ESCHBACH, Senior Circuit Judge.
When Peter Perfetti was 54 years old, he lost his job at The First National Bank of Chicago (the “Bank”). Perfetti’s unit had been disbanded as part of a reduction in force. For five months thereafter, the Bank kept Perfetti on the payroll while he searched for a new job. During this period, Perfetti applied for more than 550 jobs without success. At the same time, he was allowed to bid for positions within the Bank. Perfetti never did find another job at the Bank, although he applied for at least five separate positions. Based on these adverse employment decisions, he sued under the Age Discrimination in Employment Act (“ADEA”), 29 U.S.C. § 621
et seq.
The district court granted summary judgment for the Bank on three of the five positions. R. 89. Perfetti went to trial on the remaining counts, which concerned a position as “facilities coordinator” and a position as “operations analyst.” The only issue was whether Perfetti proved that the Bank’s proffered justifications for not rehiring him were “unworthy of credence” and therefore pretextual.
See Texas Dept. of Community Affairs v. Burdine,
450 U.S. 248, 256, 101 S.Ct. 1089, 1095, 67 L.Ed.2d 207 (1981). The jury found for Perfetti on both counts. On appeal, the Bank argues that there was insufficient evidence to support the jury’s finding that the Bank’s proffered reasons were unworthy of credence.
We agree. Because the evidence of pretext regarding both employment decisions is insubstantial, we reverse and remand for entry of judgment for the Bank.
Discussion
1. The Question on Appeal
Because Perfetti’s claim has already gone to the jury, we need not concern ourselves with the intricacies of his method of proof at trial; the only question on appeal is “whether there was sufficient evidence for a reasonable jury to find that age was a determining factor” in the hiring decisions.
Brown v. M & M/Mars,
883 F.2d 505, 507 (7th Cir.1989). In making this determination, we must resolve any evidentiary conflict and every permissible inference in favor of the party who won the verdict.
La Montagne v. American Convenience Products, Inc.,
750 F.2d 1405, 1410 (7th Cir.1984). Furthermore, we do not judge the credibility of witnesses.
Id.
“Our role is not to weigh the evidence in search of a preponderance as would a jury, but it is instead the more restrictive function of determining if the evidence in support of the verdict is
substantial;
a mere scintilla of supporting evidence will not suffice.”
Id.
Whether Perfetti’s evidence is substantial is an extremely narrow question in this case, because Perfetti’s evidence of liability went exclusively toward proving that the Bank’s proffered explanations were “unworthy of credence.”
McDonnell Douglas Corp. v. Green,
411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973), and
Texas Dept. of Community Affairs v. Burdine,
450 U.S. 248, 101 S.Ct. 1089, 67 L.Ed.2d 207 (1981), allow a plaintiff to prove employment discrimination with any of four different types of evidence. In general, a plaintiff may prove discrimination either directly, by proving that age was a determining factor in the employer’s decision, or indirectly, by proving that the employer’s proffered explanation is pretextual. Under the direct method of proof, the plaintiff’s evidence may be either direct or circumstantial. Thus, the first type of evidence in a discrimination suit is direct evidence that age was a determining factor, such as discriminatory statements uttered by the employer’s decision-maker. The second type of evidence is circumstantial evidence that age was a determining factor, such as a statistical imbalance in the employer’s work-force.
McDonnell Douglas
and
Burdine
also allow the plaintiff to prevail under an indirect method of proof. Under this method, the plaintiff must first prove his or her
prima facie
case. The burden of production then shifts to the employer, who must articulate a lawful reason for its decision. To prevail, the plaintiff retains the burden of proof and must prove that the employer’s articulated rationale is pretextual, either by showing that a discriminatory reason more likely motivated the employer or that the employer’s proffered reason is “unworthy of credence.”
See generally La Montague,
750 F.2d at 1409-10. The theory behind the indirect method of proof is that evidence of discrimination, including even circumstantial evidence, may be extremely difficult for a plaintiff to discover. Therefore, if the plaintiff proves that the employer’s proffered explanation is pretex-tual, the jury may infer discrimination from that proof alone.
Id.
