From Casetext: Smarter Legal Research

Apple Inc. v. Samsung Electronics Co., Ltd.

UNITED STATES DISTRICT COURT NORTHERN DISTRICT OF CALIFORNIA SAN JOSE DIVISION
Feb 25, 2014
Case No.: 12-CV-00630-LHK (N.D. Cal. Feb. 25, 2014)

Opinion

Case No.: 12-CV-00630-LHK

02-25-2014

APPLE, INC., a California corporation, Plaintiff and Counterdefendant, v. SAMSUNG ELECTRONICS CO., LTD., a Korean corporation; SAMSUNG ELECTRONICS AMERICA, INC., a New York corporation; and SAMSUNG TELECOMMUNICATIONS AMERICA, LLC, a Delaware limited liability company, Defendants and Counterclaimants.


ORDER GRANTING IN PART AND

DENYING IN PART MOTIONS TO

EXCLUDE CERTAIN EXPERT

OPINIONS


PUBLIC REDACTED VERSION

This Order addresses the parties' motions to exclude certain expert testimony. On October 16, 2013, Apple moved to exclude expert testimony of four Samsung experts: (1) Dr. Judith Chevalier; (2) Dr. James Kearl; (3) Dr. Daniel Schonfeld; and (4) Dr. Martin Rinard. See Apple Mot. (ECF No. 831-2). Apple further moved to preclude Samsung's experts from testifying about the opinions of Apple's experts in past cases. See id. at 24. Samsung filed an opposition. See Samsung Opp. (ECF No. 856-3). Apple filed a reply. See Apple Reply (ECF No. 945-3).

On October 16, 2013, Samsung moved to exclude expert testimony from five Apple experts: (1) Dr. John Hauser; (2) Dr. Christopher Vellturo; (3) Dr. Todd Mowry; (4) Dr. Alex Snoeren; and (5) Dr. James Storer. See Samsung Mot. (ECF No. 802-3). Apple filed an opposition. See Apple Opp. (ECF No. 857-4 ). Samsung filed a reply. See Samsung Reply (ECF No. 948-4).

The Court held a hearing on the parties' motions to exclude expert evidence on January 23, 2014. See ECF No. 1198 ("Daubert Hearing Tr."). Having considered the briefing, relevant law, and oral argument made at the hearing, the Court GRANTS in part and DENIES in part the motions to exclude expert evidence.

I. BACKGROUND

This case is the second patent dispute between the parties over which the undersigned is presiding. In case number 11-1846, which the Court in this Order refers to as "Apple I," the Court recently issued a ruling on the parties' post-trial motions following a jury verdict predominantly in Apple's favor. See Apple I, ECF No. 2947 (N.D Cal. Feb. 7, 2014). The present case, which the Court in this Order refers to as "Apple II," is set for a 12-day jury trial beginning March 31, 2014, over a different set of asserted patents and accused devices.

On February 4, 2014, the parties narrowed their lists of asserted patent claims to five each. Apple now asserts claim 9 from U.S. Patent No. 5,946,647 (the "'647 Patent"); claim 25 from U.S. Patent No. 6,847,959 (the "'959 Patent"); claim 20 from U.S. Patent No. 7,761,414 (the "'414 Patent"); claim 8 from U.S. Patent No. 8,046,721 (the "'721 Patent"); and claim 18 from U.S. Patent No. 8,074,172 (the "'172 Patent"). See ECF No. 1237. Samsung now asserts claim 10 from U.S. Patent No. 7,756,087 (the "'087 Patent"); claim 13 from U.S. Patent No. 7,551,596 (the "'596 Patent"); claim 27 from U.S. Patent No. 6,226,449 (the "'449 Patent"); and claims 1 and 15 from U.S. Patent No. 5,579,239 (the "'239 Patent"). See ECF No. 1236.

Some of the issues raised in the parties' Daubert motions have been resolved since the close of briefing. Those issues fall into three groups. First, Samsung challenges Apple's damages expert Dr. Vellturo in part because he evaluated lost profits based on design around dates that coincided with the date Samsung first received notice of the patents-in-suit. See Samsung Mot. at 13. In Apple I, this Court ruled in favor of Samsung on that issue. See Apple, Inc. v. Samsung Elecs. Co., No. 11-1846, 2013 WL 5958172, at *7 (Nov. 7, 2013) ("[T]he Court concludes that an accurate reconstruction of the hypothetical market that would have existed 'but for' an infringer's infringement must take into account actions the infringer could have taken in lieu of infringing, including designing around the patented intellectual property, as of the date of first infringement.") (emphasis added). For the reasons stated in that order, Samsung's motion to exclude Dr. Vellturo's use of notice dates rather than first-infringement dates in his lost profits analysis is GRANTED.

Second, on January 21, 2014, by stipulation, Apple withdrew its challenges to the opinions of Drs. Kearl and Schonfeld, and Samsung withdrew its challenges to the opinions of Dr. Storer. See ECF No. 1143. The Court approved the parties' stipulation. See ECF No. 1145.

Third, also on January 21, 2014, this Court denied the portion of Apple's summary judgment motion in Apple II that addressed the same issue that Apple raised in its motion to exclude the opinion of Dr. Rinard. See ECF No. 1150 at 29. As discussed at the hearing on the present motions, the Court's summary judgment ruling effectively DENIED Apple's motion to exclude with respect to Dr. Rinard. See Daubert Hearing Tr. at 127.

In this Order, the Court addresses the parties' remaining challenges to each other's experts.

II. LEGAL STANDARD

Federal Rule of Evidence 702 allows admission of "scientific, technical, or other specialized knowledge" by a qualified expert if it will "help the trier of fact to understand the evidence or to determine a fact in issue." Expert testimony is admissible pursuant to Rule 702 if it is both relevant and reliable. Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 589 (1993). A district court's decision to admit expert testimony under Daubert in a patent case follows the law of the regional circuit. Micro Chem., Inc. v. Lextron, Inc., 317 F.3d 1387, 1390-91 (Fed. Cir. 2003). When considering expert testimony offered pursuant to Federal Rule of Evidence 702, the trial court acts as a "gatekeeper" by assessing the soundness of the expert's methodology to exclude junk science. Estate of Barabin v. AstenJohnson, Inc., 740 F.3d 457, 463 (9th Cir. 2014); see Kumho Tire Co. v. Carmichael, 526 U.S. 137, 147-48 (1999); Gen. Elec. Co. v. Joiner, 522 U.S. 136, 142 (1997); Daubert, 509 U.S. at 589-90. An expert witness may provide opinion testimony if: (1) the testimony is based upon sufficient facts or data; (2) the testimony is the product of reliable principles and methods; and (3) the expert has reliably applied the principles and methods to the facts of the case. Fed. R. Evid. 702; see Sundance, Inc. v. DeMonte Fabricating Ltd., 550 F.3d 1356, 1360 (Fed. Cir. 2008). Under Daubert, courts consider (1) whether a theory or technique "can be (and has been) tested;" (2) "whether the theory or technique has been subjected to peer review and publication;" (3) "the known or potential rate of error;" and (4) whether there is "general acceptance" of the methodology in the "relevant scientific community." Daubert, 509 U.S. at 593-94.

The inquiry into admissibility of expert opinion is a "flexible one," where "[s]haky but admissible evidence is to be attacked by cross examination, contrary evidence, and attention to the burden of proof, not exclusion." Primiano v. Cook, 598 F.3d 558, 564 (9th Cir. 2010) (citing Daubert, 509 U.S. at 594, 596). "Under Daubert, the district judge is 'a gatekeeper, not a fact finder.' When an expert meets the threshold established by Rule 702 as explained in Daubert, the expert may testify and the jury decides how much weight to give that testimony." Id. (quoting United States v. Sandoval-Mendoza, 472 F.3d 645, 654 (9th Cir. 2006)).

III. DISCUSSION

A. Apple's Daubert Motions

1. Apple's Motion to Exclude Dr. Chevalier's Testimony Concerning Patent Licenses

Apple seeks to exclude the testimony of Samsung's damages expert Dr. Judith Chevalier regarding patent licenses. In her damages report, Dr. Chevalier reviewed more than Apple and Samsung licenses and a purported licensing offer from Apple to Samsung prior to this lawsuit to support her conclusion that a hypothetically negotiated lump-sum payment of $5.9 million—reflecting a per-unit royalty rate—is reasonable for a license to the Apple patents-in-suit. See Expert Report of Judith A. Chevalier, Ph.D. (ECF No. 831-3) ¶¶ 323, 408 ("Chevalier Rep."); id. Ex. 97. Apple contends that allowing the jury to hear Dr. Chevalier's analysis of those Apple and Samsung licenses, as well as Apple's purported pre-suit offer to Samsung, would flout the requirement that courts "'exercise vigilance when considering past licenses to technologies other than the patent in suit.'" Apple Mot. at 3 (quoting ResQNet.com, Inc. v. Lansa, Inc., 594 F.3d 860, 869 (Fed. Cir. 2010)). In particular, Apple challenges Dr. Chevalier's use of (i) a November 2012 agreement between Apple and HTC (the "HTC Agreement"), (ii) a 2010 proposal by Apple to Samsung regarding a global cross license to the parties' patent portfolios (the "2010 Apple Proposal"), and (iii) other licenses involving either Apple or Samsung for various types of technology. Before addressing each of these challenges below, the Court places Apple's challenges in context by examining Dr. Chevalier's overall reasonable royalty opinion.

a. Dr. Chevalier's Reasonable Royalty Opinion

Dr. Chevalier uses three approaches to reach her proposed $5.9 million lump-sum damages figure in this case, approaches she calls the "Market Approach," the "Income Approach," and the "Cost Approach," respectively. Chevalier Rep. ¶ 302. She also reviews "quantitative and qualitative evidence concerning the value contributed by the patents-in-suit, drawn, in large part, from factors identified in Georgia Pacific [Corp. v. U.S. Plywood Corp., 318 F. Supp. 1116, 1120 (S.D.N.Y. 1970)]," id., to construct a hypothetical negotiation between the parties, see id. ¶ 290. As Dr. Chevalier explains, a hypothetical negotiation "frames the analysis using a hypothetical arm's length negotiation for a license to practice the patent(s)-at-issue between a willing patent owner (here, Apple) and a willing potential licensee (here, Samsung) at the point of first alleged infringement." Id. ¶ 290 (citing Riles v. Shell Exploration & Production Co., 298 F.3d 1302, 1311 (Fed. Cir. 2002)). Apple primarily challenges Dr. Chevalier's application of the Market Approach to determine the outcome of a hypothetical negotiation in this case.

