concurring in part and dissenting in part.
Much of what the majority says'T find to be uncontroversial. I agree that the district court’s determination as to pretext is a finding of fact subject to review only for clear error, and I agree that the district court correctly defined the issue before it as whether age or race was a factor in TVA’s reseleetion process. I concur, therefore, in parts 11(A) and 11(B).
I believe, too, that the majority’s description of the purport of Saint Mary’s Honor Center v. Hicks, 509 U.S. 502, 113 S.Ct. 2742, 125 L.Ed.2d 407 (1993), and its progeny is, in large measure, accurate. Where I part company with the majority, however, is in my understanding of this court’s function as a reviewer of factual findings for clear error only. In the name of applying Saint Mary’s, the majority undertakes a de novo review of the district court’s factual findings regarding pretext. The majority opinion appears to be driven largely by a desire to reach a different end than that reached by the district court following its two-day bench trial. As a result, the majority grievously mischaracterizes the district court’s opinion and then proceeds to reject the district court’s wholly plausible and supported findings in favor of a different view. Because I believe the majority opinion represents a fundamental disregard of dear-error review, I dissent.
I.
A.
Title VII of the Civil Rights Act of 1964 provides that ,“[i]t shall be an unlawful employment practice for an employer to fail or refuse to hire .... any individual ... because of such individual’s race____” 42 U.S.C. § 2000e-2(a). Using a nearly identical formula, the Age Discrimination in Employment Act prohibits discrimination because of an individual’s age. 29 U.S.C. § 623(a)(1).
The elements of a prima facie case are similar in both contexts. Where, as here, the plaintiff alleges that the defendant failed to select him because of his race, he may make out a prima facie case by showing 1) that he is a' member of a protected class, that is, he is a racial minority; 2) that he was not selected for the position; 3) that he was qualified for the position; and 4) that á person outside the class of protected persons was selected in his stead. See McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802, 93 S.Ct. 1817, 1824, 36 L.Ed.2d 668 (1973); Mitchell v. Toledo Hospital, 964 F.2d 577, 582 (6th Cir.1992). To make out a prima facie age discrimination claim, the steps are the same, although the person selected need not be outside the protected class, that is, need not be under 40 years of age. See *353O’Connor v. Consolidated Coin Caterers Corp., — U.S. -, -, 116 S.Ct. 1307, 1310, 134 L.Ed.2d 433 (1996). An alternative method of establishing a prima facie ease is the presentation of “credible, direct evidence of discriminatory intent.” Mitchell, 964 F.2d at 582 n. 4. No such evidence was presented here, but it is not disputed that Kline successfully émployed the first method to satisfy his prima facie burden.
That Title VII and ADEA intentional-discrimination cases employ a so-called “shifting burden of production” is well-established. To summarize briefly,
[t]he plaintiff in such a case ... must first establish, by a preponderance of the evidence, a “prima facie” case of ... discrimination. ...
... “[Establishment of the prima facie case in effect creates a presumption that the employer unlawfully discriminated against the employee.” ... Thus, the ... presumption places upon the defendant the burden of producing an explanation to rebut the prima facie case — i.e., the burden of “producing evidence” that the adverse employment actions were taken “for a legitimate, nondiscriminatory reason.” ...
... [A]t that point the shifted burden of production became irrelevant: “If the defendant carries this burden of production, the presumption raised by the prima facie case is rebutted,” and “drops from the ease[.]” The plaintiff then has “the full and fair opportunity to demonstrate,” through presentation of his own case and through cross-examination of the defendant’s witnesses, “that the proffered reason was not the true reason for the employment decision,” and that [a discriminatory reason] was. He retains that “ultimate burden of persuading the [trier of fact] that [he] has been the vie-tim of intentional discrimination.”
St. Mary’s Honor Ctr., 509 U.S. at 506-08, 113 S.Ct. at 2746-48 (footnote and citations omitted). Of course, if a plaintiff can present direct evidence of discrimination, it is not necessary for a court to employ the burden-shifting approach. See Burns v. City of Columbus, 91 F.3d 836, 842-43 (6th Cir.1996); Monette v. Electronic Data Sys. Corp., 90 F.3d 1173; 1178 (6th Cir.1996).
