Plaintiff-appellant, William Speen, appeals from a district court judgment as a matter of law in favor of defendants-appellees Crown Clothing Corporation, Jack Silverman, and Richard Silverman in an age discrimination and pension rights suit involving his alleged wrongful termination in violation of federal and Massachusetts law. Because Speen failed to provide sufficient evidence to support a finding that he was a Crown employee for the purposes of his federal and state statutory claims, or evidence sufficient to support his remaining Massachusetts common law tort claims, we affirm.

Speen began his career as a men's clothing salesman following his discharge from the U.S. Army in 1945. Over the ensuing twenty-seven years, Speen served as a New England sales representative for various companies. In 1972, he became a sales representative for Crown Clothing Corporation ("Crown"). Jack and Richard Silverman respectively serve as Crown's president and treasurer.

For the next twenty years, Speen travelled throughout New England as a Crown representative hawking Crown products--sports jackets, raincoats and the like--to men's clothing stores. For some of that time, Speen, with Crown's approval, also sold non-competing lines of men's clothes from other manufacturers, most notably slacks. By 1992, however, Speen's relationship with Crown increasingly soured. Speen's immediate supervisor, Jack Silverman, often expressed his dissatisfaction with Speen's declining raincoat sales. In December 1992, Crown notified Speen--first orally and then in writing--that his service would be terminated, effective the end of the month. Crown went on to replace Speen, then 71 years old, with a new representative, aged 51.

In June 1993, unwilling to accept Crown's adverse action, Speen filed a complaint with the Massachusetts Commission Against Discrimination (MCAD). After MCAD made a preliminary determination that it had jurisdiction over Speen's claim because it found him to be an employee for purposes of the federal and state age discrimination statutes, a determination that does not carry with it preclusive effect, Speen filed this suit in federal district court in November 1994. Speen's federal action claimed that his termination amounted to unlawful age discrimination under the federal Age Discrimination in Employment Act (ADEA), 29 U.S.C. §§ 621-634, and Mass.Gen.L. ch. 151B, § 9, and also sought pension rights under the Employee Retirement Income Security Act (ERISA), 29 U.S.C. § 1140. Speen, in addition, advanced a Massachusetts common law tort claim against the Silvermans, alleging a tortious interference with his advantageous business relationship with Crown.1

The matter proceeded to trial before a jury. At the conclusion of plaintiff's evidence, the defendants filed a motion for judgment as a matter of law pursuant to Fed.R.Civ.P. 50(a) on essentially two grounds. The first was that Speen was not a Crown employee, but rather an independent contractor who enjoyed no protection under the applicable statutory provisions.2 The second was that Speen had not produced sufficient evidence to allow a jury to conclude that he was maliciously discharged because of his age in violation of Massachusetts common law. The court granted the motion, ordering a judgment for the defendants on all claims. This appeal ensued.

For the reasons set forth below, we reject Speen's arguments concerning the employee/independent contractor issue and his Massachusetts common law tort claims and thus affirm the district court's decision.

We review de novo a district court's decision to grant judgment as a matter of law pursuant to Rule 50(a). In exercising that plenary review, we use "the 'same stringent standard incumbent upon the trial court in the first instance.' " Greenberg v. Union Camp Corp., 48 F.3d 22, 26 (1st Cir.1995) (quoting Favorito v. Pannell, 27 F.3d 716, 719 (1st Cir.1994)).

We thus consider the evidence and the reasonable inferences that are to be drawn from it in the light most favorable to the party opposing the motion, in this case, the plaintiff. A motion for a judgment as a matter of law "is proper at the close of plaintiffs' case only when the plaintiffs' evidence, viewed in this light, would not permit a reasonable jury to find in favor of the plaintiffs on any permissible claim or theory." Murray v. Ross-Dove Co., 5 F.3d 573, 576 (1st Cir.1993).

Crown contends that Speen cannot sue under the ADEA or the Massachusetts anti-age discrimination statute, Mass.Gen.L. ch. 151B, because, for purposes of those statutes, he is not a covered "employee," but rather an unprotected "independent contractor." Speen vigorously disputes this contention and further argues that the issue of his proper classification, in any event, was a question for the jury that should not have been decided by the district court on a Rule 50(a) motion.

