Questions presented dispositive of the appeal are whether the Court of Appeals erred in (1) arresting judgment on defendant’s conviction of second degree rape and (2) concluding defendant was not denied his right to effective assistance of counsel at trial. We answer both questions affirmatively, reverse the decision of the Court of Appeals and award defendant a new trial in all cases.
I.
Defendant was tried on indictments charging first degree burglary, second degree rape, and second degree sexual offense at the 11 February 1985 Session of Superior Court in Wake County before Judge Bailey. He was convicted by the jury as charged in the rape and sexual offense cases and of misdemeanor breaking on the burglary indictment. Judge Bailey ordered a presentence diagnostic study before imposing sentences.
On 18 March 1985 defendant moved in writing to have his trial counsel, Mr. Jerome Paul, removed from the case. The motion was allowed on the same day, and Mr. Roger Smith then entered the case as counsel for defendant.
On 23 May 1985 defendant moved in writing to set aside the verdicts and to dismiss the charges in the rape and sexual offense cases for insufficiency of the evidence. This motion was denied at the 28 May 1985 Session of Superior Court at which session the trial court, after a sentencing hearing, imposed sentences of imprisonment as follows: two years for misdemeanor breaking; twelve years for second degree rape; and twelve years for second degree sexual offense. All sentences were ordered to run concurrently, and defendant was sentenced in all cases as a Committed Youthful Offender. Defendant appealed from these judgments to the Court of Appeals.
*389On 10 June 1985 defendant filed a Motion for Appropriate Relief by which he sought a new trial on the ground that his trial counsel, Mr. Paul, provided such ineffective assistance of counsel that defendant was convicted in violation of the federal and state constitutions. An evidentiary hearing on this motion was held, beginning 22 July 1985, before Judge Donald Stephens. Extensive evidence was taken on this motion; and on 9 August 1985 Judge Stephens, after making full findings and conclusions, denied the motion. Defendant appealed from this order to the Court of Appeals.
II.
The first question presented is whether the Court of Appeals erred in arresting judgment on the rape indictment on the ground there was a fatal variance between the indictment and the proof. We conclude that it did.
The rape indictment alleged that defendant “unlawfully, willfully and feloniously did ravish and carnally know [the victim] by force and and against her will, in violation of N.C.G.S. 14-72.3.”
At trial evidence for the state tended to show as follows:
On the evening of 31 August 1984 the victim was out with friends. She returned to her dorm room at approximately 1:00 a.m. She entered her room, closed the door, turned on the radio and fell asleep fully clothed. The victim dreamed she was engaging in sexual intercourse. She awoke to find defendant on top of her having vaginal intercourse with her. She tried to sit up, but defendant pushed her back down. Afraid her attacker might injure her, the victim offered no further resistance. Thereafter defendant engaged in anal intercourse with the victim.
The victim went to the door and turned on the light. Defendant told her not to call the police. He said, “I’m Lynn’s (the victim’s roommate) friend, I thought you were Lynn and I wouldn’t have done this if I had known it was you.” The victim told several friends about the incident, but did not report the incident to the North Carolina State Public Safety Department or make a statement until two days later.
Defendant testified in his own behalf as follows:
*390He knocked on the victim’s door. Hearing music, he believed his friend, Lynn, to be present and entered the room. Defendant observed a girl lying on the bed with her back facing him. Defendant called out the name Lynn but received no response. He then kissed the girl on the neck. The girl turned over and invited him to engage in oral sex. Defendant assisted the girl in removing her underpants. They engaged in oral sex, anal and vaginal intercourse. Following a brief rest, they engaged in sexual intercourse again. The girl then ran into the bathroom. When she returned, defendant noticed for the first time that his sexual partner was not Lynn. The victim told defendant not to worry because it could have happened to anybody. Defendant then left.
