Seth C. Oranburg

            Valuing Uncertain Trade Secrets:
Epistemic Boundaries o ---  the Reasonable Royalties Remedy

                                                 Seth C. Oranburg*

Pro

essor o — Law, University o — New Hampshire Franklin Pierce School o — Law; Director, Program on Organizations, Business and Markets at NYU Law’s Classical Liberal Institute; JD, University o — Chicago; BA, University o — Florida.

Abstract

    Reasonable royalties as a remedy  --- or trade secret misappropriation presents a paradox. The Uni --- orm Trade Secrets Act and the De --- end Trade Secrets Act authorize reasonable royalties only when actual damages and unjust enrichment cannot be proven. But courts are poorly equipped to construct the requisite counter --- actual scenario (where the parties bargained to license the trade secret) where the record lacks  --- acts su ---

icient. The result is doctrinal con — usion, where some courts treat them reasonable royalties as a de — ault remedy whenever conventional measures

all short, while others are very hesitant to award them. This Article resolves the tension by recognizing di —

erent types o — reasonable-royalty cases. Reasonable royalties work tolerably well

or “market-anchored” secrets but are deeply problematic — or “uncertain-value” secrets. Courts can use this distinction to render more consistent rulings on the appropriateness o — reasonable- royalty remedies. Market-anchored cases involve secrets that have been licensed, valued internally, or otherwise subjected to market discipline. Courts can reconstruct hypothetical negotiations by interpreting evidence o — what parties actually agreed to in comparable circumstances. Uncertain- value cases involve secrets that lack any valuation anchor. Courts asked to award reasonable royalties in such cases must engage in unconstrained price-setting rather than judicial interpretation o — market evidence. Drawing on Frank Knight’s distinction between risk and uncertainty and on Guido Calabresi and A. Douglas Melamed’s property-rule versus liability-rule — ramework, this Article shows why uncertain-value secrets present distinct challenges — or judicial valuation. It proposes a calibration principle: courts should require plainti —

o

concept, en — orcing the statutory precondition and recognizing epistemic limits courts — ace when asked to price assets that markets have never valued.

             Introduction

    In Sorrento Therapeutics, Inc. v. Mack, the Delaware Court o ---  Chancery con --- ronted a now- --- amiliar remedial move in trade secret litigation.1 The plainti ---

s’ damages case had largely

allen apart. Their expert’s lost-pro — its and unjust-enrichment models were excluded as

1 Sorrento Therapeutics, Inc. v. Mack, 2025 Del. Ch. LEXIS 195 (Del. Ch. July 31, 2025). speculative or untethered to the misappropriation. What remained was a request — or a “reasonable royalty” under Cali — ornia’s version o — the Uni — orm Trade Secrets Act, which permits courts to “order payment o — a reasonable royalty — or no longer than the period o — time the use could have been prohibited” i — “neither damages nor unjust enrichment caused by misappropriation are provable.”2 Rather than treat that language as an invitation to improvise a number, Vice Chancellor Fioravanti re — used to award any royalty at all. He held that the statutory precondition was not satis — ied and that the plainti —

’s expert had improperly anchored the proposed royalty to the very lost-pro — its and unjust-enrichment — igures the statute treated as unavailable.3 Sorrento is the latest chapter in an ongoing debate about reasonable royalties in trade secret law. Courts and commentators have long described the reasonable royalty as a “ — allback” or “sa — ety valve” — or situations in which lost pro — its are too speculative and the de — endant’s gains too di —

icult to trace.4 The Uni — orm Trade Secrets Act and the — ederal De — end Trade Secrets Act both authorize reasonable royalties when actual damages and unjust enrichment cannot be proven.5 In practice, however, courts have applied the remedy inconsistently. Some treat it as

reely available whenever conventional measures — all short. Others, like Sorrento, insist that the statutory precondition requires genuine proo — o — unavailability and that the hypothetical negotiation — ramework must rest on something more substantial than expert speculation. This Article makes a straight — orward but consequential claim: reasonable royalties work tolerably well — or market-anchored secrets but become problematic — or uncertain-value secrets. When a secret has been licensed, valued internally, or otherwise subjected to market discipline, courts can use the hypothetical negotiation — ramework to estimate what willing parties would have agreed to. When a secret has never been licensed, has no documented internal valuation, and produces no traceable revenue, courts — ace what Frank Knight — amously called “uncertainty” rather than mere “risk.”6 The hypothetical negotiation asks judges to imagine what a willing licensor and willing licensee would have agreed to at a moment when neither party knew whether the secret had value. That exercise becomes speculative in a way that runs headlong into the — oundational principle that damages may not rest on conjecture.7 The argument un — olds in three parts. Part I surveys the statutory architecture and doctrinal landscape. It describes three — amilies o — reasonable royalty cases along a spectrum: market- anchored cases where licensing history or contractual provisions supply concrete valuation evidence, thin-evidence cases where some market comparables exist but require substantial judicial gap- — illing, and uncertain-value cases where no meaning — ul anchor exists at all. The statutory precondition — unctions most coherently when applied to this spectrum. Reasonable royalties work well — or market-anchored cases, require care — ul scrutiny — or thin-evidence cases, and become problematic — or uncertain-value cases. Part II develops the theoretical case — or treating uncertain-value royalties skeptically. It begins with Knight’s distinction between risk and uncertainty and shows how that distinction applies to trade secret valuation. When secrets have market anchors, parties — ace calculable risk. When secrets lack such anchors, parties — ace Knightian uncertainty where even probability distributions cannot be speci — ied. Part II then draws on Guido Calabresi and A. Douglas 2 Cal. Civ. Code § 3426.3(b). 3 Sorrento, 2025 Del. Ch. LEXIS 195, at *37-43. 4 See 4 Roger M. Milgrim, Milgrim on Trade Secrets § 15.02[1][a] (2024) (describing reasonable royalties as available “in rare instances, such as where plainti —

cannot prove damages and there is no unjust enrichment”). 5 Uni — . Trade Secrets Act § 3(b) (Uni — . Law Comm’n 1985); 18 U.S.C. § 1836(b)(3)(B)(ii) (2016). 6 Frank H. Knight, Risk, Uncertainty and Pro — it 19-20 (1921). 7 See Bigelow v. RKO Radio Pictures, Inc., 327 U.S. 251, 264 (1946) (“The most elementary conceptions o — justice and public policy require that the wrongdoer shall bear the risk o — the uncertainty which his own wrong has created.”). Melamed’s property-rule versus liability-rule — ramework to explain why reasonable royalties convert trade secret protection — rom exclusionary rights into court-priced licenses. That conversion may be justi — ied when courts have reliable valuation in — ormation, but it becomes arbitrary when no such in — ormation exists. Part II concludes by showing why the hypothetical negotiation — ramework breaks down when applied to uncertain-value secrets. Part III proposes a calibration principle: courts should require plainti —

s to demonstrate meaning — ul price-discovery evidence be — ore authorizing reasonable royalties and should presume the remedy unavailable — or uncertain-value secrets. Sorrento provides the proo — o — concept. The opinion en — orces the statutory precondition and recognizes an epistemic boundary between what courts can interpret — rom market evidence and what they cannot know at all. Part III applies the calibration principle to paradigm cases and shows how it clari — ies the proper role o — expert testimony in reasonable royalty litigation. This Article o —

ers a modest way to organize a persistent doctrinal puzzle. Trade secret law has borrowed the reasonable royalty remedy — rom patent law without — ully adapting it to trade secret’s distinctive — eatures. Unlike patents, trade secrets have no — ixed duration, no public disclosure requirement, and no guarantee o — exclusivity. The hypothetical negotiation must account — or possibilities that patentees need not consider: independent development, reverse engineering, and law — ul discovery. More — undamentally, trade secrets o — ten reach litigation without ever being subjected to market valuation. Courts asked to award reasonable royalties in such cases must decide whether to engage in unconstrained price-setting or to acknowledge the limits o — judicial competence. The calibration principle suggests courts should choose the latter course.

