A jury found Continental liable to Inglis for violating federal and state antitrust statutes, and awarded damages. In No. 89-15412, Continental appeals the court’s denial of its motions for judgment notwithstanding the verdict (“JNOV”) and a new trial; we affirm in part and reverse in part. In No. 89-15422, Inglis appeals several of the court’s rulings affecting the amount of damages; we affirm in part and reverse in part.
I. FACTS
In the 1960s and early 1970s, William Inglis & Sons Baking Company and ITT Continental Baking Company were competing manufacturers of bread products in northern California. Both companies produced and sold, among other things, one-pound loaves of white pan bread that were marketed under the “private labels” of particular retailers. In 1971, Inglis filed this antitrust action, claiming that Continental was seeking to eliminate Inglis and other competitors by charging below-cost and discriminatory prices for private label bread.
Inglis alleged that Continental’s pricing conduct violated (i) sections 1 and 2 of the Sherman Act, 15 U.S.C. §§ 1, 2; (ii) section 2(a) of the Clayton Act, as amended by the Robinson-Patman Act, 15 U.S.C. § 13(a); and (iii) the California Unfair Practices Act, Cal. Bus. & Prof.Code §§ 17000-17101.1 A jury returned a verdict for Inglis on all three claims, and found damages in excess of $5,000,000. On post-trial motions, the court ordered JNOV for Continental, or, in the alternative, a new trial, on the federal claims; and a new trial on the state claim. On appeal, this Court reversed the award of JNOV on the federal claims and affirmed the grant of a new trial on the state claim. Thus, the entire case was tried a second time.
The jury again returned a verdict for Inglis on all three claims; it also found damages (before trebling) of approximately $8 million on the Sherman Act claim, $4 million on the Robinson-Patman Act claim, and $10 million on the state law claim. On post-trial motions, the court declined to order JNOV on any claim. It also ruled, however, that the evidence supported a maximum recovery of $5,600,000 on the Sherman Act claim, and $5,600,000 on the state law claim. Inglis agreed to remit the excess as a condition of the court’s denial of Continental’s new trial motion. Entitled to chose a recovery from among the three claims, Inglis naturally opted to accept $5.6 million (for either the Sherman Act or state law claim) over 4 million (for the Robinson-Patman Act claim). The court trebled this amount to $16,800,000, and then deducted $3,500,000, the sum Inglis received when it sold its assets.2 The final recovery figure was $13,300,000.
Continental now appeals from the denial of its motions for JNOV and a new trial; Inglis appeals from the district court’s deduction of the $3.5 million from its damage recovery against Continental, and from the *1336court’s denial of interest from the date of the first judgment.
II. CONTINENTAL’S APPEAL
A. JNOV on Claim of Attempted Monopolization
Continental argues that it was entitled to JNOV on Inglis’ claim of attempted monopolization because, inter alia, Inglis failed to produce sufficient evidence that Continental acted with anticompetitive intent. We agree. One necessary element of Inglis’ case for attempted monopolization under section 2 of the Sherman Act was proof that Continental specifically intended to control prices or destroy competition. See Thurman Industries, Inc. v. Pay ‘N Pak Stores, Inc., 875 F.2d 1369, 1378 (9th Cir.1989); Inglis I, 668 F.2d at 1027.3 Inglis’ evidence, which included an accounting study of Continental price/cost data and testimony about Continental’s sales practices, was not adequate to support a finding of such intent.
1. Inglis’ Price/Cost Study As Evidence of Intent
Inglis’ principal evidence of intent was also the centerpiece of its entire case, a study prepared by the accounting firm of Ernst & Whinney (“E & W”) that compared Continental’s price for private label bread with different measures of Continental’s cost for producing4 that bread. E & W calculated one measure of cost by (i) identifying the expenses that varied with Continental’s total output, (ii) assigning to private label bread a percentage of each of those expenses, the percentage corresponding to private label bread’s share of total Continental output, and (iii) dividing the amount resulting from step (ii) by the number of loaves of private label bread sold by Continental.5 The study showed that, between 1970 and 1976, Continental repeatedly sold private label bread at prices that were below this measure of cost. Contrary to Inglis’ contentions at trial and on appeal, this data could not serve as sufficient evidence of intent.