The Court of Appeals arrested judgment as to the charge of second degree rape. It first noted that N.C.G.S. § 14-27.3 provided for two theories of second degree rape: one theory is that the vaginal intercourse was committed “by force and against the will” of the victim, id., (a)(1); the other theory is that such intercourse was committed against one who is “mentally defective, mentally incapacitated, or physically helpless, and the person performing the act should reasonably know” it. Id., (a)(2). It then noted that N.C.G.S. § 14-27.1(3) defines “physically helpless” to mean “(i) a victim who is unconscious; or (ii) a victim who is physically unable to resist an act of vaginal intercourse or a sexual act or communicate unwillingness to submit to an act of vaginal intercourse or a sexual act.” The Court of Appeals concluded that a sleeping person is a “physically helpless” person under N.C.G.S. § 14-27.3(a)(2). It held that an indictment for the rape of one who is asleep must proceed on the theory that the victim was “physically helpless” pursuant to N.C.G.S. § 14-27.3(a)(2) and not on the theory that the rape was “by force and against the will” of the victim as provided in subsection (a)(1). The result in the Court of Appeals was that there is a fatal variance between the indictment and the proof presented at trial, and judgment was arrested.
We conclude that while the state might have elected to proceed under N.C.G.S. § 14-27.3(a)(2), it was not required to do so and that the evidence in this case supports a conviction of rape on a theory of force and lack of consent. There was, therefore, no fatal variance between the indictment and the proof.
*391At common law rape occurred when there was sexual intercourse by force and without the victim’s consent. State v. Hines, 286 N.C. 377, 380, 211 S.E. 2d 201, 203 (1975); accord, State v. Burns, 287 N.C. 102, 116, 214 S.E. 2d 56, 65, cert. denied, 423 U.S. 933, 46 L.Ed. 2d 264 (1975). Rape also occurred when there was sexual intercourse with a victim who was asleep or otherwise incapable of providing resistance or consent. Harvey v. State, 53 Ark. 425, 14 S.W. 645 (1890); Brown v. State, 138 Ga. 814, 76 S.E. 379 (1912); Territory of Hawaii v. Tatsuo Noguchi, 38 Haw. 350 (1949); State v. Lung, 21 Nev. 209, 28 P. 235 (1891); Payne v. State, 40 Tex. Crim. 202, 49 S.W. 604 (1899); 75 C.J.S. Rape § 11 (1952); 3 Wharton’s Criminal Law § 289 (1978).
In Brown v. State, 174 Ga. App. 913, 331 S.E. 2d 891 (1985), defendant had sexual relations with the victim as she lay comatose in her hospital bed. The court said that “[s]exual intercourse with a woman whose will is temporarily lost from intoxication, or unconsciousness arising from use of drugs or other cause, or sleep, is rape.” 174 Ga. App. at 913, 331 S.E. 2d at 892. An Oklahoma court held that an information which charged the accused with an act of sexual intercourse with a female while she was asleep and at the time unconscious of the nature of the act was sufficient to charge the accused with second degree rape and to give the court jurisdiction to pronounce judgment and sentence. In re Childers, 310 P. 2d 776 (Okla. Crim. App. 1957). The court said: “It is easily understood, and universally recognized, that a person who is unconscious by reason of intoxication, drugs, or sleep, is incapable of exercising any judgment in any matter whatsoever.” Id. at 778. In State v. Welch, 191 Mo. 179, 89 S.W. 945 (1905), the court said:
[T]he general, if not universal, rule is that if a man have connection with a woman while she is asleép, he is guilty of rape, because the act is without her consent. . . . We are, therefore, unanimously of opinion that the crime, which the evidence in this case tended to prove, of a man’s having carnal intercourse with a woman, without her consent, while she was, as he knew, wholly insensible so as to be incapable of consenting, and with such force as was necessary to accomplish the purpose, was rape.