   I.       The Statutory Framework and Three Families o ---  Cases

    The reasonable royalty provision in trade secret law presents a problem. The statutory text treats it as a conditional, last-resort remedy, available only when actual damages and unjust enrichment cannot be proven. Yet courts have borrowed the hypothetical negotiation  --- ramework

rom patent law, where reasonable royalties — unction as a standard measure whenever in — ringement is proven. This Part resolves the puzzle by showing that reasonable royalty cases

all along a spectrum based on the availability o — market anchors — or valuation. At one end lie cases where licensing history, contractual provisions, or documented valuations supply concrete evidence o — value. At the other end lie cases where no such evidence exists and courts must engage in unconstrained price-setting. In between lies a middle category where thin evidence provides some guidance but leaves substantial uncertainty about value. The statutory precondition — unctions most coherently when applied to this spectrum: reasonable royalties work well — or market-anchored cases, require care — ul scrutiny — or thin-evidence cases, and become problematic — or uncertain-value cases.

   A. The UTSA and DTSA Text

   The Uni --- orm Trade Secrets Act establishes a remedial hierarchy. Section 3(a) provides that damages can include both the actual loss caused by misappropriation and the unjust enrichment caused by misappropriation that is not taken into account in computing actual loss.8 This dual measure allows plainti ---

s to recover either their own losses or the de — endant’s gains,

8 Uni — . Trade Secrets Act § 3(a) (Uni — . Law Comm’n 1985). whichever is greater, and both i — they do not overlap.9 The statute thus prioritizes compensation tethered to proven economic harm. Section 3(b) then adds a conditional remedy: “I — neither damages nor unjust enrichment caused by misappropriation are provable, the court may order payment o — a reasonable royalty

or no longer than the period o — time the use could have been prohibited.”10 The text signals several important limitations. First, the reasonable royalty is available only when actual damages and unjust enrichment are not provable. Second, the remedy is discretionary (“may order”) rather than mandatory. Third, the duration is capped at the period during which injunctive relie — could have prevented use. The structure suggests that reasonable royalties — unction as a sa — ety valve rather than a primary remedy. The — ederal De — end Trade Secrets Act, enacted in 2016, mirrors this — ramework. It authorizes damages — or actual loss and unjust enrichment under 18 U.S.C. § 1836(b)(3)(B)(i), and then provides that reasonable royalty damages may be awarded “i — neither damages nor unjust enrichment caused by the misappropriation are provable.”11 The DTSA’s legislative history indicates that Congress intended to harmonize — ederal remedy provisions with the UTSA’s approach while providing a — ederal — orum — or trade secret litigation involving interstate or international commerce.12 Courts and commentators have long described the reasonable royalty as a — allback or last- resort remedy. As Milgrim on Trade Secrets explains, reasonable royalties are available “in rare instances, such as where plainti —

cannot prove damages and there is no unjust enrichment.”13 This characterization re — lects the remedy’s subsidiary role in the statutory scheme. Unlike patent law, where reasonable royalties serve as a standard measure when in — ringement is proven but lost pro — its cannot be established, trade secret law treats royalties as an exceptional remedy conditioned on the unavailability o — more direct compensation measures. The statutory precondition has generated interpretive disputes. Some courts read “not provable” to mean that plainti —

s must a —

irmatively demonstrate the unavailability o — both actual damages and unjust enrichment be — ore seeking a royalty.14 Others treat “not provable” as satis — ied whenever the evidence — ails to establish damages or unjust enrichment with reasonable certainty.15 The di —

erence matters: the — ormer reading imposes a heightened burden on plainti —

s, while the latter allows reasonable royalties whenever conventional damage measures — all short. The statutory text supports the stricter reading. By requiring that “neither” measure be provable, the statute suggests a conjunctive precondition: both avenues must be closed be — ore the court may turn to the hypothetical negotiation — ramework.

9 The dual recovery provision prevents double counting but does not limit plainti —

s to one measure or the other. See 4 Roger M. Milgrim, Milgrim on Trade Secrets § 15.02[3][a] (2024) (explaining that plainti —

s may recover both actual loss and unjust enrichment so long as the unjust enrichment “is not taken into account in computing actual loss”). 10 Uni — . Trade Secrets Act § 3(b) (Uni — . Law Comm’n 1985). 11 18 U.S.C. § 1836(b)(3)(B)(ii) (2016). 12 See S. Rep. No. 114-220, at 6-8 (2016) (explaining that DTSA remedies provisions were modeled on UTSA — ramework to provide consistency across state and — ederal trade secret law). 13 Milgrim, supra note 2, § 15.02[1][a] (emphasis added). 14 See Sorrento Therapeutics, Inc. v. Mack, 2025 Del. Ch. LEXIS 195, at *37-43 (Del. Ch. July 31, 2025) (holding that plainti —

s must a —

irmatively demonstrate that both actual damages and unjust enrichment are unavailable be — ore reasonable royalties may be considered). 15 See TXCO Res., Inc. v. Peregrine Petroleum, L.L.C., 475 B.R. 781, 821-23 (Bankr. W.D. Tex. 2012) (treating reasonable royalty as available when lost pro — its cannot be proven with reasonable certainty, without requiring a —

irmative showing that unjust enrichment is also unavailable). B. The Hypothetical Negotiation Framework

     When the statutory preconditions are satis --- ied, courts calculate reasonable royalties using a hypothetical negotiation methodology borrowed  --- rom patent law. The  --- ramework asks what a willing licensor and willing licensee would have agreed to in an arms-length transaction at the time misappropriation began.16 Courts instruct juries to imagine that the parties sat down to negotiate a license knowing what they knew then, not what became apparent later during litigation.
     The approach has deep roots in trade secret jurisprudence. In University Computing Co. v. Lykes-Youngstown Corp., the Fi --- th Circuit endorsed using “the actual value o ---  what has been appropriated” as the measure o ---  reasonable royalty damages.17 The court emphasized  --- lexibility, noting that trade secret valuation o --- ten requires consideration o ---

actors that would be irrelevant in patent cases, including the secrecy itsel — and the possibility o — independent development or reverse engineering. Patent damages assume a — ixed term o — exclusivity; trade secret damages must account — or the possibility that the de — endant could have law — ully obtained the in — ormation through alternative means. The hypothetical negotiation — ramework provides structure, but it does not eliminate indeterminacy. Courts typically instruct experts and juries to consider — actors such as: (1) the nature and value o — the secret; (2) the plainti —

’s licensing history, i — any; (3) the de — endant’s expected pro — its — rom use o — the secret; (4) the duration o — the de — endant’s competitive advantage; (5) development costs the de — endant avoided; (6) industry practices regarding similar technology; and (7) any other relevant economic evidence.18 These — actors echo the Georgia- Paci — ic — actors — amiliar in patent law, but trade secret courts have adapted them to — it a context where exclusivity is contingent rather than guaranteed. The — ramework assumes that valuation is di —

icult but not impossible. It presupposes that parties operating in good — aith could reach agreement on price even i — the negotiation would be complex. That assumption holds reasonably well when the secret has been commercialized, licensed to others, or otherwise subjected to market discipline. It becomes problematic when the secret’s value is genuinely uncertain at the time o — misappropriation, a point we develop in Part II.