Michael Wagner was Samsung's damages expert in Apple I. Dr. Chevalier's approach differs from Wagner's insofar as Wagner did not rely on a Market Approach to reach his proposed reasonable royalty number in Apple I. See Apple, Inc. v. Samsung Elecs. Co., No. 11-1846, 2013 WL 5958176, at *3 (N.D. Cal. Nov. 7, 2013) ("HTC Order").

Market Approach. Using the Market Approach, Dr. Chevalier looks at the terms of licenses and other transactions involving Apple and Samsung to determine the royalty amount on which the parties would likely have agreed in a hypothetical negotiation for a license to the patents-in-suit. See id. ¶ 323. As referenced above, Dr. Chevalier analyzes the HTC Agreement, see id. ¶¶ 309-17; the 2010 Apple Proposal, see id. ¶¶ 319-21; and other licenses from the record, see id. ¶¶ 322-27. Of the other licenses from the record, Dr. Chevalier specifically examines a subset of licenses, all of which involved payments covering U.S. rights with 20 or fewer patents, were entered into in 2007 or later, and in which Samsung was the licensee. See id. ¶¶ 328-30. Dr. Chevalier estimates the per-unit royalty amount involved in the HTC Agreement, the 2010 Apple Proposal, and the Samsung licenses ranged from . Id. ¶ 334.

Income Approach. Using the Income Approach, Dr. Chevalier separately looks at the subset of profits for the products that incorporate the patented technology at issue due to that technology, less the return Samsung would have enjoyed had it used the next-best alternative to that technology. Id. ¶ 336. Under this approach, Dr. Chevalier explains, "[a]ny profits generated on products utilizing the patented technology that are in excess of the next best alternative represent an amount up to which the licensee should be willing to pay for access to the technology." Id.

Dr. Chevalier ascertains the profits attributable to the patents-in-suit in five different ways. First, Dr. Chevalier compares the popularity of the Galaxy S III, which does not incorporate the features of the '721 and '172 patents, with products that do. Dr. Chevalier concludes from that comparison that the value of the '721 and '172 patents is "negligible":

[T]he Galaxy S III which does not incorporate either the 'image unlock' or 'word recommendations' patents is more profitable and more popular than previous phone versions (such as the Galaxy S II) which are alleged to embody those patents. . . . [This] suggests the '172 patent and '721 patent values are negligible, as the non-infringing product is more profitable than, and as equally salable as, the infringing one.
Id. ¶ 338.

Second, Dr. Chevalier looks at the price of upgrading a phone's operating system—the component of the product that incorporates the accused features' upgrades. Id. ¶ 341. "Assuming the $10 to $25 [operating-system upgrade price] reflects the price of a single upgrade with 100 new features, the average value per feature would amount to $0.10 to $0.25." Id. ¶ 342.

Third, Dr. Chevalier applies her understanding of Apple's damages theory from Apple I—a theory that she called "entirely inconsistent" with Apple's theory in Apple II —to the patents at issue in this case. Id. ¶ 343. Dr. Chevalier concludes that Apple's theory from the prior case "suggests that the [sic] each asserted patent-in-suit here would be worth less than roughly per unit." Id. ¶ 346.

Apple contends that Dr. Chevalier should not be permitted to rely on Apple's damages theory from Apple I. See infra Part III.A.2.

Fourth, Dr. Chevalier allocates Samsung's average operating profits of the accused smartphones and tablets to various identified smartphone and tablet features using the number of times those features have been mentioned in professional and consumer reviews. Id. ¶ 350. Dr. Chevalier concludes that allocating value in this way "suggests that the various asserted functionalities tend to be worth as little as $0.00 and as much as $0.41 per smartphone." Id. ¶ 351. Dr. Chevalier also concludes that, because the profits per tablet are essentially zero, "the relative share of value [for tablets] associated with the patents-in-suit would also be essentially zero." Id.

Finally, Dr. Chevalier looks at revenue from the "mobile apps marketplace." Id. ¶ 352. As Dr. Chevalier explains, "[a]pps, in some ways, are useful indicia of the value smartphone customers place on software functionality for their devices." Id. Reviewing the average price paid across all Android users for each of the top 25 apps over the relevant period, Dr. Chevalier concludes that these apps yield anywhere from $0.02 to $0.24 per smartphone. Id. ¶ 357.

Cost Approach. Dr. Chevalier also uses a "Cost Approach" to evaluate her final reasonable royalty number. Under the Cost Approach, Dr. Chevalier concludes that Samsung's relatively small "design-around cost" would have reduced the amount Samsung was willing to pay for a license to the patents-in-suit. Id. ¶ 359. Based on interviews with technical employees and experts, Dr. Chevalier concludes that Samsung could have implemented alternatives to the accused functionality in Id. ¶ 364. This "ability to design-around the patent(s) would suggest that in a hypothetical negotiation Samsung would be willing to pay less over the full time period in which Samsung is alleged to have made infringing sales." Id. ¶ 365.

Overall Conclusion. Reviewing her conclusions from the Market, Income, and Cost Approach, as well as other factors discussed below, Dr. Chevalier selects an overall per-unit royalty rate of , as represented by the dotted line in the following table from Dr. Chevalier's report: . As Samsung explains, Dr. Chevalier's "review of licenses [in the Market Approach] yielded royalty numbers which are consistent with the four independent 'Income Approaches' she also employed." Samsung Opp. at 1.

As noted, Dr. Chevalier considers a number of other factors in reaching her final damages opinion. Dr. Chevalier looks at the form of the licenses she reviewed to conclude that the parties would have agreed to a lump-sum royalty. See Chevalier Rep. ¶ 297. As Dr. Chevalier explains, a lump-sum royalty "typically involves a one-time payment, the magnitude of which is not directly tied to the extent of alleged usage of the technology at issue by the alleged user/infringer." Id. ¶ 296. Dr. Chevalier concludes that the parties would not agree to a running royalty, or "a percentage of sales or a fee per unit, tied directly to how often the licensed invention is used by or incorporated into products of the licensee." Id. Dr. Chevalier supports that conclusion with "economic considerations," namely, that Apple would not need "an ongoing incentive to help commercialize Samsung's products," that Samsung would not have wanted "to pay incremental license fees for future successes due to its own efforts," and that "a lump sum would avoid the need to audit ongoing sales of products." Id. ¶ 300. Nevertheless, Dr. Chevalier expresses her calculations on a per-unit basis, in part because "lump-sum payments typically contemplate the expected use by the licensee." Id. ¶ 301.

Dr. Chevalier looks at the Georgia-Pacific considerations "to account for differences between the Market, Income and Cost approaches and the hypothetical negotiation." Id. ¶ 366. Dr. Chevalier concludes that three of those considerations—the "legal strength" of Apple's intellectual property, id. ¶ 376, Apple's id. ¶ 383—all "provide upward pressure on the royalty," id. ¶ 407. Dr. Chevalier also concludes that the "large number of features [that] are included in smartphones and tablets," id. ¶ 397, put "downward pressure" on the royalty, id. ¶ 407. Dr. Chevalier does not explain how those "upward" and "downward" pressures quantitatively affected her conclusion. However, she does opine that, taking all the aforementioned approaches and considerations into account, the parties would have likely agreed on a lump-sum royalty of roughly $5.9 million based on a per unit royalty rate of . See id. ¶ 408.

b. Dr. Chevalier's Analysis of Licenses and the 2010 Apple Proposal

The Court now turns to Apple's challenges to Dr. Chevalier's Market Approach. For the reasons stated below, the Court agrees with Apple that Dr. Chevalier's analysis of the HTC Agreement, the 2010 Apple Proposal, and the remaining licenses produced in this litigation must be excluded on Daubert grounds.

i. HTC Agreement

This Court already identified some of the problems with the HTC Agreement as an indicator of a reasonable royalty when it excluded that agreement and related expert testimony from the Apple I retrial. See HTC Order, 2013 WL 5958176, at *4-6. Familiarity with that order is assumed. As relevant here, the Court concluded that the economic considerations surrounding the HTC Agreement—which granted each side a broad license to the other's patent portfolio in exchange for, among other things, a worldwide settlement agreement between the parties, and an anti-cloning provision that prohibits HTC from copying any patented design and related functionality from an Apple product—rendered the agreement insufficiently probative to allow it to play a role in the damages retrial in that case. Id. at *4-6.