191 Mo. at 187-88, 89 S.W. at 947.
*392As can be seen from the foregoing cases the common law implied in law the elements of force and lack of consent so as to make the crime of rape complete upon the mere showing of sexual intercourse with a person who is asleep, unconscious, or otherwise incapacitated and therefore could not resist or give consent. Our rape statutes essentially codify the common law of rape. N.C.G.S. § 14-27.2 et seq. (1986); State v. Booher, 305 N.C. 554, 290 S.E. 2d 561 (1982); State v. Locklear, 304 N.C. 534, 284 S.E. 2d 500 (1981); State v. Perry, 291 N.C. 586, 231 S.E. 2d 262 (1977). In the case of a sleeping, or similarly incapacitated victim, it makes no difference whether the indictment alleges that the vaginal intercourse was by force and against the victim’s will or whether it alleges merely the vaginal intercourse with an incapacitated victim. In such a case sexual intercourse with the victim is ipso facto rape because the force and lack of consent are implied in law.
III.
The second question presented is whether the Court of Appeals erred in concluding defendant was not denied his right to effective assistance of counsel at trial in violation of his rights guaranteed by the Sixth Amendment to the United States Constitution and Article I, §§ 19 and 23 of the North Carolina Constitution. We conclude that it did.
A hearing on defendant’s motion for appropriate relief was held at the 23 July 1985 Criminal Session of the Superior Court of Wake County, Judge Stephens presiding. At the proceeding, defendant offered evidence which may be summarized as follows:
Several attorneys experienced in defending criminal cases in Wake County testified on defendant’s behalf. According to these witnesses it is standard practice among criminal defense lawyers in the county, among other things, to locate and interview witnesses before trial; visit the physical location of any events that bear on the trial; prepare the client to testify at trial and inform him of what is to be expected on direct examination and cross-examination; adopt a defense theory; avoid promising to prove matters in opening statements, without a reasonable belief that evidence exists which supports the promises; and avoid concessions to the jury that defendant’s testimony lacks credibility.
*393Judge Bailey, who presided at defendant’s trial, testified that Mr. Paul promised in his opening statement to prove that defendant was physically and psychologically incapable of rape; yet no such evidence was forthcoming. Judge Bailey said it was unusual to make a prediction of evidence that is not produced because to do so seriously undermines your credibility and ultimately your client’s credibility with the jury. Judge Bailey also testified that Paul: (1) did not appear to be listening to the state’s case; (2) was generally disheveled or rumpled in his appearance, clothing and hair; (3) exhibited marked mood changes — there were times when he appeared alert and aggressive and other times when he appeared lethargic and even drowsy; and (4) appeared to be asleep during the cross-examination of defendant. Judge Bailey said that in his experience the combination of actions exhibited by Paul was unique.
Dorothy Moorman, defendant’s mother, testified that during the trial Paul experienced pain and ingested medication to ease the pain. She also corroborated Judge Bailey’s testimony that Paul fell asleep during the trial.
According to defendant’s testimony at the post-conviction hearing, he never told Paul that it was physically or psychologically impossible for him to commit rape. Defendant said he had no idea what Paul meant when he promised to prove defendant was incapable of rape. Paul never visited the dormitory where the incident occurred; did not locate and interview witnesses before the trial started; and never spoke to the witnesses individually. After the trial began Paul spoke with the witnesses as a group for only thirty minutes. He never advised the witnesses of what to expect in court and never discussed the questions to expect on direct and cross-examination. Paul did not prepare defendant for trial. He never discussed with defendant his testimony or the questions to expect on direct or cross-examination. Paul simply told defendant to “expect the unexpected.” At trial defendant saw Paul take medicine several times. On one occasion defendant was in Paul’s car when Paul stopped at an Eckerd Drugstore and purchased a drug. When Paul returned to the car he ingested the drug. The drug appeared to be Valium. On two or three evenings during the trial defendant saw Paul in his hotel room with various drugs. The drugs caused Paul’s speech to be slurred and Paul to fall asleep. Defendant corroborated other witnesses who said Paul fell *394asleep during the cross-examination of defendant. Defendant said he heard Paul tell Angelo Barnes that he wanted to display prejudice and racism at the trial. Paul asked Barnes to stand up and protest out loud so the media would see the protest and act upon it. In explaining this action and others, Paul told defendant that he would “have to pay the price for a lot of people, that through [his] sufferings other people will benefit.”