 C. Three Reasonable Royalty Families

    Reasonable royalty cases  --- all into three  --- amilies based on the strength o ---  the evidentiary

oundation. These — amilies represent points along a spectrum rather than rigid categories. Cases at the market-anchored end supply concrete valuation evidence. Cases at the uncertain-value end lack any meaning — ul anchor. Cases in the middle have some evidence but require substantial judicial gap- — illing.

16 See Milgrim, supra note 2, § 15.02[3][b][ii] (describing hypothetical negotiation as asking what “a willing buyer and a willing seller would have agreed to at the time o — the misappropriation”). 17 Univ. Computing Co. v. Lykes-Youngstown Corp., 504 F.2d 518, 539 (5th Cir. 1974). 18 These — actors are adapted — rom the Georgia-Paci — ic — actors used in patent reasonable royalty analysis. See Georgia-Paci — ic Corp. v. U.S. Plywood Corp., 318 F. Supp. 1116, 1120 (S.D.N.Y. 1970) (setting — orth 15 — actors — or calculating reasonable royalty in patent cases). Trade secret courts have modi — ied these — actors to account — or di —

erences between patent and trade secret law, including the absence o —

ixed exclusivity periods and the possibility o — law — ul independent development. 1. Market-Anchored Royalties

    Market-anchored cases involve secrets that have been licensed, valued internally, or otherwise subjected to market discipline be --- ore misappropriation. These cases provide concrete evidence o ---  what parties actually agreed to or would have agreed to in arms-length transactions. The plainti ---

can point to evidence o — value that existed independent o — the litigation, and the court’s role becomes interpreting and applying that evidence rather than inventing a valuation

rom whole cloth. Mid-Michigan Computer Systems, Inc. v. Marc Glassman, Inc. exempli — ies this category.19 The plainti —

licensed pharmacy management so — tware under agreements containing liquidated damages clauses. I — a licensee wrong — ully accessed source code held in escrow, it would pay $50,000 per pharmacy location using the so — tware. When the de — endant misappropriated elements o — the source code to develop a competing conversion program, the plainti —

sought a reasonable royalty based on the liquidated damages — ormula. The jury awarded $2 million, calculated as 40 pharmacies multiplied by $50,000. The Sixth Circuit upheld the award, reasoning that the parties had established their own valuation through the contractual provision and that applying it to calculate the “hypothetically agreed value o — what the de — endant wrong — ully obtained” was not clearly excessive.20 The strength o — Mid-Michigan lies in its evidentiary — oundation. The liquidated damages clause was not created — or litigation; it predated any dispute and re — lected the parties’ contemporaneous business judgment about value. The plainti —

did not ask the jury to speculate about what the source code was worth. It pointed to a contractual term that already answered that question. The court’s task was to determine whether applying that term was reasonable under the circumstances, not to construct a valuation methodology — rom scratch.21 Other market-anchored cases rely on established licensing rates, internal — inancial projections, or documented development costs. Courts have anchored reasonable royalties to — ees charged — or legitimate access to proprietary databases, to industry-standard royalty rates — or comparable technologies, or to the plainti —

’s own licensing history with unrelated third parties.22 The common thread is concrete evidence that provides an objective benchmark — or valuation. The hypothetical negotiation remains hypothetical, but it is anchored to actual market behavior.

 2. Thin-Evidence Royalties

    A middle category involves secrets with some market evidence but not enough to provide precise valuation. These cases typically rely on industry benchmarks, market comparables, or documented costs that supply rough guidance but leave substantial uncertainty. The hypothetical negotiation requires interpolation and judgment calls, but it remains tethered to external re --- erence points.

19 Mid-Michigan Computer Sys., Inc. v. Marc Glassman, Inc., 416 F.3d 505 (6th Cir. 2005). 20 Id. at 512-13. 21 The court noted that the liquidated damages clause “was designed to compensate [plainti —

] in the event that [de — endant] improperly obtained access to the source code” and that applying it to measure reasonable royalty damages was consistent with the hypothetical negotiation — ramework. Id. at 513. The clause provided objective evidence o — what the parties themselves had agreed the wrong — ul access would be worth. 22 Courts in other market-anchored cases have relied on plainti —

’s established — ee schedules — or database access, industry- standard royalty rates — or comparable technologies, and plainti —

’s licensing history with third parties. The common thread is pre- litigation evidence o — value established through market behavior. TXCO Resources, Inc. v. Peregrine Petroleum, L.L.C. illustrates this category.23 The de — endant misappropriated the plainti —

’s seismic data and geological assessments — or oil and gas exploration in Texas. The plainti —

’s lost-pro — its claim — ailed because it could not prove with reasonable certainty what pro — its it would have earned had the de — endant not drilled competing wells in the same geological — ormation. Unjust enrichment was equally di —

icult to establish because the de — endant’s gains came — rom complex drilling operations in which the seismic data was only one input among many contributing — actors. The court turned to reasonable royalty analysis and — ound evidence o — what comparable seismic surveys cost in the same region. Expert testimony established a range o — market prices

or similar data acquired in arms-length transactions. The court acknowledged that this calculation involved “rough justice” given the evidentiary limitations, but — ound su —

icient support in industry practices and market comparables to sustain the award.24 The hypothetical negotiation was not anchored to the plainti —

’s own pricing or licensing history, but it was in — ormed by what other parties in the industry actually paid — or similar in — ormation. TXCO illustrates how market anchors need not be per — ect to be serviceable. The plainti —

could not point to its own licensing agreements or contractual provisions establishing value. But it could demonstrate what similar data cost in observable market transactions. That evidence, while imper — ect, supplied a — oundation that courts could work with. The hypothetical negotiation remained tethered to external benchmarks even i — those benchmarks required interpolation and expert judgment to apply.25 The case sits on the spectrum between market-anchored and uncertain-value cases: there was enough market in — ormation to avoid pure speculation, but not enough to eliminate substantial uncertainty about the appropriate royalty rate.

