1973. Smith, a lawyer since 1975, was a lifetime resident of Floyd County.

Counsel--after concluding that the Floyd County jury would not respond well to a defense based on voluntary ingestion by Rogers of drugs or alcohol or both--avoided relying on the intoxication issue at trial and sentencing. Counsel were aware that there was evidence of PCP use and that PCP was a hallucinogen. But they did not go further to investigate its psychological or behavioral effects.

On the question of guilt or innocence, defense counsel asserted at trial that the state's evidence had not proved Rogers was the person who had, in fact, killed Perry.2 Rogers' counsel did conduct an investigation about mitigating circumstances; and, at sentencing, counsel did present mitigating evidence that, in their opinion, was less risky than evidence advancing an argument on voluntary drug use. Various members of Rogers' family testified about his difficult upbringing. And, defense counsel made use of expert testimony: a psychologist testified that Rogers suffered from some brain dysfunction and low I.Q.

After challenging the conviction and sentence in state court,3 Rogers brought this federal habeas petition in 1990. The district court granted Rogers' habeas claim for ineffective assistance of counsel at sentencing, ordering Georgia to resentence Rogers. The district court decided that the failure to investigate and to present mitigating evidence on PCP's psychological effects rendered counsel ineffective. The district court denied relief on the conviction.

Here, Rogers appeals the denial of relief on the conviction and defends the district court's conclusion that he received ineffective assistance at sentencing.4 Rogers claims that the PCP evidence should have been investigated and introduced to "explain" his actions; the evidence, he argues, would not only have negated the specific state of mind for "malice murder," it would have been valuable mitigating evidence at sentencing. Because the "undiscovered" PCP evidence might have been introduced at trial to describe Rogers' state of mind at either stage (guilt/innocence and sentencing), the claims can be considered together.

"The benchmark for judging any claim of ineffectiveness must be whether counsel's conduct so undermined the proper functioning of the adversarial process that the trial cannot be relied on as having produced a just result." Strickland v. Washington, 466 U.S. 668, 686, 104 S. Ct. 2052, 2064, 80 L. Ed. 2d 674 (1984). Because ineffectiveness claims present mixed questions of law and fact, we exercise plenary review. Id. at 698, 104 S.Ct. at 2070.

A jury trial is, by its nature, an enterprise that is filled with imponderables from the viewpoint of a trial lawyer. It is an undertaking that calls not only on the lawyer's head, but also on his heart and nerve. At times in the trial arena, audacity or imagination or patience accomplish more than pure logic might suggest is possible. The truth is that it is often hard for even a good lawyer to know what to do. Trying cases is no exact science. And, as a result, we must never delude ourselves that the fair review of a trial lawyer's judgment and performance is an activity that allows for great precision or for a categorical approach. When reviewing whether an attorney is ineffective, courts "should always presume strongly that counsel's performance was reasonable and adequate." Atkins v. Singletary, 965 F.2d 952, 958 (11th Cir.1992). And, "a court should be highly deferential to those choices ... that are arguably dictated by a reasonable trial strategy." Devier v. Zant, 3 F.3d 1445, 1450 (11th Cir.1993).

Even if many reasonable lawyers would not have done as defense counsel did at trial, no relief can be granted on ineffectiveness grounds unless it is shown that no reasonable lawyer, in the circumstances, would have done so. This burden, which is petitioner's to bear, is and is supposed to be a heavy one. And, "[w]e are not interested in grading lawyers' performances; we are interested in whether the adversarial process at trial ... worked adequately." See White v. Singletary, 972 F.2d 1218, 1221 (11th Cir.1992). Therefore, the cases in which habeas petitioners can properly prevail on the ground of ineffective assistance of counsel are few and far between.

Rogers claims that Smith and Fuller were ineffective for deciding to present no evidence and argument about Rogers' PCP use when Smith and Fuller had not investigated PCP's effects on users. The Strickland decision indicates that a decision not to investigate a potential defense, like other litigation decisions, need only be reasonable to fall within the range of professionally competent assistance:

[S]trategic choices made after less than complete investigation are reasonable precisely to the extent that reasonable professional judgments support the limitations on investigation. In other words, counsel has a duty to make reasonable investigations or to make a reasonable decision that makes particular investigations unnecessary.

466 U.S. at 690-91, 104 S. Ct. at 2066 (emphasis added). Moreover, "a particular decision not to investigate must be directly assessed for reasonableness in all the circumstances, applying a heavy measure of deference to counsel's judgments." 466 U.S. at 691, 104 S.Ct. at 2066 (emphasis added).

