These cross-appeals arise out of the sale of a defective wastewater treatment system for use in an electroplating operation. For want of a $620 part, there was a damages verdict óf over $7 million. The purchaser of the system, Cambridge Plating Co., Inc., sued the seller, Napco, Inc., for, among other things, failing to reveal that it had knowingly omitted a critical part from the system. The complaint alleged breach of contract, intentional misrepresentation, negligent misrepresentation and a violation of Mass. Gen. L. ch. 93A, §§ 2, 11 (“Chapter 9SA”). After a twelve-day trial, Cambridge Plating won on all counts, with a jury finding liability on the common law counts and the district court finding liability under Chapter 93A. Both the jury and the district court awarded Cambridge Plating significant damages. Napco now raises various challenges to the verdicts. We believe there was error in the striking of post-judgment motions and that the claims were timely filed under the Massachusetts discovery rule; we find the evidence sufficient and affirm on liability (but reverse the multiple damages under Chapter 93A), and vacate and remand the award of damages.
I. Background
We recite the facts as the jury and district court could have found them. See Sampson v. Eaton Corp., 809 F.2d 156, 157 (1st Cir.1987).
Cambridge Plating, as part of its metal plating and metal finishing operations, uses large quantities of. water for bath solutions and rinsing. This water becomes contaminated with chemicals and metals. Environmental regulations require that Cambridge Plating decontaminate the wastewater before discharging it into the sewers.
Napco manufactures and sells wastewater treatment systems for commercial users. In January 1984, Cambridge Plating entered into a contract to purchase, for approximately $398,000, a wastewater treatment system that would remove the contaminants from the water. As part of the contract, Napco provided a “performance warranty” under which Napco warranted that the system, if operated within certain defined limits, would meet all Massachusetts and federal pollution abatement requirements. The warranty, however, excluded liability for all consequential damages or business losses Cambridge Plating might incur in the event of a breach.
The system Napco sold to Cambridge Plating used a precipitation process to remove the contaminants from the water. The wastewater was fed through pipes, and injected with a polymer solution. The polymers were to attach to the contaminants and then aggregate them to form larger particles, known as “floe.” The floe was to settle out of the water and form sludge at the bottom of a clarifying tank. The clean water layer on top would be discharged into the sewer and Cambridge Plating would properly dispose of the sludge left behind in the tank. “Flocculation,” the joining of the smaller particles into bigger ones, was absolutely critical to the success of the wastewater treatment system. Absent proper flocculation, contaminants would remain suspended in the water and the water could not be discharged into the sewer.
For proper flocculation to occur, the polymer solution had to be thoroughly mixed into the wastewater stream. The system needed *757some means of creating turbulence in the stream sufficient to perform that mixing. One mechanism designed to create the necessary turbulence is a “static mixer.” A static mixer is a section of pipe containing a series of “baffles,” small metal plates placed at an angle inside the pipe which create resistance and, consequently, turbulence. The polymer solution is injected into the waste stream just before the water reaches the static mixer. Once the water with the polymer solution hits the baffles, mixing occurs.
There are alternatives to static mixers to create the required turbulence for a precipitation wastewater treatment system. As Joseph Aliota, Napco’s expert engineer, testified, proper mixing can occur if the system is designed with a series of significant bends in the piping around the area where the polymer is injected into the stream. Napco did not opt for that design. The engineering drawings (and other items) for the Cambridge Plating system clearly indicate that the system was to include a static mixer.
Napco did not install the static mixer. Nor did it tell Cambridge Plating that the static mixer had not been installed. It did, however, provide Cambridge Plating with a “tech manual” containing blueprints and operating instructions for the system. This manual, given to Cambridge Plating upon completion of the system, purported to show what had actually been built.1 It contained engineering drawings indicating that the static mixer had been installed in the system. Napco also provided a control panel that depicted the static mixer as being part of the system.
Napco’s employees were aware that the static mixer had not been installed. Bob Triplett, Napco’s plumbing subcontractor, testified that he was instructed not to install the static mixer at the direction of either Carl Bredfield, a Napco employee, or Bob DeBissehop, Napco’s project manager on the Cambridge Plating job. John Eason, Nap-co’s Manager of Pollution Abatement and the person principally responsible for the design of the system, also testified that he knew the static mixer was not there, although he claimed that the static mixer was installed at first but later removed because it had a tendency to clog.
The system was installed in late 1984. For several months after installation, the system generally met the applicable pollution limits. A series of reports from a testing laboratory that Cambridge Plating forwarded to the Massachusetts Water Resources Authority (“MWRA”) (the relevant state regulatory body) showed that from roughly March 1985 until September 1985, the system usually met the applicable discharge limits.
As time went on, however, the system regularly failed to meet the applicable pollution limits and Cambridge Plating complained to Napco about the problems. Starting in early 1986, Edward Marullo, a Cambridge Plating employee responsible for running the system, called DeBissehop at Napco to complain about the poor performance. DeBissehop told Marullo to manipulate the polymer and pH levels. In March 1986, Laurence Tosi, Cambridge Plating’s President, called DeBissehop “yelling and screaming” about the system’s failures. Tosi thought that Napco’s equipment might be at fault, but DeBissehop allayed his concerns, saying that operator error was the likely cause of the problem. Based on De-Bisschop’s assurances, Tosi took no further steps to have the system inspected for defects. Again, in 1987, Marullo called Eason at Napco, who, like his colleague DeBisschop, told Marullo to manipulate the polymer flow and pH lével. At some point, Nap-co told Tosi that it would be willing to send engineers to examine the system or to train further Cambridge Plating’s operators. But there was a price tag: Cambridge Plating had to agree to pay $1000 per day for such service. Tosi declined. At no time did Nap-co inform Cambridge Plating that the static mixer was missing.
During this period, Cambridge Plating hired a series of experts to determine what *758was wrong. In December 1986, it hired Patrick Hunt, a waste treatment operator for Hewlett-Packard who was also an instructor of a licensing course for wastewater treatment operators at the University of Lowell. Hunt inspected the system, recognized “insufficient floe formation” as a problem, and made numerous suggestions, most of which related to operation. Hunt did not discover that the static mixer was missing. In May 1987, Robert Capaceio, also a wastewater treatment expert, visited Cambridge Plating but failed to detect that the static mixer was missing. A third group of experts from Memtek, which designs and manufactures wastewater treatment systems, examined the system in September 1987 for the purpose of proposing a course of action. They recommended a substantial overhaul of the system at a cost that Cambridge Plating considered prohibitive. During their review, they did not notice that the static mixer was missing.
As Cambridge Plating was trying to identify and solve the problems with the system, it was also becoming a consistent violator of the MWRA’s regulations. In December 1988 the MWRA fined Cambridge Plating $682,250 for discharging excessive levels of contaminants. Cambridge Plating challenged the fine, which was later reduced to $128,500, but at a cost of approximately $54,000 in attorneys’ fees.