The state offered evidence tending to show that there were “problems” in defendant’s case and that these problems would have to be satisfactorily explained to the jury for defendant to be acquitted. The state also presented the testimony of several persons who were present at the trial and who did not see Paul asleep, did not think Paul was inattentive, and did not believe Paul was under the influence of drugs.
Upon this evidence the trial court made detailed findings of fact and conclusions of law. The trial court found that Paul labored under a conflict of interest, i.e., his interest in his “public cause” of establishing a racially motivated prosecution and his interest in pursuing the best defense for his client, individually. The state challenges these findings on appeal as being unsupported by the evidence. We do not consider these findings in our assessment of the case because there are ample additional findings which the state does not challenge on appeal and which are dispositive of the case. These are (paraphrased except where quoted):
1. “With regard to trial preparation, . . . Attorney Paul did little more than read the police report and meet several potential witnesses. He did not visit the crime scene and did not conduct any independent investigation of the matter. His discussions with potential witnesses were brief and failed to explore their knowledge of the incident. He failed to advise witnesses what they would be asked, what to expect, or whether or not they would even testify. This lack of preparation even extended to the defendant. Although Attorney Paul spent a considerable amount of time with Mr. Moorman, most of their conversations centered on football and other matters not related to the trial. The Court finds that Attorney Paul spent no more than an hour discussing with Moorman the specifics of his version of this incident and only briefly *395discussed his potential testimony on two occasions. He merely advised Moorman to ‘expect the unexpected.’ This limited amount of trial preparation and the investigation was well below the standard of practice routinely engaged in by attorneys who defend serious criminal cases in the Superior Court of Wake County.”
2. “The Court further finds that Attorney Paul’s conduct during trial was equally deficient. He appeared during the trial ‘disheveled and rumpled’ and demonstrated marked changes in mood from affable to lethargic, from aggressive to inattentive and drowsy. Although his defense was primarily one of consent, he engaged in rhetoric and questions, unsupported by any evidence, which suggested that the charges were racially motivated and that the defendant was the victim of a conspiracy. In his opening remarks to the jury, he advised that the prosecution witness’s account of the incident was preposterous, that a conspiracy against the defendant existed and that Moorman was physically and psychologically incapable of rape. He further stated that the defense would offer evidence regarding the victim’s prior similar encounter with another black athlete. He made reference to ‘one critical piece of evidence’ which would show that it was physically impossible for Moorman to engage in the acts which the victim would describe. In view of Attorney Paul’s failure to adequately investigate this matter, his opening statement was deficient because he was unable to produce any evidence to support the above claims. The impact of such a deficient opening statement was further aggravated by the prosecutor’s closing argument which addressed the failure of the defendant to show any of the above. The prosecutor’s closing remarks also focused on Attorney Paul’s failure to adequately interview witnesses, including the defendant. Examples of such closing comments by Prosecutor Hart are numerous.”
3. “Attorney Paul’s opening statement and wide ranging defense theories unsupported by the evidence as well as his forecasted evidence which did not materialize were practices that were deficient and failed to fall within the range of competence expected of attorneys in criminal cases.”
*3964. Counsel used and abused drugs during the trial. These drugs included Percocet, Dalmane, Fiorinal, Vicodin, Demerol, Vistaril and Phenergan. The drugs were ingested repeatedly throughout the ten-day period of the trial. The drugs were taken in combination with one another. The use and abuse of these drugs caused counsel’s judgment and mental processes to be substantially impaired during the trial. The cumulative effect of these drugs substantially impaired counsel’s sensory perceptions, reasoning, and judgment.
5. During the trial, counsel was lethargic, inattentive, and drowsy. During the defendant’s testimony, counsel dozed off briefly on at least one occasion.
6. During the trial, counsel suffered from the debilitating effects of migraine headaches. Counsel took prescription medication including Inderal, Tofranil and Librium on a daily basis as a preventive measure. Counsel never advised the defendant or the Court about the extent of this disability.
