This appeal by Richard J. Kautz from the District Court's order granting Met-Pro Corporation's motion for summary judgment in an age discrimination case requires us to decide whether Kautz met his burden of proving that his employer's reasons for laying him off, in a reduction in force situation, were pretextual.

Kautz presents no direct evidence of age discrimination. His claim must, therefore, be analyzed under the burden shifting framework provided by McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S. Ct. 1817, 36 L. Ed. 2d 668 (1973). In Stanziale v. Jargowsky, 200 F.3d 101 (3d Cir.2000), we explained this burden shifting framework in the context of an Age Discrimination in Employment Act ("ADEA"), 29 U.S.C. §§ 621-634 (2000), claim:

A plaintiff must first produce evidence sufficient to convince a reasonable factfinder as to all of the elements of a prima facie case of discrimination. If a plaintiff establishes a prima facie case, "`[t]he burden of production (but not the burden of persuasion) shifts to the defendant, who must then offer evidence that is sufficient, if believed, to support a finding that the defendant had a legitimate, nondiscriminatory reason for the [adverse employment decision].'" An employer need not prove, however, that the proffered reasons actually motivated the [employment] decision. If a defendant satisfies this burden, a plaintiff may then survive summary judgment by submitting evidence from which a factfinder could reasonably either (1) disbelieve the employer's articulated legitimate reasons; or (2) believe that an invidious discriminatory reason was more likely than not a motivating or determinative cause of the employer's action.

Id. at 105 (citations omitted).

The District Court held that Kautz established a prima facie case of discrimination under the ADEA and the Pennsylvania Human Relations Act ("PHRA"), 43 P.S. §§ 951-963 (1991). The Court determined that Met-Pro met its burden of going forward with the evidence by establishing legitimate nondiscriminatory reasons for his termination and that Kautz failed to establish that Met-Pro's reasons were pretextual. We will affirm.

This dispute arose when Met-Pro laid Kautz off from his position as a regional sales manager ("RSM") during a reduction in force which cut back the number of RSMs from six to five. Kautz was laid off by Met-Pro on February 20, 2002 after he had worked for the company as an RSM since 1987. He had just turned 64.

Met-Pro manufactures and sells industrial pumps. In October 2001, Met-Pro consolidated two of its divisions: Fybroc and Dean Pump. Kautz had previously been one of four RSMs for Dean Pump. With the consolidation, he became one of the six RSMs in charge of Fybroc and Dean Pump and his account responsibilities were revised, as were those of the other RSMs. He was assigned to the Southwest Region. Attrition, rather than layoffs, provided the vehicle for this consolidation.

In August 2001, Kautz was told that he would have to transfer from Houston, Texas to Telford, Pennsylvania to work in an office in Met-Pro's Fybroc plant located there. At this time, the five other RSMs all worked from factory locations and Kautz was the only RSM who worked from his home. Met-Pro agreed to pay Kautz's relocation expenses. Kautz gave his assent to the transfer but did not actually begin work from the new location until January 5, 2002. In February 2002, Met-Pro decided to reduce the number of RSMs from six to five because further consolidation of the sales force was necessary.

Met-Pro asserts that it decided to lay off Kautz after two statistical comparisons of the RSMs and then, after narrowing the field of possible candidates to two, a comparison of the candidates personnel files. Kautz asserts that these reasons were pretextual. We examine each of Met-Pro's proffered reasons in detail below and, therefore, will not recount them here.

When laid off, Kautz signed an agreement for a severance package which allowed him to receive 13 weeks of severance pay. The agreement provided that Met-Pro had no obligation to re-employ him. Subsequent to Kautz being laid off, two other RSMs (ages 30 and 43) were fired for cause and replaced by David Hakim, age 33, and Christopher Cousart, age 47. Kautz was not notified about these job openings or considered for either position. When these openings became available he was, and still is, working for Kirkwood Company, one of Met-Pro's distributors. His salary at this new position is significantly less than the salary he earned at Met-Pro. Met-Pro continues to employ only five RSMs.