In ány event, since there is no dispute about the existence of Kline’s prima facie case, nor about the TVA’s introduction of evidence that its actions were not caused by unlawful discrimination, it is only the final step of the burden-shifting framework, that of proving “pretext,” that is at issue here. As the majority rightly observes, the case central to the necessary analysis is the Supreme Court’s 1993 decision in Saint Mary’s Honor Center. At issue there was whether “the trier of fact’s rejection of the employer’s asserted reasons for its actions mandates a finding for the plaintiff.” 509 U.S. at 504, 113 S.Ct. at 2746. In Saint Mary’s, the district court, following a bench trial, had “found that the reasons [the employer] gave were not the real reasons for [the employee’s] demotion and discharge,” but “nonetheless held that [the employee] had failed to carry his ultimate burden of proving that his race was the determining factor in [the employer’s] decision first to demote and then to dismiss him.” Id. at 508. The court of appeals reversed, believing that once an employer’s proffered reasons are discredited, the employer is effectively in the position of having remained silent, and thus of not having rebutted the plaintiffs prima facie case.
The Supreme Court, however, held that “[t]hat is not so. By producing evidence (whether ultimately persuasive or not) of nondiscriminatory reasons, [the employers] sustained their burden of production, and thus placed themselves in- a ‘better position than if they had remained silent.’” Id. at 509, 113 S.Ct. at 2748. And once the defendant meets its burden of production, the presumption of discrimination created by the prima facie ease simply ceases to exist— irrespective of whether the defendant’s case is actually persuasive. See id.
But as the Court cautioned,
[t]he factfinder’s disbelief of the reasons put forward by the defendant (particularly if disbelief is accompanied by a suspicion of mendacity) may, together with the elements of the prima facie case, suffice to show intentional discrimination. Thus, rejection of the defendant’s proffered rea*354sons, will permit the trier of fact to infer the ultimate fact of intentional discrimination, and the Court of Appeals was correct when it noted that, upon such rejection, “[n]o additional proof of discrimination is required.”
Id. at 511, 113 S.Ct. at 2749 (citations and footnote-omitted). At the same time, disbelief of the defendant’s reasons does not require a factfinder to conclude that the plaintiff has shown intentional discrimination. See id. Thus, "disbelief of a defendant’s reasons, if accompanied by a conviction that the true reason underlying the pretextual reason is not discrimination, should not result in a finding of intentional discrimination by a factfinder. The court noted that “a reason cannot be proved to be ‘pretext for discrimination’ unless it is shown both that the reason was false, and that discrimination was the real reason.” Id. at 515, 113 S.Ct. at-2752.
B.
The majority here claims that the district court “require[d] ... the introduction of direct proof in addition to a finding of pretext,” and that this “placed an impermissible burden on the Plaintiff which is contrary to the Supreme Court’s decision in Hicks, as well as the case law of this circuit.” (Maj. op. at 347.) It asserts that the lower court “intimatted] that in order for Kline to prevail he must introduce evidence of pretext, coupled with direct proof....” (Id.) Accordingly, the majority concludes, “[t]he district court’s opinion misstates the current state of the law and the burden shifting mechanism outlined in Hicks and its progeny. The district court required Kline to prove ‘pretext plus’ a standard [sic] rejected both by the Supreme Court and this circuit.” (Maj. op. at 348.) Contrary to the majority’s representation, however, the district court below did not misapply Saint Mary’s.
Following its detailed findings of fact, the district court proceeded to make its conclusions of law, first quoting Saint Mary!s Hon- or Center to the effect that “ ‘rejection of the defendant’s proffered reasons ... will permit the trier of fact to infer the ultimate fact of intentional discrimination.’” It concluded, however, that it would not reject TVA’s reasons, because Kline “clearly failed to show that TVA’s legitimate, non-discriminatory reason for his nonselection as part of the 1991 reselection process was false.” It further concluded that “[t]here simply was no proof that discrimination was the reason Beckler was selected.” The court correctly noted that pretext could be shown either by demonstrating that the staffed reasons for the employment decision had no basis in fact, or that they were not the actual reasons, or that they were insufficient to explain the employment decision, but concluded that Kline “did not meet his burden under any of these three methods”:
TVA’s stated reasons for selecting Beckr ler are, as discussed in the findings of fact, clearly supported by the record. Moreover, defendants’ stated reason that Beckler not only had more experience in relevant areas than plaintiff, but also had better management skills, is fully sufficient to explain her selection over him. Finally, plaintiff failed to show that the stated reasons were not the actual reasons.