A redacted version of the HTC Agreement is available at ECF No. 2194-1 in Apple I.

The Court has the same concerns about the HTC Agreement here as it did in Apple I. Indeed, Dr. Chevalier herself acknowledges that Chevalier Rep. ¶ 312. Although Dr. Chevalier attempts to compensate for some of the differences between the HTC Agreement and a hypothetical negotiation here, the Court concludes that those attempts fail.

To account for the cross license in the HTC Agreement, Dr. Chevalier Id. ¶ 313. Dr. Chevalier . Id. Samsung may not rely on Apple's settlement offer to prove the amount of Apple's damages. See Fed. R. Evid. 408. This Court has already excluded the parties' attempts to rely on their settlement negotiations in Apple I, see Apple I, ECF Nos. 1695 and 2913 (order denying motion for additional discovery), and this attempt fares no better, see infra Part III.A.1 .b.iii. Dr. Chevalier presents no other way to account for HTC's cross license to Apple in the HTC Agreement. Indeed, Dr. Chevalier presents no legitimate basis for treating the HTC Agreement differently than Chevalier Rep. ¶ 324.

The HTC Agreement also includes a condition that allows for provides an additional indication that the HTC Agreement fails to quantify the benefit to Apple in a way that Dr. Chevalier can translate to this case. Dr. Chevalier does not attempt to quantify the value of this , even though she acknowledges that the hypothetical negotiation with Samsung would not include . Id. ¶ 314 n.712.

Nor does Dr. Chevalier account for the benefit that Apple received in settling 50 worldwide patent litigations and other proceedings involving HTC, some of which very likely created substantial risk and cost for Apple. Samsung does not suggest that the hypothetical negotiation in the present case encompasses foreign proceedings over foreign patents. Yet the risks and costs related to Apple's foreign litigation with HTC very likely encouraged Apple to accept a lower in the HTC Agreement than had the agreement resembled a hypothetical negotiation between Apple and Samsung over the U.S. patents-in-suit. Dr. Chevalier has not attempted to explain the status of the 50 HTC-Apple worldwide proceedings at the time of the HTC settlement or how those proceedings might have influenced the in the HTC Agreement.

Finally, Dr. Chevalier has not addressed the value to Apple of the anti-cloning provision in the HTC Agreement. As mentioned above, notwithstanding HTC's license to Apple's utility patents, the anti-cloning provision in the HTC Agreement prohibits HTC from copying the "Distinctive Apple User Experience" in Apple's products, a term that, loosely defined, appears to refer to the patented design and related functionality embodied in Apple's products. See HTC Agreement, Art. 12 & Ex. A. That anti-cloning provision appears to be a significant restriction on HTC's right to use the technology claimed in Apple's patents-in-suit, all of which relate to making portable devices easier to use. Yet Dr. Chevalier has not attempted to explain how the anti-cloning provision in the HTC Agreement relates to the unencumbered license that Samsung would have received in a hypothetical negotiation.

At the hearing on the parties' motions to exclude, Samsung listed five reasons why Dr. Chevalier should be allowed to rely on the HTC Agreement even though the Court excluded the agreement from Apple I. See Daubert Hearing Tr. 36-37. The Court concludes that none of these differences warrants allowing Dr. Chevalier to opine on the HTC Agreement in this case.

First, Samsung highlights that, unlike in this case, Apple I involved design patents. Samsung does not explain, however, how the presence of design patents in Apple I creates a meaningful difference to the relevance of the HTC Agreement. The HTC Agreement expressly carved out design patents from Apple's license to HTC, see HTC Agreement § 1.11, rendering it irrelevant to the determination of Apple's damages for design patent infringement. In the Apple I retrial, as here, Samsung sought to introduce the HTC Agreement in support of a reasonable royalty to Apple's utility patents. The parties agreed at the Apple I trial and retrial that Apple was entitled only to infringer's profits—not a reasonable royalty—for sales of Samsung devices that infringed Apple's design patents. Samsung certainly could not have relied on the HTC Agreement to establish a reasonable royalty for Apple's design patents, but that fact says nothing about the issue before the Court in both cases: whether the HTC Agreement sheds light on the appropriate royalty rate for Apple's utility patents.

Second, Samsung draws significance from the fact that, unlike in this case, its damages expert in Apple I concluded that the HTC Agreement was "'not probative'" to his primary reasonable royalty analysis. See HTC Order, at *5 (quoting Samsung's expert's updated rebuttal report). Samsung contends that the Court should reach a different outcome here because Dr. Chevalier thinks the HTC Agreement is relevant. Far from reducing the Court's concerns, however, the fact that even experts hired by the same party disagree as to the relevance of the HTC Agreement only highlights its unreliability. A "known technique which has been able to attract only minimal support within the community . . . may properly be viewed with skepticism." Daubert, 509 U.S. at 594. Of the three damages experts other than Dr. Chevalier who have opined on the HTC Agreement (Apple's Ms. Davis and Dr. Vellturo and Samsung's Mr. Wagner), none has concluded that it should play a role in a reasonable royalty determination. This uniformity of opinion against Dr. Chevalier does not help Samsung's cause.

Third, Samsung rightly notes that, in ruling on the admissibility of the HTC Agreement in the Apple I retrial, the Court addressed a unique procedural concern that is not present here. In particular, the Court excluded the HTC Agreement in advance of a retrial on damages partly out of a concern that the agreement was not part of the original trial and the Court had instructed the parties not to introduce new evidence in the retrial. See HTC Order, at *5 ("[T]he Court could not in fairness restrict Apple from revising and developing [its expert's] opinions on this subject, where Apple and [its expert] previously rightly recognized that the Court's prohibition against new evidence very likely would keep the HTC Agreement out of the case."). The Court, however, did not rest its exclusion of the HTC Agreement on this ground, but instead noted that this procedural concern supported its ruling based on other grounds. Id. For the reasons stated elsewhere in the Court's earlier opinion and repeated again here, the Court concludes that the HTC Agreement is not probative of the damages issue the jury must answer, and it therefore provides no value to the damages analysis.

Fourth, Samsung highlights that the HTC Agreement came into existence only after expert discovery in Apple I had closed, whereas here both parties' experts had time to incorporate their opinions as to the HTC Agreement in their original expert reports. This purported difference is no different than Samsung's reliance on the different procedural posture of this case, and it fails to alter the Court's conclusion for the same reason.

Fifth, and finally, Samsung asks the Court to consider that, because this case covers newer devices than the devices at issue in Apple I, the Court should be less concerned than it was in Apple I over the lapse in time between the hypothetical negotiation and the effective date of the HTC Agreement. See HTC Order, at *4 ("[T]he agreement's effective date is over two years after the date of the hypothetical negotiation, a significant lapse given the 'changing technological and financial landscape in the market' for smartphones.") (quoting LaserDynamics, 694 F.3d at 78). But the Court may not turn a blind eye to indications that a license is unhelpful merely because the dates of the license and the hypothetical negotiation are proximate. In ResQNet, the problems with the licenses the district court had erroneously considered were not eliminated simply because the dates of the licenses were apparently close in time to the date of first infringement of the patent-in-suit. Compare 594 F.3d at 877 (Newman, J., dissenting) (noting that excluded licenses were signed between 1998 and 2002) with id. at 863 (noting litigation began in 2001); U.S. Patent No. 6,295,075 (issued Sept. 25, 2001). Here, the Court has considered the proximity in time between the HTC Agreement and the hypothetical negotiation and has concluded that the proximity does not overcome the problems with Dr. Chevalier' s reliance on the HTC Agreement to form her reasonable royalty opinion.

Expert testimony based on license agreements must provide a jury enough information to "evaluat[e] the probative value of those agreements," lest the jury be given "little more than a recitation of royalty numbers" to determine the outcome of a particular hypothetical negotiation. Lucent Techs., Inc. v. Gateway, Inc., 580 F.3d 1301, 1328-29 (Fed. Cir. 2009). This requirement applies to an agreement's "economic circumstances" as much as to the agreement's technological scope. Wordtech Systems, Inc. v. Integrated Networks Sol'n, 609 F.3d 1308, 1319 (Fed. Cir. 2010); see id. at 1320 ("[C]omparisons of past licenses to the infringement must account for 'the technological and economic differences' between them.") (quoting ResQNet, 594 F.3d at 873) (emphasis added). At the hearing, Samsung expressed frustration at the prospect of not being allowed to present expert testimony related to the HTC Agreement even though it covered the patents-in-suit and "appropriate adjustments have been made" to account for the agreement's economic differences. Daubert Hearing Tr. at 38. The Court, however, rejects Samsung's premise. Dr. Chevalier has failed to appropriately adjust the running royalty found in the HTC Agreement to reach her proposed lump-sum royalty in a manner that accounts for the different economic circumstances at play here. Indeed, such appropriate adjustments appear to be impossible in light of the complicated and unsettled nature of the HTC Agreement. Dr. Chevalier's failure to adequately account for the agreement's cross license, and her failure even to attempt to make adjustments for (1) (2) the worldwide settlement agreement, and (3) the anti-cloning provision, all render her opinions related to the HTC Agreement insufficiently reliable to allow Samsung to present such testimony at trial.

ii. 2010 Apple Proposal

As with the HTC Agreement, the Court already prohibited the parties' use of Apple's pre-suit settlement negotiations with Samsung for purposes of establishing the amount of damages in Apple I. See Apple I, ECF No. 1695 at 1805, 1969. That ruling—based on Federal Rule of Evidence 408, which generally renders settlement offers inadmissible—largely disposes of Dr. Chevalier's attempt to rely on the same negotiations to establish a damages number here. Samsung describes Dr. Chevalier's use of the 2010 Apple Proposal to Samsung as "merely a data point used in Dr. Chevalier's Georgia-Pacific analysis." Samsung Opp. at 11. However, that description accurately captures the precise problem with her opinion. Dr. Chevalier's testimony regarding the relationship between the parties' pre-suit settlement negotiations and her damages number runs directly into Rule 408's prohibition of using the former to establish the latter.