Both federal and Massachusetts courts have found that the federal and Massachusetts statutes prohibiting age discrimination in employment do not reach independent contractors. See Robinson v. Overseas Military Sales Corp., 21 F.3d 502, 509 (2d Cir.1994); Hayden v. La-Z-Boy Chair Co., 9 F.3d 617, 619 (7th Cir.1993); Daughtrey v. Honeywell, Inc., 3 F.3d 1488, 1495-96 (11th Cir.1993); Oestman v. National Farmers Union Ins. Co., 958 F.2d 303, 304-05 (10th Cir.1992); Garrett v. Phillips Mills, Inc., 721 F.2d 979, 980 (4th Cir.1983); Comey v. Hill, 387 Mass. 11, 438 N.E.2d 811, 814 (1982). See generally, Francis M. Dougherty, Annotation, Who, Other Than Specifically Excluded Persons, Is "Employee" Under § 4(a)(1) Of Age Discrimination in Employment Act Of 1967 (29 USCS § 623(a)(1)), 125 A.L.R. Fed. 273, 287-89 (1995) (collecting federal cases).

The salience of the employee/independent contractor distinction in age discrimination cases thus is clear. Less easily discernible, however, are the tests federal and Massachusetts law use to distinguish a covered employee from an unprotected independent contractor.

In interpreting the Commonwealth's employment discrimination law, Mass.Gen.L. ch. 151B, Massachusetts courts use a common law test to distinguish employees who are covered by the statute from independent contractors who are not. Comey, 438 N.E.2d at 814. The parties in this case both concede this point. Where they differ is in their assessment of what factors Massachusetts courts look to in applying the common law test. Speen argues that a claimant is considered an employee under Massachusetts law if he can show that the employer enjoyed the right to control his labor. This, Speen contends, he can do. In particular, he argues he was a Crown employee because he had to call in his orders every night and fill out special Crown order forms. Crown responds that this type of activity does not make one an employee under Massachusetts law, since a mere showing of some element of control is not conclusive under the multifactored test Massachusetts courts use to determine employee status.

Speen points to older Massachusetts cases which indicate that the test of employee status is the right to control. McDermott's Case, 283 Mass. 74, 186 N.E. 231, 232 (1933) ("The exact point at issue is whether the claimant was a servant or employee, or an independent contractor. The essence of the distinction is the right of control.... Other considerations and tests are important only as they bear upon the right of control."); Khoury v. Edison Electric Illuminating Co., 265 Mass. 236, 164 N.E. 77, 78 (1928) ("Although the conclusive test of the relationship of master and servant is the right to control, other factors may be considered in determining whether the right to control exists, but they are subordinate to this primary test.").

Upon initial inspection, the language in these cases would seem to support Speen's contention that the district court misstated the relevant Massachusetts standard when it concluded that Massachusetts uses a multifactored analysis in distinguishing employees from independent contractors. A closer reading of the cases and consideration of later Massachusetts decisions, however, dispels this conclusion.

In McDermott's Case, for example, the court explained that an independent contractor is one "not subject to direction and control as to every detail of the work" to be performed. Conversely, an employee is one who "at every moment, with respect to every detail ... is bound to obedience and subject to direction and control." McDermott's Case, 186 N.E. at 232. The Khoury court explained this feature of the Massachusetts common law test as follows: "the employee must be subject to control by the employer, not only as to the result to be accomplished but also as to the means to be used." Khoury, 164 N.E. at 78.

Such language, gleaned from the decisions upon which Speen's counsel relies, indicates the great degree to which Speen and Massachusetts courts mean rather different things when they refer to "right of control" within the context of the common law test. Simply put, the level of employer control ("at every moment, with respect to every detail") necessary to conclusively establish employee status without looking to any of the additional "subordinate" factors is such that, as a practical matter, one may speak of the common law test as being a multifactored one.

Subsequent Massachusetts case law, in fact, acknowledges as much. While recognizing the vitality of the common law test in Massachusetts, for example, the Comey court explained that "[t]rial judges should carefully and fully instruct juries on all the factors that may be useful in distinguishing employees from independent contractors." Comey, 438 N.E.2d at 815. This language indicates that the common law test in Massachusetts, as in other states, while directed towards the question of right of control, involves the assessment of multiple factors. See Restatement (Second) of Agency § 220 (1957). Indeed, the Comey court immediately goes on to cite with approval federal cases which, it explains, "list[ ] factors which may distinguish employees from independent contractors." Id.

Confronted with such language, lower Massachusetts courts have proceeded on the view that

[i]n the employment context, a master-servant relationship is determined by a number of factors, including the right of the employer to control the details of the work done by the employee, the method of payment, the skill required in the particular occupation, whether the employer supplies the tools, instrumentalities and place of work, as well as the parties' own belief as to whether they are creating a master-servant relationship.