Subject matter jurisdiction over Kautz's claims under the ADEA arises pursuant to 28 U.S.C. § 1331. We exercise pendent jurisdiction over Kautz's claims arising under the PHRA pursuant to 28 U.S.C. § 1367.1 We have jurisdiction over the appeal pursuant to 28 U.S.C. § 1291.

The standard of review applicable to the District Court's order granting summary judgment is plenary. Carrasca v. Pomeroy, 313 F.3d 828, 832-833 (3d Cir.2002). We must apply the same test employed by the District Court under Federal Rule of Civil Procedure 56(c). Id. Accordingly, the District Court's grant of summary judgment was proper only if it appears "that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law." Rule 56(c), Federal Rules of Civil Procedure. Kautz, as the non-moving party, is entitled to every favorable inference that can be drawn from the record. Carrasca, 313 F.3d at 833.

The District Court's conclusion that Kautz has "made out a prima facie case," Kautz v. Met-Pro Corp., No. Civ. A. 02-CV-8610, 2004 WL 1102773, at *3 (E.D.Pa., May 17, 2004), is not disputed by Met-Pro on appeal. The sole issue of contention in this appeal, therefore, is whether Kautz has succeeded in creating an issue of fact as to whether Met-Pro's proffered non-discriminatory reasons for eliminating Kautz's position are a pretext.

In Reeves v. Sanderson Plumbing Products, Inc., 530 U.S. 133, 120 S. Ct. 2097, 147 L. Ed. 2d 105 (2000), the Court held that proof of pretext does not have to include evidence of discrimination, but rather "[i]n appropriate circumstances, the trier of fact can reasonably infer from the falsity of the explanation that the employer is dissembling to cover up a discriminatory purpose." 530 U.S. at 147, 120 S. Ct. 2097.

Although Reeves makes clear that we may not require affirmative evidence of discrimination in addition to proof of pretext, it does not change our standard for proving pretext which "places a difficult burden on the plaintiff." Fuentes v. Perskie, 32 F.3d 759, 765 (3d Cir.1994). In order to avoid summary judgment, Fuentes requires a plaintiff to put forward "such weaknesses, implausibilities, inconsistencies, incoherencies, or contradictions in the employer's proffered legitimate reasons for its action that a reasonable factfinder could rationally find them unworthy of credence." Id. (internal quotation and citation omitted; emphasis in the original).

Fuentes further explains that "to avoid summary judgment, the plaintiff's evidence rebutting the employer's proffered legitimate reasons must allow a factfinder reasonably to infer that each of the employer's proffered non-discriminatory reasons ... was either a post hoc fabrication or otherwise did not actually motivate the employment action." Id. at 764 (emphasis in the original); see also Logue v. Int'l Rehab. Assocs., Inc., 837 F.2d 150, 155 (3d Cir.1988) (holding that "the district court erred in failing to consider all of [the employer's] proffered evidence of legitimate business reasons for [the plaintiff's] termination"). In a footnote, Fuentes allowed for the possibility that in a case where a "defendant proffers a bagful of legitimate reasons," casting "substantial doubt on a fair number of them ... may impede the employer's credibility seriously enough so that a factfinder may rationally disbelieve the remaining proffered reasons." Id. at 764 n. 7; see also Sheridan v. E.I. DuPont de Nemours & Co., 100 F.3d 1061, 1074-1075 (3d Cir.1996) (en banc) (determining that where a plaintiff: (1) completely undermined the employer's main complaint by proving that she was on jury duty on several days when she was alleged to have given out free drinks; (2) called into question the credibility of central employer witnesses; and (3) presented affirmative evidence of retaliatory bias against her, there had been a showing of pretext and there was, in that situation of overwhelming evidence, no need to discuss whether pretext had been shown for each allegation that the employee was tardy or in violation of the grooming policy).

Fuentes instructs that pretext is not shown by evidence that "the employer's decision was wrong or mistaken, since the factual dispute at issue is whether discriminatory animus motivated the employer, not whether the employer is wise, shrewd, prudent, or competent." Id. at 765.