Putnam v. Henkel Consumer Adhesives, Inc., No. 05-2011, 2007 WL 4794115, at *4-5 (N.D. Ga. Oct. 29, 2007), does not help Samsung. The court in Putnam allowed the accused infringer's expert to testify about the accused infringer's settlement of a different infringement claim. As that court noted, the proposed testimony implicated neither the terms nor the policy of Rule 408. Here, Samsung seeks to use "a party's good faith settlement offer against him," id., exactly the concern underlying Rule 408.

Samsung also contends that, even if the terms of Apple's offer are inadmissible, Federal Rule of Evidence 703 still allows Dr. Chevalier to rely on them. The Court concludes that Samsung's attempt to draw a distinction between inadmissible evidence and an admissible expert opinion based on that evidence fails in this case.

Rule 703 states that "[i]f experts in a particular field would reasonably rely on those kinds of facts or data in forming an opinion on the subject, they need not be admissible for the opinion to be admitted. But if the facts or data would otherwise be inadmissible, the proponent of the opinion may disclose them to the jury only if their probative value in helping the jury evaluate the opinion substantially outweighs the prejudicial effect." Accordingly, Samsung contends that even if Apple's settlement offer is inadmissible under Rule 408, Dr. Chevalier may still rely on the settlement offer to form her opinion. Samsung therefore posits that Apple's sole recourse is to prevent the jury from hearing about the settlement offer by moving in limine to argue that the probative value of the settlement offer in assisting the jury's evaluation of Dr. Chevalier's opinion does not substantially outweigh the prejudice of presenting this information to the jury under Rule 408. See Samsung Opp. 10-12.

The problem for Samsung is that Rule 703 deals only with a situation in which the admissibility of the basis of an expert opinion is challenged. The Court has an independent responsibility to consider whether the opinion of an expert is itself admissible. Determining whether an expert opinion is admissible is no different than determining whether any other type of evidence is admissible insofar as the Court must consider whether any probative value of the opinion is outweighed by unfair prejudice under the balancing test of Rule 403. "Expert evidence can be both powerful and quite misleading because of the difficulty in evaluating it. Because of this risk, the judge in weighing possible prejudice against probative force under Rule 403 of the present rules exercises more control over experts than over lay witnesses." Daubert, 509 U.S. at 595 (internal quotation marks omitted); see United States v. Hoac, 990 F.2d 1099, 1103 (9th Cir. 1993) ("Otherwise admissible expert testimony may be excluded under Fed. R. Evid. 403 if its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or undue delay.").

The Court finds that with respect to Dr. Chevalier's opinion of the relationship between the value of the 2010 Apple Proposal and the result of a hypothetical negotiation, any probative value is substantially outweighed by the unfair prejudice to Apple. The unfair prejudicial effect of an expert opinion based on a settlement offer is high. Rule 408 is premised on a strong public policy of encouraging settlements. See Bank of Am. Nat. Trust & Sav. Ass'n v. Hotel Rittenhouse Associates, 800 F.2d 339, 344 (3d Cir. 1986) ("We acknowledge the strong public interest in encouraging settlement of private litigation. Settlements save the parties the substantial cost of litigation and conserve the limited resources of the judiciary."); Reichenbach v. Smith, 528 F.2d 1072, 1074 (5th Cir. 1976) ("A primary reason for excluding evidence of a compromise is to encourage non-litigious solutions to disputes. Admission of evidence of the settlement could work to discourage plaintiffs and defendants from settling . . . ."); see also Advisory Committee Notes to Fed. R. Evid. 408 (noting the "promotion of the public policy favoring the compromise and settlement of disputes" as underpinning Rule 408). Allowing a party to circumvent Rule 408 by admitting a damages number based on a settlement offer through an expert contravenes this central public policy of favoring settlement of disputes.

In some respects, allowing a settlement offer to support an expert's damages number, rather than simply presenting the offer to the jury through a fact witness, exacerbates the prejudice to the party who offered the settlement. Juries are likelier to credit experts, who are perceived to possess special relevant knowledge and are expected to help the jury reach the right conclusion, more than simple documentary evidence. See Jinro Am. Inc. v. Secure Investments, Inc., 266 F.3d 993, 1004 (9th Cir. 2001), amended on denial of reh'g, 272 F.3d 1289 (9th Cir. 2001) (noting, in Rule 403 balancing, that prejudicial expert testimony may "carry special weight with the jury"). Concern regarding this unfair prejudice—punishing Apple for its good faith attempt to avoid litigation—is particularly acute in light of the fact that the jury would not be able to hear the basis for Dr. Chevalier's damages number insofar as the damages number is based on Apple's settlement offer, since admitting evidence of that basis would not itself be substanitally more probative than prejudicial under Rule 703. Therefore, the Court concludes that, not only does Rule 408 prohibit Dr. Chevalier from testifying about the Apple 2010 Proposal, the unfair prejudice that would result from admitting Dr. Chevalier's damages opinion to the extent that opinion is based on Apple's settlement offer is substantially outweighed by the unfair prejudice that such admission would cause to Apple.

iii. Remaining Agreements

As noted, Dr. Chevalier started with a set of licenses that were produced in this case, in addition to the HTC Agreement. See supra p. 5. According to Samsung, Dr. Chevalier relied on only of those licenses to reach her reasonable royalty figure. See Samsung Opp. at 3. In particular, Dr. Chevalier excluded from her numerical analysis any license (regardless of the technology involved) that (a) (b) did not cover U.S. rights, (c) involved more than 20 patents, (d) were entered into before 2007, and (e) did not include Samsung as a licensee. See Chevalier Rep. ¶¶ 328-30. Dr. Chevalier adjusted the lump-sum payment in each of these agreements Id. ¶ 331. She concluded that Id.

Apple disagrees with Samsung's contention that Dr. Chevalier did not rely on the other licenses in the record to reach her reasonable royalty figure. See Apple Reply at 2. Because Dr. Chevalier did not take into account the technology of even the licenses she undoubtedly did use to reach her reasonable royalty figure (as explained herein), theuourt's conclusion regarding those licenses applies with even stronger force to the extent Dr. Chevalier relied on any of the other licenses to reach her damages figure.

The Court agrees with Apple that Dr. Chevalier's mass grouping and statistical analysis of these agreements is fundamentally flawed because it treats all the agreements the same regardless of whether any of the licensed technology has any relationship to the patents-in-suit. The Federal Circuit has repeatedly cautioned that a mere "recitation of royalty numbers" is unreliable, "particularly when it is doubtful that the technology of those license agreements is in any way similar to the technology being litigated here." Lucent, 580 F.3d at 1329; see ResQNet, 594 F.3d at 869; LaserDynamics, 694 F.3d at 79 ("When relying on licenses to prove a reasonable royalty, alleging a loose or vague comparability between different technologies or licenses does not suffice."); Utah Medical Prods., Inc. v. Graphic Controls Corp., 350 F.3d 1376, 1385 (Fed. Cir. 2003) (affirming district court's conclusion that accused infringer "had not shown that the license agreements used in its expert's analysis were in any way comparable" to the patent-in-suit).

Samsung contends that, "[f]or each of the license agreements she ultimately used to calculate a reasonable royalty, Dr. Chevalier considered . . . [the] technological terms (the technology covered by the license, the patents licensed and the technological nature of the products the license applied to)." Samsung Opp. at 5. That contention, however, does not withstand scrutiny. Dr. Chevalier included tables in her report listing certain characteristics of each of the licenses that she included in her statistical survey. See Chevalier Rep, Exs. 72 & 85. Although Dr. Chevalier listed "Technology" as one of those characteristics, she merely comments that, for of the licenses, the technology is "[n]ot specified in [the] agreement." See id, Ex. 72. In addition, although Apple's patents-in-suit relate to the user interface found in smartphones and tablets, the "Licensed Products" that Dr. Chevalier lists for the vast majority of the licenses are

The Court has reviewed Dr. Chevalier's description of these licenses and has separately reviewed the patents licensed in those agreements, even as to the licenses for which Dr. Chevalier simply said the technology was "[n]ot specified." The Court has concluded that Dr. Chevalier has improperly lumped together licenses, many of which involve , without any regard to whether the licensed technology relates at all to smartphones and tablets and, more specifically, the ease-of-use technology that Apple would seek to license in the hypothetical negotiation. The result is that Dr. Chevalier has unreliably considered many of these licenses in a way that "significantly adjust[s] [downward] the reasonable royalty without any factual findings that account[] for the technological . . . differences between those licenses" and the patents-in-suit. ResQNet, 594 F.3d at 873.