Cambridge Plating tried to manage the system’s deficiencies by rigging the system so that the wastewater would be recirculated and retreated in the system. “Closed looping,” as this practice was called, gave more time for flocculation to occur. It also slowed down production considerably. When there was closed looping, the system had to process both the retreated wastewater and the incoming wastewater generated by production. From 1985 to February 1989, the closed looping was accomplished by attaching flexible hoses to the system. In February 1989, Cambridge Plating replaced the flexible hoses with hard piping to create permanent closed looping. Cambridge Plating also shut down its zinc plating operation because, even with closed looping, the system could not remove the contaminants from that operation. Because of its slowdown in production capacity, Cambridge Plating’s business began to deteriorate. Net sales declined from a high of approximately $6.2 million in 1985 to approximately $4.8 million in 1989.
In February 1989 Cambridge Plating hired Peter Moleux, another expert in wastewater treatment systems. Moleux reviewed Nap-co’s proposal and copies of the engineering drawings. By chance, Moleux had been given a bad photocopy of the drawings. The portion of the drawings that depicted the static mixer did not appear on the photocopy. He decided to look for the static mixer in the system. He physically examined the system and inspected the area of the system where the static mixer should have been. Because of his expertise, he noticed that the piping looked different than it should have if the static mixer had been installed. He later confirmed that the static mixer had not been installed. He told Cambridge Plating that the mixer was missing.
Shortly thereafter, on March 17, 1989, Cambridge Plating sent a letter to Napco enclosing a draft complaint “concerning difficulties” Cambridge Plating had experienced with Napco. The draft complaint mentioned the missing static mixer and the letter requested an “amicable resolution.” Napco ignored the letter, never agreeing to come to a meeting to seek an amicable resolution nor agreeing to fix the problem.
Despite Moleux’s discovery, Cambridge Plating did not order the static mixer until December 1989. It arrived at the plant in January 1990 but was not installed until May 1990. Installation required the plant to be shut down for one day. Once the static mixer was installed, the system’s performance improved dramatically, and Cambridge Plating was able to discontinue, for the most part, closed looping. The static mixer cost $620.
Cambridge Plating sued Napco on June 22, 1990 charging breach of contract (including willful repudiation of warranty), intentional misrepresentation, negligent misrepresentation and violation of Chapter 93A. Napco subsequently moved for summary judgment. The district court granted the motion, holding that Cambridge Plating’s claims were time-barred. This court reversed and remanded for trial, see Cambridge Plating Co., *759Inc. v. Napco, Inc. (Cambridge Plating I), 991 F.2d 21, 22 (1st Cir.1998), holding that a genuine issue of material fact existed as to-whether Cambridge Plating could benefit from the discovery rule.
On remand, the ease was tried to a jury in September 1994. The district court submitted Fed.R.Civ.P. 49(b) special interrogatories to the jury on the statute of limitations question, the three common law counts, and the Chapter 93A count. The jury answered all of the interrogatories in Cambridge Plating’s favor, returned a general verdict on each of the common law counts, and awarded Cambridge Plating $12,183,120. The district court treated as advisory the jury’s answers to the Rule 49(b) interrogatories on the Chapter 93A count, and, on February 7, 1995, the district court issued findings of fact and conclusions of law on the Chapter 93A count. See Cambridge Plating Co., Inc. v. NAPCO, Inc. (Cambridge Plating II), 876 F.Supp. 326 (D.Mass.1995). In that opinion, the district court concluded that the Chapter 93A count was timely, that Napco had violated Chapter 93A, and that Cambridge Plating was entitled to compensatory damages in the amount of $3,363,120.2 The district court also concluded that the violation of Chapter 93A was “willful or knowing” and ordered a punitive award of double damages.
Napco filed post-judgment motions for judgment as a matter of law, a new trial, remittitur, and amended findings on the Chapter 93A claim. After Cambridge Plating argued that the post-judgment motions failed to set forth with sufficient specificity the grounds for relief, the district court struck all of the motions, except the motion for remittitur. The district court then granted a remittitur in the amount of $7,839,000 and gave Cambridge Plating the option of accepting the remittitur, thereby accepting a reduced damage award on the common law counts in the amount of $4,344,120, or submitting to a new trial. See Cambridge Plating Co., Inc. v. NAPCO, Inc. (Cambridge Plating III), 890 F.Supp. 55, 59 (D.Mass.1995). Cambridge Plating accepted the remittitur.
Napco now challenges the sufficiency of the evidence, both as to liability and damages. Alternatively, it seeks a new trial due to instructional error. It also argues that the district court should have limited the damages even more. We turn first to the question of our scope of review.
II. Scope Of Review
The district court’s decision to strike Napco’s post-judgment motions affects the scope of our review. Napco challenges the sufficiency of the evidence to show willful or intentional misconduct, seeking a judgment as a matter of law or, in the alternative, a new trial on the “intentional” counts: intentional misrepresentation, willful repudiation of warranty and Chapter 93A. This court will not, however, review sufficiency challenges absent a proper motion in the district court for judgment as a matter of law or a new trial. See Johnson v. New York, New Haven & Hartford R.R. Co., 344 U.S. 48, 54, 73 S.Ct. 125, 128, 97 L.Ed. 77 (1952) (motion for j.n.o.v.); Pinkham v. Burgess, 933 F.2d 1066, 1070 (1st Cir.1991) (motion for new trial); cf. Hammond v. T.J. Litle & Co., Inc., 82 F.3d 1166, 1171 (1st Cir.1996) (“It is beyond peradventure that in order to challenge the sufficiency of the evidence on appeal, a party must first have presented the claim to the district court, either by moving for judgment as a matter of law before the case is submitted to the jury and renewing that motion after the verdict, Fed.R.Civ.P. 50(a), (b), or by moving for a new trial pursuant to Fed.R.Civ.P. 59.”). If the district court acted properly in striking the motions, the motions are nullities, and, under Johnson and Pinkham, Napco is barred from challenging the sufficiency of the evidence. We believe, however, that the district court understandably but improperly struck the post-judgment motions.
A Procedural Background
The district court entered judgment on February 8, 1995. The next day, Napco *760moved for an extension of time for filing its memoranda in support of its post-judgment motions, stating:
Plaintiff has prevailed on four separate and distinct legal claims. Therefore, in order to obtain postjudgment relief, Napco must challenge all four bases for the judgment. This will require Napco to argue several substantial legal and factual issues including, for example, the recoverability of lost profits for negligent misrepresentation, the sufficiency of the evidence of intentional misrepresentation and of repudiation of warranty, the statute of limitations (three-year and four-year), as well as issues relating to Chapter 93A and damages.
On February 14, the district court granted the motion; giving Napco until March 1,1995 to file the post-judgment memoranda. On February 17, six days before the 10-day time limit for filing post-judgment motions expired, Napco filed a motion pursuant to Fed. R.Civ.P. 50(b) and 59. In summary fashion, the motion outlined its subject matter and said the grounds would be set forth in the March 1 memorandum to be filed later in accord with the Court’s extension.3 Also on February 17, Napco filed a similar motion under Fed.R.Civ.P. 52(b) and 59 seeking either to amend the district court’s findings of fact and conclusions of law or to have a new trial on the Chapter 93A claim. This motion also said that the grounds for the motion would be set forth in the March 1 memorandum.