 3. Uncertain-Value Royalties

    At the  --- ar end o ---  the spectrum lie cases where no meaning --- ul market anchor exists. These cases typically involve technical know-how, preliminary research, process improvements, or strategic insights that were never commercialized, never licensed, and never valued be --- ore misappropriation. The plainti ---

proves that the de — endant obtained something o — competitive value, but cannot quanti — y that value through conventional means. The hypothetical negotiation asks courts to imagine what parties would have agreed to when there is no evidence o — what either party thought the in — ormation was worth at the time. AirFacts, Inc. v. Amezaga illustrates judicial skepticism toward uncertain-value royalties.26 The plainti —

operated a subscription service providing real-time in — ormation about private aircra — t movements. A — ormer employee took customer lists and proprietary — light- tracking methods to a competitor. The plainti —

could not establish lost pro — its with reasonable certainty because market volatility and causation problems made damage calculations speculative. Unjust enrichment — oundered on similar di —

iculties: the de — endant’s competing service used di —

erent technology and served overlapping but distinct customer bases, making it

23 TXCO Res., Inc. v. Peregrine Petroleum, L.L.C., 475 B.R. 781 (Bankr. W.D. Tex. 2012). 24 Id. at 822-23. 25 The court emphasized that “rough justice” was appropriate given the di —

iculty o — proving lost pro — its or unjust enrichment with certainty, but still required evidence that comparable market transactions existed and could in — orm the valuation. Id. The plainti —

’s burden was to provide enough market evidence to distinguish the case — rom pure speculation, even i — that evidence did not permit precise calculation. 26 AirFacts, Inc. v. Amezaga, 30 F.4th 359 (4th Cir. 2022). impossible to trace what portion o — the de — endant’s revenues came — rom the misappropriated in — ormation. The plainti —

’s damages expert proposed a reasonable royalty calculated as a percentage o — the de — endant’s revenues — rom the competing service. The district court excluded the testimony under Daubert — or — ailing to tie the proposed royalty to concrete valuation evidence. On appeal, the Fourth Circuit a —

irmed, emphasizing that “damages may not be based on mere speculation, guess, or conjecture” and that expert testimony “must be grounded in the methods and procedures o — science and must be more than unsupported speculation or subjective belie — .”27 The court remanded — or — urther proceedings but signaled that substantially more evidence would be required to support any reasonable royalty award. AirFacts does not categorically prohibit reasonable royalties — or customer lists or business intelligence. It holds that when plainti —

s seek such royalties, they must provide evidence beyond the — act o — misappropriation and the de — endant’s subsequent revenues. The expert needed to explain why a willing licensee would have paid the proposed percentage, what comparable licenses existed in the industry, or what other objective — actors supported the calculation. Without such evidence, the hypothetical negotiation becomes untethered — rom any observable market behavior.28 The case illustrates the problem o — uncertain-value secrets: when neither party placed a value on the in — ormation be — ore misappropriation, and when no comparable transactions provide guidance, asking a jury to construct a hypothetical license agreement amounts to asking it to guess.

 D. The Doctrinal Puzzle

    These three  --- amilies reveal how the statutory precondition operates in practice and why it matters. Market-anchored cases satis --- y the precondition almost by de --- inition: i ---  concrete valuation evidence exists, courts can use it  --- or reasonable royalty calculations even when lost pro --- its and unjust enrichment remain di ---

icult to prove with precision. The evidence provides a substitute measure o — value that avoids speculation. Uncertain-value cases — ail the precondition: i — no valuation anchor exists, “not provable” means that courts lack the evidentiary — oundation to calculate any monetary remedy reliably. Thin-evidence cases — all in between and present the hardest line-drawing problems. They have enough evidence to avoid pure guesswork but not enough to eliminate substantial uncertainty. The statutory text treats reasonable royalties as conditional, available only when actual damages and unjust enrichment cannot be proven. That condition makes sense when understood as requiring meaning — ul price-discovery evidence. I — courts can identi — y what the secret was worth through licensing history, internal valuations, documented costs, or market comparables, the hypothetical negotiation has evidentiary support. The court interprets market in — ormation rather than inventing it. I — courts cannot identi — y such evidence, awarding a reasonable royalty requires judicial creation o — value rather than judicial interpretation o — existing market signals.

27 Id. at 367 (internal quotation marks omitted). 28 On remand, the case would need to address whether plainti —

could provide evidence such as: what plainti —

charged — or legitimate data access, what competitors paid — or similar customer in — ormation, or what industry standards suggested about the value o — customer lists in the private aviation services sector. The Fourth Circuit’s opinion signaled that merely pointing to the de — endant’s revenues and proposing a percentage royalty, without evidence linking that percentage to market behavior, would be insu —

icient. Sorrento Therapeutics, Inc. v. Mack en — orces this understanding.29 The Delaware Court o — Chancery re — used to award reasonable royalties under Cali — ornia’s UTSA where the plainti —

had not demonstrated that both actual damages and unjust enrichment were unavailable. The court held that the statutory precondition requires an a —

irmative showing that both measures are not provable be — ore reasonable royalties become available. The court also held that the plainti —

’s damages expert had improperly derived the proposed royalty — rom lost-pro — its and unjust- enrichment — igures that had been excluded as speculative. I — those measures lacked evidentiary support su —

icient — or direct recovery, they could not supply inputs — or the hypothetical negotiation. The opinion signals that the statutory precondition — unctions as a substantive limitation on speculative damages rather than a mere procedural hurdle that plainti —

s can satis — y by nominally attempting to prove actual damages be — ore pivoting to reasonable royalties. Part II explains why uncertain-value secrets present distinct theoretical challenges — or judicial valuation and why the statutory precondition serves important — unctions in limiting courts’ engagement with speculative price-setting.

 II.      Why Uncertain-Value Secrets Are Di ---

erent

    Part I showed that reasonable royalty cases  --- all along a spectrum based on the availability o ---  market anchors  --- or valuation. This Part explains why uncertain-value secrets present distinct challenges that market-anchored and thin-evidence cases do not. When parties can point to licensing history, internal valuations, or market comparables, the hypothetical negotiation interprets existing market in --- ormation. When no such anchors exist, courts must create value rather than discover it. Three interconnected problems arise. First, uncertain-value secrets involve what Frank Knight called “uncertainty” rather than calculable “risk,” making probability-based valuation impossible. Second, reasonable royalties convert property-rule protection into liability-rule protection,  --- orcing exchanges at court-determined prices when courts lack reliable valuation in --- ormation. Third, the hypothetical negotiation  --- ramework assumes parties could have reached agreement, but that assumption breaks down when neither party knew what the secret was worth at the time o ---  misappropriation.

