Kenneth Lamont Slocum was convicted of capital murder and sentenced to life imprisonment without parole. We affirmed the conviction. Slocum v. State, 325 Ark. 38, 924 S.W.2d 237 (1996). Mr. Slocum sought post-conviction relief pursuant to Ark. R. Crim. P. 37 on the ground that his counsel was ineffective in that he failed to request an instruction (AMCI 2d 403) to the effect that he could not be convicted solely on the uncorroborated testimony of an accomplice. See Ark. Code Ann. § 16-89-111 (e)(1) (1987). In response, the Trial Court granted a new trial. The State appeals from that decision, and we reverse it.
At the trial on the capital-murder charge, Vernon Scott testified that Mr. Slocum gave him a rock of cocaine worth forty dollars in return for luring the victim, Willie Simpkins, to the home of a man named Hattison. Mr. Scott testified he did not know why Mr. Slocum wanted it done. Mr. Scott said that, while he and Mr. Simpkins were at Hattison’s, Elgin King and Mr. Slocum, who was brandishing a .45 caliber pistol, entered and abducted Mr. Simpkins. Mr. Scott said that, despite the fact that Mr. King and Mr. Slocum were wearing masks, he was able to identify Mr. Slocum whom he had known for most of his life. Mr. Simpkins’s body was found with both .45 and .38 caliber bullets in it, and a rubber mask was found nearby.
*210Mr. King was convicted of first-degree murder. We reversed and remanded that conviction on the ground that the Trial Court refused Mr. King’s proffer of AMCI 2d 403 with respect to Mr. Scott’s testimony. King v. State, 323 Ark. 671, 916 S.W.2d 732 (1996). In Mr. Slocum’s direct appeal, however, we declined to reverse on the failure of the Trial Court to give the instruction because it had not been proffered to the Trial Court.
At the hearing on Mr. Slocum’s claim that his counsel’s failure to proffer AMCI 2d 403 resulted in his counsel being ineffective, the lawyer who represented Mr. Slocum at the trial testified as follows. He and his co-counsel sought to have Mr. Scott declared an accomplice as a matter of law. See Williams v. State, 328 Ark. 487, 944 S.W.2d 822 (1997). That request was denied. Counsel then agreed that AMCI 2d 403 would be requested, but, through inadvertence, it was not done. It is clear, however, that counsel’s trial strategy was not to depend on evidence that Mr. Scott was an accomplice whose testimony was uncorroborated. Rather, their strategy was to challenge the State’s evidence that Mr. Slocum had participated in the crime.
Although Mr. Slocum did not testify, he had consistently denied his guilt in conversations with his counsel. Counsel’s testimony on the point at the Rule 37 hearing was as follows:
And I specifically recall my conversation with [co-counsel] where we decided we would not argue to the jury that Vernon Scott was an accomplice, so even if you believed him, you could not convict because we felt that that was not a beneficial argument to make. But, and during that same discussion, we agreed that it was crucial that we submit that issue to the jury.
At a later point, counsel was questioned about whether it might have been inconsistent to have argued to the jury that Mr. Slocum was not even present at the crime scene and that Mr. Scott’s testimony was uncorroborated. Counsel responded that it would not have been inconsistent but that it would not have been a “winning argument.”
The only evidence produced that might have been considered corroborative of Mr. Scott’s testimony was: (1) a .45 caliber bullet like the ones found in Mr. Simpkins’s body that was found *211at Mr. Slocum’s grandmother’s home, which was frequented by Mr. Slocum; and (2) testimony that Mr. Slocum had a motive to kill Mr. Simpkins because Mr. Simpkins was supposed to testify in a trial against one of Mr. Slocum’s relatives. Thus, the “corroborating evidence” was hardly substantial, but it is also questionable whether the jury would have found Mr. Scott to be an accomplice, given his protestation that he knew nothing of the plan to abduct and kill the victim.
It is difficult to understand counsel’s statement that the giving of AMCI 2d 403 was “crucial” to their client’s case in light of their concession that they had no plan to argue the accomplice-testimony issue to the jury but were following the tack of their client’s complete innocence. Taking counsel at his word that there was an intention to seek the instruction, we are nonetheless left with the defense strategy, which was to argue to the jury that the State had failed to prove Mr. Slocum’s participation and not to rely on the accomplice-corroboration point. That is so because of counsel’s clear testimony that there was no plan to argue the accomplice-corroboration point to the jury, even if the instruction had been given.