The indirect method of proof thus allows a plaintiff to prevail by bringing two additional types of evidence. The third type of evidence, therefore, is direct evidence of pretext, such as a contradiction between the employer’s proffered justification at trial and documentary evidence from the time of the decision, or a contradiction among the witnesses to the employment decision at issue. Last, the fourth type of evidence is circumstantial evidence of pretext, such as evidence that the proffered justification is not a genuine job requirement or has been inconsistently applied to other employees.
In the present case, almost all of Perfet-ti’s evidence fell into the fourth and weakest category, circumstantial evidence of pretext. He employed the indirect method of proof under the ADEA. He did not present even a scintilla of evidence that the Bank had a discriminatory attitude regarding his or anyone else’s age, or that the Bank’s policies had a discriminatory age-based impact on his or anyone else’s employment. Furthermore, Perfetti’s proof of pretext
was almost entirely indirect; he did not present even a scintilla of evidence to indicate that the Bank’s reasons were pretexts for age discrimination. Age has been raised in this case only twice: Perfetti was in his mid-fifties at the time of the adverse employment decisions, and the persons who were hired in his stead were younger than he.
Finally, the Bank’s employees did not contradict one another or the documentary evidence. Perfetti’s only direct evidence of pretext is his own testimony contradicting a single detail of the testimony of one of the Bank’s witnesses. As a result, Perfetti’s case is virtually as indirect and circumstantial as possible; we have before us a “pure” unworthy of credence claim.
We emphasize that Perfetti’s failure to bring any evidence regarding age does not defeat his claim
per se;
“[t]he special virtue of the indirect method of proof is that it allows victims of age discrimination to prevail without presenting
any
evidence that age was a determining factor in the employer’s motivation.”
La Montague,
750 F.2d at 1409-10.
Holding to the con
trary would violate one of the purposes of
McDonnell Douglas
and
Burdine,
which is to permit plaintiffs to recover, even if they are unable to discover evidence of discrimination, simply by proving that the employer’s proffered justification is unworthy of credence.
Once we accept that Perfetti’s reliance on circumstantial “unworthy of credence” evidence does not defeat his claim, it is tempting to go to the other extreme. Because Perfetti’s evidence goes only to the question whether the Bank’s proffered justification is “unworthy of credence,” the jury, in finding for the plaintiff, necessarily found the Bank’s witnesses and explanations “unworthy of credence.” But we cannot judge the credibility of the Bank’s employees. Can we reverse a jury verdict based solely on a lack of
credence
without infringing on the jury’s exclusive right to judge
credibility l
The answer is, “We can.” Perfetti’s lack of evidence regarding age cannot insulate his jury verdict from review for sufficiency of the evidence. Such a holding would require that virtually every unworthy of credence claim go to the jury. Let us backtrack for a moment. The standards for granting summary judgment, a directed verdict, and a judgment notwithstanding the verdict (“JNOV”) are essentially the same; all of these motions “call upon the court to make basically the same determination — that there is no genuine issue of fact and that the moving party is entitled to prevail as a matter of law.” Wright & Miller,
Federal Practice and Procedure
§ 2713.1 at 613, § 2537 at 599 (1983). If we could never reverse a jury verdict for the plaintiff in an “unworthy of credence” claim on the ground of insufficiency of the evidence, then a district court could never grant summary judgment or a directed verdict for the employer; in every claim, the jury would have to be given its opportunity simply to disbelieve the employer. This result does not follow from
Burdine
and would overrule the basic requirement that a jury verdict be supported by “substantial” evidence. The purpose of
Burdine
was not to allow plaintiffs to go to the jury without presenting any evidence at all; its purpose was to allow plaintiffs to prove discrimination without presenting any evidence
of discrimination,
and solely with evidence of pretext. Furthermore, such a holding would be perverse. Cases based on indirect and circumstantial evidence of pretext should not be less subject to dismissal for insufficiency than cases based on evidence which is
more
probative of discrimination.