On February 24, 1995, one day after the ten-day period expired, Cambridge Plating moved to strike Napco’s post-judgment motions, arguing that they lacked sufficient “particularity” under Fed.R.Civ.P. 7(b)(1) and that, accordingly, no “motion” had been timely filed within the ten-day period prescribed by Rules 50(b), 52(b) and 59. In granting Cambridge Plating’s motion, except on the remittitur issue, the district court refused to take into consideration Napco’s extension motion or any of the other surrounding circumstances.
B. Analysis
Rule 7(b)(1) requires that motions “state with particularity the grounds therefor.” Fed.R.Civ.P. 7(b)(1). Napco’s post-judgment motions are subject to the requirements of Rule 7(b)(1). The particularity requirement, however, is to be read flexibly in “recognition of the peculiar circumstances of the case.” Registration Control Sys., Inc. v. Compusystems, Inc., 922 F.2d 805, 808 (Fed.Cir.1990). This is because Rule 7 is designed “to afford notice of the grounds and prayer of the motion to both the court and the opposing party, providing that party with a meaningful opportunity to respond and the court with enough information to process the motion correctly.” Id. at 807. When a motion is challenged for lack of particularity the question is “whether any party is prejudiced by a lack of particularity or ‘whether the court can comprehend the basis for the motion and deal with it fairly.’ ” Id. at 807-08 (quoting 5 C. Wright & A. Miller, Federal Practice & Procedure § 1192, at 42 (1990)).
While Napco’s motion was at best sloppy practice, we believe that it was sufficiently particular when read in conjunction with the extension motion and prior filings. Although the extension motion was not filed simultaneously with the Rule 50(b), 59 and 52(b) motions, it was filed only a week before, within the ten-day period, and was obviously closely related to the Rule 50(b) motion. Compare Lac Du Flambeau Band of Lake Superior Chippewa Indians v. Wisconsin, 957 F.2d 515, 517 (7th Cir.) (supporting memorandum filed with insufficiently particular motion), cert. denied, 506 U.S. 829, 113 *761S.Ct. 91, 121 L.Ed.2d 53 (1992). The extension motion specified the bases of the judgment that Napco “must challenge,” including the sufficiency of the evidence on the intentional misrepresentation claim and the willful repudiation of warranty claim, as well as issues relating to Chapter 93A and damages. Napco thus represented to both the court and Cambridge Plating the grounds for its post-judgment motions. No claim is made that there was any intervening event that would have made the representations in the extension motion unreliable.
Cambridge Plating makes a passing argument in its brief that it was unable to respond to, or the district court to process, Napco’s motions. If the Rule 50(b), 59 and 52(b) motions are viewed in isolation, Cambridge Plating has a point. But the motions cannot be viewed in isolation. In addition to the closely filed extension motion, significant briefing on the Chapter 93A issues had just been completed and Napco had earlier made quite a detailed Rule 50(a) motion, of which the Rule 50(b) motion was a “renewal.” In short, the record shows that Napco was taking steps specifically to make evidentiary challenges to the verdict on all of the major issues litigated at trial. The grounds Napco would press in its post-judgment motions were sufficiently known. The motions under Rules 50(b), 59 and 52(b) were adequate, although barely, under the circumstances.4
The district court premised its decision on the belief that the law prevented it from looking beyond the four corners of the motion to determine whether the motion had stated its grounds with sufficient particularity. While understandable, such a view of Rule 7(b)(1) is, in our view, too narrow. “Overly technical” evaluations of particularity are disfavored. Wright & Miller, supra, § 1192, at 43. Courts routinely take into consideration other closely filed pleadings to determine whether sufficient notice of the grounds for the motion are given and the opposing party has a fair opportunity to respond. See Chippewa Indians, 957 F.2d at 517 (motion failing to state grounds is sufficiently particular where supporting memorandum adequately discusses the grounds); Brown v. United States Postal Serv., 860 F.2d 884, 887 (9th Cir.1988) (motion for reconsideration was adequate under Rule 7 even though a particular ground was omitted, where the parties had already briefed and argued the issue and no prejudice would result); see also King v. Mordowanec, 46 F.R.D. 474, 477 (D.R.I.1969) (where grounds for Rule 60(b) motion were stated at oral argument previous day and were discussed after filing, during in-ehambers argument, motion did not run afoul of Rule 7(b)(1) particularity requirement). Accordingly, we reach Napco’s various challenges to the sufficiency (and weight) of the evidence.5
III. Liability
A Statute Of Limitations
1. Sufficiency of the evidence.
Napco first argues that none of the claims survive the statute of limitations. Cambridge Plating filed suit on June 22, 1990. A four-year statute of limitations governs the Chapter 93A and breach of warranty claims. See Mass. Gen. L. ch. 106, § 2-725 (contract for sale of goods); Mass. Gen. L. ch. 260, § 5A (Chapter 93A). A three-year statute of limitations governs the intentional and negligent misrepresentation claims. See Mass. Gen. L. eh. 260, § 2A. As established in Cambridge Plating I, 991 F.2d at 27, Cambridge Plating must rely on the *762discovery rule to prove its claims were timely filed. The discovery rule ‘“prescribes as crucial the date when a plaintiff discovers, or any earlier date when she should reasonably have discovered, that she has been harmed or may have been harmed by the defendant’s conduct.’ ” Id. (quoting Bowen v. Eli Lilly & Co., Inc., 408 Mass. 204, 557 N.E.2d 739, 740-41 (1990)). The discovery rule here “turns on when the company should have known that Napco might be responsible for the water treatment system’s fading performance.” Cambridge Plating I, 991 F.2d at 29. The question to be resolved at trial was whether Cambridge Plating “knew or should have known of its claims before June 22,1987 [or June 22, 1986].” Id. Napco argues that the evidence was insufficient for a jury to conclude that Cambridge Plating should not have known of its claims before the pertinent cut-off dates. We disagree.
The question of the timeliness of the breach of warranty and Chapter 93A claims under the four-year statute is, we believe, largely answered by the admirable description of the evidence provided by the district court in light of our previous opinion in Cambridge Plating I. The evidence at trial showed that Cambridge Plating was aware that the system was complex and that the performance warranty Napco provided contained the implicit condition that the system be operated properly. At least at first, Cambridge Plating could reasonably have believed that its own inexperience, rather than Napco’s defective equipment, was to blame.
Of course, at some point, when things were no better, Cambridge Plating should have thought that something other than operator error was causing the problem. And as Nap-co points out, there was evidence that such a point came for Cambridge Plating either in early 1986, when Marullo complained to Debisschop about the system, or March 1986, when Tosi “kept yelling and screaming” at Debisschop that the problems were due to the equipment.
But the critical question is whether Cambridge Plating reasonably relied on Napeo’s responses to those early 1986 inquiries that operator error, rather than equipment defect, was the cause of the trouble. The fact that Tosi complained vociferously that the equipment was at fault is not dispositive if DeBisschop was reasonably able to convince Tosi he was wrong. Napco had considerably superior expertise in this area. The absence of the static mixer was not something Cambridge Plating, with its level of expertise, should have detected. We believe that under such circumstances, when Napco gave Tosi its “assur[ances]” that the problem had to be with Cambridge Plating’s operators, Cambridge Plating reasonably took Napco at its word.