Chase v. Independent Practice Ass'n, 31 Mass.App.Ct. 661, 583 N.E.2d 251, 253 (1991).

The district court thus did not err when it determined this multifactored approach to be the applicable legal test in Massachusetts. Contrary to Speen's assertion that the district court found that "subordinate" factors might outweigh the existence of a right of control, the state cases tell us that Massachusetts courts make the employee determination in this way only when a right of control is not conclusively established and other factors need to be examined. Given how Massachusetts precedent discusses "right of control" in its technical sense, this would seem to mean the multifactored test is triggered when employer control does not encompass the person hired "at every moment, with respect to every detail." McDermott's Case, 186 N.E. at 232. It is thus not so much the case that additional "subordinate" factors might outweigh the existence of a right of control (as Speen wrongly contends was the district court's view) as it is that the failure to demonstrate a "right of control" in the narrowly-defined technical sense of that term serves as the gateway to a multifactored analysis. This analysis, in turn, does not ignore but takes into account the level of control present in the employment relationship despite the fact that this control, taken alone, would not be enough to establish employee status.

Federal courts have used at least three different tests to determine whether a claimant is a covered employee rather than an unprotected independent contractor under antidiscrimination acts such as the ADEA. The first test is the traditional common law test of agency which focuses on the employer's right of control using a multifactored analysis. See Frankel v. Bally, Inc., 987 F.2d 86 (2d Cir.1993). The second test--typically more expansive--is the "economic realities" test, which holds that "employees are those who as a matter of economic reality are dependent upon the business to which they render service." Bartels v. Birmingham, 332 U.S. 126, 130, 67 S. Ct. 1547, 1550, 91 L. Ed. 1947 (1947); Doty v. Elias, 733 F.2d 720, 722-23 (10th Cir.1984). The third test is a "hybrid" test, which considers the economic realities of the employment relationship but retains a focus on the employer's right to control. See Oestman v. National Farmers Union Ins. Co., 958 F.2d 303 (10th Cir.1992).

The First Circuit has not previously decided which test to apply to the ADEA. In view of the Supreme Court's unanimous decision in Nationwide Mut. Ins. Co. v. Darden, 503 U.S. 318, 112 S. Ct. 1344, 117 L. Ed. 2d 581 (1992), we now adopt the common law test for determining who qualifies as an "employee" under the ADEA and expressly hold that covered employees under the ADEA are those who are employees under traditional agency law principles.

While the Supreme Court has not directly determined this issue, the Court in Darden faced the task of interpreting a definition of "employee" found in ERISA, 29 U.S.C. § 1002(6) ("any individual employed by an employer"), that is virtually identical to that found in the ADEA, 29 U.S.C. § 630(f) ("an individual employed by any employer"). See Darden, 503 U.S. at 323, 112 S.Ct. at 1348. The Court found this to be a "nominal definition" that "is completely circular and explains nothing." Id. In the absence of any provision suggesting a contrary congressional design or an indication that "absurd results" would follow, the Court took the view that the term "employee" should be interpreted in accordance with traditional agency law principles:

"[w]here Congress uses terms that have accumulated settled meaning under ... the common law, a court must infer, unless the statute otherwise dictates, that Congress means to incorporate the established meaning of these terms.... In the past, when Congress has used the term 'employee' without defining it, we have concluded that Congress intended to describe the conventional master-servant relationship as understood by common-law agency doctrine."

Darden, 503 U.S. at 322-23, 112 S.Ct. at 1348 (internal citations omitted) (quoting Community for Creative Non-Violence v. Reid, 490 U.S. 730, 739-40, 109 S. Ct. 2166, 2172, 104 L. Ed. 2d 811 (1989) (internal quotation marks omitted)).

To help avoid any confusion on the matter, the Darden Court went on to summarize the operative common law test with the following language:

"In determining whether a hired party is an employee under the general common law of agency, we consider the hiring party's right to control the manner and means by which the product is accomplished. Among the other factors relevant to this inquiry are the skill required; the source of the instrumentalities and tools; the location of the work; the duration of the relationship between the parties; whether the hiring party has the right to assign additional projects to the hired party; the extent of the hired party's discretion over when and how long to work; the method of payment; the hired party's role in hiring and paying assistants; whether the work is part of the regular business of the hiring party; whether the hiring party is in business; the provision of employee benefits; and the tax treatment of the hired party."