Looking closer at some of the agreements in Dr. Chevalier's statistical analysis illustrates the problem. The highest payment in Dr. Chevalier's list of licenses is from an agreement between Chevalier Rep. ¶ 332 n. 741 (internal quotation marks omitted). This technology appears to be a relevant and important element of a smartphone user's experience, yet the license to it involves a amount that is more than double—in most cases much more—than of the remaining licenses that Dr. Chevalier used to estimate her royalty.

Notwithstanding the apparent similarity of the agreement to the hypothetical negotiation, Dr. Chevalier gives equal weight in her analysis to licenses to clearly unrelated technology owned by , such as technology for For all these specific agreements, as well as others, Dr. Chevalier does not address the technology at all, but says merely that the technology is "[n]ot specified in the agreement." Other agreements Dr. Chevalier treats as equal are much broader than the subject of the hypothetical negotiation, such as

By indiscriminately summarizing lump-sum license agreements without reference to their technological subject matter, Dr. Chevalier marginalizes agreements that appear to provide important indicators of the contours of the hypothetical negotiation. The agreement discussed above provides one example of Dr. Chevalier's failure to acknowledge the relevance of technology to the hypothetical negotiation. In addition, Dr. Chevalier reviewed but disregarded an agreement between Id. ¶ 333. Dr. Chevalier's cursory treatment of a license with a structure and rates that discredit her conclusions, based on reasons unrelated to the economic and technical similarities of the agreement to the hypothetical negotiation, is contrary to the Federal Circuit's requirement that, when relying on licenses, parties and the Court must review the licenses' terms with "vigilance." See ResQNet, 594 F.3d at 869.

Without some attempt to address the technological differences and similarities of the agreements in the record, Dr. Chevalier's use of some of those agreements and disregard of others to calculate her lump-sum amount is unreliable, irrelevant, and unhelpful to the jury's task of evaluating the result of the hypothetical negotiation. Accordingly, the Court excludes it.

c. Conclusion Regarding Dr. Chevalier

For the foregoing reasons, Apple's motion to exclude portions of Dr. Chevalier's testimony is GRANTED in part and DENIED in part. Dr. Chevalier may not present damages figures derived from her analysis of the HTC Agreement, the 2010 Apple Proposal, or the remaining license agreements in the record. Dr. Chevalier may still rely on her Income and Cost Approaches, as well as her analysis of the Georgia-Pacific factors, insofar as that analysis does not rely on the HTC Agreement, the 2010 Apple Proposal, or the remaining license agreements. As to the form of the reasonable royalty (lump-sum versus running royalty), Dr. Chevalier may continue to rely on her overall review of the license agreements in the record. Apple has not shown that Dr. Chevalier's choice of a lump-sum form based on is improper or otherwise unfairly prejudicial.

2. Apple's Motion to Exclude Samsung's Experts' Testimony About Opinions of Apple's Experts From Past Cases

Apple claims that "Samsung's experts should be precluded from testifying about the opinions of Apple's experts expressed in past cases, including experts that are not testifying in this case, as such testimony from other cases is not probative of any issue in this case and would cause confusion of the issues and a mini-trial on expert opinions not given in this case." Apple Mot. at 24. In response, Samsung argues that Apple's expert opinions are relevant and that Apple has provided no basis for which to strike such expert testimony under Daubert. For the reasons explained below, the Court agrees with Samsung and declines to preclude the testimony Apple cites.

The Court first provides an overview of the testimony that Apple seeks to strike. While Apple's request is broad and not limited to any particular Samsung expert, Apple specifically cites in its motion three aspects of Dr. Chevalier's testimony. First, Apple cites her opinion that the analysis undertaken by Apple's damages experts in Apple I, Terry Musika and Davis, is "inconsistent" with the analysis undertaken by one of Apple's damages expert in this case, Dr. James Vellturo. Apple Mot. at 24 (citing Chevalier Rep. ¶¶ 343-47). Second, Apple cites Dr. Chevalier's express reliance on "Apple's patent apportionment approach from its last case," which is described above, in the Market Approach section of her reasonable royalty analysis. Id. (citing Chevalier Rep. ¶ 317). Third, Apple cites various portions of Dr. Chevalier's report that reference Dr. Vellturo's purportedly inconsistent testimony in prior litigation. Id. at 30-31 (citing Chevalier Rep. ¶¶ 289, n.674, 301 n.690, 322 n.725, 342 n.761).

The Court rejects Apple's argument that such testimony must be excluded under Daubert on irrelevancy grounds. Apple has not cited, nor has this Court found, any authority in support of Apple's position. Although the Court acknowledges that damages experts may indeed use different damages theories in different cases, see Grain Processing Corp. v. Am. Maize-Prods. Co., 185 F.3d 1341, 1350 (Fed. Cir. 1999) (listing examples where "courts have given patentees significant latitude to prove and recover lost profits for a wide variety of foreseeable economic effects of the infringement"), Dr. Chevalier's reliance on testimony from other cases is nonetheless relevant, both to impeach Apple's expert and to support her own opinions. As to the former, the revelation that Dr. Vellturo (or any other Apple expert) previously relied on different methodologies may help the jury assess the reliability and credibility of Dr. Vellturo's methodologies and conclusions in this case. As to the latter, the fact that Dr. Chevalier relies on methods used by Apple's own experts in past cases could bolster Dr. Chevalier's credibility and mitigate any argument by Apple that her methods are unreliable.

As Samsung notes, neither Daubert nor Federal Rule of Evidence 702 may be used as a vehicle to strike relevant evidence from the fact-finder unless that evidence is scientifically unreliable or otherwise will not help the jury. Samsung Opp. at 32. Apple has given this Court no basis to conclude that the cited portions of Dr. Chevalier's testimony are unreliable or unhelpful.

Apple also appeals to Federal Rule of Evidence 403, arguing in the alternative that "even if the opinions regarding the methodology and conclusions of experts in other cases have some relevance, they are far more prejudicial than probative and would cause substantial confusion of the issues." Apple Mot. at 25. Apple claims that if "Samsung plans to attack Apple's experts regarding differences between their opinions in this case and opinions of other experts in a different case, Apple will be entitled to put forth the evidence of what theories were pursued in those previous cases," thereby resulting "in a mini-trial regarding economic analysis that is not pertinent to this case" and "muddl[ing] already complicated issues." Id.; see also Apple Reply at 15 ("Allowing Dr. Chevalier to opine on supposed inconsistencies in the testimony of Apple's experts in different cases would inevitably result in a number of mini-trials to compare different opinions from different experts regarding different accused products and different patents").

The Court rejects this argument. The Court disagrees because the probative value of such testimony is not substantially outweighed by any speculative danger of "confusing the issues" or "misleading the jury." Fed. R. Evid. 403. As discussed above, the past methodologies of Apple's experts are highly probative impeachment evidence that a fact-finder will consider in assessing the weight a fact-finder may choose to give to the experts in the instant litigation. Moreover, Apple's concerns about jury confusion and wasting time are unpersuasive. The parties' conduct in the damages retrial in Apple I shows that exploring expert opinions from other cases does not raise the types of irrelevant forays of which Apple complains about in this case. During that retrial, Samsung's counsel cross examined Ms. Davis, Apple's damages expert, regarding lost profit methodologies she had employed in prior patent cases and implied that she erred by not applying those same methodologies in the instant case. See Apple I, Retrial Tr. at 724, 733-34. Apple did not object to that brief line of questioning during the retrial, nor did it feel compelled to rehabilitate its expert with additional questioning or a mini-trial on her prior opinions. In light of this past behavior, the Court finds that the dangers identified in Rule 403 would not substantially outweigh the probative value for impeachment purposes of past methodologies. In fact, the Court is persuaded that granting Apple's request would unfairly prejudice Samsung because Apple would be allowed to attack Dr. Chevalier's methodologies while not allowing Samsung to rehabilitate Dr. Chevalier on the basis that Apple's own damages experts used the same methodologies in prior cases. Samsung Opp. at 32.

Accordingly, Apple's motion to exclude Samsung's experts from testifying about Apple's experts' opinions from past cases is DENIED.

B. Samsung's Daubert Motions

1. Samsung's Motion to Exclude Expert Opinion Based on Apple's Conjoint Survey Evidence

Samsung moves to exclude Dr. John Hauser, Apple's survey expert who designed and performed conjoint surveys to determine the demand for the patented features at issue in this case. See Samsung Mot. at 1-12. Samsung's challenges to Dr. Hauser's conjoint surveys, and Apple's damages expert's reliance on these surveys, falls into three broad categories. First, Samsung contends that conjoint surveys cannot be used to quantify the hypothetical reduction in sales of Samsung's allegedly infringing smartphones (and tablets) had Samsung developed noninfringing alternatives to those smartphones (and tablets). Id. at 1-7, 10-11. Second, Samsung contends that the survey question Dr. Hauser used to quantify the hypothetical reduction in sales ignored the realities of the marketplace. Id. at 11-12. Third, Samsung contends that Dr. Hauser's survey questions did not accurately capture the invention claimed by the patents, rendering the results unreliable. Id. at 8-10.