In order for such a price/cost comparison to support a finding of specific intent to monopolize, the cost measure would have had to include only expenses that were “uniquely incurred” in the production of private label bread. See Marsann Co. v. Brammall, Inc., 788 F.2d 611, 612, 613 (9th Cir.1986). By a “uniquely incurred” expense, we mean one that Continental would not have incurred had it not produced private label bread. See id. at 613. The methodology of the E & W study, however, was to include expenses that Continental may have incurred regardless of its private label bread production. Inglis’ cost expert, who supervised the study, testified at trial that he included in private label cost a portion of some expenses, such as those for heating, telephone, supervisors, display space, sanitation labor and supplies, even though these expenses may have been exactly the same had Continental produced no private label bread. In this way, the study failed to identify the unique cost of producing private label bread, and consequently failed to show that Continental’s prices were below that cost. This flaw rendered the accounting study insufficient evidence of intent to monopolize.6
*13372. Other Evidence of Intent
Virtually the only other evidence of specific intent that Inglis offered was testimony about an incident in which a Continental salesman offered a certain price to a potential customer on the condition that the customer state in writing that he had received an offer at the same price from another seller. There was, however, no indication that the salesman urged the customer to misstate the truth. Hence, although Inglis interpreted this behavior as improper, it was no more than an effort to verify that Continental offered the price in order to meet competition. The testimony about this incident did not—either by itself or in combination with other evidence—establish that Continental specifically intended to control prices or destroy competition.7
B. JNOV on Claim of Price Discrimination
The fate of Inglis’ price discrimination claim follows that of its attempted monopolization claim. To make out a case of illegal price discrimination under the Robinson-Patman Act, Inglis needed to prove that Continental “discriminate[d] in price between different purchasers of [bread] of like grade and quality” and that this difference may have tended “to lessen competition or ... to create a monopoly.” 15 U.S.C. § 13(a); see Inglis I, 668 F.2d at 1040; Texaco, Inc. v. Hasbrouck, — U.S. -, 110 S.Ct. 2535, 2542-43, 110 L.Ed.2d 492 (1990). Inglis succeeded in showing that Continental offered the same bread to different customers at different prices, but failed to introduce sufficient evidence of the anticompetitive effect of these differences.
Inglis was entitled to demonstrate the necessary effect through a showing of anti-competitive intent. See Inglis I, 668 F.2d at 1040.8 Inglis attempted to show that intent by means of the Ernst & Whinney study. However, to serve as sufficient evidence of intent for a price discrimination claim, a price/cost comparison must meet the same standard required to establish intent for an attempted monopolization claim. See id. at 1041. Because the E & W study failed to calculate the cost uniquely incurred in producing private label bread, it could not demonstrate that Continental’s price was below that figure. Unable to demonstrate such pricing, the study did not suffice as evidence of anticompeti-tive intent. Nor was Inglis’ other evidence of intent adequate to support a finding of intent. See section A.2 supra.
C. JNOV on the California Unfair Practices Claim
Continental contends that it was entitled to JNOV on Inglis’ claim under California’s Unfair Practices Act, Cal. Bus. & Prof. Code sections 17000-17101, because Inglis failed to prove that Continental’s pricing had an anticompetitive purpose or caused injury to Inglis, as required under the statute, and because Inglis’ damages calculations were below the minimal standard for *1338probative value. We disagree, and affirm the court’s denial of JNOV to Continental on this claim.
1. Anticompetitive purpose
To establish its claim under the California Unfair Practices Act, Cal. Bus. & Prof. Code sections 17000-17101, Inglis needed to prove that Continental sold bread at “below ... average total cost for the purpose of injuring competitors or destroying competition.” Inglis I, 668 F.2d at 1048 (citing Cal.Bus. & Prof.Code §§ 17043, 17026, 17029); see Turnbull & Turnbull v. ARA Transportation, Inc., 219 Cal.App.3d 811, 820-23, 268 Cal.Rptr. 856, 861-63 (1990). The statute permitted Inglis “to create a presumption of unlawful purpose by introducing evidence of sales below [average total] cost plus proof of injury to competitors or competition.” Id. at 1049 (citing Cal.Bus. & Prof.Code § 17071); see Turnbull, 219 Cal.App.3d at 825, 268 Cal. Rptr. at 864.
Continental does not challenge Inglis’ figures for Continental’s average total cost; nor does it deny that its prices were below average total cost or that Inglis suffered injury. Instead, Continental argues that the statutory presumption was unavailable because Inglis admitted at trial that the market would not have allowed Continental to sell at its average total cost.9 Contrary to Continental’s position, this fact did not render the presumption ineffective under state law or place it in conflict with federal law.
First, the California legislature did not expressly limit use of § 17071 to situations in which the market would have permitted sales at or above the defendant’s average total cost. Even though pricing below average total cost may be legitimate under certain circumstances, see Inglis I, 668 F.2d at 1035, the statute allows the fact-finder to infer unlawful intent from such pricing (plus injury to competition or competitor) without regard to the peculiar market conditions of the case. Nor are we aware of any California judicial decision creating the restriction Continental urges. Indeed, a handful of California cases treat prices charged by the defendant’s competitors (albeit relevant to defenses) as irrelevant to the operation of § 17071, which suggests that state courts may well consider market price to be irrelevant also. See Page v. Bakersfield Uniform & Towel Supply Co., 239 Cal.App.2d 762, 770, 49 Cal.Rptr. 46 (1966); Kofsky v. Smart and Final Iris Co., 131 Cal.App.2d 530, 531-31, 281 P.2d 5 (1955); Sandler v. Gordon, 94 Cal.App.2d 254, 258, 210 P.2d 314 (1949). As a., federal court sitting in diversity, we are unable to read so clear a state statute as § 17071 in the restrictive way Continental urges without any indication that the state legislature or state courts have adopted or are tending toward such a position.