The misrepresentation claims, governed by a three-year statute, were also timely. Cambridge Plating did not sit idle during the next twelve months. When Napeo’s suggestions did not solve the problem, Cambridge Plating hired Patrick Hunt, in late 1986, to evaluate the system. In January 1987 Hunt gave Cambridge Plating his recommendations, which suggested operational changes but not that the static mixer was either missing or not working properly. If Hunt were unqualified to examine the system, then Cambridge Plating might have difficulty arguing that it was acting reasonably. But Hunt was an expert in wastewater treatment systems and competent to evaluate the system.
Was Hunt’s failure to discover that the static mixer was missing reasonable? See Cambridge Plating I, 991 F.2d at 29-30. We think the jury was entitled to think so. First, the absence of the static mixer was not something that could be easily detected from a physical observation of the system. Indeed, other experts after Hunt who looked over the system were also fooled. Second, Napco had supplied to Cambridge Plating drawings representing that the static mixer had been installed. According to Cambridge Plating, these drawings were “as built” drawings. Napco disputes Cambridge Plating’s characterization of the drawings, pointing to testimony that “as built” drawings, as the term is technically used, were never prepared. Yet regardless of whether the drawings technically could be called “as built,” Napco admitted supplying the drawings to Cambridge Plating in the tech manual “[t]o show what was built.”
*763Perhaps, given the problems with the system, the accuracy of the drawings should have been called into question at this time. The jury was entitled to think otherwise. The system was extremely complex and the cause of the malfunction could have been any number of problems, including, as Napco points out, problems of design. The factfinder could reasonably conclude on this record that the probability the drawings were false was sufficiently low that questioning the accuracy of the drawings would be low on Cambridge Plating’s (or Hunt’s) diagnostic checklist. The evidence was sufficient to support a finding that the claims were timely. Nor was such a finding against the weight of the evidence.
2. Special interrogatory.
Napco alternatively argues that a new trial should be granted on the misrepresentation counts because the special interrogatory submitted to the jury on this point was defectively worded. Over Napeo’s objection, the court submitted interrogatory 1(b), which asked: “Should plaintiff Cambridge Plating reasonably have known before June 22, 1987 of defendant Napeo’s failure to install the static mixer?” (Emphasis in original.)
Review of this interrogatory is for abuse of discretion. See Frank Briscoe Co., Inc. v. Clark County, 857 F.2d 606, 614 (9th Cir.1988), cert. denied, 490 U.S. 1048, 109 S.Ct. 1957, 104 L.Ed.2d 426 (1989). Abuse of discretion may be found if the interrogatories are worded in such a way that they are likely to mislead or confuse the jury or inaccurately state the issues. Id. Although Napeo’s argument has some merit, the district court did not abuse its discretion.
Cambridge Plating I stated that the appropriate test for the discovery rule was whether Cambridge Plating “should have known that Napco might be responsible for the water treatment system’s failing performance.” Cambridge Plating I, 991 F.2d at 29. Napco points out that the opinion drew no specific distinction among the claims and argues that it was improper to direct the jury’s focus onto the failure to discover the static mixer specifically.6 Napco also relies on the passage from the opinion stating that “the statute of limitations will begin to run once the plaintiff has enough information to target the defendant as a suspect, though not necessarily to identify the defendant as a culprit.” Id. at 29-30. Under Napeo’s reading, all claims — warranty, negligence and fraud — would have the same trigger date under the discovery rule: when Cambridge Plating should have thought Napco might be responsible for the problems with the system.
Cambridge Plating I did not say that. The opinion does not address the question specifically of what information is needed to “target the defendant as a suspect” for a breach of contract as compared to that information needed to “target the defendant as a suspect” for fraud. The information needed to target the defendant as a suspect is, in fact, different for each claim. Although Cambridge Plating may have had reason to know that Napco might be in breach of its warranty when the system did not perform up to snuff and Cambridge Plating had ruled out operator error, that does not mean that Cambridge Plating had reason to know that Napco had deceived it. Cf. Childers Oil Co., Inc. v. Exxon Corp., 960 F.2d 1265, 1275 (4th Cir.1992) (Luttig, J., dissenting) (“[Although appellants knew or should have known when construction of the station began that Exxon had breached its contract, appellants did not have reason to know of the possibility of deception until they learned in 1988 that Exxon had always intended to build a competing station.”).
Under Massachusetts law, a cause of action for deceit accrues when the plaintiff knew or should have known of the misrepresentation. See Friedman v. Jablonski, 371 Mass. 482, 358 N.E.2d 994, 997 (1976) (cause of action for misrepresentation in the sale of real estate accrues when the plaintiff knew or reasonably should have known of the misrepresentation); see also Tagliente v. Him*764mer, 949 F.2d 1, 5 (1st Cir.1991) (“The burden is on the plaintiff to prove that in the exercise of reasonable diligence she could not have known of the misrepresentation within the statute of limitations.”).7 In this case, the misrepresentation was in the failure to disclose that the static mixer had not been installed. Thus, under Massachusetts law, the cause of action for misrepresentation did not accrue until Cambridge Plating should have known that the static mixer had not been installed. The special interrogatory was not erroneous.8
B. Intentional Misrepresentation
Cambridge Plating had the burden of proving that Napco had engaged in an intentional misrepresentation. Cambridge Plating alleged in its complaint that Napco fraudulently induced it to purchase the wastewater system by falsely promising that the system would contain a static mixer. At trial, however, the district court did not believe that Cambridge Plating had presented sufficient evidence to show such fraudulent inducement, and did not submit that theory to the jury. The district court, nevertheless, believed that Cambridge Plating had presented sufficient evidence to show fraudulent nondisclosure 9 and, accordingly, charged the jury on that theory. The jury subsequently found, as framed in the special interrogatories, that Napco had “intentionally eonceal[ed]” its failure to install the static mixer “while aware of the System’s failure to meet the applicable discharge limits” and was thus hable for intentional misrepresentation.
Napco raises two challenges to this verdict. First, it argues that the evidence was insufficient to show fraudulent nondisclosure. Second, it argues that, regardless of the evidence, the intentional misrepresentation verdict was tainted by a defective jury instruction. It requests that either judgment be entered in its favor or a new trial granted.
1. Sufficiency of the evidence.
Napco argues that there was insufficient evidence to support the jury’s finding of fraudulent nondisclosure. To set aside a jury verdict and enter a contrary verdict for Napco, we must examine the evidence in the light most favorable to Cambridge Plating, drawing all possible inferences in its favor. See Havinga v. Crowley Towing and Transp. Co., 24 F.3d 1480, 1483 (1st Cir.1994). To set aside a verdict and remand for a new trial based on the evidence, Napco must show that the verdict was against the great weight of the evidence, viewed in the light most favorable to Cambridge Plating, or would work a clear miscarriage of justice. See id. at 1482-83. Napco cannot meet either standard.
