This is the second round of appeals in an antitrust case brought by a health insurer against a medical clinic. See 65 F.3d 1406 (7th Cir.1995). The defendant, the Marshfield Clinic in rural Wisconsin, is the dominant provider of medical services in the north-central part of that state; indeed, despite its remote location, it is one of the largest medical clinics in the nation, comparable in size to the Mayo and Cleveland clinics. Blue Cross of Wisconsin (together with its HMO) sued the Marshfield Clinic (and its HMO), alleging a variety of antitrust violations centered on the contention that the Clinic had illegally monopolized the provision of certain medical services in its region, enabling it to jack up its prices for those services above competitive levels to the detriment of Blue Cross, which pays the Clinic to treat patients insured by Blue Cross. The plaintiffs won a $20 million damages judgment (most of it in favor of Blue Cross's HMO rather than Blue Cross itself) and an injunction. The Clinic appealed, and we threw out all but one of the charges. That was the charge that the Clinic had divided markets with some of its competitors, that is, had agreed with them to keep out of each other's service areas. Id. at 1415-16. The evidence in support of this charge was sufficient for us to "conclude that the jury verdict on liability must stand insofar as the charge of a division of markets is concerned," to direct that the district court revise the injunction to confine it to that practice, and to order "a new trial limited to damages for dividing markets." Id. at 1416. (Neither HMO is involved in the division of markets charge, so we shall not refer to them any more.) We explained that at the new trial the "burden will be on Blue Cross to show how much less it would have paid the Clinic had the Clinic refrained from that illegal practice." Id.

On remand, following additional discovery, the district judge granted summary judgment for the Clinic on the ground that Blue Cross had produced no evidence that it had suffered any injury as a result of the division of markets. 980 F.Supp. 1298 (W.D.Wis.1997). In the same order she ruled that Blue Cross was not entitled to an award of attorneys' fees, since it was not the prevailing party; and later she awarded the Clinic its court costs of some $220,000. So this time Blue Cross has appealed. It argues that it is entitled to an injunction even if it cannot prove damages; that it can prove damages; and that it should be adjudged a prevailing plaintiff and hence awarded a reasonable attorney's fee and court costs (see Clayton Act, § 4, 15 U.S.C. § 15(a)) and not have to pay its opponent's costs.

Blue Cross is clearly right with regard to the injunction. We held when this case was last before us that the jury's finding on liability for dividing up the market with its competitors must be upheld and that Blue Cross was entitled to an injunction against that practice. This holding established the law of the case, binding the district judge on remand and us on this subsequent appeal unless we have good reasons to depart from the previous decision. Agostini v. Felton, 521 U.S. 203, ----, 117 S.Ct. 1997, 2017, 138 L.Ed.2d 391 (1997); Messenger v. Anderson, 225 U.S. 436, 444, 32 S.Ct. 739, 56 L.Ed. 1152 (1912) (Holmes, J.); Best v. Shell Oil Co., 107 F.3d 544, 546 (7th Cir.1997). We don't. Even though, as we shall see, the district judge was correct that Blue Cross has failed to come up with evidence that would authorize an award of damages for the division of markets, this does not justify withholding an injunction--rather the contrary. Inadequacy of a plaintiff's remedy at law, that is, his damages remedy, is normally (and so under section 16 of the Clayton Act, see 15 U.S.C. § 26, which authorizes an injunction in a private antitrust suit "when and under the same conditions" in which it would be granted by "courts of equity") a prerequisite to the entry of an injunction. Walgreen Co. v. Sara Creek Property Co., 966 F.2d 273, 274 (7th Cir.1992). And a common reason why the damages remedy is inadequate is that the plaintiff is unable to quantify the harm that the defendant's practice has inflicted or will inflict on him. Miller v. LeSea Broadcasting, Inc., 87 F.3d 224, 230 (7th Cir.1996); Roland Machinery Co. v. Dresser Industries, Inc., 749 F.2d 380, 383 (7th Cir.1984).