Second, the inference permitted by § 17071 does not become irrational by virtue of the fact that Continental could not have sold at or above its average total cost.10 There is nothing inherently impossible about selling below average total cost for the purpose of injuring a competitor in a market that would not permit the seller to recover average total cost. For example, a seller may choose to sell below a market price that is itself below average total cost in order to weaken or destroy a competitor. This is, in fact, Inglis’ claim: *1339although Continental could not have sold at average total cost, it could have sold at a price higher than the one it actually charged.11 On these facts, we find nothing anomalous in the operation of section 17071.12
Finally, although Continental introduced evidence tending to justify its price as designed to meet competitors’ offers, it did not prevail as a matter of law on this issue. Given that the jury rejected Continental’s explanation, we may not accept it unless the evidence “supported only one reasonable conclusion — that [Continental’s] lower prices were a good faith response to competition.” Inglis I, 668 F.2d at 1045 (emphasis removed). Testimony at trial tending to show that Continental led at least one price decrease and failed carefully to verify competitors’ prices before “meeting” them prevents us from adopting Continental’s explanation as the only reasonable conclusion. See id. at 1047-48, 1050.
2. Causation13
In order to establish liability against Continental, Inglis was required to prove “ ‘the existence of some causal connection between defendant’s wrongful act and some loss of anticipated revenue.’ ” See Inglis I, 668 F.2d at 1051 (quoting Flintkote Co. v. Lysfjord, 246 F.2d 368, 392 (9th Cir.), cert. denied, 355 U.S. 835, 78 S.Ct. 54, 2 L.Ed.2d 46 (1957)). To satisfy this requirement, Inglis introduced expert testimony that Inglis’ profit on private label bread from 1970 to 1976 as well as the company’s “going concern” value at the time it was sold in 1976 would have been higher, by specifically stated amounts, had Continental not engaged in below-cost pricing. According to Continental, this testimony was insufficient evidence of causation because other factors, such as the prices charged by other bread producers, influenced the extent of the claimed aggregate loss. Having accepted arguendo Continental’s assertion about these additional influences, we nevertheless reject this argument.
Continental argues as follows: because Inglis claimed that all of the stated lost profits and lost going concern value were caused by Continental, it may establish the causation element of its case only by proof that all of these losses were caused by Continental; failure to prove that some particular part of the claimed loss was caused by Continental negates the causation element altogether and warrants judgment for Continental.14 Continental’s theory is erroneous. According to our caselaw, Inglis was able to satisfy the causation requirement by “establishpng] that the anticompetitive activities were a material cause of some of its injury”; whether In-glis proved that Continental caused all of the claimed injury is properly an issue affecting only the amount of damages *1340proved. Dolphin Tours, Inc. v. Pacifico Creative Service, 773 F.2d 1506, 1510 (9th Cir.1985) (emphasis supplied); see also Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 114 & n. 9, 89 S.Ct. 1562, 1571 & n. 9, 23 L.Ed.2d 129 (1969) (“[The plaintiffs] burden of proving the fact of damage under section 4 of the Clayton Act is satisfied by its proof of some damage flowing from the unlawful conspiracy; inquiry beyond this minimum point goes only to the amount and not the fact of damage.” (emphasis in original)).
Continental does not dispute that Inglis’ evidence was sufficient to show that Continental’s pricing was a material cause of at least some of Inglis’ claimed injury. For purposes of this appeal, then, Inglis has succeeded in establishing the causation element of its case.15
3. Damages calculations
Having shown a causal connection between the unlawful conduct' and an injury to itself, Inglis was required to provide “sufficient evidence to permit ‘a just and reasonable estimate of the damages.’ ” Los Angeles Memorial Coliseum Comm’n v. National Football League, 791 F.2d 1356, 1360 (9th Cir.1986) (quoting Hand-gards, Inc. v. Ethicon, Inc., 743 F.2d 1282, 1297 (9th Cir.1984), cert. denied, 469 U.S. 1190, 105 S.Ct. 963, 83 L.Ed.2d 968 (1985). This is a “liberal proof of damages standard,” Handgards, 743 F.2d at 1297; it is designed to ensure a quantification of injury that is “approximat[e]” rather than highly precise and certain, Dolphin Tours, Inc. v. Pacifico Creative Service, Inc., 773 F.2d 1506, 1511 (9th Cir.1985) (citing Story Parchment Co. v. Paterson Parchment Paper Co., 282 U.S. 555, 563, 51 S.Ct. 248, 250, 75 L.Ed. 544 (1931); see also Texaco, Inc. v. Hasbrouck, — U.S.-, 110 S.Ct. 2535, 2551, 110 L.Ed.2d 492 (1990). The economic calculations prepared by Inglis’ cost expert satisfied that standard in establishing a portion, but far from all, of the $5.6 million damages figure.