Napco’s sufficiency challenge largely rests on a single proposition: that there was insufficient evidence from which a jury could conclude that Napco knew or believed that the static mixer was responsible for the problems with the system.10 In Napco’s words, “every piece of proof bearing on the issue confirmed that Napco never believed the static mixer was necessary, and in fact, believed the mixer might impair system performance.” (Emphasis in original.) But there was such evidence: Napco designed the system to include the static mixer. It knew that the mixer was originally included in the system to create the proper mixing of the polymer solution, without which flocculation would be hindered. This evidence, when combined with the evidence that Napco knew *765the mixer had not been installed (so that the omission was not a mere oversight) and that it knew the system was not working, adequately supports the conclusion that Napco intentionally faded to tell Cambridge Plating that the static mixer was missing, knowing it was responsible for the system’s problems.
Napco protests that it simply made a “good faith (and, at worst, negligent) professional judgment that a mixer was not material to the system.” After all, it argues, what did it have to gain from omitting a $620 part from the system or from hiding a problem that could be inexpensively cured. Napco relies on testimony from Triplett, Napco’s subcontractor, Aliota, Napeo’s expert, and Eason, Napco’s manager of pollution abatement. But Triplett’s testimony and Aliota’s testimony do not much help Napco. Although both Triplett and Aliota testified that the use of “elbows,” or pipe bends, could create the necessary turbulence in lieu of the static mixer, Napco did not show that such “elbows” were specifically designed into the system. More to the point, Aliota, who was not a fact witness, did not testify as to what the people at Napco were thinking in leaving out the static mixer, and Triplett, at best, had only a “vague recollection” of “some discussion that when the pumps came on they hit the two — they hit a 90 and a T, and that was enough turbulence to mix with the polymer.”
And, unfortunately for Napco, the jury could have found Eason’s testimony in a number of respects not to be credible. His account of the events surrounding the decision to omit the static mixer conflicted somewhat with that of other Napco witnesses. Triplett testified that the mixer was never installed, while Eason said it was installed but taken out. Eason’s testimony that the static mixer was omitted because it was “minor” was damaged by his admission on cross-examination that he could not find even one other component depicted on the drawings that was “minor.” Napco does not point to other evidence presented to the jury (for example, contemporaneous memos showing that Napco had made a judgment that the mixer was irrelevant) that corroborates Ea-son’s testimony. In short, the “professional judgment” theory was largely a credibility question. In light of the shaky aspects of Eason’s testimony, the jury was entitled to conclude that Eason was not believable and that Triplett’s “vague recollection” was not enough to rebut the circumstantial evidence (particularly the drawings) that Napco knew the absence of the static mixer was the problem.
Napco properly makes the point that it “strains credulity” to suppose that Napco would expose itself to such drastic liability over a $620 part that took one day to install. Why Napco would do so is something of a puzzle and could raise questions about the reliability of the jury’s finding that Napco intentionally concealed the absence of the mixer. But Napeo’s behavior is not wholly inexplicable. It might be explained in terms of the theory of “agency costs”: the aberrant conduct occurred when Napco’s interests in avoiding exposure to drastic liability diverged from those of its employees. Cf. AM-PAT,/Midwest, Inc. v. Illinois Tool Works, Inc., 896 F.2d 1035, 1043 (7th Cir.1990) (Posner, J.) (seemingly irrational behavior on the part of a corporation may be explained by the “divergence of objectives” between the corporation and its employees). For example, the designer Eason and the project manager DeBisschop may each have wished to avoid individual blame for the defects in the system.
Finally, Napeo’s current focus on plaintiffs proof of scienter represents a shift from its defense at trial.11 Its defense to the misrepresentation count principally focussed on *766whether the static mixer was in fact material to the problem. Moreover, Napco’s defense to the breach of warranty claim, which rested in large part on the assertion that it never had any idea that Cambridge Plating’s system was not meeting the effluent requirements, was in considerable tension with any defense made on the misrepresentation claims that, once it heard about the problems, Napco made a judgment that the static mixer was unimportant. Not surprisingly, then, there was little specific testimony that Napco seriously considered Cambridge Plating’s complaints and then made a judgment that the static mixer was the problem. Nap-co principally chose to take the position at trial that it did not know there was a problem, rather than that it thought there was a problem but the cause was something else. Since there was competent evidence showing that Eason and DeBisschop were contacted about the problems with the system on numerous occasions, and that the system worked after the static mixer was installed, Napco was vulnerable on the intentional misrepresentation count.
Cambridge Plating needed to show intentional misrepresentation only by a preponderance of the evidence. See Compagnie De Reassurance D’Ile De France v. New England Reinsurance Corp., 57 F.3d 56, 72 (1st Cir.), cert. denied, — U.S. -, 116 S.Ct. 564, 133 L.Ed.2d 490 (1995). Under this standard, there was sufficient evidence of fraudulent nondisclosure, and the district court was entitled to submit the theory to the jury. The verdict was also not against the weight of the evidence and a new trial is not warranted.
2. Instructional error.
In its initial charge, the district court instructed the jury that Napco’s intentional misrepresentation was in its “silence.”12 It instructed that “[t]he question to be decided here is whether ... [the] defendant willfully concealed from the plaintiff the absence of the static mixer when it knew ... that the discharge limits were not being met and that the inclusion of the static mixer would, at least to some extent, enable the system to perform as it was intended.” Following this initial charge, plaintiffs counsel requested an instruction specifying that materiality was to be measured against an objective standard. In response,the court gave the following supplemental instruction:
[I]n discussing the state of mind of the defendant with regard to the importance of the static mixer to the system ... the standard is objective. It’s not what the defendant did or did not subjectively think about it. It’s what would a reasonable manufacturer or seller under the same circumstances have thought about it.
Napco argues that this instruction “conflated the torts of intentional and negligent misrepresentation” and was “manifestly erroneous.” 13 In Napeo’s view, it cannot be held liable unless it actually knew the static mixer was material.
The standard under which we review this argument depends upon whether Napco properly preserved it. Cambridge Plating protests that Napco has not preserved this argument because it failed to lodge a proper objection before the jury retired to consider its verdict. Under Fed.R.Civ.P. 51, objections must “stat[e] distinctly the matter objected to and the grounds of the objection.” Fed.R.Civ.P. 51. An objection based on one ground does not preserve appellate review of a different ground. See Wells Real Estate, Inc. v. Greater Lowell Bd. of Realtors, 850 F.2d 803, 809 (1st Cir.), cert. denied, 488 U.S. 955, 109 S.Ct. 392, 102 L.Ed.2d 381 (1988).
*767Napco states that the instruction was given over its “opposition.” But Napeo’s objection to the charge only identified three problems with the instruction, and not the issue it now raises. Napco complained that (1) the complaint did not fairly disclose the nondisclosure theory, (2) the theory was not supported by the evidence, and (3) there was no obligation on the part of the seller to disclose. Napco did not object that the instruction was incorrect because it failed to make clear that actual knowledge of materiality was required for liability, nor did Napco object that the instruction conflated negligent and intentional misrepresentation. Thus, Napco failed to “state[ ] distinctly” the argument it makes now, i.e., that the instruction imposed an incorrect scienter requirement.