We have applied the principles explained in Fuentes to require plaintiffs to present evidence contradicting the core facts put forward by the employer as the legitimate reason for its decision. See Stanziale, 200 F.3d at 106 (upholding summary judgment where the plaintiff attempted to show pretext by disputing the importance of the difference in educational qualifications between himself and the person hired rather than challenging the disparity itself or proving that the qualifications at issue bore no actual relationship to the employment being sought); Keller v. Orix Credit Alliance, Inc., 130 F.3d 1101, 1110 (3d Cir.1997) (en banc) (determining that summary judgment was appropriate notwithstanding the plaintiff's contention that his failure to meet or approach his goal of raising $1.5 billion in financing was due to factors beyond his control, stating that "the relevant question is not whether Keller could have done better; instead, the relevant question is whether the evidence shows that it was so clear that Keller could not have done better that ORIX Credit Alliance could not have believed otherwise").

An employer may not use evaluating criteria which lacks any relationship at all to the performance of the employee being evaluated because to do so would be inconsistent with and contradictory to the employer's stated purpose. See Fuentes, 32 F.3d at 765. Absent this type of violation of the Fuentes standard, we will not second guess the method an employer uses to evaluate its employees. See Simpson v. Kay Jewelers, Div. of Sterling, Inc., 142 F.3d 639, 647 (3d Cir.1998) ("Whether sales quotas or evaluation scores are a more appropriate measure of a manager's performance is not for the court (or factfinder) to decide."); Keller, 130 F.3d at 1109 ("The question is not whether the employer made the best or even a sound business decision; it is whether the real reason is discrimination."); Healy v. New York Life Ins. Co., 860 F.2d 1209, 1216 (3d Cir.1988) ("[O]ur inquiry must concern pretext, and is not an independent assessment of how we might evaluate and treat a loyal employee."); Logue, 837 F.2d at 155 n. 5 ("[O]ur task is not to assess the overall fairness of [the] ... employer's actions.").

Determining pretext is a fact-based inquiry. See Simpson, 142 F.3d at 646. We must, therefore, look carefully at each of Met-Pro's proffered reasons as well as Kautz's claim of pretext regarding each of these reasons. See Ezold v. Wolf, Block, Schorr and Solis-Cohen, 983 F.2d 509, 524-525 (3d Cir.1992) (concluding that a district court is obligated to focus on the employer's articulated reasons and citing Logue in support of this conclusion).

Met-Pro offered two groups of reasons for its decision to eliminate Kautz's RSM position after it had decided to downsize from six to five sales regions. First, Met-Pro analyzed two different sales statistics and on the basis of that analysis determined that Kautz and one other RSM were the lowest performers.

Second, upon reviewing the files of Kautz and the other RSM selected for further review, Met-Pro determined that there was a record of several specific instances of inadequate performance by Kautz: (1) a June 18, 2001 memo from Edward Murphy, Kautz's immediate supervisor, related to poor performance on a specific project; (2) a December 12, 2001 memo from Murphy faulting Kautz for the loss of a $38,000 job due to his "serious failure in judgment;" (3) a February 13, 2002 memo by James Board, vice president and general manager of the Fybroc and Dean Pump divisions, that memorialized a complaint by Gary Cauble, president of one of Met-Pro's distributors, that, after receiving adequate advanced notice of a job, Kautz had failed to inform the distributor of Met-Pro's desire to bid on it until the morning the bid was due, which was too late; (4) a January 16, 2002 memo from Board asserting that, at a year-end review meeting, Kautz "offered that Dean Pump had lost a $40,000 quote;" and (5) January 18, 2002 memos from both Board and Murphy criticizing Kautz for his inability to present any details about what was going on in his territory during a weekly production meeting. There was no record of comparable derelictions from the other RSM whose record was considered.2

We first address the two sets of statistics offered by the employer as evidence of substantial non-discriminatory reasons for selecting Kautz as the RSM to be let go.

The first statistic relied on in Met-Pro's decision presents the greatest problem. The statistic measured number of sales by region and the southwest region, Kautz's territory, was the lowest. Kautz does not dispute that this statistic was used or the accuracy of the numbers. He does argue that the statistic was not actually relevant to a comparison of individual RSM performance. Importantly, he supports this argument with the statement of DeHont who, in deposition testimony, admitted that the RSMs "didn't have control over these numbers." DeHont did not admit that these numbers were irrelevant but he did concede that the other statistical method used was "the more relevant one."