Id., 503 U.S. at 323-24, 112 S. Ct. at 1348-49 (quoting Reid, 490 U.S. at 751-52, 109 S.Ct. at 2178-79 (footnotes omitted)).

The Court went on to stress that the common law test requires that "[']all of the incidents of the relationship must be assessed and weighed with no one factor being decisive.' " Darden, 503 U.S. at 324, 112 S.Ct. at 1349 (quoting NLRB v. United Ins. Co. of America, 390 U.S. 254, 258, 88 S. Ct. 988, 991, 19 L. Ed. 2d 1083 (1968)).

We conclude that the Court's opinion in Darden is sufficiently clear to remove doubt as to the identity of the proper standard and its contours. We therefore disregard decisions in those circuits that have employed standards other than the common law test in determining whether a claimant was a covered employee under the ADEA. See, e.g., Oestman, 958 F.2d at 305 (the Tenth Circuit applying the hybrid test in determining whether an insurance agent is an employee under the ADEA). But see Frankel v. Bally, Inc., 987 F.2d 86, 90 (2d Cir.1993) (holding that, in the wake of Darden, the traditional common law test for agency must be applied to the ADEA instead of the hybrid standard).

The Darden decision also circumscribes otherwise suggestive language in First Circuit case law interpreting federal employment legislation such as the Fair Labor Standards Act (FLSA). In earlier decisions, this court has looked to a line of Supreme Court precedent interpreting the FLSA and Social Security Act to reach the view that "[i]n determining employer status, 'economic reality' prevails over technical common law concepts of agency." Donovan v. Agnew, 712 F.2d 1509, 1510 (1st Cir.1983) (FLSA case) (citing Goldberg v. Whitaker House Co-op., 366 U.S. 28, 33, 81 S. Ct. 933, 936, 6 L. Ed. 2d 100 (1961) (FLSA case) (citing United States v. Silk, 331 U.S. 704, 713, 67 S. Ct. 1463, 1468, 91 L. Ed. 1757 (1947) (Social Security Act case) and Rutherford Food Corp. v. McComb, 331 U.S. 722, 729, 67 S. Ct. 1473, 1476, 91 L. Ed. 1772 (1947) (FLSA case))).

This line of cases essentially adopted the non-common law view we considered above, namely that "employees are those who as a matter of economic reality are dependent upon the business to which they render service." Bartels v. Birmingham, 332 U.S. at 130, 67 S.Ct. at 1550. The Darden Court, however, explicitly differentiated the definitions of employee found in the FLSA from that in ERISA, which virtually mirrors the ADEA in this regard. See Darden, 503 U.S. at 325-26, 112 S.Ct. at 1349-50. The Court's analysis of the difference between these two pieces of legislation suggests that this circuit's earlier pronouncements in FLSA cases like Donovan concerning use of the "economic reality" test in determining employee status may need to be confined to the FLSA context in which they were first enunciated. In any event, in view of the Court's express reasoning in Darden, we feel confident in reasoning that the "economic reality" test cannot be readily imported into the ADEA context, either on its own or as part of some "hybrid" test that amalgamates the "economic reality" standard and the traditional common law approach.

3. Speen's Status Under the ADEA and Mass.Gen.L. ch. 151B

Based on our review of the relevant federal and state precedent, we are of the view that federal and Massachusetts law use roughly identical tests based on traditional agency law principles to determine whether a claimant in an age discrimination suit is a protected employee.

There may be some question whether the federal and state tests are employed in exactly the same way in view of some of the language in older Massachusetts decisions we considered above. These earlier opinions speak of the right of control as if it were a predominant factor that is considered before and above others, at least in some instances. This way of employing the multifactor test runs counter to instructions the Supreme Court reiterated in Darden, 503 U.S. at 324, 112 S.Ct. at 1349 ("all of the incidents of the relationship must be assessed and weighed with no one factor being decisive.") (internal quotations omitted). To the extent any divergence between the federal and Massachusetts multifactor test might exist,3 it would seem to involve cases in which a hired party is subject to the "direction and control" of the hiring party "at every moment, with respect to every detail." McDermott's Case, 186 N.E. at 232.

A review of the record, however, reveals that Speen's claim does not present such a case. The record does not contain evidence that even remotely suggests Speen was subject to the "direction and control" of Crown "at every moment, with respect to every detail." We are thus convinced that federal and Massachusetts law confront Speen with multifactored tests concerning employee status that are, as a practical matter, indistinguishable.