Having established that Perfetti’s narrowly focused evidence neither defeats nor insulates his claim as a matter of law, we come to what the nature of his evidence does do. Perfetti’s reliance solely on the fourth type of evidence, save his own contradiction of a single detail of one witness’s testimony, narrows our determination whether there was substantial evidence to support the verdict. Because Perfetti did not have any proof regarding age at trial, his entire burden of having “substantial evidence” falls on his “unworthy of credence” evidence. Under these circumstances, Perfetti must have presented sufficient evidence for a reasonable jury to infer,
solely on the basis of indirect evidence that the Bank’s proffered explanations were unworthy of credence,
that the Bank discriminated against him on account of his age. Thus, we reverse not because Perfetti failed to present direct or circumstantial evidence regarding age, but because he presented almost exclusively indirect evidence of pretext
and even that evidence was insubstantial.
2. The Evidence at Trial
A. Facilities Coordinator Position
The Bank’s rules regarding employment decisions stipulate that before an external candidate for any position can be considered, all internal candidates must be considered and eliminated. Because Perfetti had been fired by the time he applied for the facilities coordinator position, he was considered an external candidate for that job. The Bank’s proffered explanation for
not hiring Perfetti was that an internal candidate, 41-year-old Thomas Jarecki, filled the position first. As a result, the human resources recruiter, Robert Chapman, never even forwarded Perfetti’s application to, or discussed the application with, the person who made the hiring decision, Ray Altmann. Tr. at 77, 173. Moreover, Altmann did not even know who Perfetti was. Tr. at 290. If Altmann did not even know that Perfetti was applying for the job, the Bank argues, how could he have denied Perfetti the job on the basis of his age?
At trial, Perfetti attempted to puncture this justification by hypothesizing a “flimflam.” To support this theory, Perfetti relied on certain irregularities in Jarecki’s promotion. The Bank did not have any computer-generated documentation to specify the date on which Jarecki took the job. At first, the only documentation regarding Jarecki’s promotion the Bank could produce was a hand-written note from Chapman. Then, a month before trial, the Bank produced a type-written memo from Altmann that Altmann had not remembered having written. And last, two days before trial, the Bank produced a typed evaluation form which included references to the time Jar-ecki spent in the facilities coordinator position. Perfetti also noted that Jarecki remained in the job for only a few weeks and never received the raise that ought to have accompanied the promotion. Perfetti thus argued that because Jarecki left the position so quickly, he was obviously unqualified.
Initially, Perfetti wove this highly circumstantial evidence into an hypothesis that Jarecki had never actually moved into the position; the Bank had been dissembling all along. Plaintiff’s Opening Statement, Tr. at 6. Later, Perfetti apparently recognized that there was no reason for the jury to believe that Jarecki had not filled the position; unrebutted testimony from several individuals, as well as the aforementioned note, memo, and evaluation form, were all to the contrary. Thus, Per-fetti reformulated his theory. Jarecki did move into the position, but not legitimately. Altmann promoted Jarecki as a “flim-flam” for the purpose of preventing Perfetti from getting the job. Plaintiff's Closing Statement, Tr. at 575. A few weeks later, with that goal accomplished, Jarecki vacated the position.
This theory is reminiscent of the plaintiff’s “sham” theory in
La Montagne v. American Convenience Products, Inc.,
750 F.2d 1405, 1410-11 (7th Cir.1984). La Montagne argued that his employer had discriminated against him by firing him on account of his age. The employer in that case, like the Bank in this one, justified discharging La Montagne on account of his interpersonal skills; the president simply did not get along with La Montagne. In its defense, the employer noted that it first offered La Montagne’s job to a 47-year-old man only four years younger than La Mon-tagne himself. After the 47-year-old rejected the offer, the job eventually fell to a 34-year-old. La Montagne attempted to escape the force of the initial offer by arguing that it was a “sham.” The employer offered the job to the 47-year-old, knowing that he would decline, merely to mask its true motivation, which was to hire the 34-year-old. We rejected this argument, noting that “there is no direct evidence in support of the theory, and the circumstantial evidence is too insubstantial to support it.”
Id.
at 1410-11.
Perfetti’s flim-flam is hardly more supported than La Montagne’s sham.