 A. Risk, Uncertainty, and the Limits o ---  Valuation

   Economists distinguish “risk”  --- rom “uncertainty” in ways that matter  --- or trade secret damages.30 Risk re --- ers to situations where outcomes are unknown but probabilities can be estimated. A coin  --- lip involves risk: we cannot predict any individual  --- lip, but we know the probability distribution. Uncertainty, by contrast, re --- ers to situations where neither outcomes nor probabilities can be reliably estimated. Frank Knight developed this distinction to explain entrepreneurial pro --- it: entrepreneurs earn returns by success --- ully bearing true uncertainty rather than merely insuring against calculable risk.31
   When courts apply the hypothetical negotiation  --- ramework to market-anchored secrets, they work with risk. Licensing history may not reveal the precise value, but it provides data  --- rom which parties can estimate ranges and probabilities. Internal valuations re --- lect management’s

29 Sorrento Therapeutics, Inc. v. Mack, 2025 Del. Ch. LEXIS 195 (Del. Ch. July 31, 2025). 30 Frank H. Knight, Risk, Uncertainty and Pro — it 19-20 (1921). 31 Knight argued that measurable uncertainty (risk) can be managed through insurance and diversi — ication, while unmeasurable uncertainty cannot. Entrepreneurs earn pro — its by success — ully bearing unmeasurable uncertainty, while workers and capital providers receive — ixed returns because they bear only calculable risk. Id. at 269-70. probability-weighted projections. Market comparables show what similar parties paid in analogous transactions. Uncertainty remains, but it is the manageable uncertainty o — estimation error rather than the deep uncertainty o — unknowability. Parties operating in good — aith could negotiate a license even i — they disagreed about precise values, because both sides could articulate their valuations and explain their reasoning. Uncertain-value secrets present Knightian uncertainty rather than calculable risk. Consider a concrete scenario: a de — endant misappropriates preliminary research notes documenting experiments that might lead to a marketable pharmaceutical product. At the time o —

misappropriation, the research is incomplete and its commercial prospects are genuinely unknown. Neither plainti —

nor de — endant knows whether the experiments will yield a viable drug candidate, whether that candidate would survive clinical trials, whether regulatory approval would be obtained, or what market share a success — ul drug might capture. What would a willing licensor demand and a willing licensee pay — or access to these notes at the time o — misappropriation? A risk-averse licensor might demand substantial up — ront payment plus royalties on any — uture revenues, reasoning that the research has option value even i — its probability o — success is low. A risk-averse licensee might re — use to pay anything signi — icant

or speculative leads, reasoning that most pharmaceutical research — ails and that investing in unproven concepts would be economically irrational. A risk-neutral party might try to calculate expected value by probability-weighting potential outcomes. But i — neither party can estimate the relevant probabilities with any con — idence, expected value calculations become impossible.32 Both parties — ace Knightian uncertainty, and the hypothetical negotiation has no determinate answer. Patent law con — ronted similar problems and developed doctrines to constrain speculative damages. Courts prohibit calculating reasonable royalties as a percentage o — de — endants’ total product revenues when the patented — eature contributes only part o — that value.33 Apportionment requirements mandate that royalty calculations isolate the value attributable to the patented invention rather than to unpatented — eatures.34 These constraints re — lect judicial recognition that juries should not be invited to award damages untethered to what was actually misappropriated. Trade secret law lacks these guardrails, yet the same concerns about speculation apply with equal

orce.

 B. Property Rules, Liability Rules, and Judicial Price-Setting

    Guido Calabresi and A. Douglas Melamed distinguished “property rules”  --- rom “liability rules” in ways that illuminate reasonable royalty problems.35 Property rules protect entitlements through injunctions. Entitlement holders can re --- use any transaction, and others may acquire the entitlement only through voluntary exchange at prices the holder accepts. Liability rules protect entitlements through damages. Entitlements can be taken without consent so long as takers pay

32 This problem is distinct — rom situations where parties disagree about probabilities but each can articulate a probability estimate. Under Knightian uncertainty, parties lack su —

icient in — ormation to — orm probability estimates at all. See Itzhak Gilboa & David Schmeidler, Maxmin Expected Utility with Non-Unique Prior, 18 J. Math. Econ. 141 (1989) ( — ormalizing decision-making under Knightian uncertainty). 33 See VirnetX, Inc. v. Cisco Sys., Inc., 767 F.3d 1308, 1326-27 (Fed. Cir. 2014) (holding that reasonable royalty must be apportioned to re — lect value o — patented — eature rather than entire product). 34 Id. at 1327 (requiring that damages be commensurate with the value that the patented — eature adds to the in — ringing product). 35 Guido Calabresi & A. Douglas Melamed, Property Rules, Liability Rules, and Inalienability: One View o — the Cathedral, 85 Harv. L. Rev. 1089 (1972). court-determined compensation. Property rules give entitlement holders veto power; liability rules permit — orced exchanges at judicially set prices. Traditionally, trade secret law relied on property rules. Courts enjoined misappropriation and le — t parties to negotiate voluntary licenses i — they wished.36 Plainti —

s could re — use to license at any price or could demand terms de — endants — ound unacceptable. I — negotiations — ailed, de — endants had to develop the in — ormation independently or do without. This arrangement placed the burden o — negotiation on parties who possessed in — ormation about their own valuations and had incentives to reveal that in — ormation truth — ully through bargaining. Reasonable royalties convert this property-rule protection into liability-rule protection. Courts — orce licenses on plainti —

s at prices courts set retrospectively. De — endants obtain entitlements without plainti —

s’ consent, and plainti —

s receive compensation courts determine rather than amounts plainti —

s would have accepted voluntarily.37 Calabresi and Melamed cautioned that liability rules work well only when courts have reliable in — ormation about value. Without such in — ormation, — orced exchanges at court-set prices risk systematic over- compensation or under-compensation.38 This risk intensi — ies — or uncertain-value secrets. A court that awards substantial reasonable royalties — or preliminary research that never produces commercial results may over- compensate plainti —

s. A court that awards nominal damages — or research that eventually yields valuable products may under-compensate. Neither party can appeal to market evidence to demonstrate error because no market existed at the relevant time. Courts must make normative judgments about appropriate compensation rather than empirical — indings about market value. Nuisance cases provide instructive analogies. In Whalen v. Union Bag & Paper Co., the New York Court o — Appeals upheld an injunction requiring a pulp mill to stop discharging pollutants into a stream, even though the mill represented an investment exceeding one million dollars and the plainti —

’s damages were comparatively small.39 Property rights deserved property-rule protection, and “balancing o — injuries cannot be justi — ied by the circumstances o —

this case.”40 Decades later, Boomer v. Atlantic Cement Co. awarded permanent damages in lieu o — an injunction, reasoning that relative hardship and public interest counseled against shutting down a cement plant.41 Yet even Boomer required proo — o — actual harm; it authorized damages based on proven losses rather than speculative estimates.42 Similar principles should constrain trade secret reasonable royalties. Courts can protect entitlements with injunctions. They can award damages based on proven losses. They can trace de — endants’ actual gains through unjust enrichment measures. What ordinary damages law resists is awarding compensation based on speculation about value when neither party could have determined that value at the relevant time.