Looking at the record with this common multifactored test in mind--even through a lens that requires us to consider the evidence and the reasonable inferences that are to be drawn from it in the light most favorable to Speen as the non-moving party--we conclude that the district court correctly granted Crown's motion for a judgment as a matter of law on the federal ADEA and Massachusetts statutory age discrimination claims.

We do not see how the jury was presented with evidence sufficient to support a finding that Speen was an employee rather than an independent contractor. Speen's counsel vigorously argues that the evidence presented established that Crown kept Speen on a "short leash" and that he had to obey "onerous work rules." The evidence overwhelmingly shows, however, that Speen was kept on a rather long leash, if not actually allowed to run free in a rather large yard, and was allowed to follow procedures that afforded him the type of independence for which employees typically yearn.

We first call attention to the substantial number of factors that, as the district court rightly noted, weigh in favor of a finding that Speen's relationship with Crown was that of an independent contractor. The evidence reveals that Speen himself decided where he went and how long he worked on any particular day. How and in what order he covered his territory was something he determined. Speen was not required to report to a Crown place of business on a daily basis; in fact, he appeared at a Crown location infrequently during the year.

Furthermore, Speen was not required to carry anything, do anything, or say anything in particular as he went about trying to sell Crown (and other) products. Although Crown provided Speen with some business cards that announced him as a Crown representative and also provided him with samples that he was free to use in attempting to make sales, the evidence does not indicate that Crown compelled Speen to do anything in particular or somehow controlled the manner in which Speen attempted to sell men's clothing. The fact that Speen was paid on a commission basis also weighs in favor of a finding of independent contractor status, as does the fact that he received Form 1099s rather than W-2s for federal tax purposes.

Moreover, Speen had early on told his Crown supervisors he wished to be treated as an employee and in particular wished to be enlisted in Crown's employee retirement pension plan. Crown refused and gave Speen a take-it-or-leave-it response, but Speen continued his relationship with Crown. He did not reject what was offered; rather, he accepted it and worked under this regimen for many years. The parties' understanding and Speen's exclusion from Crown's employee pension plan thus represent two additional factors that weigh in favor of a finding of independent contractor status.

Several other salient factors were also present in the Speen-Crown relationship, but the district court correctly noted that they are all compatible with either an independent contractor or employee relationship. Thus, although Speen was required to phone Crown daily and report his sales and the calls he had made, typically by leaving information on an answering machine, this arrangement is equally compatible with the status of either an independent contractor or employee.

Crown required Speen to attend two sales meetings a year that featured the introduction of the new season's line of clothing. He also was required to fill out orders he obtained on forms that Crown provided. Speen also decided to stop selling non-Crown items once he reached one million dollars in sales on Crown's line of products. The district court correctly noted that these features of the Speen-Crown relationship are essentially neutral in terms of the multifactor test, since they are equally compatible with either an employee or independent contractor status. The evidence developed at great and tedious length during the plaintiff's case on these points thus did not significantly advance Speen's view of his relationship with Crown.

On the other hand, the evidence presented reveals not only that Speen did accept the take-it-or-leave-it proposition Crown offered him, but also that he went on to form a corporation, Newton Company, Inc. ("Newton"), of which he became an employee. Some dispute exists in the record as to the reason behind Speen's decision to establish the corporation, but its function is uncontroverted. Crown issued checks made out jointly to Newton and Speen for the commissions Speen's services earned, and Newton, in turn, paid Speen. The district court correctly noted that this fact, while not conclusive in any specific sense, constituted an additional factor militating against a finding of employee status.

Under the multifactored test, we conclude that there was not enough evidence from which a factfinder could make a reasoned determination in favor of Speen on the question of whether he was a Crown employee for purposes of the ADEA or Massachusetts law. As the district court ably determined, a finding for the plaintiff would not comport with the applicable legal standard governing employee status.

The preceding analysis also disposes of Speen's ERISA claim. In view of the Supreme Court's unanimous interpretation of the term "employee" in Darden, 503 U.S. at 323, 112 S.Ct. at 1348, Speen can be considered an employee for ERISA purposes only if we so find using the same test we have just used to determine that he is not an employee for ADEA purposes. We conclude, therefore, that Speen's ERISA claim fails for lack of standing.

Speen finally appeals the district court's disposition of his Massachusetts common law claims. Specifically, he alleged that the Silverman brothers tortiously interfered with his advantageous relationship with Crown. The district court entered a judgment as a matter of law in favor of the defendants on the tortious interference claim as well.