Before the Court turns to these criticisms, the Court briefly summarizes Dr. Hauser's methodology and compares Dr. Hauser's methodology in the instant case to Dr. Hauser's methodology in Apple I, in which Dr. Hauser's conjoint survey survived a Daubert challenge. See Apple, Inc. v. Samsung Elecs. Co., No. 11-1846, 2012 WL 2571332, at *9 (N.D. Cal. June 30, 2012). The conjoint surveys, which Apple's damages expert's (Dr. Vellturo) used to calculate lost profits and reasonable royalties, asked two types of questions.

First, Dr. Hauser presented survey respondents, who all owned an accused Samsung device, with four profiles at a time. Each profile consisted of a hypothetical smartphone, and the survey asked respondents to choose which of the four profiles they preferred. See Expert Report of John Hauser (ECF No. 1182) ¶ 85 ("Hauser Rep."). These profiles varied the price, camera features, call initiation and screening features, input assistance features, screen size, and data accessibility features of hypothetical smartphones. Id. ¶ 45. The surveyed features included the patented features and "distraction features," which were chosen because they were featured in Samsung guides and manuals. For any feature not listed (e.g., battery life), respondents were asked to assume that each profile offered the same level of functionality as their current smartphones. Id. ¶ 83. Each respondent chose sixteen hypothetical smartphones from sixteen sets of four profiles. Id. ¶ 77.

Dr. Hauser also asked respondents questions related to tablets. These questions differed somewhat in focus from the questions related to smartphones. For purposes of understanding Samsung's challenge to Dr. Hauser's methodology, however, those differences are immaterial. For simplicity, the remainder of this Order primarily refers to Dr. Hauser's smartphone survey, but the Order applies equally to the tablet survey.

Second, survey respondents, after selecting each of the sixteen hypothetical smartphones, were asked whether they would purchase each smartphone embodied by the profile at the given price. In answering this question, respondents were asked to consider other smartphones on the market, including other Samsung smartphones. Id. ¶ 92. Dr. Hauser used a "yes" response as an indication that the respondent would prefer to purchase a smartphone with the features of the selected profile, called the "inside option," over other available choices in the marketplace, called the "outside option." Id. ¶¶ 85-86. To provide a further incentive for respondents to make choices consistent with their true preferences, respondents were told that 1 in every 20 survey takers would be selected to receive a free smartphone based on the winning respondent's survey choices and, if the smartphone offered was priced at less than $300, a cash award equal to the difference. Id. ¶ 70.

Dr. Hauser uses the results of the conjoint surveys for two purposes. First, Dr. Hauser uses the first set of survey answers (selections from among the four-profile sets) to determine a willingness to pay for the patented features. Id. ¶¶ 125-31. Second, Dr. Hauser uses the second set of survey answers (the inside/outside option selections) in conjunction with the first set of answers to quantify the proportion of Samsung customers who would not have purchased Samsung smartphones if they lacked one or more of the patented features. Id. ¶¶ 114-24. To calculate this proportion, Dr. Hauser relies on the notion that respondents will value the tested features and the outside option in varying amounts (called "partworths"), which the survey measures, and that a respondent will purchase a smartphone with certain features only if the partworths for those features exceed the partworth for the outside option (called the "minimum level of attractiveness"). Id. ¶ 36. To estimate the proportion of Samsung customers who would not have purchased Samsung smartphones if they lacked the patented features, Dr. Hauser compares the percentage of respondents who would purchase a phone with all the tested features (i.e., the patented features and the distraction features) to the percentage of respondents who would purchase a phone with the distraction features but without one or more of the patented features. Id. ¶¶ 114-15.

The conjoint survey and its use in the instant litigation differ from the use of conjoint surveys in Apple I in one critical respect. In Apple I, Dr. Hauser performed only the first step of the conjoint survey discussed above. That is, he asked survey respondents to choose only between four profiles sixteen times. Accordingly, he was able to conclude only that there was demand for the patented features. He did not use the "outside option," and thus did not quantify the proportion of customers Samsung would have lost if its smartphones did not contain the patented features at issue in that case. Dr. Hauser did not need to make this calculation, because in Apple I, Apple sought lost profits damages only during the design around period (not also during the period that Samsung would have had a noninfringing alternative on the market). Apple theorized in Apple I that, if the infringing Samsung device had been off the market, all the real-world purchasers of those devices would have been "up for grabs" by Apple, Samsung (using its other noninfringing products), and other market participants. In contrast, here, Apple also seeks lost profits damages after the design around period. This presents a situation in which the previously infringing Samsung device is back on the market without the patented feature. In this case, Apple theorizes that not all the Samsung customers who bought the infringing Samsung device after the design around period were "up for grabs." Rather, only the portion of Samsung customers who would not have purchased the Samsung device in the absence of the patented feature would be available to purchase an Apple device. Dr. Hauser uses the outside option step of the conjoint survey to quantify this portion.

To summarize, in this case, Dr. Hauser's methodology with respect to the use of conjoint surveys to conclude that there is some demand for the patented features—Dr. Hauser's first step—is identical to the methodology permitted in Apple I. In Apple I, as here, Dr. Hauser presented several profiles of hypothetical devices to survey respondents (all of whom were current Samsung customers) to gauge the demand for the patented feature (also called "willingness to pay"). Moreover, the distribution of the customers who are "up for grabs" on the basis of market participants' market share is also similar to the analysis conducted in Apple I. Dr. Vellturo, like Apple's experts in Apple I, Terry Musika and Ms. Davis, redistributed customers that were up for grabs on the basis of market share calculated from survey data provided by third parties (not on the basis of the conjoint surveys) with some accounting for carrier limitations.

With respect to Dr. Hauser's willingness to pay methodology in this case, which was identical to Dr. Hauser's methodology in Apple I, Samsung recycles arguments that this Court rejected in Apple I. For example, Samsung adds up the willingness-to-pay amounts for four smartphone patents that were surveyed to contend that the survey results suggested that consumers were willing to pay more for the four features than the value of the phone itself. Samsung Mot. at 3. However, both Samsung and its expert recognize that the willingness-to-pay results of the conjoint survey cannot be added in this way. See id. ("[W]illingness to pay numbers are not strictly additive . . . ."). Expert Report of David Reibstein (ECF No. 1182) n. 301 ("Reibstein Rep.") ("[T]he willingness to pay for individual features is not strictly additive because correlation between participant's valuation of individual features and varying price sensitivity outside of the modeled conjoint price range may result in the aggregate willingness to pay to be either higher or lower than the sum of the willingness to pay estimates for the individual features."). Recognizing this, in Apple I, Samsung used the additive calculations to challenge only the weight—not the admissibility—of Dr. Hauser's conjoint analysis. Apple I, Retrial Tr. at 389:5-390:1 (Samsung counsel's opening statement regarding the conjoint survey, in which counsel stated, "what the evidence is going to show is that if you do [the conjoint] analysis on every feature in the phone, you're going to have a phone that would cost, that people would be willing to pay thousands and thousands of dollars for, because these aren't the main features on the phone.").

Samsung extensively attacks the use of the conjoint studies to calculate "market share." However, just as in Apple I, market share here was calculated on the basis of third-party data, not Hauser's conjoint survey. Apple I, 2012 WL 2571332, at *9. Accordingly, the Court rejects Samsung's reiteration of its prior market share argument.

Therefore, the critical issue before the Court is whether Dr. Hauser's quantification of the proportion of Samsung customers who would be "up for grabs" after the design around period—through use of conjoint surveys and the outside option—is sufficiently reliable under Daubert. Samsung contends that no court has allowed an expert to use conjoint surveys to quantify demand as Dr. Hauser does here. Apple, on the other hand, while conceding that no case has yet accepted conjoint surveys for quantifying demand, as Dr. Hauser does here, contends that the literature suggests that conjoint surveys can accurately predict market demand for complicated products analogous to smartphone and tablet devices.

a. Reliance on Conjoint Surveys to Quantify Decreased Demand

The Court begins by noting that the test for admission of expert opinion is not whether a court has admitted such an opinion in a previous case. Rather, Daubert and its progeny explicitly contemplate the admission of expert opinions in new fields, as long as such opinions are rooted in a methodology generally accepted in the scientific field at issue. See Daubert, 509 U.S. at 594 (noting that widespread acceptance in the relevant scientific community is an important factor in determining the admissibility of expert evidence).