One component of Inglis’ damages was lost profits during the period of Continental’s below-cost pricing. Ernst & Whinney demonstrated that in the second quarter of 1970, just prior to the onset of below-cost pricing, Continental’s price for private label bread was 3.1 above a certain measure of its cost16 for producing that bread. E & W then calculated what Continental’s price would have been had it maintained that spread throughout the period of below-cost pricing, and projected the profit that Inglis would have made had it been able to sell bread at that price (rather than having to match Continental’s below-cost price). The result was a lost profits figure of $2 million.17 We agree with Continental that the evidentiary base underlying this calculation was insufficient.
Continental’s primary objection is to the use of the second quarter of 1970 as the base period; it contends that no evidence showed that quarter to be “ ‘similar [to the liability period] but for the impact of the *1341violation.’ ” Inglis I, 668 F.2d at 1062 (quoting Pacific Coast Agricultural Export Ass’n v. Sunkist Growers, Inc., 526 F.2d 1196, 1207 (9th Cir.1975), cert. denied, 425 U.S. 959, 96 S.Ct. 1741, 48 L.Ed.2d 204 (1976)). We agree.
Inglis’ expert testified that he selected the second quarter of 1970 because it was immediately prior to the alleged below-cost pricing. He also testified that subsequent quarters differed from the base period because pricing was below average variable cost. Neither he nor anyone else, however, offered evidence that market conditions existing during the base period remained the same (or even comparable after appropriate adjustments) during the liability period ending in 1975. Continental offered considerable evidence that conditions differed; prices of ingredients and fuel increased, captives expanded their share of the market, and consumer tastes continued to move away from white pan bread.18 In light of the total absence of contrary evidence by Inglis, its calculation of lost profits is wholly speculative. The district court accordingly erred in failing to award Continental judgment notwithstanding the verdict on Inglis’ claim for $2 million in lost profits.19
The other component of Inglis’ requested damages was lost going concern value. E & W computed this component by restating Inglis’ sales for 1975 to show what they would have been had it maintained its 1970 market share, and multiplying that amount by 1.7% to reflect projected pre-tax profits; E & W then discounted this amount to a present value of $3.6 million. We accept Continental’s contention that no sufficient evidence supported the assumption of a 1.7% profit margin.
The E & W expert testified that Inglis’ actual performance in 1975 was a function of depressed market conditions, and that he therefore consulted a survey of baking company profit margins published in the 1975 edition of the “Almanac of Business & Industrial Financial Ratios.” The profit margin for surveyed companies that he assumed were “comparable” to Inglis was 1.7%, and he selected it as appropriate for Inglis’ projected performance in a nonde-pressed market. The 1.7% figure, he explained, was below both the average for companies in the survey, and the Northern California industry standard profit. However, the expert was unable to identify any particular significant respect in which In-glis was “comparable” to the surveyed companies. In light of the absence of any meaningful economic similarity between In-glis and these other companies, the use of their profit margins to project Inglis’ earnings was mere speculation.
Nevertheless, despite the illegitimacy of the 1.7% profit figure, Inglis was entitled to some recovery for lost going concern value. At trial, Inglis offered, as an alternative to the 1.7% figure, its actual 1975 profit margin of .8%; on this basis, it calculated lost going concern value, discounted to present value, at $1.8 million. Continental does not challenge the accuracy of the .8% figure, and it obviously is not an entirely speculative basis for projecting In-glis’ earnings after 1975. Thus, the evidence supported a recovery for lost going concern value of $1.8 million.
In summary, then, the total damages for which Inglis has shown adequate support in the record are $1.8 million, representing lost going concern value. On remand, the district court should therefore reduce the untrebled damages award from $5.6 million to $1.8 million.
*1342D. New Trial on California Claim
Continental urges that the district court’s denial of its motion for a new trial was reversible error because the jury verdict was against the weight of the evidence. We cannot agree. The court’s denial of Continental’s new trial motion displayed none of the abuses for which we may reverse such a decision: the record did not lack evidence supporting the verdict; the court did not believe that it was without power to grant a new trial; the court did not fail to weigh the evidence, or weigh it according to the wrong standard; and the court did not deny the motion despite finding that the verdict was against the weight of the evidence. See Landes Constr. Co., Inc. v. Royal Bank of Canada, 833 F.2d 1365, 1372 (9th Cir.1987). The court’s decision, therefore, was within its discretion. See id. at 1371-72.