36 See Milgrim, supra note 2, § 15.02[2] (discussing the traditional primacy o — injunctive relie — in trade secret litigation). 37 The UTSA also permits ongoing royalty injunctions in exceptional circumstances, where the court orders an ongoing royalty in lieu o — prohibiting — uture use. See Uni — . Trade Secrets Act § 2 cmt. These “reasonable royalty injunctions” present similar valuation challenges when no market anchor exists. 38 Calabresi & Melamed, supra note 28, at 1106-10. 39 Whalen v. Union Bag & Paper Co., 208 N.Y. 1 (1913). 40 Id. at 4-5. 41 Boomer v. Atlantic Cement Co., 26 N.Y.2d 219 (1970). 42 The dissent in Boomer emphasized this point, noting that even under the majority’s permanent damages approach, plainti —

s still had to prove actual harm rather than speculative — uture harm. Id. at 229 (Jasen, J., dissenting). The majority authorized substituting permanent damages — or an injunction but did not authorize awarding damages — or harms that could not be proven. C. When Hypothetical Negotiations Become Indeterminate

    Hypothetical negotiation methodology assumes parties could have reached agreement had they negotiated at the time misappropriation began. For market-anchored secrets, this assumption holds tolerably well. Licensing history, internal valuations, and market comparables reveal what parties likely would have agreed to by showing what similar parties actually agreed to in comparable circumstances. Courts interpret market evidence rather than invent it.
    Uncertain-value secrets strain this assumption to the breaking point. When no market anchor exists, courts must imagine what parties would have agreed to when the parties themselves would not have known what to agree to. I ---  parties  --- aced Knightian uncertainty, there may be no determinate price both would have accepted. Without evidence o ---  what parties  --- acing similar uncertainty actually agreed to, courts cannot ground the hypothetical negotiation in market behavior. Any price selected becomes arbitrary in the sense that it cannot be tied to how actual markets resolved similar valuation problems.
    This indeterminacy di ---

ers — rom bilateral monopoly. In bilateral monopoly situations, parties know the value but — ace strategic bargaining challenges. Each party has incentives to misrepresent its reservation price to capture more surplus. Game theory provides tools — or analyzing such strategic behavior, and courts can sometimes in — er likely outcomes by examining how similar bilateral monopolies were resolved. Under Knightian uncertainty, however, there is no “true” value to discover and no strategic bargaining problem to solve. Parties genuinely do not know what the secret is worth, and examining how other parties resolved similar problems provides no guidance because those parties — aced the same indeterminacy.43 Identi — ying — actors relevant to valuation becomes equally problematic. For market- anchored secrets, courts can evaluate licensing history, development costs, industry standards, and comparable transactions. Experts can testi — y about how these — actors in — ormed actual licensing negotiations. For uncertain-value secrets, however, identi — ying relevant — actors requires speculation. Should the hypothetical royalty be based on development costs, probability- weighted expected value, or de — endants’ avoided costs? Without market evidence showing how actual parties weighted these — actors, courts must make normative choices about how value ought to be measured rather than empirical — indings about how it was measured.44 Problems compound when plainti —

s’ damages experts work backwards — rom litigation- generated valuations. I — experts start with de — endants’ actual revenues or pro — its and calculate royalties as percentages, the analysis assumes the secret’s value equals its ex post commercial impact. But this assumption is precisely what the hypothetical negotiation should test. At the time o — misappropriation, neither party knew what commercial impact the secret would have. Using hindsight to determine royalty rates converts the exercise — rom hypothetical negotiation into penalty based on actual results, which may bear no relationship to what parties would have agreed to ex ante.45

43 The inability to — orm probability distributions distinguishes Knightian uncertainty — rom standard risk scenarios. Under risk, parties can disagree about probabilities but both can calculate expected values. Under Knightian uncertainty, even the attempt to calculate expected values becomes indeterminate because no probability distribution exists. 44 This is the problem o — “deep” versus “shallow” indeterminacy. Shallow indeterminacy involves disagreement about how to apply agreed-upon — actors. Deep indeterminacy involves disagreement about what — actors are relevant at all. Uncertain-value secrets present deep indeterminacy that cannot be resolved through better expert testimony or more sophisticated methodology. 45 Courts in patent cases have similarly cautioned against using hindsight to in — late reasonable royalty awards. See Whitserve, LLC v. Computer Packages, Inc., 694 F.3d 10, 32 (Fed. Cir. 2012) (noting that “a — undamental precept o — the hypothetical negotiation is that the results o — the in — ringement cannot be considered in the analysis”). Sorrento Therapeutics identi — ied precisely this methodological — law. The court held that the plainti —

’s expert had improperly anchored the proposed royalty to lost-pro — its and unjust- enrichment — igures already excluded as speculative. I — those measures lacked su —

icient evidentiary support — or direct recovery, they could not supply inputs — or the hypothetical negotiation.46 The opinion signals that courts should re — use to award reasonable royalties when the only available evidence consists o — speculative projections untethered to contemporaneous market behavior. Part III applies these insights to develop a calibration principle: reasonable royalties should be available only when plainti —

s can demonstrate meaning — ul price-discovery evidence, and courts should presume the remedy unavailable — or uncertain-value secrets.

 III.     The Calibration Principle

    Parts I and II established that reasonable royalty cases  --- all along a spectrum and that uncertain-value secrets present distinct challenges  --- or judicial valuation. This Part proposes a practical  --- ramework: courts should reserve reasonable royalties  --- or cases with meaning --- ul price- discovery evidence and presume the remedy unavailable  --- or uncertain-value secrets. Sorrento Therapeutics provides the proo ---  o ---  concept. Properly understood, the opinion en --- orces the statutory precondition and recognizes an epistemic boundary between what courts can interpret

rom market evidence and what they cannot know at all.

 A. Sorrento and Judicial Valuation Under Uncertainty

    Sorrento Therapeutics, Inc. v. Mack arose  --- rom a  --- ailed business relationship in the pharmaceutical industry.47 Anthony Mack served as president o ---  Scilex Pharmaceuticals while simultaneously diverting development opportunities to Virpax Pharmaceuticals, a competing entity he controlled.48 The Delaware Court o ---  Chancery  --- ound liability  --- or breach o ---

iduciary duty, breach o — contract, and trade secret misappropriation under Cali — ornia’s UTSA.49 The remedial phase, however, proved more di —

icult. Plainti —

s sought reasonable royalty damages through a hypothetical negotiation

ramework. Their expert, Dr. Darius Lakdawalla, constructed a “zone o — potential agreement” bounded by plainti —

s’ projected lost pro — its and de — endants’ avoided development costs.50 Working — rom this range, Lakdawalla recommended a royalty approaching $6.7 million.51 On its

ace, the methodology tracked conventional approaches borrowed — rom patent law. It reconstructed a negotiation and applied — amiliar valuation — actors. Vice Chancellor Fioravanti re — used to award any reasonable royalty. His analysis proceeded through two distinct but related steps. First, Cali — ornia’s UTSA conditions reasonable royalties on an a —

irmative showing that “neither damages nor unjust enrichment caused by misappropriation are provable.”52 Plainti —

s had not made that showing. They had abandoned their lost-pro — its theory a — ter trial rulings excluded portions o — their evidence and dropped their

46 Sorrento Therapeutics, Inc. v. Mack, 2025 Del. Ch. LEXIS 195, at *37-43 (Del. Ch. July 31, 2025). 47 Sorrento Therapeutics, Inc. v. Mack, 2025 Del. Ch. LEXIS 195 (Del. Ch. July 31, 2025). 48 Id. at *5-10. 49 Id. at *11-24. 50 Id. at *31-33. 51 Id. 52 Id. at *37-38 (quoting Cal. Civ. Code § 3426.3(b)). unjust-enrichment theory a — ter settling with the corporate de — endant Virpax.53 Rather than demonstrate that both conventional measures were genuinely unavailable, plainti —

s treated the reasonable royalty as their pre — erred option once other approaches proved inconvenient. Vice Chancellor Fioravanti read Cali — ornia’s conjunctive “neither…nor” language literally: both avenues must be closed be — ore reasonable royalties become available.54 Second, even i — the statutory precondition were satis — ied, plainti —

s’ methodology was

undamentally — lawed. Lakdawalla had derived his royalty range — rom the very lost-pro — its and unjust-enrichment — igures that lacked su —

icient evidentiary support — or direct recovery.55 Plainti —

s con — ronted what the opinion aptly termed a “Catch-22”: the statute makes reasonable royalties available only when other measures are not provable, yet the expert’s analysis required those unprovable measures as inputs.56 Using hindsight to determine royalty rates converts the exercise — rom hypothetical negotiation into penalty calculation. At the time o — misappropriation, neither party knew what — uture outcomes would be. Reconstructing a license based on ex post knowledge o — what happened trans — orms judicial valuation into speculation.57