The jury found on sufficient evidence that the Marshfield Clinic entered into agreements with competitors to stay out of each other's territories. The purpose was to enable each of the firms, prominently including the Clinic, to charge higher prices to its customers. It is true that a by-product of such an agreement might be to reduce the advertising and other marketing expenses of the conspirators, and the reduction might even be so great that the optimal monopoly price of each of the noncompetitors would be lower than the former competitive price, since a monopolist will charge a lower price the lower his costs are. The agreement would still be illegal; divisions of markets are per se illegal, just like price-fixing agreements, Timken Roller Bearing Co. v. United States, 341 U.S. 593, 71 S.Ct. 971, 95 L.Ed. 1199 (1951); but as it would be harmless, there would be no basis for a consumer's seeking an injunction. Indeed, a rational consumer would not seek an injunction that would lead to his having to pay higher prices.

Blue Cross was not, however, required to negate the remote possibility that the division of markets was a form of benign cartel (especially since the point was not argued by the defendant). And if this possibility is therefore set to one side, we have a major customer of the Clinic that is likely to have to pay higher prices unless the division of markets is enjoined. It is also possible, of course, that the defendant has abandoned the unlawful practice, perhaps induced to do so by the filing of the lawsuit; and in that event Blue Cross wouldn't need an injunction. But this possibility would not justify the withholding of the injunction without a degree of proof (of abandonment with no likelihood of resumption), see, e.g., United States v. Oregon State Medical Society, 343 U.S. 326, 333, 72 S.Ct. 690, 96 L.Ed. 978 (1952); Wilk v. American Medical Ass'n, 895 F.2d 352, 367 (7th Cir.1990), that the Clinic has not attempted.

What is true and misled the district judge is the principle that there is no tort without an injury. E.g., Janmark, Inc. v. Reidy, 132 F.3d 1200, 1202 (7th Cir.1997); Rozenfeld v. Medical Protective Co., 73 F.3d 154, 156 (7th Cir.1996). A private suit under the antitrust laws is a suit seeking relief against a statutory tort, and the principle that there is no tort without an injury is applicable to it. See Clayton Act, § 4, 15 U.S.C. § 15(a). But all that this implies, so far as equitable relief is concerned, is that a plaintiff has to prove that he is likely to be harmed by the defendant's wrongful conduct unless that conduct is enjoined. This is clear from the text of the Clayton Act, which, evoking traditional principles of equity, Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 130, 89 S.Ct. 1562, 23 L.Ed.2d 129 (1969), requires proof only of "threatened loss or damage." Clayton Act, § 16, 15 U.S.C. § 26 (emphasis added); see California v. American Stores Co., 495 U.S. 271, 282 n. 8, 110 S.Ct. 1853, 109 L.Ed.2d 240 (1990); Ohio-Sealy Mattress Mfg. Co. v. Sealy, Inc., 585 F.2d 821, 844 (7th Cir.1978). There is sufficient evidence of that here. Blue Cross bought services from a seller that had agreed with other sellers to refrain from competing for customers, and it is likely, whether or not provable with the degree of precision required to ground an award of damages, that unless the defendant is enjoined Blue Cross will have to pay Marshfield Clinic more than if the Clinic were not a member of an anticompetitive conspiracy directed in part against the plaintiff, the conspiracy that the suit seeks to destroy.

A more difficult question is whether the judge was right to conclude that Blue Cross could not prove what damages it had sustained as a result of the division of markets. The usual way to measure damages in such a case would be to compare the prices that the Marshfield Clinic charged Blue Cross before and during the conspiracy, or inside and outside the region covered by the conspiracy, or during the conspiracy and after it ended (if it has ended--the injunction issued by the district court before the first appeal was in effect for less than six weeks before we stayed it pending the appeal), correcting by various statistical techniques for any nonconspiratorial factors that might have caused the prices that are being compared to be different from each other. See Isaksen v. Vermont Castings, Inc., 825 F.2d 1158, 1165 (7th Cir.1987); New York v. Hendrickson Bros., Inc., 840 F.2d 1065, 1078 (2d Cir.1988); Home Placement Service, Inc. v. Providence Journal Co., 819 F.2d 1199, 1205 n. 7 (1st Cir.1987); Phillips v. Crown Central Petroleum Corp., 602 F.2d 616, 632 (4th Cir.1979); 1 American Bar Association Section of Antitrust Law, Antitrust Law Developments (Fourth) 787 (4th ed.1997). This method or congeries of methods was unavailable to Blue Cross, however, because the division of markets embraced the entire period and region in which the necessary data are obtainable; at least Blue Cross made no effort to show otherwise. Instead it compared the Marshfield Clinic's prices for medical services between 1988 and 1995 with the prices charged by other providers of medical care for the same services during the same period elsewhere in Wisconsin, on the theory that those other prices, properly adjusted, are what Blue Cross would have had to pay the Clinic and the Clinic's competitors had it not been for the conspiracy.