The district court's opinion suggests that "thorough investigation" must precede all strategic decisions or the decision will probably be ineffective in the constitutional sense. Cf. Palmes v. Wainwright, 725 F.2d 1511, 1521 (11th Cir.1984) ("a strategic decision to pursue less than all plausible lines of defense will rarely, if ever, be deemed ineffective if counsel first adequately investigated the rejected alternatives"). But this view seems incorrect to us. By its nature, "strategy" can include a decision not to investigate. Strickland indicates clearly that the ineffectiveness question turns on whether the decision not to make a particular investigation was reasonable. 466 U.S. at 691, 104 S. Ct. at 2066. This correct approach toward investigation reflects the reality that lawyers do not enjoy the benefit of endless time, energy or financial resources. See Atkins, 965 F.2d at 959-60 (noting that "[a]t some point, a trial lawyer has done enough," and that "[a] lawyer can almost always do something more in every case"); Gates v. Zant, 863 F.2d 1492, 1498 (11th Cir.1989) (as long as decision was reasonable under circumstances, counsel may elect to forego line of defense without first investigating it substantially).

Sometimes, a lawyer can make a reasonable decision that no matter what an investigation might produce, he wants to steer clear of a certain course. In this case, counsel were longtime local lawyers who knew their community. And, they also investigated the persons on the pertinent jury list. Defense counsel knew, at least, four important things when they investigated no further about PCP: (1) they knew that there was some evidence of PCP use; (2) they knew that PCP was a hallucinogen; (3) they knew that a Floyd County jury would likely react hostilely to an assertion that a person should in some way be excused from the consequences of his acts because he had voluntarily taken drugs; and (4) they knew there were other possible defenses and mitigating factors that might be advanced. This information was enough to make going no further down the path toward a voluntary intoxication defense a reasonable step. Because declining to spend time and energy delving into Rogers' PCP use was a reasonable course, we hold that Rogers' counsel performed in a way that was constitutionally acceptable when they did not investigate and did not present evidence of PCP's effects.5

A sample of Smith's testimony at the state habeas hearing describes counsels' justifiable concern:

[We] are struggling with the definition of mitigating. We had been with that jury for quite awhile and I had lived in that county all of my life. I don't believe [the drug evidence] was going to be perceived by anybody as mitigating.... In fact, I felt it would have gone the other way.

Record Vol. 11, B4-63. Fuller testified similarly:

I don't believe that would have been mitigating. Our circuit is ... very conservative. And voluntary intoxication doesn't play well and I don't think voluntary hallucination does either.

Record Vol. 11, B4-177. These views are within the range of reasonable judgment for lawyers. No evidence in the record shows that the lawyers' worries about the counterproductive nature of a drug-based defense in Floyd County were unrealistic. Counsel could have reasonably believed that developing Rogers' drug use as a defense would have (1) damaged their own credibility as advocates of good sense before the Floyd County jury; (2) drawn attention away from other kinds of evidence and argument that the lawyers thought might be better received; and (3) at worst, been perceived by the jury as aggravating instead of mitigating.

To avoid being branded ineffective, defense lawyers need not assert every nonfrivolous defense. We have accepted in earlier cases that stacking different defenses can undercut with the jury the defense team's credibility, which is essential to a likelihood of success. See Harich v. Dugger, 844 F.2d 1464, 1471 (11th Cir.1988) (en banc); White v. Singletary, 972 F.2d 1218, 1221, 1225-26 (11th Cir.1992); Funchess v. Wainwright, 772 F.2d 683, 689 (11th Cir.1985). And, as the Supreme Court has stressed, we know good advocacy requires the winnowing out of some arguments in favor of stressing others: multiplicity of arguments or defenses hints at the lack of confidence in any one. See Jones v. Barnes, 463 U.S. 745, 751-53, 103 S. Ct. 3308, 3313, 77 L. Ed. 2d 987 (1983) (discussing appellate advocacy in criminal cases). The lawyers in this case did not pursue voluntary intoxication as a defense. They knew enough about their case, about the potential jury, and about PCP to make that course an informed and reasonable one.

Once we conclude that declining to investigate further was a reasonable act, we do not look to see what a further investigation would have produced. Nor do we need to decide whether the information that would have resulted from further investigation would have necessarily been used at trial by every reasonable lawyer who was armed with such information. If the act of conducting no investigation of a particular defense is reasonable, the matter is closed; there can be no question about the reasonableness of having failed to present evidence of which the lawyer was unaware as a result of a reasonable decision to investigate no further. Cf. Middleton v. Dugger, 849 F.2d 491, 493 (11th Cir.1988) (where lack of investigation is inadequately explained, court looks at information that more investigation would have produced and determines whether a reasonable trial counsel might not have used it).

Perhaps, other reasonable lawyers in this case would have chosen to investigate and to introduce drug evidence; but Rogers' lawyers were not constitutionally ineffective for making the professional judgment that they would avoid asserting Rogers' drug use as a defense, either to guilt or to the death penalty, before the jury. Because their choice was "arguably dictated by a reasonable trial strategy," we defer to the lawyers' judgment. Devier, 3 F.3d at 1450.

The district court grant of relief for ineffective assistance of counsel at petitioner's sentencing is REVERSED. All other denials of relief are AFFIRMED.