 B. The Epistemic Boundary

    Sorrento brings into  --- ocus a deeper question about what courts can know. Market- anchored cases present valuation problems: evidence exists but requires interpretation. Uncertain-value cases present epistemological problems: the evidence needed to value the secret may not exist at all. Distinguishing between these requires courts to recognize an epistemic boundary.
    Consider what parties negotiating at the time o ---  misappropriation actually knew. In market-anchored cases, they could point to licensing history, internal valuations prepared  --- or business purposes, or market comparables. Disagreements about precise values would remain, but both sides could articulate positions grounded in observable evidence. Courts reconstructing such negotiations interpret market in --- ormation rather than invent it. In Mid-Michigan, the liquidated damages clause in the escrow agreement supplied exactly this kind o ---  anchor: the parties had already negotiated a value  --- or wrong --- ul access be --- ore any dispute arose.58
    Uncertain-value cases present a di ---

erent situation. Plainti —

s in Sorrento alleged misappropriation o —

ive documents related to pharmaceutical development.59 No licensing history existed — or those documents or similar materials. No internal valuations assessed their worth. No market comparables showed what pharmaceutical companies pay — or early-stage development documents. The expert’s hypothetical negotiation rested on projected revenues

rom products not yet approved by FDA regulators.60 At the time o — misappropriation, neither party could have speci — ied probability distributions — or the relevant outcomes. The value depended on whether drug candidates would clear clinical trials, obtain regulatory approval, secure reimbursement, and compete success — ully in their markets. These contingencies were not

53 Id. at *33-35. 54 Cali — ornia’s UTSA provides that reasonable royalties are available only “i — neither damages nor unjust enrichment caused by misappropriation are provable.” Cal. Civ. Code § 3426.3(b) (emphasis added). 55 Sorrento, 2025 Del. Ch. LEXIS 195, at *40-41. 56 Id. at *41 n.123. 57 The court emphasized that allowing this methodology would permit plainti —

s to recover through the reasonable royalty provision damages that the statute explicitly makes unavailable when other measures cannot be proven. Id. 58 Mid-Michigan Computer Sys., Inc. v. Marc Glassman, Inc., 416 F.3d 505, 512-13 (6th Cir. 2005). 59 Sorrento, 2025 Del. Ch. LEXIS 195, at *32-33. 60 Id. at *32-35. merely uncertain in the sense that outcomes were unknown; they were uncertain in Frank Knight’s sense that even the probability distributions could not be speci — ied.61 When courts con — ront this kind o — uncertainty, the hypothetical negotiation asks them to imagine what parties would have agreed to when the parties themselves could not have agreed to anything determinate. Sometimes the law’s most honest answer is that the price o — a never- bargained- — or license is not di —

icult to ascertain; rather, it is genuinely unknowable based on available evidence. Asking courts to proceed anyway invites arbitrary price-setting that ordinary damages law prohibits.62

 C. The Calibration Principle Articulated

     Sorrento’s reasoning suggests a general principle: courts should require plainti ---

s to demonstrate meaning — ul price-discovery evidence be — ore authorizing reasonable royalty damages. Where such evidence exists, the hypothetical negotiation can be grounded in observable market behavior. Where it does not, courts should presume reasonable royalties unavailable and rely on alternative remedies. Price-discovery evidence includes licensing history — or the secret or substantially similar secrets, internal valuations prepared be — ore misappropriation — or business purposes, documented development costs that establish value — loors, market comparables — rom arms-length transactions in the relevant industry, or expert testimony grounded in established methodologies and tied to speci — ic market evidence. This evidence need not determine a precise value, but it must provide an anchor that distinguishes judicial interpretation — rom judicial invention. Several considerations support this approach. First, it respects statutory text. Both the UTSA and DTSA condition reasonable royalties on unavailability o — other measures.63 That condition prevents reasonable royalties — rom becoming a general substitute whenever damages are merely di —

icult rather than genuinely impossible to prove. Without price-discovery evidence, courts lack — oundations to distinguish the two. Plainti —

s who cannot establish actual loss or unjust enrichment with reasonable certainty likely cannot establish what hypothetical licensing negotiations would have produced either. Second, it acknowledges institutional limits. Courts excel at interpreting market evidence and applying legal standards to — acts. They do not excel at creating markets where none existed. When parties have not valued a secret be — ore misappropriation, when no comparable market exists, and when value depends on unknown — uture contingencies, courts cannot reliably reconstruct hypothetical negotiations. As Part II explained, Knightian uncertainty makes probability-based valuation impossible. Reasonable royalties ask courts to solve problems that parties themselves could not solve, using in — ormation that parties themselves did not possess. Third, alternative remedies remain available. Re — using reasonable royalties in uncertain- value cases does not leave plainti —

s without recourse. Injunctions prevent ongoing misappropriation and can be coupled with nominal damages to vindicate rights. Development costs provide value — loors even when precise quanti — ication proves impossible. Unjust enrichment measures capture de — endants’ actual gains when those gains can be traced. Fee-

61 Frank H. Knight distinguished measurable uncertainty (risk) — rom unmeasurable uncertainty, arguing that entrepreneurs earn pro — its by bearing the latter. Frank H. Knight, Risk, Uncertainty and Pro — it 19-20, 269-70 (1921). 62 See Bigelow v. RKO Radio Pictures, Inc., 327 U.S. 251, 264 (1946) (“The most elementary conceptions o — justice and public policy require that the wrongdoer shall bear the risk o — the uncertainty which his own wrong has created.”). Bigelow assumes, however, that some quantum o — damages can be proven even i — the precise amount remains uncertain. 63 See Uni — . Trade Secrets Act § 3(b) (Uni — . Law Comm’n 1985); 18 U.S.C. § 1836(b)(3)(B)(ii). shi — ting provisions under the DTSA deter wrongdoing and compensate plainti —

s — or litigation costs.64 Sorrento itsel — awarded injunctive relie — and damages — or breach o —

iduciary duty measured by Mack’s salary during the period o — disloyalty.65 Fourth, preserving property-rule protection matters. When courts grant injunctions, they preserve plainti —

s’ ability to negotiate voluntary licenses or re — use licensing altogether. Reasonable royalties convert property-rule protection into liability-rule protection by permitting continued use at court-determined prices.66 That conversion may be justi — ied when courts have reliable valuation in — ormation. Without such in — ormation, — orced exchanges at judicially set prices risk systematically undercompensating plainti —

s while encouraging de — endants to misappropriate — irst and pay court-set royalties later.