Alt-
mann never received Perfetti’s application. There is no evidence to indicate that Alt-mann even knew that Perfetti existed, let alone knew that he wanted the job, let alone knew Perfetti’s age, let alone wanted to deny him the job, let alone denied him the job on account of his age. To the contrary, the series of speculations that would be required to fill these gaps ends at square one, because Altmann’s testimony that he did not even know Perfetti existed was unrebutted. Nor is there any evidence to indicate that Chapman ever discussed with Altmann the possibility of Perfetti interviewing for that job, let alone discussed with Altmann the possibility of denying Perfetti the job, let alone wanted to deny Perfetti the job himself. Again, the necessary conjecture stalls at the outset, because Chapman’s testimony that he never discussed Perfetti with Altmann was unrebutted.
Not only does the evidence lend little support to Perfetti’s theory, but it also undercuts it. Perfetti places great stock in the fact that the three independent documents, all verifying Jarecki’s promotion, were tardily produced and not computer-generated. The Bank’s tardy production of these documents does not necessarily fit into Perfetti’s theory, however; these documents were entirely corroborative of the Bank’s defense and, had they been located earlier, might have justified summary judgment on this count. These are not the sort of documents the Bank would have wanted to suppress. If Perfetti is suggesting that the Bank has fabricated all three documents, we do not believe that the minor irregularities in the Bank’s production of them, without any affirmative evidence from Perfetti, can support the weight of his elaborate theory. The jury could not reasonably have disbelieved all of the Bank’s uncontradicted oral and documentary evidence,
see Jang v. Biltmore Tire Co., Inc.,
797 F.2d 486 (7th Cir.1986) (upholding directed verdict for employer when plaintiff’s only argument was that he was qualified and that jury could have disbelieved employer’s proffered justification because it had produced no supporting documentation), nor could the jury reasonably have inferred from disbelieving all that evidence that the “flim-flam” actually occurred. The circumstances under which Jarecki vacated the facilities coordinator position also undercut Perfetti’s theory. According to Perfetti’s theory, Jarecki moved into the job to foreclose Perfetti, presumably to allow the Bank eventually to hire a younger candidate. But the uncontradicted evidence indicates that no younger candidate ever did fill the position. After Jarecki left the job a few weeks later, the Bank simply eliminated it and merged its responsibilities into another position. Tr. at 290. We hold that no reasonable jury could have inferred from the slim and ambiguous evidence adduced by Perfetti that the Bank’s proffered justification was pretextual.
B. The Operations Analyst Position
To fill the operations analyst position, the Bank posted a “Job Opportunity Bulletin” describing the position and its requirements. The bulletin stated that “strong interpersonal skills” were necessary “to function successfully in this position.” Plaintiff’s Ex. 12. The position required strong interpersonal skills because it involved consulting with numerous employees within the firm. Tr. at 199. Evaluation of candidates’ interpersonal skills fell on Maria McSweeney, who worked for the head of operations for the consumer bank area. McSweeney interviewed Perfetti for almost an hour. After the interview, she noted on his Job Opportunity Request Form:
Work experience appropriate. Consensus opinion was that other candidates had skill/experience/education packages that might be preferable. Candidate will not be pursued further at this time.
Plaintiff’s Ex. 23. At trial, McSweeney testified that she decided not to hire Perfet-
ti because another candidate, Rose Carter, had a superior "skill, experience, education package[].” Tr. at 196. When pressed, McSweeney specified that based on her hour-long interview with Perfetti, she concluded that he did not have the excellent interpersonal skills necessary for the position, as specified in the original bulletin. She testified that an internal consulting job requires the consultant to be a “team player.” McSweeney decided that Perfetti was not a team player based on his “whole communication skills and demeanor throughout the interview,” Tr. at 209, as well as a particular remark he had allegedly made. According to McSweeney, Per-fetti told her that he had once had a disagreement with a vice president; Perfetti insisted that the vice president had been wrong. Tr. at 200.