This--the so-called "yardstick" method of computing antitrust damages--was not improper. In re NASDAQ Market-Makers Antitrust Litigation, 169 F.R.D. 493, 521 (S.D.N.Y.1996). But the qualification "properly adjusted" is at least as vital as when the plaintiff is trying to prove damages by comparing the defendant's prices at one period or in one area with its prices in another period or another area rather than (as under the yardstick approach) comparing the defendant's prices with the prices of other sellers. See, e.g., Package Shop, Inc. v. Anheuser-Busch, Inc., 675 F.Supp. 894, 947-52 (D.N.J.1987). Any nonconspiratorial factors likely to have made the prices charged by the Marshfield Clinic higher than the prices charged by other health-care providers had to be taken into account in order to make a responsible estimate of the prices that Blue Cross would have paid had it not been for the conspiracy. Isaksen v. Vermont Castings, Inc., supra, 825 F.2d at 1165.

The most important factors were the amount and quality of the Marshfield Clinic's service and its market share. The significance of market share is that even though the Marshfield Clinic was not proved to have monopoly power, it does have a large market share throughout north-central Wisconsin, which might confer enough market power on it to enable it all by itself, without dividing markets with its competitors, to charge a price somewhat above the average for the state. The larger a firm's market share is, the larger is the percentage increase that the other firms in the market must make in their output to offset the effect of the firm's curtailing its output in order to drive the market price above the competitive level. For example, if a firm has 50 percent of the market and as a corollary to jacking its price above the existing, competitive level reduces its output by 20 percent (say from 1,000 to 800 units), the other firms will have to increase their output by an average of 20 percent (from 1,000 to 1,200 units) in order to offset completely the reduction in the output of the dominant firm. This may be difficult for them to do, at least in the near term, and so the dominant firm, though not a monopolist, will be able to get away, at least for a time, with its price hike. See George J. Stigler, "The Dominant Firm and the Inverted Umbrella," in Stigler, The Organization of Industry 108 (1968). That may have been the situation of the Marshfield Clinic.

To make the necessary corrections and thus establish that there was enough evidence to enable a jury to make a responsible estimate of damages, Blue Cross submitted to the district court multiple reports by two economic experts, John Beyer and Thomas McGuire. Beyer's two reports, which compute a range of damages exceeding $7 million from the division of markets, are worthless. They attribute the entire difference between the prices of the Marshfield Clinic and the prices of other Wisconsin providers of medical services to the division of markets, with no correction for any other factor except differences in the treatment mix. Statistical studies that fail to correct for salient factors, not attributable to the defendant's misconduct, that may have caused the harm of which the plaintiff is complaining do not provide a rational basis for a judgment. Milwaukee Branch of NAACP v. Thompson, 116 F.3d 1194, 1198 (7th Cir.1997); Kuhn v. Ball State University, 78 F.3d 330, 332-33 (7th Cir.1996); Schiller & Schmidt, Inc. v. Nordisco Corp., 969 F.2d 410, 415-16 (7th Cir.1992); Tagatz v. Marquette University, 861 F.2d 1040, 1044-45 (7th Cir.1988); MCI Communications Corp. v. American Tel. & Tel. Co., 708 F.2d 1081, 1164-65 (7th Cir.1983); cf. Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 242, 113 S.Ct. 2578, 125 L.Ed.2d 168 (1993); J. Truett Payne Co. v. Chrysler Motors Corp., 451 U.S. 557, 563-67, 101 S.Ct. 1923, 68 L.Ed.2d 442 (1981); Navarro v. Fuji Heavy Industries, Ltd., 117 F.3d 1027, 1031 (7th Cir.1997).