E. Postjudgment Interest
On February 4, 1987, after the jury returned verdicts on liability and damages, the district court entered a “judgment” for three times the amounts specified in the verdicts, but provided that Inglis was “entitled to receive only the highest of such judgments, after appropriate set-offs, if any.” Eventually, the court reduced the jury’s verdict through remittitur and a set-off, and on February 24, 1989, entered another “judgment,” and indicated that Inglis should receive interest from February 4, 1987.
Emphasizing that the court retained jurisdiction after February 4, 1987 in order to entertain motions that could affect the amount of damages, Continental claims that the order of that date could not properly trigger interest accrual. The applicable California Code provision states that “interest commences to accrue on a money judgment on the date of entry of the judgment.” Cal.Code Civ.P. § 685.020.20 Continental has offered us no further enlightenment on this subject, and our own research indicates that the California courts have not confronted the issue presented here. Left with only the bare statute for guidance, we conclude that the court's award of interest from the earlier order was not error. Inasmuch as that decision allowed interest to accrue from the entry of what was by its own terms a “judgment” for money damages, it was consistent with the plain language of the statute. Cf Kaiser Aluminum & Chemical Corp. v. Bonjorno, 494 U.S. 827, 110 S.Ct. 1570, 1573, 1576, 108 L.Ed.2d 842 (1990) (holding that interest on federal claim began to accrue when district court entered judgment soon after verdict, even though court later entertained JNOV motion and reduced amount of judgment).
F.Setoff After Trebling
Campbell-Taggart paid Inglis $3.5 million to purchase Inglis’ business and settle the antitrust claims that Inglis had brought against Campbell-Taggart.21 Continental argues that the district court should have deducted this amount22 before, rather than after, trebling Inglis’ judgment against Continental. Our precedent, however, is to the contrary.
In Flintkote Co. v. Lysfjord, 246 F.2d 368 (9th Cir.1957), the plaintiff sued two defendants for conspiracy in violation of antitrust statutes. Before trial, the plaintiff settled with one defendant for $20,000. The jury returned a verdict against the other defendant for $50,000. The district court trebled the amount to $150,000, then deducted the $20,000 settlement recovery. We upheld this procedure, reasoning that the jury verdict, in combination with a treble damages provision, entitled the plaintiff to a total of $150,000; and, unless the *1343$20,000 was deducted from the trebled amount, the plaintiff would not recover what he was entitled to.
We apply Flintkote’s reasoning to this case as follows. The verdict, as amended by the remittitur and further adjusted in Section C, supra, set the damages against Continental at $1.8 million. State law provides for trebling that figure, entitling In-glis to a recovery of $5.4 million. Only by deducting the Campbell-Taggart recovery after trebling can we allow Inglis the full benefit of its entitlement. We therefore so direct.
Continental urges that Los Angeles Memorial Coliseum Comm’n v. National Football League, 791 F.2d 1356 (9th Cir. 1986), is dispositive of the issue in the present case. Coliseum, however, dealt with a critically different situation. There, the jury’s verdict was found to have aggregated two alternative measures of the amount of antitrust injury suffered by the plaintiff. See id. at 1368, ' 1375 (“[T]he [plaintiffs] have been awarded [before trebling] double the compensatory relief to which they are entitled. In order to treble only a single measure of compensatory relief, this double recovery must be corrected before trebling.”). In contrast, the (remitted) verdict in this case derives from only one measure of Inglis’ injury. Hence, Coliseum is inapposite and does not dictate that the Campbell-Taggart payment should have been set off before trebling.
III. INGLIS’ CROSS APPEAL
A. Effect of Remittitur on Inglis’ Right To Appeal
Continental argues that Inglis relinquished its right to appeal the rulings on setoff and postjudgment interest by accepting a remitted verdict. Both the Supreme Court and this court, however, have described the waiver implicit in remittitur as a narrow one that involves only the right to appeal the reduction of damages effected by the remittitur. See Donovan v. Penn Shipping Co., 429 U.S. 648, 649, 97 S.Ct. 835, 836, 51 L.Ed.2d 112 (1977) (approving the appellate court’s enforcement of “the settled rule that a plaintiff who has accepted a remittitur may not appeal to seek reinstatement of the original verdict” (emphasis supplied)); 999 v. C.I.T. Corp., 776 F.2d 866, 873 (9th Cir.1985) (“[T]he plaintiff cannot contest the validity of a remittitur to which he has consented.” (emphasis supplied)); see also 6A Moore’s Federal Practice 59.08[7] (2d ed.1989) at 203-05 (observing that plaintiff may not appeal from remittitur order he has accepted, but may appeal from “other parts of the judgment.” Id., at 205.).