Here, as described above, the novelty of Dr. Hauser's opinion stems from his use of the outside option to quantify the decrease in Samsung customers' demand for Samsung products without the infringing features. The Court finds that this type of use of the outside option to study demand is adequately supported by marketing literature. Specifically, Sawtooth Software, which produces the software that Dr. Hauser uses to conduct the analysis of the choice-based conjoint surveys in the instant case, notes that in choice-based conjoint surveys, "[b]y selecting the [outside] option, a respondent can contribute information about the decrease in demand to be expected if, for example, prices of all offered products increased or the products became unattractive in other ways." Sawtooth Software, The CBC System for Choice-Based Conjoint Analysis, at 2 (2013). As one study regarding the use of conjoint analysis in the context of HMO enrollment states, "[d]ata from a conjoint measurement study can be used to simulate market responses to changes in the levels of attributes or to the addition of new attributes." Michael Rosko et al., Strategic Marketing Applications of Conjoint Analysis: An HMO Perspective, 5 J. Health Care Mktg. 27, 34 (1985); see also Paul E. Green et al., Thirty Years of Conjoint Analysis: Reflections and Prospects, 31 Interfaces S56, S68 (2001) (noting that a seven-attribute conjoint study was able to adequately predict percentage demand for the EZ-Pass toll collection system). Another publication on conjoint studies, by the president of Sawtooth Software, suggests that "the price elasticity of demand (defined as a percentage change in quantity divided by percentage change in price) can be easily calculated for each brand in a choice-based conjoint survey." Bryan Orme, Getting Started with Conjoint Analysis: Strategies for Product Design and Pricing Research, at 25 (2010). In light of these studies in the relevant field of marketing research, the Court finds that Dr. Hauser's general methodology, using a choice-based conjoint survey to quantify decrease in demand for the product absent the patented feature, is sufficiently reliable to survive Daubert scrutiny because Dr. Hauser's methodology is substantially similar to those employed in these studies.

The Court finds these studies persuasive in light of the fact that they come from well-respected sources or are peer-reviewed. Sawtooth Software is among the most-used marketing research software. See Catherine Arnold, Self-Examination: Researchers Reveal State of MR in Survey, Marketing News, at 55 (2005). The Journal of Health Care Marketing is promulgated by the American Marketing Association, a leading organization of marketing professionals and academics. Interfaces is a peer-reviewed journal.

Samsung contends that even if a choice-based conjoint survey with an outside option could be used to quantify decrease in demand, such a survey is unreliable with a product as complex as a smartphone or tablet. Specifically, Samsung contends that Dr. Hauser's survey, in which he tests only six smartphone and tablet attributes, cannot be used to predict demand of an entire smartphone or tablet, both of which contain hundreds of features. The literature on conjoint analysis and the case law recognize that there are limitations on the number of distraction features that can be surveyed in a conjoint survey. See TV Interactive Data Corp. v. Sony Corp., 929 F. Supp. 2d 1006, 1025-26 (N.D. Cal. 2013) (noting on a Daubert motion that "the literature on conjoint analysis condones testing six or fewer variables to produce research with a better predictive value" and that "long-standing peer reviewed literature [suggests] using six or fewer variables leads to better predictive results because survey respondents are not overwhelmed by too much data"). Dr. Hauser himself has testified to this limitation of his survey. Retrial Tr. at 523:17-25 (Q: "Okay. Did you include each and every possible other feature of a smartphone in your survey?" A: No. That would not be feasible." Q: "Why not?" A: "Well, if I did, it would be a very difficult survey to take, and it's important there that the consumers do keep all the other features constant in their mind. And this is the way that we do it in the industry. It's a very accepted form of doing a conjoint analysis."). In fact, Samsung's expert's criticism is largely based on the premise that Dr. Hauser's survey was excessively complicated and too long. Implicit in this criticism is an acknowledgment that the scope of a conjoint survey must be limited. Reibstein Rep. ¶¶ 224-27 (criticizing complexity of Dr. Hauser's conjoint survey).

Samsung's contention that the particular distraction features that Dr. Hauser chose to test (and the features he chose not to test) render the survey unreliable is not persuasive. Dr. Hauser selected the distraction features to be surveyed based on features highlighted in Samsung's own manuals. Whether Dr. Hauser chose the correct distraction features, or whether he should have instead relied on other distraction features, goes to weight, not admissibility. Samsung's reliance on Oracle is misplaced. See Samsung Mot. at 5 (citing Oracle Am., Inc. v. Google Inc., 2012 WL 850705, at *9 (N.D. Cal. Mar. 13, 2012)). In Oracle, the chosen distraction features led to irrational results. See 2012 WL 850705 at In contrast, there are no irrational results that stem from the surveys in this case. Samsung's sole contention that the results are irrational is contained in a footnote in Samsung's motion, where Samsung, relying on its own expert, Dr. Reibstein, contends that Dr. Hauser's survey suggested that 68% of survey respondents preferred a higher price smartphone over an identical phone that was priced lower and that therefore Dr. Hauser's survey contains irrational results. Samsung Mot. at 5 n.5. A brief summary of Dr. Hauser's methodology with respect to his model is helpful to explain the dispute. Dr. Hauser utilized a price monotonicity constraint, that is, a constraint that instructed the model that consumers would prefer a lower price over a higher price if all else were held equal. Hauser Rep. ¶ 103. Dr. Hauser imposed this constraint using a standard tool that is included in the Sawtooth Software. Hauser Dep. at 313. To reach his 68% figure, Dr. Reibstein reran the model without the monotonicity constraint to conclude that the monotonicity assumption is not borne out by the underlying data (that is, respondent choices). Reibstein Rep. ¶ 235. However, Dr. Hauser also ran his own tests without the monotonicity constraint and found that fewer than 5% of predicted choices were inconsistent with principles of monotonicity. Hauser Dep. at 315-16. Dr. Hauser testified at his deposition that Dr. Reibstein's results stemmed from the fact that Dr. Reibstein did not take into account the uncertainty of the estimates by ensuring statistical significance at the 95 percent confidence level. See id. Samsung has not rebutted Dr. Hauser's testimony regarding the source of the difference in the monotonicity tests. Furthermore, Dr. Hauser testified that no survey respondent in the raw data (that is, not in the created models) stated a preference for a higher price smartphone over an identical, lower-priced product. Id. at 316. Samsung also does not rebut this contention, and accordingly, this battle of the experts is best resolved by the jury.

Samsung's contention that the limitations of conjoint surveys with respect to the number of attributes that can be tested renders choice-based conjoint surveys unreliable is belied by the literature on conjoint surveys. Specifically, studies have demonstrated that conjoint surveys were adequately able to quantify consumer demand with respect to complex products such as HMOs and hotel chains. See Rosko et al., supra; Green et al., supra. Moreover, conjoint surveys have been admitted in multiple complex patent cases, including Apple I. See, e.g., TV Interactive Data Corp., 929 F. Supp. 2d at 1020-27 (rejecting Daubert challenge to use of conjoint analysis in patent case regarding autoplay feature of DVD player, Blu-Ray player, and PlayStation 3 console); Microsoft Corp. v. Motorola, Inc., 803 F. Supp. 2d 1109, 1119-20 (W.D. Wash. 2012) (rejecting Daubert challenge to conjoint survey in a case concerning calculation of royalty rates for video decoding and Wi-Fi capabilities of Xbox 360). While the conjoint surveys in Apple I, TV Interactive, and Microsoft were not used to quantify demand as in this case, Samsung provides no basis for why this distinction is material. Neither the literature nor Samsung provides any theoretical basis for concluding that the quantification of demand as Dr. Hauser does in this case is any more troublesome for complex products than identifying demand in a more abstract sense (as the willingness to pay study in Apple I did). Thus, the Court concludes that Dr. Vellturo's use of Dr. Hauser's choice-based conjoint survey with an outside option to quantify the diminished demand that Samsung would have experienced after it designed around the patented features is sufficiently supported by the literature and is therefore admissible.

The conjoint surveys in the HMO and EZ-Pass context are not choice-based conjoint surveys. However, the Court could find no basis to distinguish choice-based conjoint surveys from the conjoint surveys used in those articles. In fact, marketing research suggests that choice-based conjoint surveys are similar to ratings-based conjoint surveys (the type of survey used in the HMO study). See Sawtooth Software, supra, at 6.

There is also nothing in the literature to suggest that an external validity test is required, as Samsung contends. Samsung Mot. at 10. Moreover, the external validity test that Samsung's expert performs itself contains significant methodological problems. For example, Samsung contends that its expert shows that Dr. Hauser's model effectively but erroneously predicted See id. at 10. However, Dr. Hauser's survey does not provide a basis to conduct such a product-to-product comparison, because the survey asked respondents to hold all the features other than those tested constant to their most recent smartphone. Hauser Rep. ¶ 83.

b. Improper Description of the Outside Option

Samsung also attacks the "outside option" in Dr. Hauser's study. As described above, the outside option consists of the market choices available to respondents other than each chosen device profile. Because the second set of survey questions asked respondents to compare each chosen profile to other market choices, those other market choices would naturally include choices such as keeping a respondent's current phone and purchasing other devices on the current market. Samsung contends that the outside option did not allow Dr. Hauser to measure accurately customers' willingness to buy devices with the infringing features because the outside option included accused products (which have the infringing features) and products that came to the market in some cases more than two years later than the infringing devices. (Dr. Hauser performed his survey in the summer of 2013, whereas the accused products were introduced as early as the spring of 2011.) In light of the myriad of choices available to each respondent, Samsung contends that respondents' selection of the outside option over the chosen hypothetical phone was likely influenced by factors unrelated to the removal of a patented feature. Samsung Mot. at 11-12.