 D. Application and Implications

    Applying the calibration principle requires courts to evaluate the quality and speci --- icity o ---  pro ---

ered evidence rather than merely its existence. Plainti —

s o —

ering internal valuations must show those valuations were prepared — or business purposes be — ore any dispute arose, not created in anticipation o — litigation. Plainti —

s o —

ering licensing history must demonstrate that prior licenses involved su —

iciently similar secrets in comparable contexts. Plainti —

s o —

ering market comparables must establish that compared transactions involved arms-length bargaining rather than settlement or special relationships. Sorrento illustrates how this works in practice. Plainti —

s alleged misappropriation o —

documents related to pharmaceutical development but pointed to no licensing history, no internal valuations, and no market comparables. Their expert’s hypothetical negotiation relied on speculative projections about — uture FDA approval, market entry, and commercial success.67 Under the calibration principle, reasonable royalties were unavailable because the evidentiary record provided no anchor — or judicial price-setting. Future courts — acing similar — acts can cite Sorrento — or the proposition that uncertain-value secrets presumptively do not support reasonable royalty awards. Conversely, Mid-Michigan shows how the principle permits reasonable royalties when proper anchors exist. Contractual liquidated damages clauses supplied price-discovery evidence: parties had negotiated value — or unauthorized use be — ore any dispute arose.68 Courts could interpret the contractual term rather than invent a valuation. Similarly, cases where industries have established royalty rates — or particular technologies, where plainti —

s have documented licensing programs, or where internal valuations prepared — or business purposes provide benchmarks allow reasonable royalty analysis to proceed because courts interpret market in — ormation. Thin-evidence cases present harder questions. TXCO Resources involved seismic data with some market comparables but no direct licensing history — or the speci — ic in — ormation misappropriated.69 I — expert testimony establishes that comparable data trades at identi — iable rates in observable transactions, the case moves toward the market-anchored category. I — experts can only gesture toward vague analogies without pointing to actual market behavior, the case

64 See 18 U.S.C. § 1836(b)(3)(D) (authorizing attorney — ee awards in DTSA cases). 65 Sorrento, 2025 Del. Ch. LEXIS 195, at *15-25. 66 See discussion supra Part II.B (analyzing property rule versus liability rule — ramework). 67 Sorrento, 2025 Del. Ch. LEXIS 195, at *32-35. 68 Mid-Michigan, 416 F.3d at 512-13. 69 TXCO Res., Inc. v. Peregrine Petroleum, L.L.C., 475 B.R. 781, 822-23 (Bankr. W.D. Tex. 2012). remains uncertain-value and courts should be skeptical. The calibration principle provides a

ramework — or disciplined analysis without mechanical answers. The principle also clari — ies expert testimony’s proper role. Experts in market-anchored cases translate observable data into royalty rates. Experts in uncertain-value cases — ace greater constraints. Courts should subject royalty models built on speculative projections to heightened scrutiny and should be willing to exclude such testimony when it lacks grounding in price- discovery evidence. This gatekeeping — unction preserves the boundary between judicial interpretation o — markets and judicial creation o — prices where no markets existed.

Conclusion

    Reasonable royalties serve an important  --- unction in trade secret law when properly cabined. Market-anchored cases demonstrate that the remedy can work: licensing history, contractual provisions, internal valuations, and market comparables provide anchors that allow courts to interpret rather than invent value. Mid-Michigan Computer Systems illustrates the paradigm. Parties negotiated liquidated damages  --- or wrong --- ul access to escrowed source code, and courts applied that contractual term to measure reasonable royalty damages.70 The hypothetical negotiation was genuinely hypothetical—parties had not actually negotiated a reasonable royalty—but it was grounded in their actual agreement about value.
    Uncertain-value cases present  --- undamentally di ---

erent challenges. When secrets have never been commercialized, never been licensed, and never been valued, courts cannot reconstruct hypothetical negotiations without engaging in speculation. Sorrento Therapeutics brings this problem into sharp relie — . Plainti —

s alleged misappropriation o — pharmaceutical development documents but could point to no licensing history, no internal valuations, and no market comparables. Their expert’s proposed royalty rested on projections about — uture FDA approval and commercial success. Vice Chancellor Fioravanti recognized that authorizing such awards would convert the reasonable royalty provision — rom conditional remedy into judicial price-setting unconstrained by market evidence. The calibration principle proposed here o —

ers courts a — ramework — or navigating between these extremes: require plainti —

s to demonstrate meaning — ul price-discovery evidence; presume reasonable royalties unavailable — or uncertain-value secrets; and reserve the reasonable-royalties remedy — or cases where the hypothetical negotiation can be grounded in observable market behavior. This approach respects statutory text, acknowledges institutional limits, preserves alternative remedies, and maintains trade secret law’s intellectual-property-law alignment. Several limitations warrant acknowledgment. First, this Article — ocuses on damages — or past misappropriation rather than ongoing royalty injunctions — or — uture use. The latter raise distinct questions about whether courts should authorize continued misappropriation at judicially determined prices rather than enjoin — uture use altogether.71 Second, the analysis treats the three-

amily spectrum as a use — ul heuristic without claiming that courts can always classi — y cases cleanly. Thin-evidence cases present genuinely hard questions about when market comparables are su —

iciently similar to provide meaning — ul guidance. Third, the calibration principle requires courts to exercise judgment about evidence quality rather than apply mechanical tests. Some indeterminacy at the margins is inevitable. 70 Mid-Michigan Computer Sys., Inc. v. Marc Glassman, Inc., 416 F.3d 505, 512-13 (6th Cir. 2005). 71 The UTSA permits courts to order ongoing royalties in lieu o — injunctions in “exceptional circumstances.” Uni — . Trade Secrets Act § 2 cmt. Whether such remedies should be available — or uncertain-value secrets raises similar but distinct questions about prospective price-setting. Fourth, this Article makes normative claims about how courts should interpret the statutory precondition without predicting how they will behave in practice. Plainti —

s have strong incentives to seek reasonable royalties when other measures — ail, and courts may — ind it di —

icult to leave plainti —

s with only injunctive relie — when misappropriation is proven. Empirical research could test whether courts actually en — orce the statutory precondition, whether jurisdictions that treat reasonable royalties skeptically see di —

erent settlement patterns, and whether thin-evidence cases correlate with higher variance in award amounts. Such research would complement the doctrinal and theoretical analysis o —

ered here. Finally, this Article — ocuses on domestic trade secret law without addressing how reasonable royalties — unction in other jurisdictions or in international trade secret disputes. The UTSA and DTSA provide the relevant — ramework — or American courts, but trade secret litigation increasingly crosses borders. Whether the calibration principle should apply when — oreign law governs or when parties negotiate licenses across jurisdictions with di —

erent legal regimes remains an open question. Within these limits, the analysis points toward a general insight. Sometimes the law’s most honest answer is that the price o — a never-bargained- — or license is not just di —

icult to ascertain. Sometimes this counter — actual is — undamentally unknowable. Courts asked to award reasonable royalties — or uncertain-value secrets — ace a choice between unconstrained price-setting and acknowledging epistemic limits. The calibration principle counsels — or the latter approach because reasonable royalties serve as a supplement when market evidence permits meaning — ul valuation. Where such evidence is absent, courts should resist the temptation to treat judicial price-setting as mandatory and should instead recognize the boundaries o — what damages law can reliably accomplish.