Continental contends further that, regardless of the general rule on appealing from remitted judgments, the particular order in this case embodied an understanding by Inglis and the court that Inglis would not appeal the rulings on setoff and interest. To the contrary, however, Inglis has demonstrated that no such understanding existed. In the postjudgment proceedings below, Inglis and Continental waged a battle of proposed orders on just this point, with Continental urging the court to issue a remittitur that barred Inglis from appealing these rulings. Ultimately, the court rejected Continental’s proposal and accepted Inglis’ proposed order, which contained no such bar.
B. Interest From Date of Judgment After First Trial
Inglis maintains that it was entitled to have postjudgment interest accrue from the date judgment was entered after the first trial.23 This contention is contrary to the Supreme Court’s decision in Bonjorno.24 In that case, judgment (“I”) was *1344entered after the jury returned a verdict for the plaintiff; subsequently, the district court concluded that the evidence as to damages was insufficient, and ordered a new trial. On a limited retrial, the jury awarded even more damages to the plaintiff, and judgment (“II”) was entered; subsequently, the court ordered JNOV to the defendant on a portion of the damages. The appeals court reversed the JNOV and reinstated judgment II. The Supreme Court ruled that interest did not accrue from judgment I because it was “legally insufficient.” 110 S.Ct. at 1576-78. In the present case, the 1978 judgment for Inglis also was legally insufficient, requiring a new trial (not only on damages but on liability as well). Following the reasoning in Kaiser, we hold that it was proper for the district court not to award interest from that judgment.
C. The Deduction of the Campbell-Tag-gart Payment
Inglis argues that the court erred in deducting the $3.5 million Campbell-Taggart payment from the award of damages. In-glis makes two points: (1) only $800,000 of the payment was in settlement of Inglis’ antitrust claims; (2) not even the $800,000 should have been deducted since it was not compensation for injury for which Continental was found liable.
1. The settlement portion
The trial court found that the entire $3.5 million paid by Campbell-Taggart to Inglis was paid in settlement of the lawsuit. That finding is clearly erroneous. It would attribute no value at all to Inglis’ land, plant and equipment. The agreement between Inglis and Campbell-Taggart clearly indicated that $800,000 was being paid in settlement of Inglis’ antitrust claims, and that $2,700,000 was being paid for the land and buildings of Inglis’ baking operation. The $800,000 settlement of the lawsuit was expressly made binding whether or not the sale of assets for $2.7 million went through. Furthermore, at the trial Inglis’ expert specifically excluded the value of the baking plant from the damage calculations he proposed to the jury. See RT at 2046. Cf. Standard Oil Co. of California v. Moore, 251 F.2d 188, 219-20 (9th Cir.1957), cert. denied, 356 U.S. 975, 78 S.Ct. 1139, 2 L.Ed.2d 1148 (1958) (recognizing lost “good will” as separate from tangible assets in assessing damages). Thus, at a minimum, the court erred by deducting any more than $800,000 of the Campbell-Taggart payment from the damage award.
2. Deduction of the settlement amount
Inglis argues that the damage figure adopted by the jury was for injury caused by Continental and no one else, and that any deduction because of CampbellTaggart’s settlement is consequently improper.
There is more than one reason why we cannot accept Inglis’ argument. The first is that the claims asserted by Inglis against Campbell-Taggart and Continental in this case were coextensive. Indeed, they both were alleged to have combined and conspired with each other and others to bring about all of the injuries to Inglis’ business for which relief was sought. Campbell-Taggart settled those claims without narrowing them in any way; Continental chose to litigate and lost. The prospect of a double recovery is clear if no deduction is made.
This case is consequently distinguishable from Convoy Corp. v. Sperry Rand Corp., 601 F.2d 385 (9th Cir.1979). In Convoy, we agreed that equity required an offset to prevent double recovery, but determined that the offset should be only partial because the claims against the two successive, independent tortfeasors were not coextensive; the claim that was settled included damages for lost profits and punitive damages, relief which was not sought against the non-settling defendant. Here there is no basis for excluding a part of the Campbell-Taggart settlement as inapplicable to the claim asserted against Continental. Cf. Franklin Supply Co. v. Tolman, 454 F.2d 1059, 1072 (9th Cir.1972) (recovery *1345of damages for over-evaluation of shares impermissible when equivalent relief had already been obtained in settlement of “same claims” asserted against another).
Inglis argues that the settlement payment by Campbell-Taggart could not duplicate any of the award against Continental because the verdict was for damages caused by Continental alone. Inglis points to the jury instruction that Inglis was not to be held liable for loss resulting from the actions of others. Other instructions, however, informed the jury that injuries were proximately caused by an act if that act “played a substantial part in bringing about or actually causing the injury or damage;” that an antitrust violation proximately caused an injury if the violation “was a material cause of any injury to plaintiffs, even though other factors may have contributed to plaintiffs injury;” and that plaintiffs could recover for injuries suffered “because of” violations by Continental. Taken as a whole, these instructions permitted the jury to award damages against Continental for wrongs committed by Continental even though other parties such as Campbell-Taggart might have been contributed somewhat to the same injuries.