Under established Massachusetts jurisprudence, a plaintiff suing for relief on a claim of tortious interference must prove the existence of the following: "(1) a business relationship or contemplated contract of economic benefit; (2) the defendant's knowledge of such [a] relationship; (3) the defendant's intentional and malicious interference with it; (4) the plaintiff's loss of advantage directly resulting from the defendant's conduct." Comey, 438 N.E.2d at 816 (citing Owen v. Williams, 322 Mass. 356, 77 N.E.2d 318 (1948)).4

Importantly, for our purposes here, Massachusetts case law indicates that this tort claim does not require a finding that the plaintiff was an employee, but rather encompasses independent contractors as well. See Comey, 438 N.E.2d at 816-17. This lifts the barrier that proved fatal to Speen's federal and state statutory claims.

The tort of interference with an advantageous relationship, of course, does not recognize a right to lifetime tenure or a perpetual business relationship. Massachusetts case law discussing the claim in the context of discharge cases explains that companies and their supervisors have the right to fire or terminate the services of hired parties so long as they do not do so "malevolently, i.e., for a spiteful, malignant purpose, unrelated to the legitimate corporate interest." Wright v. Shriners Hosp., 412 Mass. 469, 589 N.E.2d 1241, 1246 (1992) (quoting Sereni v. Star Sportswear Mfg., 24 Mass.App.Ct. 428, 509 N.E.2d 1203, 1206 (1987)). Under Massachusetts law, corporations and corporate officers thus possess both a qualified privilege and a corresponding "duty" to shareholders to discharge hired parties when those hired "d[o] not measure up to the job." Sereni, 509 N.E.2d at 1206.

This qualified privilege and concomitant duty, necessarily, are not unbounded. The privilege does not excuse unlawful malevolence or malice in connection with a decision to discharge a hired party. Whether the requisite malice exists for a defendant to be held liable under this cause of action "depends on the evidence in each case and on what the trier of fact may reasonably infer from that evidence." Gram v. Liberty Mut. Ins. Co., 384 Mass. 659, 429 N.E.2d 21, 24 (1981). For our purposes, it is only important to note that Massachusetts courts treat a showing of intentional age discrimination as sufficient to meet the proof of malice needed for recovery under this tort claim. See Comey, 438 N.E.2d at 816-17.

Thus, our inquiry turns to whether Speen presented sufficient evidence of age discrimination to require the question to be put to the jury. For reasons that follow, we conclude he did not.

In reaching this result, we focus both on Speen's proffered statistical arguments regarding the treatment of other Crown salesmen and his alleged direct evidence of discriminatory motive. On the issue of use of statistical evidence, our cases5 establish that a plaintiff need not and "should not be required to produce 'smoking-gun' evidence before prevailing in a discrimination suit. There are many veins of circumstantial evidence that may be mined by a plaintiff to this end. These include ... statistical evidence showing disparate treatment by the employer of members of the protected class." Mesnick v. General Elec. Co., 950 F.2d 816, 824 (1st Cir.1991). Massachusetts courts have allowed the use of indirect evidence of disparate treatment, including evidence concerning "the employer's general practices and policies concerning employment of [protected classes]." Lewis v. Area II Homecare For Sr. Citizens, 397 Mass. 761, 493 N.E.2d 867, 872 (1986).

The difficulty with Speen's attempts to use statistical evidence was not in what he was trying to accomplish but rather how. In particular, Speen attempted to rely on evidence that compared Crown's treatment of Speen and Speen's sales figures with other members of Crown's sales force. Other testimony, however, established that Crown expected different results from differently positioned sales representatives, depending on their territory, whether they were new to a territory, and other factors. In turn, Speen failed to explain why the group that he selected for treatment was an appropriate and representative sample. Judge Keeton correctly characterized Speen's statistical method as one involving "simply picking out whatever employees the plaintiff wants to pick out from all the evidence before the court, treat[ing] those as if they were the only instances, and urg[ing] the jury to draw inferences of age discrimination from that comparison." Judge Keeton was right in saying,

That won't do....[I]t doesn't take an expert on statistical method to understand that of course [it]'s not appropriate to limit yourself [to some of the population] when you're looking at evidence for the purpose of drawing an inference from a statistical distribution to pick only a few [people] rather than some [number] that can be shown in some way on a reasoned basis to be at least a representative sample if not a consideration of all the evidence.