It was then, after repeatedly stating that Kline had failed to demonstrate that the TVA’s reasons were pretextual, that the court concluded as follows: “This complete lack of proof as to pretext, coupled with the lack of any direct proof of discrimination, are fatal to plaintiffs case.” The majority relies on this statement for its claim that the district court required Kline to prove pretext and to introduce direct evidence of discrimination. The claim defies reality. It is clear, when read in context, that the district court meant that" Kline could either have proved pretext or presented direct proof of discrimination as an alternative — but the fact that he proved neither meant that he had not proved his case. In other words, given the district court’s conclusion that there was no pretext, Kline could only prevail by offering direct proof of discrimination, and thus circumventing the burden-shifting framework. It is simply false to suggest that the district court erroneously required the dual task of demonstrating pretext and direct evidence of discrimination. Rather than requiring Kline to *355produce direct evidence of discrimination, the district court simply noted that he had not done so.
It is true, however, that the district court added a problematic footnote:
Although plaintiff could, perhaps, prevail under the analysis set forth by the dissent in Saint Mary’s, that is not the law of the land. The dissent in Saint Mary’s focused, like the appellate panel did in remanding this ease, on the majority’s statement about the factfinder’s disbelief of reasons accompanied by a suspicion of mendacity, which, for the Uninitiated, means deceptive or false. It is my opinion that the dear import of the Saint Mary’s case is that a Title VII (orAJDEA) plaintiff must ultimately show that the stated reason for the employment decision was false, and that discrimination was the real reason.
(Emphasis added.) Although less than perfectly clear — arguably, the court meant only that if the “pretext only” position advocated by the Saint Mary’s dissent had been the law, then the plaintiffs argument would at least have had legal, if not factual, support— it certainly appears that this footnote misunderstands the import of Saint Mary’s. But even assuming that the district was indeed expressing its erroneous belief that Saint Mary’s had placed an imprimatur on the “pretext plus” school of thought, the misunderstanding does not merit reversal. Indeed, any such misunderstanding is of no moment, in light of the court’s clear finding that there was no pretext; if that finding is not clearly erroneous, then it does not matter if the district court wrongly believed pretext alone would not be enough to justify a verdict. Thus, as the TVA points out, since the district court found that there was no .evidence of pretext, it was unnecessary, if not impossible, for the court then to proceed to a determination whether the proof of pretext was sufficient to prove intentional discrimination.
In short, the district court concluded that the TVA had articulated a legitimate, nondiscriminatory explanation for its selection of Beekler, and that Kline failed to rebut that explanation. Those two findings, if not clearly erroneous, effectively end the case.
II.
A.
Using its condemnation of the district court’s legal reasoning as a springboard, the majority proceeds to “evaluate the reasons which ... could have caused a jury to disbelieve the reasons offered by the TVA and therefore establish pretext.” (Maj. op. at 349.) Following its evaluation, the majority concludes that it “is not convinced that TVA’s reselection was undertaken in good faith---Thus, the district court’s conclusion that Kline failed to establish pretext is clearly erroneous.” (Maj. op. at 351.) With all due respect, the majority asks the wrong question: it does not matter if the majority is convinced. Instead, the pertinent question is whether the district court was convinced, and if so, whether it was clearly erroneous. The fact that the majority would reach a different conclusion than the district court does not mean that the district court was clearly erroneous. The majority’s flawed perspective leads it, wrongly, to conduct a de novo review of the factual findings that led the district court to conclude that Kline had failed to show pretext.
It is beyond question, however, that when this court reviews a record of a bench trial in the district court, “[fjindings of fact, whether based on oral or documentary evidence, shall not be set aside unless clearly erroneous, and due regard shall be given to the opportunity of the trial court to judge the credibility of the witnesses.” Fed.R.Civ.P. 52(a). While the majority opinion acknowledges that clear error is the applicable standard, its execution of the review demonstrates that the acknowledgment is hollow.