Two witnesses corroborated McSwee-ney’s testimony. The manager of the Bank’s career counseling center, Nancy Seever, testified that Perfetti was acting “like a jerk” at the time and had an “attitude problem.” Tr. at 429, 431. His attitude was that the bank owed him a job. Tr. at 429. A human resource recruiter for the Bank, Robert Chapman, testified that “at that point in time Mr. Perfetti was somewhat angry, hostile, came across being somewhat egocentric.” Tr. at 190. These witnesses also supported McSwee-ney’s testimony regarding Perfetti’s unwise remark. Seever testified that Perfetti often discussed his interviews with her. Perfetti once told her that in a particular interview, he had told the interviewer that if he ever disagreed with a manager in the office, he would do things his way. Tr. at 427. In addition, Chapman testified that McSweeney had related the substance of this remark to him within a few days of the interview. Tr. at 554-55.
To impeach McSweeney’s justification of “interpersonal skills,” Perfetti argues that, based on two former co-workers’ testimony, he did in fact have good interpersonal skills. Tr. at 140-44; 490-93. Perfetti also casts doubt on the veracity of McSwee-ney’s justification with his testimony that he never made the offending remark to her. And last, Perfetti argues that McSweeney’s testimony was inherently incredible. Aside from some clearly insufficient contentions,
this evidence constituted the entirety of Perfetti’s case regarding the operations analyst position.
The testimony that Perfetti had good interpersonal skills utterly failed to rebut McSweeney’s proffered justification. The other employees’ opinions about Perfetti concerned his interpersonal skills in general and over time. But McSweeney based her decision on a particular interview and a particular point in time. Two other witnesses testified that, at that particular time, Perfetti was bitter, disaffected, and acting like a “jerk.” The plaintiff offered no testimony to rebut this testimony. All that remains is testimony about the plaintiff’s qualities in the past. Such testimony does not prove that reasons based on the present are pretextual.
See Jang,
797 F.2d at 490 (even if jury could find that employee’s past performance was good, that finding would not rebut employer’s justification that “changing business and changing job description had outgrown” employee’s capabilities). In addition, other employees’ impressions of Perfetti were never communicated to McSweeney; she testified that
she never spoke to Perfetti’s co-workers about his interpersonal skills, Tr. at 210, and neither of these witnesses had ever related their opinions to McSweeney. Tr. at 147-48, 497. The only person who communicated her opinion to McSweeney was Nancy Seever, who tried to convince McSweeney that Perfetti was a nice guy. Seever did this, however,
precisely because
she recognized that Perfetti was behaving poorly at that time. Although testimony regarding the plaintiffs abilities may sometimes be relevant to the question of pretext,
Graefenhain v. Pabst Brewing Co.,
827 F.2d 13, 18-19 (7th Cir.1987), in this case the co-workers’ testimony failed to rebut McSweeney’s conclusion that Perfetti made a poor impression at the time. And even if McSweeney’s conclusion was mistaken, a mistaken business decision is not a pretext.
Pollard v. Rea Magnet Wire Co., Inc.,
824 F.2d 557, 559 (7th Cir.),
cert. denied,
484 U.S. 977, 108 S.Ct. 488, 98 L.Ed.2d 486 (1987).
Perfetti also testified that he did not make the offending remark to McSweeney. Tr. at 544-45. Even if we resolve this credibility dispute in favor of Perfetti, as we must, this scintilla of evidence lends little support to the verdict. McSweeney testified that she made her decision based on the entire interview with Perfetti, not on a single remark. Two witnesses other than McSweeney, a career counselor and a human resource recruiter within the Bank, also testified that Perfetti had a bad attitude at that particular point in time. This testimony is unrebutted and is sufficient to justify McSweeney’s hiring decision.
See La Montagne,
750 F.2d at 1415 (holding that reasonable jury could not have found that employer’s proffered justification regarding interpersonal skills was pretextual when president testified without rebuttal that he was annoyed with plaintiff). Even if Perfetti did not make the offending remark, no reasonable jury could have concluded that Perfetti had not genuinely made a poor impression on McSweeney.
All that remains of Perfetti’s case is his general assertion that he proved pretext because McSweeney's testimony “lacks credibility.” McSweeney may have appeared somewhat evasive at trial. But to avoid a directed verdict or a JNOV, “[a] plaintiff must do more than merely argue that the jury might have chosen to disbelieve all of the defendant’s evidence.... [A] plaintiff must offer substantial evidence to support the argument.”