7. “Attorney Paul’s closing argument was also deficient for his crude language and his suggestion to the jury that his client’s testimony was unworthy of belief as it related to Moorman’s claim of misidentification of the victim. Although that statement by Paul appears accurate, in making the statement he abandoned his client’s interest.”*
8. Counsel’s performance significantly impaired his effectiveness. Counsel’s performance was below the routine standard of practice.
Upon the foregoing findings the trial court concluded:
“Based upon the foregoing findings of fact, the Court concludes as a matter of law that the pretrial and trial performance of Jerome Paul was significantly deficient and fell well below the minimum standard of professional competence expected and required of attorneys handling serious criminal cases in the Superior Courts of Wake County. The quality of Mr. Paul’s representation was far below that standard of *397practice routinely engaged in by members of the Wake County Bar who practice criminal law in the Superior Court.”
Notwithstanding these findings and this conclusion the trial court ultimately concluded that defendant suffered no actual prejudice from trial counsel’s deficiencies; there was no reasonable probability or possibility that absent these deficiencies the jury would have had a reasonable doubt regarding defendant’s guilt; and defendant failed to show that his attorney’s conduct so affected the verdict that “defendant’s trial cannot be relied on as having produced a just result.” The court therefore denied defendant’s motion for appropriate relief based on ineffective assistance of counsel.
The trial court’s conclusion that defendant suffered no actual prejudice from his trial counsel’s deficiencies was based on the following factual findings (paraphrased except where quoted):
1. The court has carefully examined the testimony at the post conviction hearing and at trial.
2. Defendant has not shown how the testimony of any witness would have been different had Paul properly interviewed and prepared them.
3. No additional witnesses which could have added any material evidence were revealed and “all witnesses with relevant evidence testified to the full extent of their knowledge” at trial.
4. Defendant has not shown how his trial testimony would have been any different had Paul properly prepared him as a witness.
5. Paul’s cross-examination of the victim and other state’s witnesses was “thorough and aggressive.”
6. Although Paul was deficient in presenting the defendant’s case, he “was not deficient in attacking the prosecution’s case.”
7. “[T]he jury verdicts were based upon a determination by the jury that [the victim] testified truthfully and that [the defendant] did not. . . . Although [the victim’s] testimony . . . could be characterized as unusual, the testimony of the de*398fendant was simply not credible. [The victim’s] testimony was corroborated by evidence of physical injuries to her neck and rectum consistent with an assault.”
8. Absent all errors and deficiencies by Paul “there is no likely possibility that the factfinders would have had a reasonable doubt as to the defendant’s guilt.”
9. “The Court is satisfied that the jury in this case returned a verdict based upon the law and the evidence without regard to and unaffected by the conduct of Attorney Paul.”
In reviewing an order entered on a motion for appropriate relief, the findings of fact made by the trial court are binding on us if they are supported by evidence, even though the evidence is conflicting. State v. Stevens, 305 N.C. 712, 291 S.E. 2d 585 (1982). Our inquiry as an appellate court is to determine whether the findings of fact are supported by evidence, whether the findings of fact support the conclusions of law, and whether the conclusions of law support the order entered by the hearing court. Id.
We find all of the foregoing facts regarding Paul to be supported by the evidence. Indeed, the state, as noted, does not challenge these findings on appeal. We determine, however, for the reasons which follow that the trial court’s findings do not support its conclusion that Paul’s failure to provide effective assistance of counsel had no probable effect on the trial’s outcome.
The trial court based its ultimate conclusion that the trial result was unaffected by Paul’s deficiencies upon its finding that the evidence would have been essentially the same had the deficiencies not been present; yet it recognized that the ultimate question for the jury was the credibility of the principal witnesses —the victim and the defendant. We agree with this assessment, and, for this reason, must disagree with the trial court’s ultimate conclusion that on the facts as found there is no reasonable probability that Paul’s failure to provide effective assistance affected the outcome of the trial. We are satisfied the facts found demonstrate the existence of such a probability.
Neither does the state challenge, as unsupported by the evidence or the findings, the trial court’s overall conclusion that Paul’s performance “was significantly deficient and fell well below *399the minimum standard of professional competence expected and required of attorneys handling serious criminal cases in the Superior Courts of Wake County.”