Napco urges that colloquies occurring days before the charge satisfy Rule 51. Without considering whether such colloquies may be used to determine compliance, the colloquies Napco directs us to do not contain any specific statement that Napco needed to have “actual knowledge” of the static mixer’s materiality to be liable. Napeo’s contention that this lack of specificity should be excused because the fraudulent nondisclosure theory was “novel” and came as a “surprise” to counsel is contrary to what the record shows.14
In any event, Napco concedes that the district court charged the jury correctly on the scienter element of intentional misrepresentation during its initial charge. It was only after plaintiff objected to the charge, suggesting that the court make clear that materiality be viewed from an objective point of view, that the court gave the supplemental instruction. The supplemental instruction thus created a new issue, independent of the supposed surprise over the fraudulent nondisclosure theory. Despite this, Napeo’s only objections to the supplemental charge related to the charging of the fraudulent nondisclosure theory generally, not to the new ambiguity the instruction created over the scienter requirement. On the question of scienter, Napco remained silent.15
Absent a proper objection, review is for plain error. On this record, the plain error hurdle is too high for Napco to overcome. “Plain error [ ] is a rare species in civil litigation.” Gay v. P.K. Lindsay Co., Inc., 666 F.2d 710, 712 n. 1 (1st Cir.1981), cert. denied, 456 U.S. 975, 102 S.Ct. 2240, 72 L.Ed.2d 849 (1982); see also Clausen v. Sea-3, Inc., 21 F.3d 1181, 1196 (1st Cir.1994) (“[T]he plain error standard, high in any event, ... is near its zenith in the Rule 51 milieu.” (internal quotations omitted)). “[I]t applies only where the error results in a clear miscarriage of justice or seriously affects the fairness, integrity or public reputation of judicial proceedings.” Clausen, 21 F.3d at 1196 (internal quotations omitted). Napco cannot show on this record that a miscarriage of justice will result if the error is not corrected. There was sufficient evi*768dence adduced at trial to conclude that Ñapeo knew that the static mixer was the problem with the system. Nor can Napco show that the error seriously affects the integrity or impairs public confidence in the proceedings. Under the circumstances, the supplemental instruction did not reach the “pinnacle of fault” envisioned by the plain error standard. See id.16
C. Willful Breach Of Warranty
Napco’s warranty excludes liability for consequential damages. By its terms, this damages limitation provision bars Cambridge Plating from recovering consequential damages and, under usual circumstances, would be enforceable. See Mass. Gen. L. ch. 106, § 2-719(3); Deerskin Trading Post, Inc. v. Spencer Press, Inc., 398 Mass. 118, 495 N.E.2d 303, 306 (1986). Massachusetts law provides, however, that the damages limitation provision is not enforceable if Napco either willfully repudiated or was willfully dilatory in performing its warranty obligations. Cf. Canal Elec. Co. v. Westinghouse Elec. Corp., 406 Mass. 369, 548 N.E.2d 182, 186 (1990). The jury concluded that Napco willfully repudiated or was dilatory in performing its warranty obligations. Napco claims this finding was not supported by the evidence and, in any event, was fatally tainted by the district court’s “state of mind” supplemental instruction.
These arguments are worth only brief comment. Napco concedes that evidence it knew the static mixer was to blame “might well amount to ‘willful repudiation.’ ” Since Cambridge Plating adduced sufficient evidence of this fact, the willful breach of warranty verdict stands.17 As for Napco’s *769claim of instructional error, Napco’s challenge suffers the same fate as it did on the intentional misrepresentation count: there was no objection and no plain error.
D. Chapter 93A
Upon making independent findings of fact following the jury verdict, the district court held that Napco had violated Chapter 93A and that punitive damages were warranted. Napco challenges both aspects of the district court’s decision. Review of the district court’s findings of fact is for clear error, see Fed.R.Civ.P. 52(a); review of its conclusions of law is de novo. Pullman-Standard v. Swint, 456 U.S. 273, 287, 102 S.Ct. 1781, 1789, 72 L.Ed.2d 66 (1982).
1. Chapter 93A liability.
Section 2 of Chapter 93A makes it unlawful to engage in “unfair methods of competition and unfair or deceptive acts or practices.” Mass. Gen. L. ch. 93A, §§ 2, 11 (section 11 makes section 2 applicable to businesses). Perhaps by design, the dimensions of Chapter 93A liability are difficult to discern with precision. Neither “unfair” nor “deceptive” is specifically defined in the statute; nor has the case law supplied precise definitions. There is a rubric: “The objectionable conduct must attain a level of rascality that would raise an eyebrow of someone inured to the rough and tumble of the world of commerce.” Levings v. Forbes & Wallace, Inc., 8 Mass.App.Ct. 498, 396 N.E.2d 149, 153 (1979); see also Quaker State Oil Refining Corp. v. Garrity Oil Co., Inc., 884 F.2d 1510, 1513 (1st Cir.1989) (quoting Levings). But, as the Supreme Judicial Court has recently observed, the rhetoric of “rascality” is “uninstruetive.” Massachusetts Employers Ins. Exch. v. Propac-Mass, Inc., 420 Mass. 39, 648 N.E.2d 435, 438 (1995).
Chapter 93A liability may exist if the defendant’s conduct falls “within at least the penumbra of some common-law, statutory, or other established concept of unfairness” or is “immoral, unethical, oppressive or unscrupulous.” PMP Assoc., Inc. v. Globe Newspaper Co., 366 Mass. 593, 321 N.E.2d 915, 917 (1975). Thus, proof of a. common law tort, while not necessary for liability, see Massachusetts Farm Bureau Federation, Inc. v. Blue Cross of Massachusetts, Inc., 403 Mass. 722, 532 N.E.2d 660, 664 (1989) (“a violation of G.L. c. 93A, § 11, need not be premised on a violation of an independent common law or statutory duty”), may be sufficient. See Anthony’s Pier Four, Inc. v. HBC Assoc., 411 Mass. 451, 583 N.E.2d 806, 822 (1991) (breach of implied contractual duty of good faith and fair dealing gave rise to Chapter 93A liability); VMark Software, Inc. v. EMC Corp., 37 Mass.App.Ct. 610, 642 N.E.2d 587, 594 (1994) (intentional misrepresentation provided basis for Chapter 93A liability).
Under this precedent, Napco’s liability under Chapter 93A follows almost as a matter of course. The district court appropriately found that:
NAPCO, which knew throughout that the static mixer was not installed, learned that the system was not performing as warranted and nevertheless 1) failed to reveal knowledge within its possession which would have stemmed the tide of damages being caused by its own misconduct and 2) misdirected plaintiffs attention to operator error as the source of the System’s shortcomings.
Cambridge Plating II, 876 F.Supp. at 337. This finding, which was not clearly erroneous, supports the conclusion that Napco behaved both “deceptively” and “unfairly” under Chapter 93A.18
Moreover, Napco’s argument — that Chapter 93A liability is inappropriate because its decision to omit the static mixer was simply a professional judgment — misses the mark. Even if at the time of installation Napco believed that the probability was small that the static mixer would make a difference to the system’s performance, the probability was large -enough that it should not have been ignored once Napco learned that Cambridge Plating was having problems. Yet even after Cambridge Plating had com*770plained about the system’s performance, Napco never seriously reconsidered its professed judgment about the static mixer. Napco also knew all along that it had provided Cambridge Plating with inaccurate drawings. So Napco also knew that Cambridge Plating would be handicapped in reaching its own conclusion on the matter. Where Napco knew the system was not performing as warranted, Napco was not free to ignore the fact that Cambridge Plating might benefit from knowing that the mixer had been omitted. Napco’s silence became sufficiently “unscrupulous” to fall within a “penumbra ... of [an] established concept of unfairness.” PMP Assoc., 321 N.E.2d at 917. We therefore affirm the district court’s judgment of liability for single damages under Chapter 93A.