I find it instructive to consider the Supreme '.Court’s, authoritative exegesis on clear-error review, found in Anderson v. City of Bessemer City, 470 U.S. 564, 105 S.Ct. 1504, 84 L.Ed.2d 518 (1985). As the Court explained, there are abundant reasons underlying our deference to the district courts:
The rationale for deference to the original finder of fact is not limited to the *356superiority of the trial judge’s position - to make determinations of credibility. The trial judge’s major role is the determination of fact, and with experience in fulfilling that role comes expertise. Duplication of the trial judge’s efforts in the court of appeals would very likely contribute only negligibly to the accuracy of fact determination at a huge cost in diversion of judicial resources. In addition, the parties to a case on appeal have already been forced to concentrate their energies and resources on persuading the trial judge that their account of the facts is the correct one; requiring them to persuade three more judges at the appellate level is requiring too much. As the Court has stated in a different context, the trial on the merits should be “the ‘main event’ ... rather than a ‘tryout on the road.’ ” For these reasons, review of factual findings under the clearly-erroneous standard— with its deference to the trier of fact — is the rule, not the exception.
Id. at 574-75, 105 S.Ct. at 1512 (citation omitted). The Court then went to some trouble to explain, at length, a reviewing court’s function and role:
Although the meaning of the phrase “clearly erroneous” is not immediately apparent, certain general principles governing the exercise of the appellate court’s power to overturn findings of a district court may be derived from our cases. The foremost of these principles ... is that “[a] finding is ‘clearly erroneous’ when although there is evidence to support it, the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed.” This standard plainly does not entitle a reviewing court to reverse the finding of the trier of fact simply because it is convinced that it would have decided the case differently. The reviewing court oversteps the bounds of its duty under Rule 52(a) if it undertakes to duplicate the role of the lower court. ... If the district court’s account of the evidence is plausible in light of the record viewed in its entirety, the court of appeals may not reverse it even though convinced that had it been sitting as the trier of fact, it would have weighed the evidence differently. Where there are two permissible views of the evidence, the fact-finder’s choice between them cannot be clearly erroneous.
Id. at 573-74, 105 S.Ct. at 1511 (emphasis added) (citations omitted). And when a finding is based on a credibility determination, as the Anderson Court made clear, it is due a special deference:
When findings are based on determinations regarding the credibility of witnesses, Rule 52(a) demands even greater deference to the trial court’s findings; for only the trial judge can be aware of the variations in demeanor and tone of voice that bear so heavily on the listener’s understanding of and belief in what is said____
[WJhen a trial judge’s finding is based on his decision to credit the testimony of one of two or more witnesses, each of whom has told a coherent and facially plausible story that is not contradicted by extrinsic evidence, that finding, if not internally inconsistent, can virtually never be clear error.
Id. at 575,105 S.Ct. at 1512.
B.
As the district court recognized, the standard for assessing whether an employer’s explanation for an adverse employment action is pretextual was established by this court in Manzer v. Diamond Shamrock Chemicals Co., 29 F.3d 1078 (6th Cir.1994). The court clarified the three ways in which a plaintiff can.produce sufficient evidence to .allow a factfinder to reject the employer’s explanation: ■
To make a submissible case on the credibility of his employer’s explanation, the plaintiff is “required to show by a preponderance of the evidence either (1) that the proffered reasons had no basis in fact, (2) that the proffered reasons did not actually 'motivate [the adverse employment action], or (3) that they were insufficient to motivate [the adverse employment action.]” The first type of showing is easily recognizable and consists of evidence that the proffered bases for the plaintiffs discharge never happened, i.e., that they are *357“factually false.” The third showing is also easily recognizable and, ordinarily, consists of evidence that other employees, particularly employees not in the protected class, were not fired even though they engaged in substantially identical conduct to that which the employer contends motivated its discharge of the plaintiff. These two types of rebuttals are direct attacks on the credibility' of the employer’s proffered motivation for firing plaintiff and, if shown, provide an evidentiary basis for what the Supreme Court has termed “a suspicion of mendacity.” ... [S]uch a showing permits, but does not require, the factfinder to infer illegal discrimination from the plaintiffs prima facie case.