We are aware of only one other State appeal from the granting of Rule 37 relief. State v. Manees, 264 Ark. 190, 569 S.W.2d 665 (1978). Our decision in that case was that the Trial Court lacked jurisdiction to alter a sentence being served and that error occurred because of failure to recite formal findings of fact and conclusions of law. There was no need to state a standard of review for cases such as the one we now consider. In considering the standard of review to be applied, we see no need to vary from the one used when Rule 37 relief has been denied, i.e., that we will not reverse the decision unless it is clearly erroneous. Catlett v. State, 331 Ark. 270, 962 S.W.2d 313 (1998) (per curiam), citing Thomas v. State, 330 Ark. 442, 954 S.W.2d 255 (1997). The question we must decide is whether the Trial Court clearly erred in holding that counsel’s performance was ineffective, applying the standard set in Strickland v. Washington, 466 U.S. 668 (1984).
*212We recently discussed the principles relating to ineffective assistance of counsel in Thomas v. State, 330 Ark. 442, 447-48, 954 S.W.2d 255, 257-58 (1997):
The criteria for assessing the effectiveness of counsel were enunciated by the Supreme Court in Strickland v. Washington, 466 U.S. 668 (1984). Strickland provides that when a convicted defendant complains of ineffective assistance of counsel, he must show that counsel’s representation fell below an objective standard of reasonableness and that counsel’s deficient performance prejudiced his defense. Judicial review of counsel’s performance must be highly deferential, and a fair assessment of counsel’s performance under Strickland requires that every effort be made to eliminate the distorting effects of hindsight, to reconstruct the circumstances of counsel’s conduct, and to evaluate the conduct from counsel’s perspective at the time. Missildine v. State, 314 Ark. 500, 863 S.W.2d 813 (1993). A reviewing court must indulge a strong presumption that the conduct falls within the wide range of reasonable professional assistance. Id.
To prevail on any claim of ineffective assistance of counsel, the petitioner must show first that counsel’s performance was deficient. Thomas [v. State], 322 Ark. 670, 911 S.W.2d 259 [(1995)]. This requires a showing that counsel made errors so serious that counsel was not functioning as the “counsel” guaranteed the petitioner by the Sixth Amendment. Id. Secondly, the petitioner must show that the deficient performance prejudiced the defense, which requires a showing that counsel’s errors were so serious as to deprive the petitioner of a fair trial. Id. Unless a petitioner makes both showings, it cannot be said that the conviction resulted from a breakdown in the adversarial process that renders the result unreliable. Id. In reviewing the denial of relief under Rule 37, this court must indulge in a strong presumption that counsel’s conduct falls within the wide range of reasonable professional assistance. Id. The petitioner must show that there is a reasonable probability that, but for counsel’s errors, the factfinder would have had a reasonable doubt respecting guilt in that the decision reached would have been different absent the errors. Id.; Huls v. State, 301 Ark. 572, 785 S.W.2d 467 (1990). A reasonable probability is a probability sufficient to undermine confidence in the outcome of the trial. Strickland, 466 U.S. 668; Thomas, 322 Ark. 670, 911 S.W.2d 259.
*213We have written on many occasions that a lawyer’s choice of trial strategy that proved ineffective is not a basis for meeting the Strickland test. See, e.g., Vickers v. State, 320 Ark. 437, 898 S.W.2d 26 (1995); Monts v. State, 312 Ark. 547, 851 S.W.2d 432 (1993). If counsel had proffered AMCI 2d 403 and it had been presented to the jury, and if counsel had then testified at the Rule 37 hearing that he declined to argue the accomplice issue to the jury because he did not consider the argument to have been “beneficial” or a “winning argument,” we most assuredly would have held that the Strickland test had not been met. The distinction here is that the instruction was not requested. Given the strategy choice of not asking the jury to follow that instruction, even if it had been given, we must say it is a distinction without a difference. True, the jury might possibly have seized upon the instruction and acquitted Mr. Slocum; however, we cannot say that the mistake made in this instance resulted in counsel’s performance being so deficient as to have denied a fair trial. It is difficult to say there was prejudice to Mr. Slocum in view of counsel’s choice, based on Mr. Slocum’s statements to him, to defend on what counsel referred to as “general denial.”
Counsel testified at the Rule 37 hearing that the failure to request the instruction did not result from a tactical decision on his part. It was, however, clearly a tactical decision not to argue the requirement of accomplice corroboration, for which counsel thought he had requested an instruction, to the jury. Given that tactical decision, it cannot be said that the result of the trial would have been different had the instruction been requested. Given the stringent standards set by the Strickland decision, which we have followed consistently, we hold that the Trial Court clearly erred in granting a new trial.
Reversed.