2. Punitive damages.
We part company with the district court on the question of punitive damages. Punitive damages are awarded only for “willful or knowing” violations of section 2 of Chapter 93A. Mass. Gen. L. ch. 93A, § 11 (providing for up to three, but not less than two, times actual damages for willful or knowing violations of section 2). Here, the district court imposed double damages based on essentially the same finding upon which it imposed substantive liability under Chapter 93A: that Napco failed to disclose what it knew about the static mixer while knowing that it was material to the problem. See Cambridge Plating II, 876 F.Supp. at 346. At first blush, this conclusion may seem sound, given the “willful or knowing” language of the statute. Napco’s conduct, which also amounts to intentional misrepresentation (and willful breach of warranty), clearly involves a certain level of deliberateness.
But shades of culpability are supposed to matter in applying the punitive damages provision of the statute. See Kansallis Fin. Ltd. v. Fern, 421 Mass. 659, 659 N.E.2d 731, 738 (1996) (“[T]he Legislature- envisaged multiple damage awards against those defendants with a higher degree of culpability than that sufficient to ground simple liability.”); Heller v. Silverbranch Constr. Corp., 376 Mass. 621, 382 N.E.2d 1065, 1070 (1978) (only “callous and intentional violations” merit multiple damages); VMark Software, 642 N.E.2d at 596 (court refused to multiply damages in intentional misrepresentation case stating that section 11 multiple damages are an “extraordinary remedy” not applicable to a case of “dogged bumbling”); cf. International Fidelity Ins. Co. v. Wilson, 387 Mass. 841, 443 N.E.2d 1308, 1317 (1983) (“The Massachusetts legislature consciously enacted a rule whereby defendant’s [Chapter 93A] liability is measured by the degree of his culpability.”).
Liability under Chapter 93A for conduct amounting to intentional misrepresentation (or breach of warranty like that here) does not automatically trigger punitive damages. There must be something more. See VMark Software, 642 N.E.2d at 595 (liability for intentional misrepresentation supported Chapter 93A liability, but misrepresentations were not “made so ‘knowingly’ as to warrant the punitive sanctions of double damages under Chapter 93A”); International Totalizing Sys., Inc. v. PepsiCo, Inc., 29 Mass.App.Ct. 424, 560 N.E.2d 749, 757 (1990) (defendant liable for “knowing” misrepresentation and failure to disclose also violated Chapter 93A, but was not liable for multiple damages because of the absence of “willful or intentional conduct within the purview of [Chapter 93A]” (internal quotation omitted)).
The district corat appropriately recognized these principles. See Cambridge Plating II, 876 F.Supp. at 346. It believed, however, that its findings supported the conclusion that Napco’s conduct was sufficiently egregious to warrant a punitive sanction. We do not think so. If Napco’s conduct was fraud and a willful repudiation of warranty, it was only marginally so. Napco had reason to believe that operator error was the cause of Cambridge Plating’s problems. The system, after all, had worked for a period of time after its installation. Indeed, Cambridge Plating itself took nearly a year and a half to install the static mixer once it found it was missing. Napco did not stand to profit by its actions, and there was no evidence that Napco acted maliciously towards Cambridge Plating or remained silent so that it could watch Cambridge Plating go into distress. Rather, as the district court noted, this was a *771case where Napco simply ignored the problem hoping that it would somehow resolve itself. Id.; cf. VMark Software, 642 N.E.2d at 596 (“The inept blend of hopeful dissembling and dogged bumbling displayed by VMark does not, however, reflect the culpable state of mind required for imposition of § ll’s extraordinary damage penalty.” (internal quotation omitted)). This evidence does not rise to the level of callousness or meretriciousness that would justify multiple damages.19 See Wasserman v. Agnastopoulos, 22 Mass.App.Ct. 672, 497 N.E.2d 19, 24-25 (setting aside award of multiple damages because facts as found by the trial court did not rise to the “purposeful level of culpability” contemplated by the statute), rev. den., 398 Mass. 1105, 499 N.E.2d 298 (1986).
IV. Damages
The district court awarded Cambridge Plating $3,363,120 in compensatory damages for the Chapter 93A violation. The district court then remitted the $12,183,120 jury award on the common law counts to $4,344,-120, reasoning that no rational jury could award more than $4,161,000 in lost profits.20 See Cambridge Plating III, 890 F.Supp. at 59. Napco argues that the award of lost profits still “exceeds any rational appraisal or estimate of the damages that could be based upon the evidence.” See Eastern Mountain Platform Tennis, Inc. v. Sherwin-Williams Co., Inc., 40 F.3d 492, 502 (1st Cir.1994), cert. denied, — U.S. -, 115 S.Ct. 2247, 132 L.Ed.2d 256 (1995). Napco argues that the entire award of lost profits should be set aside or be further remitted to reflect (1) Cambridge Plating’s failure to mitigate damages and (2) Cambridge Plating’s failure to account for selling, genéral and administrative (“SG & A”) expenses associated with the lost profits. Napco separately argues that the award of lost profits for the negligent misrepresentation count must be set aside because such damages are not cognizable under Massachusetts law.
A Sufficiency Of The Evidence On Lost Profits
Cambridge Plating had the burden of providing a reasonably certain basis to believe that Napeo’s wrongful conduct caused the loss of anticipated profits, cf. Augat, Inc. v. Aegis, Inc. (Augat I), 409 Mass.165, 565 N.E.2d 415, 421 (1991) (plaintiff must prove losses would not have occurred but for the wrongful conduct), and proving with sufficient certainty the amount of those anticipated profits. “The nature of the business or venture upon which the anticipated profits are claimed must be such as to support an inference of definite profits grounded upon a reasonably sure basis of facts.” Augat, Inc. v. Aegis, Inc. (Augat II), 417 Mass.484, 631 N.E.2d 995, 998 (1994) (internal quotations omitted). Because every calculation of lost profits has some element of uncertainty, a plaintiff need not calculate lost profits with “mathematical exactness.” Id. But they cannot be “remote, speculative [or] hypothetical.” Id. Napco argues that Cambridge Plating failed to establish the critical connection between the defective wastewater treatment system and Cambridge Plating’s inabil*772ity to do plating work. On this point Napco is wrong.
Cambridge Plating provided evidence of causation principally through three witnesses — Mssrs. Tosi, Moleux and Joseph Finn, Cambridge Plating’s damages expert. Tosi testified that, in the plating business, customers insist on quick turn-around time. He also testified that because the wastewater treatment system was not working properly Cambridge Plating had to employ closed looping, which slowed down the amount of wastewater fed through the system, consequently slowing down the plating process. Moleux confirmed that closed looping “required Cambridge Plating to either partially or totally shut down.” Additionally, Tosi testified that the zinc plating operation closed because the system was unable to remove sufficiently the contaminants. Further, Debissehop directly linked the wastewater treatment system to a plating operation’s profitability.