Id. at 1084 (citations omitted); see EEOC v. Yenkin-Majestic Paint Corp., 112 F.3d 831, 834 (6th Cir.1997). Kline attempted to introduce evidence showing the first type of pretext: that the reasons given by the TVA for its reselection of Beckler had no basis in fact. The district court, however, made lengthy and detailed findings of fact leading it to conclude that Kline’s evidence was insufficient to make this showing.
The majority, however, states that “[tjhere are ample undisputed facts in the record which' support the conclusion that TVA’s proffered reasons are pretextual.” (Maj. op. at 349.) As should be clear from the Supreme Court’s description, this statement is not even close to a finding of clear error. It is, instead, simply a recognition that evidence can often support more than one version of reality, and an impermissible choice between two supported realities. It is true that there arguably was evidence that the TVA’s reselection of Carol Beckler was pretextual — but it is equally true that there was evidence that the reselection was not pretextual. The district court concluded the latter was the case, and that conclusion' should be left undisturbed by this court.
In any event, the majority fails to call the conclusion into serious question. It focuses on two narrow components of the evidence— the job description and the failure to interview — ignoring huge portions of the district court’s carefully reasoned and thorough opinion. Moreover, it is plain that there was ample evidence to support the district court’s findings with regard to those two items. Its findings, therefore, easily withstand dear-error review.
1.
I will consider first, the job description. A large part of Kline’s argument on appeal is directed at having this court infer that the TVA acted in bad faith because of some confusion about the job description applicable to Beckler’s position at the time Beckler was hired. In accepting this argument, the mav jority states that there is “conflicting testimony concerning the draft job description used in the reselection.” (Maj. op. at 349.) The acknowledgment of a conflict alone should forestall any further inquiry, for a finding based on a choice of one of “two permissible views of the evidence ... cannot be clearly erroneous.” Anderson, 470 U.S. at 574, 105 S.Ct. at 1511. But instead, the majority reasons that the conflicting evidence is the “crux” of the case “because it makes this court question whether good faith was used in the reselection process.” (Maj. op. at 350.)
The district court concluded that Sharon Tousek, the individual charged with reconstructing the selection process, derived the selection criteria for the position from a September 1988 job description, which was the job description, issued to Tousek when she began her job, and which was “essentially the same as the draft job description used by [Randy] Cole in making the original selection.” The job description, the court found, had- “[s]ix functional areas”: “récruitment and employment, salary policy and trades and labor classification systems, salary policy and trades and labor salary administration, labor .relations, affirmative employment/equal employment opportunity, and training [of personnel].”
Kline’s two complaints are that the functional criteria were not derived from the September 1988 job description, and that the September job description was not essentially the same as the one actually used. In particular, he asserts that the job description actually used did not contain a criterion for “salary policy and trades and labor classifica*358tion systems,” a category in which Beckler was more qualified, and the inclusion of which accordingly favored Beckler. According to Kline, the draft job description was improperly destroyed by Cole, and then replaced with a fake draft job description that included the labor category.
Of the three job descriptions in the record, however — one from June 1988, one from September 1988, and the undated job description that Kline claims was actually used — two (June and September) list the “labor classification” criterion. Since descriptions from both before and after the time of the actual August selection list this criterion, Kline’s convoluted conspiracy theory is facially implausible. Furthermore, irrespective of when the labor category got into the job description, there is no question that at the time of the initial selection, Cole was very concerned that the appointee have extensive labor experience; he testified accordingly, and Kline offered nothing to challenge this testimony. In other words, there is no question that Tousek was correct to consider the candidates’ labor experience. Thus, the majority’s Sturm, und Drang about which job description was used, and whether it contained the labor category is, at bottom, of no moment.
In any event, the district court explicitly rejected this theory when it found that “there was absolutely no credible evidence that TVA used or manipulated any of these criteria in an effort to justify a discriminatory selection.” The majority employs the completely- inconsequential confusion over 'which job description was used as a smokescreen, and simply never confronts the fact that there is absolutely no justifiable basis for throwing out this factual finding premised on credibility determinations. This amounts to a fundamental disregard of this court’s role as a reviewer for clear error only.
2.