Inglis argues that the evidence does not support a suggestion that Campbell-Tag-gart contributed to the post-1970 injuries for which Continental was held liable; Campbell-Taggart appeared from the evidence to follow Continental’s lead, as did others, in the post-1970 period. The short answer to this argument is that the question at issue concerns the scope of the claim against Campbell-Taggart as it was actually settled, not the scope of that claim as it might later be derived from evidence in litigation of which Campbell-Taggart was no longer a part. We have no way of knowing what the evidence against Campbell-Taggart would have been if it had not settled. We cannot assume, and the evidence cannot be viewed as establishing, that prices charged by Campbell-Taggart in the post-1976 period were necessarily in compliance with California law. They were claimed not to be, and that claim was settled.
We conclude, therefore, that the trial court did not err in deducting from the trebled verdict $800,000 of the settlement payment made by Campbell-Taggart to In-glis.
IV. CONCLUSIONS
In Continental’s appeal, we reverse the district court’s denial of JNOV on the Sherman Act and Robinson-Patman Act claims. We reverse the denial of JNOV on the state law claim only as to the award for lost profits; in all other respects, we affirm the denial of JNOV and new trial on the state law claim. We accordingly reduce the untrebled damages total to $1.8 million. We also affirm the court’s decision to award interest from the date of entry of the earlier order of judgment (February 4, 1987) entered in this, second trial, and to set off the reduced Campbell-Taggart payment after, rather than before, trebling.
In Inglis’ cross-appeal, we affirm the district court’s refusal to award interest from the date of judgment on the first trial, and amend its setoff decision so that Inglis’ recovery, after trebling, is reduced by only $800,000.
The order of the district court is AFFIRMED IN PART, REVERSED IN PART, AND AMENDED IN PART. The case is REMANDED for entry of judgment in accordance with this opinion.
Each party will bear its own costs on this appeal.
. In addition, Inglis brought two conspiracy claims against Continental; Inglis eventually dropped one and the district court awarded summary judgment to Continental on the second. Inglis’ original Complaint also named two other defendants, American Bakeries Company and Campbell-Taggart Incorporated. At trial, the jury found American not liable. Campbell-Taggart settled the Inglis claims as part of a transaction in which Campbell-Taggart purchased Inglis’ assets. As a result of its sale to Campbell-Taggart, Inglis ceased production and withdrew from the bread market. William Inglis & Sons Baking Co. v. ITT Continental Baking Co., 668 F.2d 1014, 1023-24 (9th Cir.1981), cert. denied, 459 U.S. 825, 103 S.Ct. 57, 58, 74 L.Ed.2d 61 (1982) (hereinafter "Inglis I”).
. See note 1 supra.
. To prevail on this claim, Inglis needed to prove, in addition, that Continental engaged in anticompetitive conduct to accomplish its unlawful end, and that there was a dangerous probability that Continental would succeed in acquiring a monopoly. See Thurman at 1378; Inglis I, 668 F.2d at 1027.
. We mean "production costs” to include expenses for distribution and marketing and any other expenses required to bring the bread to market.
. Inglis and E & W referred to this measure as the "average variable cost” of Continental’s private label bread.
. To constitute sufficient evidence of intent, a price/cost comparison must employ a measure of cost that limits includable expenses to those that not only are "uniquely incurred," but also “vary with output" of the product alleged to have been sold at a predatory price. See Inglis I, 668 F.2d at 1037. Step (ii) in Inglis’ cost methodology shows that Inglis attempted to meet this second requirement. Because the cost *1337computation was otherwise flawed, we do not reach the issue of whether the study correctly determined which expenses were variable.
. Inglis argues that the district court erred in excluding as inadmissible a report, prepared for Continental by McKinsey & Co., that advised Continental to "maintain prices to hasten wholesaler exit pace.” Inglis argues that the court also erred by redacting Continental’s own reports to remove references to "eliminat[ing competitors]" and "gear[ing] efforts toward market dominance." We express no opinion as to whether the district judge abused his discretion in ruling these items inadmissible because we find that even if admitted they would not have been sufficient evidence of specific intent. Without any evidence of Continental’s reaction to McKinsey’s recommendation, the report, while possibly relevant to the issue, could not establish Continental’s intent. The sales reports were equally inadequate because they contained no indication that Continental did, or should, seek to achieve its objectives by means of anti-competitive strategies. Even in the aggregate, this evidence was insufficient to prove specific intent to monopolize.
. Inglis argues that it was also entitled to demonstrate the necessary effect by showing a substantial possibility of injury to competition through market analysis. Our circuit has not yet accepted or rejected this alternative means of proving anticompetitive effect. See Inglis I, 668 F.2d at 1042 & n. 49. We decline to reach the question here because we find that Inglis’ market analysis failed to show the substantial possibility of injury to competition.