In Board's deposition testimony, he admitted that this statistic showed "territory snapshots ... not snapshots of individual performance." He went on, however, to explain that even though territories had been swapped and things had been shifted around, he, knowing where everybody was and where they had been, was able to evaluate individual performance based on this statistic. Even in view of Board's attempt to explain the relevance of this statistic, we determine that Kautz has put forward evidence which creates a dispute of fact on this issue. DeHont's testimony that the RSMs "didn't have control over these numbers" would give a jury a reasonable basis for concluding that the statistical method used was not at all relevant to Met-Pro's purported purpose: individual evaluation of RSM performance.

Here, the Fuentes standard has been satisfied by Kautz because it is implausible, inconsistent, incoherent and contradictory for an employer to use a method of evaluation that has nothing to do with individual performance in order to measure individual performance. 32 F.3d at 765.

The second statistical method requires a closer analysis. This statistical method compares two years of what are described as booking numbers, raw sales numbers computed at the time of sale that are not adjusted to later variances or profit margins, and then set forth the percentages between the two years. The figures compare booking numbers in fiscal 2002 with fiscal 2001 for each RSM.3

Bookings for fiscal year 2002 were measured as a percentage of the RSM's bookings for 2001, the previous fiscal year. Of the six RSMs, Kautz had a percentage of 84.73%, the second lowest. This means that in the fiscal year of 2002 he booked only 84.73% of what he had booked in the previous fiscal year. John Chenault was the lowest, with 76.9%. Here again, Kautz does not dispute that this statistic was used nor does he dispute the accuracy of the numbers, but asserts that the methodology itself was pretextual.

Kautz makes much of the circumstance that the number was handwritten on the bottom of the page. He also asserts that this ratio does not appear on any of Met-Pro's computer runs and that a comparison of two years of production by percentage had never been used before to evaluate RSM performance.

With each of these assertions Kautz fails to create a question of fact on the issue of pretext. The handwritten notation at the bottom of each RSM's bookings chart simply tabulates the booking numbers from a portion of the year—from October through January—and recalculates the ratio already set forth in the computer printout. Kautz actually fares better when the full fiscal year is taken into account. He goes from 80.88% in the computer generated partial year comparison to 84.73% in the full year comparison. At the same time, the percentages of some of the other RSMs are reduced by the added handwritten notation.

Kautz does not explain what he means by his contention that this percentage does not appear on any of Met-Pro's computer runs. If his assertion is that this number only appears as a handwritten notation, it is demonstrably false. The computer printout for each RSM, marked exhibit 4, calculates "% versus Prior Year" for both shipments and bookings. In any event, Kautz provides no citation to the record for his claim that this statistic is something not normally calculated.

Similarly, Kautz provides no supporting evidence for his claim that this statistical method had never been used before to evaluate the performance of RSMs. Even if there was some evidentiary support for the allegation, a naked assertion that a method of evaluation is new would not, by itself, support a finding that it is implausible, inconsistent, incoherent or contradictory. See Fuentes, 32 F.3d at 765.

Kautz relies heavily on the contention that Met-Pro should have looked at total bookings rather than the comparison between the two fiscal years. He asserts, correctly, that he was the second highest in total bookings for the 2001 and 2002 fiscal years. This argument fails because it is axiomatic that the mere fact that a different, perhaps better, method of evaluation could have been used is not evidence of pretext unless the method that was used is so deficient as to transgress the Fuentes standard. See Simpson, 142 F.3d at 647; Keller, 130 F.3d at 1109; Healy, 860 F.2d at 1216; Logue, 837 F.2d at 155 n. 5.