McGuire's reports are more promising, though as his bottom line is a damages figure only $11,000 below the lower of Beyer's highest two estimates, it is a little hard to take seriously. For how is it that after making the subtractions that Beyer failed to make, McGuire came up with essentially the same figure? And is it purely a coincidence that this number, after trebling, would give Blue Cross the same $20 million judgment (actually a little more) that it won in the first trial, even though the division of markets was among the least important of the many antitrust violations charged in this suit? Of course it is possible that the alleged violations were redundant; like the assassins of Rasputin, who drowned him after poisoning and shooting him in order to make sure he was really dead, the Marshfield Clinic may have stacked the division of markets on top of other practices any one of which would have had the same effect on its customers' prices. But this observation cannot help Blue Cross. For only one of the practices was illegal, the division of markets. If it added nothing to the price effects of the legal practices, it did not cause Blue Cross any harm.

The main reason that McGuire came up with such a high estimate of his client's damages from the division of markets is that he compared the prices charged by the Marshfield Clinic with the prices charged by other providers on the basis of price per patient (correcting for differences in the treatment mix), rather than price per service. The result of this procedure was that if Marshfield Clinic provided on average more treatment per patient than other providers, the difference was attributed to the division of markets. This is a strange inversion of the usual logic of cartelization, which is that cartelists drive up the market price by (or with the effect of) restricting their output, as in our numerical example. In defense of his procedure, McGuire pointed out that one way in which the Marshfield Clinic may be exploiting market power conferred upon it by the conspiratorial division of markets is by inducing patients to undergo unnecessary procedures paid for by Blue Cross. This is a possibility, but there is no evidence of it, and so it cannot provide the basis for calculating the damages due to the division of markets. To see this, imagine that in a suit for the breach of a building contract the plaintiff proved that the defendant had substituted an inferior grade of roofing. The plaintiff could not obtain damages for the defendant's having installed an inferior grade of plumbing, on the ground that the defendant might have done that as well. The plaintiff would have to present evidence that the defendant did do it. It is not as if fraud were a conventional method of exploiting monopoly power and so could be presumed; and if Blue Cross's real gripe is that the Marshfield Clinic defrauded it, why isn't there a fraud count in the complaint?

McGuire did try to correct for differences in the quality of the services rendered by the Marshfield Clinic compared to the statewide average; but quality and quantity are not the same. If the Clinic because of its reputation (emphasized by Blue Cross itself) for high quality gets referred to it patients who are sicker than average and so require longer treatment, the average price per patient will be higher simply as a function of the more extensive or protracted care required on average by a sicker patient. As far as the record discloses, this is all there is to the higher average price per patient charged by the Marshfield Clinic. McGuire also failed to correct for the effect of market share on the Clinic's prices. In sum, no reasonable jury could estimate the plaintiff's damages from the reports of the plaintiff's experts.

In addition to those reports, there is some nonexpert evidence of damages, consisting of discounts of 15 or 20 percent that the Clinic gave to some of its coconspirators. But this evidence would not enable a reasonable jury to estimate the plaintiff's damages either. The discounts were given in exchange for bulk referrals, and there is no evidence that customers of the Marshfield Clinic, even so large a customer as Blue Cross, would have gotten equivalent discounts had the clinic been competing rather than conspiring. They might have, but no evidence was presented that they probably would have.

We are not saying that Blue Cross did not in fact lose any money as a result of the division of markets. It may well have, as we suggested in discussing the issue of the injunction. (But remember that there is a difference between an actual and a threatened harm. And there is also, and here critically, a difference between an actual and a quantifiable harm and also between a quantifiable and a quantified harm--and only the last supports an award of damages.) The Clinic's own economic experts estimated that the division of markets may have caused the Clinic's prices to be between .4 and .9 percent higher than they would otherwise have been. Blue Cross, however, did not cite that estimate in its briefs to us, and this raises an interesting question of waiver or forfeiture.