Finally, Finn testified that until 1985, the first full year the wastewater treatment system was operational, Cambridge Plating had generally experienced an increase in sales. He also testified that for the years following the installation of the system, Cambridge Plating’s revenues decreased from approximately $6.2 million in 1985 to $4.8 million in 1989, and net income declined from approximately $284,000 in 1985 to a net loss of approximately $131,000 in 1989. During this time period, the plating industry as a whole averaged modest growth.
Cambridge Plating provided a simple before-and-after financial picture of an established company. It also provided testimony from people expert in plating and in waste-water treatment that the difference in profits was due to a slowdown in production and failure to meet effluent limitations, both of which could be traced back to the malfunctioning wastewater treatment system. Ñapeo chose not to present an expert of its own to break the connection between the financial decline and the malfunctioning system. Instead, Napco was content to try to poke holes in Cambridge Plating’s damages testimony. “This [was] a risky strategy, and it failed.” AMPAT/Midwest, 896 F.2d at 1046 (internal citation omitted). There was sufficient evidence of causation to support an award of lost profits.
B. Mitigation
“The general principle is well settled that a party cannot recover for harms that its own reasonable precautions would have avoided.” Knapp Shoes, Inc. v. Sylvania Shoe Mfg. Corp., 72 F.3d 190, 204-05 (1st Cir.1995), cert. denied, — U.S. -, 116 S.Ct. 2500, 135 L.Ed.2d 191 (1996). Cambridge Plating did not install the static mixer until 15 months after it knew in February 1989 that the mixer was missing. Installation took one day. This was an inexcusable failure to mitigate damages.
The district court recognized that Cambridge Plating had failed to mitigate its damages. See Cambridge Plating II, 876 F.Supp. at 345 (“Once Moleux informed Cambridge Plating that the System was missing a vital part.... [t]he obvious next step was to buy and install a mixer immediately.”). Nevertheless, in its Chapter 93A decision, the district court awarded damages for both 1990 and 1991 and then discounted them. It also did not adjust the jury award in its remittitur to take account of this failure to mitigate. We agree with Napco that both the Chapter 93A and the remittitur rulings were in error.
Cambridge Plating argues that it should be absolved for its failure to mitigate after February 1989 because there would have, in any event, been recovery time. The recovery time, it says, was needed in order to offset the damage resulting from its new reputation as a polluter, and to get word out to customers that it could fill their needs while complying with the environmental regulations. This is a plausible theory. But Cambridge Plating did not provide the type or quantity of proof that Massachusetts law requires to support damages for this “reputational” injury sufficient to overcome its failure to mitigate.
Cambridge Plating points to no evidence from customers that it would not have been able to recover business because of its repu*773tatíon as a polluter. The one customer who testified, Alfred Jacques of General Electric, did not support Cambridge Plating’s reputational injury theory. The evidence Cambridge Plating relies upon — a statement from DeBisschop agreeing that companies “tend” to lose business when they are found to be polluters, Tosi’s testimony that Cambridge Plating became “known as a polluter,” and an advertisement from Cambridge Plating touting its system — provides scant support for the proposition that there was any reputational harm. This evidence of reputational injury is also in tension with Finn’s testimony that if Cambridge Plating had a working wastewater treatment system, it could have replaced its business “almost immediately.” Even Cambridge Plating states in its brief that “demand for plating and metal finishing services such as those provided by Cambridge Plating was strong throughout the late 1980s and to the time of trial.”21
On this record, Cambridge Plating’s theory of reputational injury is, to say the least, “speculative.” Augat II, 631 N.E.2d at 998. Cambridge Plating can find no comfort in the case law that allows for some uncertainty in proving damages in tort cases. See, e.g., Computer Sys. Eng., Inc. v. Qantel Corp., 740 F.2d 59, 67 (1st Cir.1984). Those cases reason that some uncertainty is allowed because it has been created by the defendant’s wrongful conduct. Id. Here, however, the uncertainty for the period following February 1989 was largely caused by Cambridge Plating’s own wrongful conduct in failing to mitigate.
Still, Napco has conceded that the period through November 1989, or nine months after Cambridge Plating should have installed the static mixer, is a reasonable recovery period. In the absence of evidence showing Cambridge Plating mitigated its damages, the outer limit for such damages on the record before the trial court was November 1989. Cf. Augat II, 631 N.E.2d at 1000 (reducing period of lost profits to six months).
We also believe that both the Chapter 93A damages and the remittitur should equally take account of Cambridge Plating’s failure to mitigate. We therefore vacate the district court’s award of Chapter 93A damages and remand to eliminate all damages occurring after November 1989; we also vacate the order of remittitur and remand with directions that the district court grant a further remittitur to eliminate damages occurring after November 1989. See 28 U.S.C. § 2106 (appellate court may vacate any judgment and remand to require such further proceedings as may be just); Anthony v. G.M.D. Airline Servs., Inc., 17 F.3d 490, 495 (1st Cir.1994) (remanding with directions to district court to grant a remittitur).22 Of course, plaintiff has the option to seek a new trial on damages in lieu of the remittitur.
C. SG & A Expenses
Cambridge Plating had to prove lost profits in terms of net profits, not gross profits. Jet Spray Cooler, Inc. v. Crampton, 377 Mass. 159, 385 N.E.2d 1349, 1359 n. 15 (1979). Generally, this requires that gross profits be adjusted for SG & A expenses. In this case, however, Cambridge Plating argued that SG & A expenses would not have increased as its sales increased and thus the anticipated gross profits should not be reduced by SG & A expenses. The district court accepted Cambridge Plating’s argument, as do we.
Cambridge Plating relies on the testimony of Finn that “lost gross profit for Cambridge Plating is the exact same number as the lost net profit for the years in question.” Although, as Napco points out, the consolidated financial statements from 1980 to 1985 show a general increase in both sales and SG & A expenses, Finn testified that SG & A expenses increased only 10%-11% as sales in*774creased 30% for the period 1982 to 1984. Moreover, in 1984, SG & A expenses decreased slightly as sales increased. From this evidence, Finn believed that there was “no dependent relationship between S G and A and sales” and that SG & A expenses would not have necessarily increased as gross profits increased. Although these inferences from the evidence are weak, they are sufficiently plausible to survive Napco’s challenge on appeal.
D. Negligent Misrepresentation
Lost profits of $4,161,000 were awarded on the negligent misrepresentation count. This was error. Massachusetts law does not allow “benefit of the bargain” damages for negligent misrepresentation. Danca v. Taunton Sav. Bank, 385 Mass. 1, 429 N.E.2d 1129, 1134 (1982). Lost profits, which are a species of benefit of the bargain damages, are therefore prohibited. See also Redstone v. Goldman, Sachs & Co., 583 F.Supp. 74, 76-77 (D.Mass.1984) (lost profits not available for negligent misrepresentation). The award of lost profits on the negligent misrepresentation count is reversed.
V. Conclusion
For the foregoing reasons, we affirm on liability (save for 98A multiple damages), but vacate and remand the Chapter 93A single and multiple damages award and the remittitur for further proceedings consistent with this opinion. Parties to bear their own costs. It is so ordered.