Moreover, Kautz has not shown Met-Pro's focus on comparative rather than total booking numbers to be implausible, inconsistent, incoherent or contradictory. See Fuentes, 32 F.3d at 765. Kautz's superiors at Met-Pro gave reasonable business-oriented reasons why they felt that total booking numbers were not the best basis for comparison. Kautz has not rebutted the logic of these explanations and has not even offered a sound basis for his own conclusion that total booking numbers are a better method of evaluation. He failed to rebut the basic premise that total bookings is a faulty indicia of performance, because the employer's marketing regions are not fungible in the sense that each region has the same sales potential. Specific unrebutted evidence was presented by Met-Pro that the marketing potential varied from region to region.

Even if Kautz had succeeded in showing that total bookings would provide a better basis for comparing the RSMs, that would not be enough. Evidence that the method of evaluation an employer used was not the best method does not amount to evidence that the method was so implausible, inconsistent, incoherent or contradictory that it must be a pretext for something else. See Simpson, 142 F.3d at 647; Keller, 130 F.3d at 1109; Fuentes, 32 F.3d at 765; Healy, 860 F.2d at 1216; Logue, 837 F.2d at 155 n. 5.

We now turn to the contention that the two-year percentage methodology was skewed deliberately and intentionally to discriminate against Kautz because of his age. The argument seems to be that people who have been on the job for a long time will be more likely to have reached their full sales potential and have fairly stable sales numbers. These veterans will be more affected by a down turn in the market, like the one that apparently happened in the industrial pump industry post 9/11/2001, than newer RSMs who are more likely to improve from year to year, and therefore compensate for the market downturn with improved performance.

This argument is riddled with assumptions and lacking evidentiary support in the record. The argument assumes that more experienced RSMs will have higher total bookings and that older RSMs are the most experienced. Next, the argument assumes that it takes a long period of time for an RSM to reach his or her sales potential, as measured by total bookings, and that continued improvement is not a viable or expected goal.

By its very nature a sound legal argument must contain a conclusion supported by evidentiary premises. It is a formal inference in which the conclusion is arrived at and affirmed on the basis of one or more propositions, which are accepted as the starting point of the process. Its key is the reasonable probability that the conclusion flows from evidentiary datum because of past experience in human affairs. The passage cannot be made by mere speculation, intuition or guessing.

The record shows that Kautz had been an RSM since 1987; we are not told how long the other RSM's held their positions, and it is quite possible that some have been working for Met-Pro as long or longer.

Kautz relies on the the testimony of David Gutt, one of the RSMs fired for cause after Kautz had been laid off. Gutt identified himself as inexperienced, and said that Gene Silvers, another RSM, did not have the same experience as Kautz. He identified Kautz, John Chenault and Lloyd Hill as "veteran sales guys."

This information does not support the argument that using a bookings comparison rather than total bookings is deliberately skewed to prejudice older workers. Gene Silvers, an inexperienced RSM and 31 years-old, had the fourth best bookings comparison percentage. Ronald Aceto, whose length of experience is not revealed on our reading of the record and was 42 years-old, had the second best bookings comparison percentage. This example casts doubt on the premise that longevity as an RSM inevitably produces a higher and more static production level.

The explanation offered by both Board and DeHont for the use of the comparison percentages rather than some other measurement is important. These corporate officers explained that bookings expectations depend on the market conditions and other factors in the area. One RSM might have higher raw booking numbers than another but not be performing as well because of the greater potential of his territory. Board explained that booking 2 million in a territory that would typically generate 4 million would be poor performance compared to booking 1.5 million in a territory that would typically generate 1.25 million. DeHont made a very similar statement in his testimony using different numbers.

Board specifically stated: "You have to compare [booking numbers] to historical data coming out of that same region to judge whether the performance is better or worse than you would expect based on historical data and based on current economical [sic] conditions." Moreover, Board testified that bookings numbers from the previous year are a key factor in setting the budget for a particular region.

Kautz's contention that the discrimination-free test for productivity of RSMs is simply to tabulate total bookings by region is demonstrably faulty because market demand varies by region. Instead, Met-Pro decided that, in a falling market that required a reduction in force, the better method was a comparison of bookings in 2002 to bookings in 2001. This, Met-Pro reasoned, would eliminate market demand differences that existed region by region, and instead show a more neutral and more accurate picture of sales production of the six RSMs.