Numbers selected in such an unreasoned fashion are not sufficient to support a reasoned inference of impermissible discrimination. Ironically, the evidence presented weighs against the finding of disparate treatment that Speen sought to prove. Out of a sales force of twenty people, all five salesmen over the age of 70 at the time Speen was discharged in December 1992 were still selling for Crown at the time of the trial of Speen's claim in March 1996. Combined with the fact that no other salesmen experienced a larger decline in sales than Speen during the five year period 1987-1992, this part of the evidence does not permit a reasoned inference either of age discrimination or that Crown's proffered reason for terminating Speen (declining and unsatisfactory sales figures) was pretextual.

Contrary to Speen's assertions, a different result does not obtain if we consider Speen's testimony about the alleged statement that Jack Silverman made when Speen allegedly asked Silverman to justify the decision to fire him: "Why do I need a 71 year old when I can have a 51 year old?" This piece of evidence, which we must credit as true in view of the requirement that we review the evidence in the light most favorable to the nonmoving party, would still not enable a jury to draw a reasonable inference that Speen was fired due to his age.

In reaching this conclusion, we first call attention to the line of Massachusetts and federal cases which indicate that " 'isolated or ambiguous remarks, tending to suggest animus based on age, are insufficient, standing alone, to prove an employer's discriminatory intent.' " Blare v. Husky Injection Molding Sys., 419 Mass. 437, 646 N.E.2d 111, 118 n. 9 (1995) (quoting Fontaine v. Ebtec Corp., 415 Mass. 309, 613 N.E.2d 881, 885 n. 7 (1993) (citing Gagne v. Northwestern Nat'l Ins. Co., 881 F.2d 309, 314 (6th Cir.1989) and Leichihman v. Pickwick Int'l, 814 F.2d 1263, 1271 (8th Cir.), cert. denied, 484 U.S. 855, 108 S. Ct. 161, 98 L. Ed. 2d 116 (1987))). See Lehman v. Prudential Ins. Co. of America, 74 F.3d 323, 329 (1st Cir.1996).

Speen's counsel argues that the statement was not isolated or ambiguous but rather constitutes direct evidence of age discrimination and reveals that the protected characteristic--age--was a motivating factor in the decision to fire him. Speen's counsel thus urges us to conclude that "even standing alone the plaintiff's quotation of Jack Silverman's comment to him, 'Why do I need a 71 year old when I can have a 51 year old' is, if credited by the jury, sufficient to prove ... [Speen] was fired due to his age."

We reach a contrary result because the relevant case law instructs us not to consider the statement standing alone but instead to look at all the evidence presented in the totality of the circumstances. In other words, a fact finder looking solely at the statement "Why do I need a 71 year old when I can have a 51 year old" could reach a reasonable inference that Speen was fired because of his age. But that is not this case. If we consider this statement in the context of all the evidence presented, which we must do, since we are not afforded the luxury of selectively picking and choosing what evidence we will consider, we conclude that Judge Keeton was correct in ruling that there was insufficient evidence for a jury to draw a reasonable inference that Speen was fired due to his age or permit a reasonable inference that Crown's proffered reason for terminating Speen (declining and unsatisfactory sales figures) was pretextual.

To the extent that we reach a different outcome than the one Speen urges, Speen's mistaken view of what precedent requires a court to do in a case like the one at bar explains this difference. Speen correctly notes that proffered direct evidence of unlawful employment discrimination removes a claimant's case from the well-known McDonnell Douglas three-part test for discrimination (prima facie case, legitimate business justification, and rebuttal) operative in essentially the same way in both Massachusetts and federal courts. We have indicated as much. See Smith v. F.W. Morse & Co., 76 F.3d 413, 421 (1st Cir.1996) ("On the relatively rare occasions when a smoking gun is discernible--that is, when a plaintiff produces direct evidence that the protected characteristic was a motivating factor in the employment action--the McDonnell Douglas framework is inapposite."); see also Smith v. Stratus Computer, Inc., 40 F.3d 11, 15 (1st Cir.1994).

Notwithstanding what Speen would have us believe, the fact that the familiar framework that guides cases involving indirect, circumstantial evidence of discrimination may be inapposite here does not conclude the matter. Specifically, whether Silverman's alleged statement actually constitutes direct evidence of discriminatory motive remains somewhat of an open question, since the line in the case law between what constitutes direct and indirect evidence of discriminatory motive is blurred rather than clearly drawn. See Smith, 76 F.3d at 421. References to "smoking guns" can thus be less than fruitful to the extent they obscure the fact that this Circuit has yet to define clearly what constitutes direct evidence of discrimination. See Ayala-Gerena v. Bristol Myers-Squibb Co., 95 F.3d 86, 96 (1st Cir.1996) (citing Smith, 76 F.3d at 431 (Bownes, J., concurring)).