The essence of the state’s argument is: notwithstanding the trial court’s unchallenged factual findings regarding Paul’s deficiencies, its conclusion that Paul’s substandard representation did not prejudice defendant should be sustained under the standard set out in Strickland v. Washington, 466 U.S. 668, 80 L.Ed. 2d 674 (1984); and State v. Braswell, 312 N.C. 553, 324 S.E. 2d 241 (1985).
A defendant’s right to counsel includes the right to the effective assistance of counsel. State v. Braswell, 312 N.C. 553, 324 S.E. 2d 241. The test for ineffective assistance of counsel is the same under both the federal and state constitutions. Id. The test of effective assistance of counsel has two components. First, defendant must show counsel’s performance fell below an objective standard of reasonableness. Id., citing Strickland v. Washington, 466 U.S. 668, 80 L.Ed. 2d 674. Second, defendant must show that “counsel’s errors were so serious as to deprive the defendant of a fair trial, a trial whose result is reliable.” Id. at 687, 80 L.Ed. 2d at 693; State v. Braswell, 312 N.C. at 562, 324 S.E. 2d at 248. The question becomes whether a reasonable probability exists that, absent counsel’s deficient performance, the result of the proceeding would have been different. Strickland v. Washington, 466 U.S. at 695, 80 L.Ed. 2d at 698.
Where, as here, ineffective assistance of counsel has been established, the full ramifications of counsel’s deficient performance can never be completely reconstructed. Strickland and Braswell do not place on defendant the burden of proving that the trial outcome would have been different. Rather, defendant must show that “there is a reasonable probability that, but for counsel’s ineffective performance, the result of the proceedings would have been different. A reasonable probability is a probability sufficient to undermine confidence in the outcome.” Strickland v. Washington, 466 U.S. at 694, 80 L.Ed. 2d at 698; accord, State v. Braswell, 312 N.C. at 563, 324 S.E. 2d at 248.
We are satisfied there is a reasonable probability that but for Paul’s acts of substandard performance not challenged by the state the result at trial would have been different and that these acts are sufficient to undermine confidence in the trial’s reliabili*400ty. The principal issue at trial, as the trial court correctly concluded, was one of credibility. The victim testified that defendant engaged in vaginal intercourse with her while she was asleep; and when she awoke, he forcibly and against her will engaged in anal intercourse. Defendant testified both incidents were with the victim’s consent, given while she was awake and aware of what was transpiring. The question for the jury, then, was which witness to believe. This translates into a relatively close case on the facts, especially when, as the trial court noted, the testimony of the victim, herself, “could be characterized as unusual.”
A cardinal tenet of successful advocacy is that the advocate be unquestionably credible. If the fact finder loses confidence in the credibility of the advocate, it loses confidence in the credibility of the advocate’s cause. When in a trial such as the one before us the whole defense rests on the credibility of the defendant as a witness, it is particularly crucial that the defendant’s advocate not only retain credibility for himself but also that he do nothing which undermines the credibility of his client.
Paul undoubtedly undermined his own credibility with the jury when he promised in his opening statement to produce “one critical piece of evidence” which would demonstrate that defendant was physically and psychologically incapable of engaging in sexual acts, yet failed to produce a single bit of evidence remotely suggesting this fact. About this incident the presiding judge at defendant’s trial testified at the post conviction hearing as follows:
I was surprised at the time by his [Paul’s] statement that he would prove that the defendant was physically and psychologically incapable of rape. I interpreted that to indicate that he . . . probably would prove impotency.
When at a later time the defense lawyer called as a witness a young woman who candidly stated that she had intercourse with Mr. Moorman on three different occasions one afternoon prior to the rape it seemed to me that we had sort of lost contact with physical inability.
The record before us is void of suggestion that Paul had any basis for thinking that defendant was physically unable to commit the *401crimes charged against him. Defendant testified Paul never discussed the matter with him in pretrial preparation and he had no knowledge of any plans to proffer such a defense. We are confident the inconsistency in counsel’s promised defense of psychological and physical inability and the only defense supported by the evidence, consent, was not lost on the jury. This promised defense severely undercut the credibility of the actual evidence offered at trial, including defendant’s own testimony, all of which supported only the defense of consent.