Kautz offered no evidence that this method discriminated against him because of his age. The idea that Met-Pro discriminated against veteran sales managers because older men reach a plateau in sales production is totally devoid of proof. The record contains no empirical data of the existence of this plateau, no evidence of reports or studies supporting this theory, no expert witnesses. It is a thesis totally devoid of data to substantiate it.

To avoid summary judgment, the teachings of Fuentes require a plaintiff to put forward "such weaknesses, implausibilities, inconsistencies, incoherencies, or contradictions in the employer's proffered legitimate reasons for its action that a reasonable factfinder could rationally find them unworthy of credence." 32 F.3d at 765 (internal quotation and citation omitted; emphasis in the original). Kautz failed to meet this burden.

To the extent that Kautz relies on the teachings of Showalter v. University of Pittsburgh Medical Center, 190 F.3d 231 (3d Cir.1999), and Potence v. Hazleton Area School District, 357 F.3d 366 (3d Cir.2004), we conclude that the facts in those cases are fundamentally dissimilar to the facts before us and, therefore, do not serve as proper analogues.

In Showalter, the employer determined that it would lay off the security guard with the least seniority in a downsizing situation. 190 F.3d at 237. The plaintiff proffered evidence that the employer was presented with three different methods of determining seniority-seniority in the job, department or hospital. Id. at 233. Plaintiff also offered substantial proof that before the employer decided which method to use, it knew which employee would be affected by each test. Id. at 237-238 ("A reasonable factfinder could conclude that [the employer] had the discretion to choose any of the three forms of seniority; that he knew in advance the result that each choice would produce; and that he selected department seniority because he knew it would result in the layoff of the oldest employee, Showalter."). The employer countered that "as far as he knew, this form of seniority was always used in a RIF." Id. at 237. We reversed the District Court's grant of summary judgment because "[b]ased on this evidence, a reasonable factfinder could find" that the decision maker's explanation "was pretextual." Id.

In the case at bar, there is no evidence that Board and DeHont, the corporate officers who made the decision to lay off Kautz, considered different methods of statistical analysis for evaluating the RSMs knowing in advance which RSMs would benefit under each method. Showalter depended for its holding on the employer's advanced knowledge that the method selected would result in the termination of the oldest employee. Id. In Potence, a major issue was whether the employer's requirement of refrigerant certification was a pretext. 357 F.3d at 370. The plaintiff had proffered evidence that refrigerant certification was not listed in an advertisement for the position which listed two other certification requirements, plaintiff had been given several different explanations of why he was not hired, the refrigerant exam was given to him at the end of the hiring process after other candidates had already been interviewed and several ageist remarks had been made by a person involved in the hiring process. Id. at 370-371. This evidence created a question about the plausibility, consistency and coherence of the refrigerant certification requirement. As we have explained, Kautz has not succeeded in creating a similar question concerning Met-Pro's use of the bookings comparison percentage.

Accordingly, the facts in these two cases indicated a discriminatory animus not present in the case at bar.

We turn now to Met-Pro's second set of proffered nondiscriminatory reasons for its selection of Kautz as the RSM to lay off in the reduction in force circumstances. This involved the comparison between the personnel files of Kautz and John Chenault, the RSM who scored last in the statistic comparing bookings in 2001 and 2002. It is important to note that Chenault's file undisputedly contained no complaints.

Kautz complains about the timing of the complaints against him because all the complaints specifically pointed to by Met-Pro occurred in the eight months which immediately preceded his termination. Questioning the timing of these complaints, without more, cannot suffice to establish pretext.4 Kautz must dispute the factual basis of each negative document in his file offered by Met-Pro as a basis for his termination.

Kautz also asserted that these documents were fabricated long after the events took place. But this assertion is unsupported by the record. Kautz's claim that he had periodically reviewed his file and never found any negative information about his performance fails to create an issue of fact as to fabrication of documents because he never claimed to have reviewed his file after the incidents which were the subject of more recent negative performance reviews.

In a final attempt to make a global showing of pretext for all the negative documents in his file, Kautz points to his record of basically positive performance reviews over the course of his time at Met-Pro. The attempt to use past positive performance reviews to show that more recent criticism was pretextual fails as a matter of law. See Ezold, 983 F.2d at 528 ("Pretext is not established by virtue of the fact that an employee has received some favorable comments in some categories or has, in the past, received some good evaluations.").