. What Continental characterizes as Inglis’ "admission” occurred during an extended argument over the significance of a damages exhibit. Both Inglis’ expert witness and counsel stated that the intensity of market competition would have compelled Continental to price its private label bread below its average total cost. Inglis now argues that these statements were intended as part of the damages case only, and should have no relevance to liability. We do not reach this issue because we conclude that the alleged admission is irrelevant to the evidentiary burdens established by the California statute.
. The rationality or wisdom of § 17071 as applied to "normal" market conditions, in which sellers are able to sell at or above total cost, is not at issue here. We accept, for purposes of this appeal, the soundness of § 17071’s eviden-tiary presumption under those circumstances, and concern ourselves only with whether the particular market condition identified by Continental renders the presumption irrational in that application.
. With respect to the fourth quarter of 1970, for example, Inglis alleged that Continental’s average total cost was 22®, that the market would have allowed sales at a price somewhere between 18.2® and 22®, and that Continental sold at 18®. See Plaintiffs Exhibit 501 and Transcript at 1852-54.
. Continental also argues that § 17071 conflicts with federal law by allowing a jury to condemn a price below average total cost where it was not possible to obtain a higher price. We need not reach this issue, however, because this case does not present that scenario. Inglis admitted only that the market would not have permitted Continental to sell at or above average total cost; it did not admit — and expressly denied — that the market would not have allowed Continental to sell at a price higher than the price Continental actually charged.
. Continental bases its argument here entirely on federal law. This fact may indicate that Continental does not intend to challenge the causation element of the state law claim, in which case there is no need for us to consider it. Alternatively, Continental may intend for this argument to apply to the state claim, thus impliedly contending that state standards on causation are identical to federal standards. Granting Continental the benefit of the doubt, we construe the argument in the second way. We thus assume arguendo that the federal standard applies, and reject Continental’s challenge because Inglis satisfied the federal standard for causation.
. Continental intimated at trial, for example, that during particular periods between 1970 and 1976 Inglis may have suffered a loss in profits regardless of Continental’s pricing. See Transcript at 1912-13.
.McGlinchy v. Shell Chemical Co., 845 F.2d 802 (9th Cir.1988), does not support Continental’s theory on causation. In McGlinchy, we upheld the district court's decision to exclude the plaintiffs evidence on damages because it failed to meet the standard for competent evidence as to any of the claimed injury. Among the several serious flaws in the McGlinchy evidence that distinguish it from Inglis’ evidence in this case are the following: failure to relate asserted lost profits to any specific acts by the defendants; concession that the decline in sales, which was responsible for lost profits, could have been caused by "anything”; failure to state which particular lines of plaintiffs business had suffered declining sales, even though plaintiffs theory was that defendant's unlawful conduct had affected particular lines. 845 F.2d at 806-07.
. E & W used the measure it called "average variable cost". See note 4 and accompanying text supra.
. Continental claims that it was wrong to include the entire period from late 1970 to early 1976 in this calculation, arguing that not every price charged by Continental through this period was unlawful, and citing E & W’s own data showing that there were occasions when Continental priced above cost. However, the particular data to which Continental refers involves only what E & W called "average variable cost." With respect to liability on the state claim, only average total cost is relevant; and E & W's study showed that Continental's prices were below that measure throughout the entire period involved.
. Continental also introduced evidence that federal price controls were implemented during the liability period, but Inglis countered with evidence that Continental could have sought relief from price controls. The jury could have accepted Inglis’ theory that price controls were not a substantial factor in pricing during the liability period.
. Our conclusion that the evidence was insufficient to support use of the base period figures makes it unnecessary to address two further arguments Continental offers against Inglis’ calculation of lost profits: (1) that it was improper to assume that, if Continental had increased its price, others would have followed, and (2) that it was improper to assume that, if prices had been increased, the quantity sold would not have declined.
. With respect to the state claim, the district court and we sit in diversity and must base decisions affecting postjudgment interest on state law. See Michael-Regan Co., Inc. v. Lindell, 527 F.2d 653, 659 (9th Cir.1975).
. See note 1 supra.
.Inglis’ cross-appeal challenges the amount of the deduction resulting from the Campbell-Tag-gart settlement. We deal with that question in Section III, C, infra. We are concerned here only with the question whether the deduction should be made before or after trebling of the damages.
. See section I supra.
. Inglis bases its argument here entirely on federal law. This fact may indicate that Inglis does not intend to challenge the causation element of the state law claim, in which case we need not consider the issue. Alternatively, In-glis may intend for this argument to apply to the state claim, thus impliedly contending that the state approach to interest accrual is identical to the federal approach. Granting Inglis the benefit of the doubt, we construe the argument in the second way. We therefore assume ar-guendo that the federal approach applies, and reject Inglis’ challenge because the court acted consistently with federal law by awarding inter*1344est only from the date of judgment on the second trial’s verdicts.