The omission to cite a specific item of evidence to an appellate court is ordinarily not a waiver of anything; parties have no obligation to bombard us with record citations. But it is not clear that a party's approach to calculating damages should be treated as evidence, rather than as theory or argument and hence as waived unless mentioned. And in any event silence about a critical piece of evidence can be and here is eloquent. Blue Cross vigorously defends its own experts. It does not say or imply that if the judge was right to throw out their reports (a possibility it seems to think unthinkable), it is nevertheless entitled to go to trial on the basis of the methodology used and results produced by its opponent's experts. A party can, in resisting summary judgment, rely on his opponent's evidence, theories, or arguments. The question is whether Blue Cross wants to. Apparently it does not; it is represented by sophisticated lawyers who would not keep mum if they thought the Clinic's own method of computing damages gave their client a fair shot at substantial damages. The bottom of the .4 to .9 percent range computes to less than $200,000. Maybe Blue Cross doesn't think this amount is worth further litigation, despite the prospect of trebling. Maybe it doesn't want to have a modest fallback position, but would rather go for broke. Maybe it thinks that to acknowledge any merit in its opponent's experts would fatally undermine the credibility of its own experts, of whom the Clinic's were, naturally, critical. And bear in mind that while we have been treating the defendants' experts' reports as if they were evidence, they are not. They are merely discovery materials. See Kirk v. Raymark Industries, Inc., 61 F.3d 147, 164 (3d Cir.1995); Bryan v. John Bean Division of FMC Corp., 566 F.2d 541, 544-47 (5th Cir.1978). To get them into evidence, Blue Cross would probably have to subpoena the experts (though conceivably the reports could come into evidence directly under the catch-all exception to the hearsay rule, Fed.R.Evid. 807). It could subpoena these experts, Kaufman v. Edelstein, 539 F.2d 811, 821 (2d Cir.1976) (Friendly, J.), but of course they might be reluctant to testify in a way damaging to their client and might be able to squirm out of doing so. Their depositions, as distinct from their reports, do not acknowledge that Blue Cross sustained damages from the division of markets.

All this is speculation; for whatever reason, Blue Cross has waived the right to seek damages on the basis of its opponent's expert reports. And the clincher is what its lawyer told the district judge: that those reports had found "that the market shares of the coconspirators did not affect the price charged" (emphasis added). This was a misinterpretation; but by telling the judge not to rely on the defendant's experts for evidence concerning Blue Cross's damages, the plaintiff waived reliance on such evidence should its own evidence fail, as it has done.

So the district judge was right to throw out the damages claim on summary judgment but wrong to throw out the injunction as well and therefore premature in pronouncing the defendant the winner of this lawsuit. The plaintiff has prevailed on a part, although only a small part, of its claim, and this gives it a prima facie entitlement to an award of some of the attorneys' fees and court costs that it incurred. Ohio-Sealy Mattress Mfg. Co. v. Sealy Inc., 776 F.2d 646, 650-53 (7th Cir.1985); Sciambra v. Graham News, 892 F.2d 411, 415 (5th Cir.1990); Royal Crown Cola Co. v. Coca-Cola Co., 887 F.2d 1480, 1485-86 (11th Cir.1989); United States Football League v. National Football League, 887 F.2d 408, 411-12 (2d Cir.1989); Home Placement Service, Inc. v. Providence Journal Co., supra, 819 F.2d at 1209-12. The contrary suggestion in In re Multidistrict Vehicle Air Pollution, 481 F.2d 122, 130 n. 12 (9th Cir.1973), is not persuasive. For it makes no difference that the only relief that Blue Cross has obtained is equitable; the statute is explicit in allowing an award of attorney's fees when only injunctive relief is sought or awarded. Clayton Act, § 16, 15 U.S.C. § 26; Home Placement Service, Inc. v. Providence Journal Co., supra, 819 F.2d at 1210. Blue Cross is not, however, entitled to fees and costs allocable to claims and theories that neither succeeded, nor contributed to the limited victory that it has obtained. Ohio-Sealy Mattress Mfg. Co. v. Sealy Inc., supra, 776 F.2d at 652-53; H.J., Inc. v. Flygt Corp., 925 F.2d 257, 260-61 (8th Cir.1991); see also Reazin v. Blue Cross & Blue Shield of Kansas, 899 F.2d 951, 980 (10th Cir.1990). That would be unreasonable, and the Clayton Act entitles a prevailing plaintiff only to "a reasonable attorney's fee." Clayton Act, §§ 4, 16, 15 U.S.C. §§ 15(a), 26; cf. Hensley v. Eckerhart, 461 U.S. 424, 103 S.Ct. 1933, 76 L.Ed.2d 40 (1983).

We vacate the district court's decision relating to attorneys' fees and to costs, as well as its denial of injunctive relief, but we affirm the denial of damages, and we remand the case to the district court for further proceedings consistent with this opinion.

AFFIRMED IN PART, VACATED IN PART, AND REMANDED.