Turning to a specific examination of the adverse documents, we first consider the June 18, 2001 memo from Edward Murphy, whom we previously identified as Kautz's immediate supervisor. This memo faulted Kautz for depending entirely on information he was getting from his distributor and not establishing his own line of communication. This practice resulted in the loss of ability to bid successfully on a specific job because of inability to "receive any competitive pricing feedback." Kautz also, according to the memo, had similar performance problems in a subsequent project.

Kautz claims this memo is a pretext because it is inconsistent with another memo sent from Murphy to Board on the same day. Review of the memo from Murphy to Board reveals no inconsistency. Murphy's memo to Board does not directly blame Kautz but it lists many of the same problems for which Murphy faulted Kautz in the other memo: "not being aware of what the competition was;" "lack of direct contact;" "allowing a distributor to quote a job of this size." No reasonable factfinder could conclude that the second memo is inconsistent with the first and that therefore the first was a pretext. Kautz has not pointed us to any other dispute of the facts contained in this memo.

The second document in the file is a December 12, 2001 memo from Murphy. This memo faulted Kautz for the loss of a $38,000 job due to his "serious failure in judgment" regarding pricing needs of the customer and failure to be informed as to the status of the job. Kautz makes no effort to show that this memo is a pretext. Nor was it rebutted before the District Court. Kautz, 2004 WL 1102773, at *7 ("Plaintiff does not address Murphy's criticism of his performance in December of 2001.").

The third item in the file is a memo by Board memorializing the complaint of Gary Cauble, president of one of Met-Pro's largest distributors.5 Cauble also gave deposition testimony concerning this complaint. The complaint was that after receiving plenty of advanced notice of a job, Kautz had failed to inform the distributor of Met-Pro's desire to bid on it until the morning the bid was due, which was too late. Kautz attempts to show that Cauble's testimony was not believable and was therefore a pretext. He points to Cauble's company's financial interest in keeping Met-Pro happy and the fact that during Kautz's tenure with Met-Pro, Allesco's (Cauble's distribution company) bookings consistently rose. Kautz does not specifically assert that Cauble was lying about Kautz's performance, though he does say that based on the foregoing "a jury could easily disregard all of Mr. Cauble's testimony as not worthy of belief." (Br. at 27.) Kautz asks us to assume that Cauble's testimony is false because he is an interested party without coming forth with specific evidence contradicting the testimony.

Kautz points also to the testimony of David Kirkwood, another distributor who was satisfied with his performance. Quite obviously, the fact that one distributor is pleased does not in any way create a question of fact about the opinion of a different distributor related to specific instances of deficient performance. To create a question of fact, Kautz must assert that there is no factual basis for Cauble's specific complaints and then present some evidence supporting this claim. He has not done this.

The fourth document in the file is the January 16, 2002 memo from Board asserting that, at a year-end review meeting, Kautz "offered that Dean Pump had lost a $40,000 quote." Kautz did not know any of the pertinent information about this loss: "who the job was lost to;" "what level the Dean distributor had quoted the job at;" etc. He was told that this was an unacceptable lack of awareness of his territory. In order to show pretext, Kautz claims no memory of Board's criticism at the meeting or the bid itself. Lack of memory is not a denial of the truth of the memo and therefore does not show pretext; a specific denial of the truth or relevance of the employer's proffered reason is required.

The fifth document is a January 18, 2002 memo from Board that criticizes Kautz for his inability to present any details about what was going on in his territory at a weekly production meeting. A similar memo with identical date and criticism is in the file from Murphy. Kautz does not deny that these criticisms were made. He also fails to assert that he was well prepared for this meeting. Rather, he claims that he was not able to be as well prepared as the other RSMs because he, unlike them, did not have a computer sitting in front of him. An explanation of the reason he was less prepared than he should have been for the meeting does not suffice as a showing of pretext.