The effect of the unfulfilled promise of a defense of psychological and physical inability was exacerbated by similar references in counsel’s opening statement that defendant was the victim of some kind of racially motivated conspiracy. Again the record contains no suggestion that Paul had any basis for this assertion, and he produced no evidence to support it.
The defense’s failure to produce any evidence to support the theories proffered at the outset of the trial formed the basis of one of the principal closing arguments made by the state in favor of conviction. As the trial court found, “The impact of such a deficient opening statement was further aggravated by the prosecutor’s closing argument which addressed the failure of the defendant to show any of the above. . . . Examples . . . are numerous:
(a) ‘. . . it was real hard to tell . . . what the defense was.’
(b) ‘It is hard to tell what Mr. Paul wants you to believe.’
(c) T would ask you to think about all the different things that Mr. Paul said in his opening statement that he was going to show you . . . that were never shown through the evidence ....’”
Paul’s assertion in closing argument that his client’s testimony that he mistook the victim for someone else was not worthy of belief was devastating to the defense of consent and must have undermined defendant’s credibility in the jury’s eyes. This aspect of defendant’s testimony formed the underpinning of his consent defense. According to this testimony defendant believed the victim to be his friend, Lynn, who consented to his sexual advances. It was not until the sexual episodes were over that defendant, according to his testimony, discovered his mistake. If defendant’s own advocate suggests to the jury that this testimony *402is not credible, why should the jury believe anything else the defendant has said?
The only defense supported by the evidence —consent —depended for its success on the jury’s acceptance of defendant’s credibility as a witness. Under this circumstance we conclude that Paul’s wide-ranging opening assertions, which had no foundation in his pretrial investigation and were never remotely supported by any evidence proffered at trial, undercut the only defense supported by the evidence and defendant’s own testimony. When Paul’s opening statement and closing argument that an important part of his client’s testimony was not credible are coupled with his regular use of a variety of pain killing drugs, his frequent migraine headaches, and his drowsiness, lethargy, and inattentiveness during portions of the trial, a reasonable probability is created that had all these things not occurred the trial outcome might have been different. Confidence in the trial’s reliability is, therefore, undermined.
Courts of other jurisdictions have set aside trials upon findings of ineffective assistance of counsel based in part on counsel’s failure to produce evidence promised in the opening statement. In People v. Zaborski 59 N.Y. 2d 863, 465 N.Y.S. 2d 927, 452 N.E. 2d 1255 (1983), counsel raised the defense of entrapment but never produced evidence to support it. As could be expected, the prosecutor was able to use this omission against defendant. Based largely on counsel’s failure in this regard, the court ordered a new trial. The same result obtained for similar reasons in People v. Corona, 80 Cal. App. 3d 684, 145 Cal. Rptr. 894 (1978) (unfulfilled promises resulted in “devastating comments” by prosecutor); People v. LaBree, 34 N.Y. 2d 257, 357 N.Y.S. 2d 412, 313 N.E. 2d 730 (1974); and Commonwealth v. Lambeth, 273 Pa. Super. 460, 417 A. 2d 739 (1979). See also Javor v. United States, 724 F. 2d 831 (9th Cir. 1984) (new trial ordered when attorney was asleep or dozing during a substantial part of the trial); White v. State, 414 N.E. 2d 973 (Ind. App. 1981) (new trial ordered when attorney ill and taking five different medications); and Ex Parte Love, 468 S.W. 2d 836 (Tex. Crim. App. 1971) (new trial ordered when attorney impaired due to physical injury).
The result is that the decision of the Court of Appeals is reversed and the case remanded to that court for further remand to the Superior Court for Wake County for a new trial.
*403Reversed and remanded.
The state argues on appeal that there is no evidence to support the finding that Paul “abandoned” his client; but it does not challenge the fact that Paul made the argument as set out in this finding.