Jang,
797 F.2d at 490;
see also
Wright & Miller,
Federal Practice and Procedure
§ 2527, at 563 (1971) (A party cannot meet its burden of proof “by relying on the hope that the jury will not trust the credibility of the witnesses- There must be some affirmative evidence_”). In some circumstances, of course, a decision-maker’s testimony may be so riddled with blatant inconsistencies or negative admissions that it could lend much support to a verdict for the plaintiff. If at the time of the adverse employment decision the decision-maker gave one reason, but at the time of the trial gave a different reason which was unsupported by the documentary evidence, the jury could reasonably conclude that the new reason was a pretextual after-the-fact justification.
E.g., Graefenhain,
827 F.2d at 16, 21 (supervisor’s unsupported attack on Graefenhain’s qualifications). In this case, however, McSweeney did not bring up a new justification for refusing to hire Per-fetti. At the time of his interview, she noted that his “skill/experience/education” package was inferior to that of other candidates, and specified to Seever that Perfetti had interviewed poorly. She simply elaborated on that justification at trial. A jury may also base a finding of pretext on a decision-maker’s testimony if that testimony is internally inconsistent or contradicts other aspects of the employer’s case.
E.g., id.
at 16, 22. In Perfetti’s case, McSwee-ney did not remember every detail of her interview with Perfetti. But the absence of these minor details is very different from an outright consistency. Finally, if a decision-maker made damaging admissions at trial, these could certainly support a finding of pretext.
E.g., Tye v. Polaris Joint Voc. Sch. Dist. Bd. of Educ.,
811 F.2d 315, 319 n. 2 (6th Cir.) (decision-maker admitted that he did not have any reasons for adverse employment decision, and that
proffered justification had been constructed for litigation),
cert. denied,
484 U.S. 924, 108 S.Ct. 285, 98 L.Ed.2d 246 (1987). The Bank’s employees’ testimony contained no such admissions, however. Indeed, the testimony of the Bank’s employees contained no post hoc rationalizations, inconsistencies, or admissions that would support the verdict. Perfetti’s only affirmative evidence was his denial of one detail in McSweeney’s testimony. We hold that no reasonable jury could have found the Bank’s justification pretextual based on this scintilla of evidence.
In so holding, we recognize the ease with which employers may use subjective factors to “camouflage discrimination.”
Namenwirth v. Bd. of Regents of U. of Wis. System,
769 F.2d 1285, 1243 (7th Cir.1985), ce
rt. denied,
474 U.S. 1061, 106 S.Ct. 807, 88 L.Ed.2d 782 (1986). To prevent employers from masking discrimination behind an incantation of subjective factors, we have held that, under many circumstances, subjective factors may be insufficient to justify the employer’s hiring decision. Perfetti has not presented sufficient evidence or authority to avail himself of any of those circumstances, however. Per-fetti did not present evidence to indicate that the Bank’s requirement of excellent interpersonal skills was disingenuous or inconsistently applied.
Cf. Coston v. Plitt Theatres, Inc.,
831 F.2d 1321, 1326 (7th Cir.1987) (upholding jury’s finding that proffered justification of seniority was pre-textual when employer presented no evidence that it actually had seniority system or that system would produce claimed result),
vacated on other grounds,
486 U.S. 1020, 108 S.Ct. 1990, 100 L.Ed.2d 223 (1988);
Christie v. Foremost Ins. Co.,
785 F.2d 584, 586-87 (7th Cir.1986) (upholding jury’s finding that proffered justification of “standard policy” was pretextual when plaintiff presented evidence that employer’s own managers did not know of policy and that policy would not have produced claimed result). Nor did Perfetti adduce evidence that McSweeney could and should have evaluated Perfetti’s interpersonal skills by objective criteria.