* * * * * *

In summary, the only proffered nondiscriminatory reason offered by Met-Pro which Kautz has succeeded in materially disputing is the sales by region statistic because of DeHont's admission that the RSMs "didn't have control over these numbers."

Our Court has held that the plaintiff must demonstrate that each of the employers proffered nondiscriminatory reasons are pretextual. Fuentes, 32 F.3d at 764. We have also said that this can be done by showing that some of the employers proffered reasons are a pretext in such a way that the employer's credibility is seriously undermined, therefore throwing all the proffered reasons into doubt. Id. at 764 n. 7. We conclude that an issue of fact as to the genuineness of one of the two statistical methods used by Met-Pro to narrow the field, when the other method is not suspect, does not create such an issue concerning Met-Pro's credibility as to cast all its proffered reasons into doubt.

Accordingly, the judgment of the District Court will be affirmed.

SLOVITER, Circuit Judge, dissenting.

I respectfully disagree with the majority's disposition of this case because I believe that the reason advanced by Met-Pro Corporation for Richard J. Kautz's termination could be found by a trier of fact to be pretextual.

Of course, an employer is, and should be, free to evaluate an employee's performance according to its business judgment. But when it uses a method of evaluation that it is inherently unreasonable, the factfinder may infer that the decision was based on some other consideration. In Aka v. Washington Hosp. Ctr., 156 F.3d 1284 (D.C.Cir.1998) (en banc), the United States Court of Appeals for the District of Columbia Circuit held that: "[i]f a factfinder can conclude that a reasonable employer would have found the plaintiff to be significantly better qualified for the job, but this employer did not, the factfinder can legitimately infer that the employer consciously selected a less-qualified candidate — something that employers do not usually do, unless some other strong consideration, such as discrimination, enters into the picture." 156 F.3d at 1294. The United States Court of Appeals for the Second Circuit has reached a similar conclusion, recognizing that "facts may exist from which a reasonable jury could conclude that the employer's `business decision' was so lacking in merit as to call into question its genuineness." Dister v. Cont'l Group, Inc., 859 F.2d 1108, 1116 (2d Cir. 1988); see also Ryther v. KARE 11, 108 F.3d 832, 840 (8th Cir.1997) (holding that factfinder was allowed to consider whether the basis purportedly relied upon by defendant in its decision to fire plaintiff was "actually a sound — as opposed to pretextual — basis upon which to make employment decisions"); Loeb v. Textron, Inc., 600 F.2d 1003, 1012 n. 6 (1st Cir.1979) ("The reasonableness of the employer's reasons may of course be probative of whether they are pretexts. The more idiosyncratic or questionable the employer's reason, the easier it will be to expose it as a pretext....").

Kautz argued that the statistical formula on which Met-Pro relied was tilted to the disadvantage of the older worker, such as Kautz. Rather than rely on the sales statistics, the traditional evaluator used by Met-Pro, which showed that Kautz's bookings for fiscal year 2001 and fiscal year 2002 were the second highest of the six regional sales managers, Met-Pro chose to rely on a statistical formula that calculated the percentage of fiscal year 2002 bookings to the 2001 bookings. Under this formula, the total amount that an employee sold was not considered; rather, the determinative figure was the difference between an employee's sales in fiscal year 2001 and fiscal year 2002. Notably, the three top earners of 2001 had the three worst percentages under the formula adopted. Met-Pro's formula confers the worst scores to the best salespersons and the best scores to the less successful salespersons. Under this formula, it is much more likely that the youngest sellers will have the highest percentages while the older, more experienced, employees will have the lowest, when 2002 sales are viewed as a percentage of those employees' 2001 sales. When use of this formula is considered in light of the bad economy which Met-Pro acknowledges was experienced in 2002, the employees most likely to have the highest percentages are those who were the least able to take advantage of the prosperous economic conditions of 2001. Not surprisingly, two of the three top earners for 2001 were also the oldest employees. This formula therefore is geared to the disadvantage of Met-Pro's older employees, such as Kautz.

A factfinder could reasonably determine that the use of this method, rather than the sales as such, was so unreasonable that it was a pretext for age discrimination. Because that flawed formula was the basis of Kautz's termination, I would reverse the grant of summary judgment.