Given the relevant jurisprudence and the approach the district court took in resolving this case, we need not decide whether or not Silverman's alleged statement constituted a "smoking gun" because the result here would be the same either way. As we have previously noted in a similar case involving appellate review of a directed verdict, "[d]iscretion is sometimes the better part of valor, and courts often wisely decide to sidestep difficult theoretical questions if answers to them are not essential to the proper resolution of a case." Smith, 76 F.3d at 421. As in Smith, "[w]e have here a good example of such a prudential approach. The trial court largely bypassed any differential direct evidence/circumstantial evidence tamisage, preferring to go directly to a finding that, on the totality of the evidence presented, [Crown and the Silvermans] had proven that [age] discrimination did not trigger the firing." Id.

The evidence presented in the instant case resolves the age discrimination issue in favor of the defendants, whether we find Jack Silverman's alleged statement to be direct evidence of discrimination (a "smoking gun") or not. In particular, the evidence presented reveals that no other salesman experienced a larger decline in sales for the five year period 1987-1992. The evidence does show that some salesmen who were not terminated had sales that declined more than Speen's in absolute dollar terms in the year or two prior to Speen's termination. But other evidence shows that Crown salesmen each faced different expectations in terms of year-to-year sales, depending on the location of their territory, how long they had covered it, and other factors.

The evidence further indicates that Speen's supervisors were unhappy with Speen's performance and that Jack Silverman had complained about Speen's declining sales figures and attitude on numerous occasions in the two years or so prior to Speen's termination. In at least one instance, Silverman did so in front of other Crown salesmen, much to Speen's embarrassment. Speen's performance, however, did not improve. The evidence also demonstrates that at least two of Crown's more valued customers contacted Crown on their own volition to complain about Speen or to advise Crown to replace Speen with "a real salesman." The evidence further reveals both that Crown had terminated the services of several salesmen and that other salesmen had retired. No discernible age-related pattern, however, emerges from this evidence. Those who were fired included young, middle-aged, and older salesmen. We again note that out of a sales force of twenty people, all five salesmen over the age of 70 at the time of Speen's discharge in December 1992 still sold for Crown at the time of Speen's trial in March 1996. One of those active salesmen was over 80 years old.

This proffered evidence--considered in the light most favorable to Speen, but also in its entirety--cannot be said either to permit a reasonable factfinder to reach the conclusion that Speen's firing was triggered by age discrimination or to permit a reasonable inference that Crown's proffered reason for terminating Speen (declining and unsatisfactory sales figures) was pretextual. Even if the jury credited Jack Silverman's alleged statement, therefore, Speen was not entitled to a jury verdict in his favor.

To be sure, our duty in this appeal from a judgment as a matter of law is to review the evidence and the reasonable inferences extractable from it in the light most favorable to the nonmovant, namely, Speen. While "this approach does not allow the court to consider the credibility of witnesses, resolve conflicts in testimony, or evaluate the weight of the evidence, neither does it pave the way for every case, no matter how sketchy, to reach the jury." Smith, 76 F.3d at 425 (internal quotations and citations omitted). Put another way, "a mere scintilla of evidence is not enough to forestall a directed verdict, especially on a claim or issue as to which the burden of proof belongs to the objecting party." Id. at 425-26 (internal quotations omitted). These time-worn principles of law support the district court's disposition of this part of Speen's action.

Nothing in the Massachusetts cases leads us to question this result given the fact that the age discrimination issue we are considering is embedded in Speen's state common law tort claim. As we noted earlier, Massachusetts courts have explained that the requisite malice required for finding liability under a claim for tortious interference exists "depend[ing] on the evidence in each case and on what the trier of fact may reasonably infer from that evidence." Gram, 429 N.E.2d at 24. "Any reasonable inference of malice must, however, be based on probabilities, rather than possibilities." Id. 429 N.E.2d at 24-25 (internal quotations omitted). In view of the language in Gram, we thus conclude that Massachusetts courts would not view Speen's claim (that he was fired due to his age) as one supported by reasonable inferences drawn from the evidence presented.

Speen failed to provide sufficient evidence to support a finding that he was a Crown employee who enjoyed protection under the applicable federal and state statutory provisions governing age discrimination and pension rights. Nor did he produce evidence sufficient to support his remaining Massachusetts common law tort claims. On the evidence presented, we conclude that the district court's entry of judgment as a matter of law for the appellee-defendants was correct.

Affirmed.