Cf. Christie,
785 F.2d at 586-87 (upholding jury verdict of pretext when employer relied on subjective judgment about plaintiff’s capabilities even though proffered explanation was objective and plaintiff was objectively superi- or to employee who was not fired). Last, Perfetti did not present evidence that McSweeney’s decision was arbitrary because of a failure to investigate more thoroughly. Cf
. Tye,
811 F.2d at 320 (“Where a decision-maker does not possess any information on which to choose between two individuals, ... [wjhere
no ...
attempt is made [to discover such information], and [where] the employer gives evasive and contradictory testimony regarding the choice,” plaintiff may be able to prove pretext). Granted, McSweeney did not explore Perfetti’s application in great detail. But Perfetti does not argue that an interviewer must explore an applicant’s credentials thoroughly, even if that applicant has made a poor impression in his or her interview. Nor does Perfetti cite any authority for that proposition.
If Perfetti had adduced any of the above types of evidence, his jury verdict might have sufficient evidentiary support. Of course, in requiring such evidence we are not back-handedly requiring ADEA plaintiffs to bring evidence tying the employment decision to an impermissible consideration of age. Many of the types of evidence that Perfetti lacks have no connection to age. Perfetti’s verdict might be supported if the record indicated that the job requirement of interpersonal skills was disingenuous or inconsistently applied; that McSweeney could and should have investigated his application objectively or more thoroughly; that the testimony of the Bank’s employees was inconsistent with their deposition testimony, with one another’s testimony, or with the documentary evidence. Evidence like this is the weakest form of evidence in an employment discrimination case; it is circumstantial and goes only to the question of pretext. Nevertheless, we reiterate that evidence of this nature may be sufficient to support a jury verdict. Because even the cumulative effect of Perfetti’s circumstantial .evidence of pretext is insubstantial, however, no rea
sonable jury could have found McSweeney unentitled to rely on a personal interview in choosing the best person for the position.
A contrary result would make personal interviews an extraordinarily hazardous hiring tool. It is hard to imagine a human being so heinous that he or she could not muster a single witness to testify to his or her favorable interpersonal skills. In Per-fetti’s case, the only evidence beyond that testimony is Perfetti’s uncorroborated statement contradicting a portion of what happened in the personal interview; McSweeney’s asserted lack of credibility is not, of course, evidence. If Perfetti’s self-serving contradiction of a single detail of McSweeney’s testimony could support the verdict, then it would be virtually impossible for a judge to direct a verdict or to grant JNOV in an unworthy of credence case. And, as explained above, it would be similarly impossible to grant summary judgment for the employer in such a case. We refuse to hold that every “unworthy of credence” claim must go to the jury.
Perfetti’s case is very similar to
Namenwirth v. Bd. of Regents of U. of Wis. System,
769 F.2d 1235 (7th Cir.1985),
cert. denied,
474 U.S. 1061, 106 S.Ct. 807, 88 L.Ed.2d 782 (1986). In that case, the magistrate found that the plaintiff, an assistant professor, was qualified for tenure. But the university denied the plaintiff tenure on the ground that other candidates were even more qualified because they had a greater academic potential. We held that the university’s justification was not pretextual. When a number of qualified persons apply for a few open positions, mere qualification does not mean that the applicant must get the job. And in choosing between different candidates, all of whom are qualified, an employer may legitimately use subjective qualifications.
Id.
at 1242. Although allowing the employer to base employment decisions on subjective factors “would ordinarily defeat the purpose of the discrimination laws,” we recognized that “in the case of tenure decisions we see no alternative.” Tenure decisions are traditionally and necessarily subjective.
Id.
at 1243. Thus, we upheld the entry of judgment for the university.
See also Nellis v. Brown County,
722 F.2d 853, 860 (7th Cir.1983) (upholding reliance on subjective factors when process “inherently subjective”). Like
Namen-wirth,
Perfetti had the objective qualifications for the position, but so did another applicant, Rose Carter. Also like
Namen-wirth,
the position for which he applied required subjective judgments. McSwee-ney chose between two qualified candidates based on a personal interview with them. The jury could not reasonably have considered this business decision pretextual.
Conclusion
Because Perfetti did not present substantial evidence that the Bank’s proffered justifications for not hiring him were unworthy of credence, we hold that no reasonable jury could have found the Bank guilty of age discrimination. Accordingly, the judgment is Reversed. We Remand for entry of judgment in the Bank’s favor.