The majority next turns its focus to the TVA’s failure to interview Kline or Beckler as part of the reselection process, stating that “[t]he lack of an interview for Kline also questions [sic] the good faith of the reselection process.”. (Maj. op. at 351.) The district court rejected, however, Kline’s argument that “the absence of an interview [is] proof of a discriminatory animus,” noting that the initial TVA finding of discrimination that had led to the reselection process, had “mandated that Beckler’s ‘experience in the position cannot be considered’ in the reselection,” and that an interview would have resulted in just that. “Beckler’s experience in the job ... would very likely have given her an advantage if an interview of the candidates had been included in the reselection process.” .The district court finally noted that Kline “failed to adduce any proof that Tousek avoided using interviews with the intent to discriminate against him or to deny him an opportunity to fully describe his experience. ”
The majority disregards these findings, on the basis of an argument never even put forward by Kline. According to the majority, although Beckler should not have been interviewed because her experience would have given her an advantage, Kline nonetheless should have been:
Beckler’s interview during her initial selection allowed her to demonstrate and strengthen her credentials for the position of HRO. Kline should have been given a similar opportunity. Thus, the appropriate inquiry was not whether interviewing the candidates during the reselection process would have given Beckler an unfair advantage, as the district court notes, but whether Kline was entitled to the same degree of consideration that Beckler received. It would have been unnecessary to interview Beckler a second time to determine whether she was qualified for the job, but it would have been helpful to determine whether Kline had qualifications that were not adequately represented in his paper credentials.
(Maj. op. at 351.)
This reasoning defies logic: Beckler’s successful interview performance during the initial phase would simply not be a piece of information that Tousek would have been privy to in conducting the reselection. Therefore, if Kline were interviewed during the reselection and Beckler were not, then Kline would have an unfair advantage over *359Beckler, assuming his interview was a success. The purpose of the reselection process in part was to reproduce the original selection to the extent possible; it was not meant to give the plaintiff every possible benefit. The only way to level the playing field for Kline and Beckler would be, as the district court found, to conduct no interview of either candidate, rather than to interview Kline only.
But even if it would have been possible to conduct a fair interview of Kline, that does not mean that the TVA’s failure to do so indicates pretext. As the district court emphasized, “there was no proof that interviews had ever been conducted with a reselection process at TVA”; interviews are not even required as part of an initial selection process. There is, moreover, no indication in the record that Kline ever requested an interview, nor does he ever specify what information an interview would have revealed that was not contained elsewhere in the voluminous material considered by the TVA. All of these facts strongly suggest the reasonableness of the TVA’s decision not to interview the candidates. And most significantly, the district court explicitly found that the failure to conduct interviews was based solely on the desire to be fair to Kline, rather than on discriminatory animus. The majority fails to confront this finding, and its post hoc invention of a potential interview process simply does nothing to lead one to the conclusion that the finding was clearly erroneous.
III.
I make one final observation. Although the majority repeatedly acknowledges that Saint Mary’s instructs that “[t]he evidence that established Kline’s prima facie case, along with a rejection of the reasons offered by the defendant, would have been sufficient to allow the district court to infer discrimination on the part of TVA,” (maj. op. at 350) (emphasis added), it nonetheless réasons as if the rule of Saint Mary’s were that these facts mandate an inference of discrimination. Even if the district court were clearly erroneous in finding that pretext had not been shown, that does not mean that judgment should be entered for Kline. A finding of pretext, coupled with the prima facie showing, is only the bare minimum necessary to support a judgment for the plaintiff. Given, however, that the district court made independent findings that Kline was not the victim of any discrimination — which findings the majority nowhere challenges — it is completely inappropriate to enter judgment for Kline.
As the majority itself acknowledges, the point of the Title VII undertaking is to determine whether discrimination was at work; we do not sit as TVA’s supermanager, simply to ensure that its hiring decisions are scrupulously fair in every particular. Kline' is not entitled to the best possible management decision by the TVA; he is entitled only to one free of improper motive. See Chappell v. GTE Prods. Corp., 803 F.2d 261, 266 (6th Cir.1986). The district court’s finding that Kline presented no evidence of discrimination is not clearly erroneous and therefore Kline is not entitled to judgment in.his favor.