The Court disagrees with Samsung's criticism of the outside option. Samsung's challenge appears to misapprehend Dr. Hauser's willingness to buy calculation. If a respondent highly values the outside option, for whatever reason, the partworth of the outside option will be high. Moreover, if the respondent cares little for the tested features (the patented features and the distraction features), the partworths of those tested features will be lower than the partworth of the outside option. This discrepancy will reflect a conclusion that the respondent will not purchase a phone even if it comes fully loaded, or if it comes fully loaded but for the patented features. As he explained at his deposition, by first calculating the percentage of respondents who would be willing to buy a fully loaded device, Dr. Hauser creates a control that accounts for any untested values that the respondent might attach to the outside option. See ECF No. 859-2 at 271:5-276:24.

The Court uses the term "fully loaded" to mean a smartphone with all the tested features, including the patented features (which a respondent's current phone will have) and the distraction features (which a respondent's current phone may not have). Dr. Hauser refers to a fully loaded smartphone as a "baseline smartphone." Hauser Rep. ¶ 115.

Samsung's critique seems to suggest that Dr. Hauser relies merely on a comparison of a respondent's choices between the inside option and the outside option for each question. But Dr. Hauser does more than that. He first compares the respondent's partworths of a fully loaded phone to the respondent's partworth of the outside option. See Hauser Rep. ¶¶ 114. Dr. Hauser uses this comparison as a baseline for all his later comparisons. Dr. Hauser then compares the respondent's partworths of a phone that is missing one or more of the patented features to the respondent's partworth of the outside option. See id. ¶ 115. The difference between the results of the first comparison (between a fully loaded phone and the outside option) and the second comparison (between the fully loaded phone but for one or more of the patented features and the outside option) is the critical factor in calculating the percentage of respondents who are willing to buy the patented features. See id. ¶ 116. Therefore, this comparison should render any untested considerations regarding the outside option immaterial.

Samsung does not address this control in its Daubert motion. Yet Dr. Hauser's approach is similar to that used in the published literature. See Sawtooth Software, supra at 9 (Example 2); Jeff Brazell et al., The No-Choice Option and Dual Response Choice Designs, 17 Marketing Letters 255 (2006). Because the Court is satisfied that Dr. Hauser's approach has peer support and makes logical sense, the Court concludes that Samsung's challenge goes to the weight of Dr. Hauser's survey evidence, not its admissibility.

c. Overbroad Characterizations of the Patented Features

Samsung finally contends that, even if Dr. Hauser's approach is acceptable as a general matter, particular flaws in Dr. Hauser's survey questions render his opinion faulty in this case. In particular, Samsung argues that Dr. Hauser's characterization of the patents in his survey questions is at odds with Apple's own technical experts' characterization of the patents and Samsung's experts' characterization of the patents. Samsung Mot. at 8-9. In Apple I, this Court rejected Samsung's argument that Dr. Hauser's "descriptions of features in the survey do not match Apple's expert's descriptions of the patented features." Apple I, 2012 WL 2571332, at *9. As this Court held, "Samsung's dissatisfaction with the description of the patented features in the survey . . . goes to weight, not admissibility." Id. at *10.

The Court continues to recognize that the framing of questions for purposes of surveys is generally an issue of weight, not admissibility. See Fortune Dynamic, Inc. v. Victoria's Secret Stores Brand Mgmt., Inc., 618 F.3d 1025, 1038 (9th Cir. 2010). Nevertheless, the Court also recognizes that there must be outer limits to this principle. At some point, a description of a patent in a survey may vary so much from what is claimed that the survey no longer "relate[s] to any issue in the case" and is "not relevant and, ergo, non-helpful." Daubert, 509 U.S. at 591. Such survey evidence would not "help the trier of fact" and therefore must be excluded under Rule 702(a). Moreover, at some point, discrepancies between the scope of the patent claims and the survey questions may be so confusing to the jury as to substantially outweigh the survey's probative value, thus requiring the Court to exclude such material under Rule 403. See Daubert, 509 U.S. at 595; Fractus S.A. v. Samsung, No. 09-203, 2011 WL 7563820 (E.D. Tex. Apr. 29, 2011) (excluding a survey regarding demand for internal antennas in cell phone where the patent claimed only one type of internal antenna because the survey would "confuse[] the issues"). The precise line between when a survey question's description of patented technology is "close enough" to the asserted claim as to be an issue of weight and when a survey question so departs from the asserted claim as to be excluded under Rules 702 and 403 has not been defined.

In the instant case, Samsung contends that with respect to the '414 Patent, Dr. Hauser's conjoint survey is flawed because Dr. Hauser's description of background synchronization (the subject matter of the '414 Patent) failed to capture an essential limitation of the asserted claim. Samsung Reply at 6-7. Apple and its experts have recognized that the asserted claim requires three background synchronization components. See Rebuttal Expert Report of Alex C. Snoeren (ECF No. 948-14) ¶ 493 ("[C]laim 20 requires at least three distinct synchronization software components each of which provides its own synchronization processing thread(s) to synchronize a corresponding data class."); Apple Opp. at 9 ("Claim 20 is narrower in technical scope than claim 11 by virtue of the additional requirements of three or more separate, data class specific software synchronization components.") (internal quotation marks and alterations omitted); ECF No. 1133 at 120 (Apple counsel's statement that "Claim 20 requires multiple synchronization software components").

In contrast with these statements describing the scope of claim 20, however, Dr. Hauser's survey description of background synchronization suggested to survey respondents that the patent covered only one synchronization component:

The Background Synching feature allows you to continue to use an app while data related to that app that is stored on your smartphone [or tablet] synchronizes with data stored elsewhere, such as on a remote computer. For example, you can access and edit data, such as the list of contacts on your smartphone [or tablet], even as your phone [or tablet] is sending data for those contacts to your work computer. Without this feature, you would have to wait to use an app, for example the contacts app, while the data for the app is synchronizing with a remote computer, and that wait might be long or short.
Hauser Rep. ¶ 74 (emphases added). As such, Samsung contends that Dr. Hauser's survey is premised on a broader conception of the asserted claim of the '414 Patent than that claim's scope and in no sense could be understood by the respondents to capture the limitation of claim 20.

The Court does not agree with Samsung that Dr. Hauser's description of claim 20 is overbroad. In light of the purpose for which Apple will use this survey evidence at trial, the Court concludes that Dr. Hauser's description of claim 20 of the '414 Patent is sufficiently tailored to the requirements of that claim. The Court agrees with Apple that the accuracy of Dr. Hauser's survey question must be evaluated from the perspective of a Samsung customer. See Daubert Hearing Tr. at 118 ("The question the Court's trying to answer here is, . . . were the benefits of claim 20 adequately captured to the survey recipients'") (emphasis added). That perspective is appropriate because Dr. Vellturo uses Dr. Hauser's survey evidence to evaluate the demand that Samsung customers have for the technology of the asserted claim in Samsung's products. See, e.g., Expert Report of Christopher A. Vellturo, Ph.D. (ECF No. 1156-3) ¶ 284 ("Vellturo Rep.") ("Dr. Hauser's research demonstrates that there are significant numbers of Samsung customers who do not find the posited non-infringing alternatives acceptable.") (emphasis added); id. n.507 ("Dr. Hauser's research also demonstrates that the patented inventions play a material role in demand for the Samsung smartphones.").

Thus, the Court must ask whether, by failing to incorporate a three-component synchronization concept in the survey question, Dr. Hauser's request that Samsung customers choose between a single-component background synchronization feature and no background synchronization was improper. The Court answers that question in the negative. Importantly, the record here will allow Apple to contend at trial that Samsung could not have implemented a single-component noninfringing alternative to claim 20. Apple's technical expert Dr. Snoren explained in his report that the nature of the code in Samsung's devices precluded a background synchronization system with only one synchronization component. See id. ECF No. 859-26 ¶ 510-11 (Apple's technical expert Dr. Snoeren describing "significant technical drawbacks" to altering Samsung's architecture to use "a single, combined Sync Adapter"). Samsung has presented no evidence to rebut that contention. Accordingly, Apple can argue that, for Samsung consumers, the technical limitations of Samsung's devices rendered the relevant choice to be between a device with background synchronization and a device without background synchronization (not a choice between three-component background synchronization and one-component background synchronization). Given the apparent limitations of Samsung's devices, the Court concludes that Dr. Hauser's description of claim 20 of the '414 Patent was not improper.

With respect to the remaining patents, Samsung's briefing is wholly inadequate to draw the line between admissible survey evidence that is sufficiently faithful to the claimed patent and inadmissible evidence that is unreliable. Samsung merely string cites a series of expert reports in a footnote with no explanation of the similarities and differences between Dr. Hauser's description of the patents and the asserted claims. Samsung Mot. at 8 n. 14. The footnote attempts to incorporate by reference several of Samsung's experts' reports, which Samsung contends explain how Dr. Hauser's survey questions are flawed. See id. Such citations are an impermissible circumvention of the page limits, which are in place to ensure that both the Court and the non-moving party can comprehend and analyze the contentions the moving party is making. Incorporating by reference several expert reports greatly expands the scope of the contentions of the moving party, here, Samsung.

Accordingly, the Court cannot on the basis of Samsung's briefing make an appropriate determination under Rule 403 and Rule 702. The Court finds that because Samsung has failed to appropriately brief the issue by providing argument as to the differences between the survey questions and asserted patents, Samsung has forfeited any Rule 702 argument that Dr. Hauser's survey questions depart from the asserted claims, except with respect to the '414 Patent as explained above.

d. Conclusion Regarding Dr. Hauser

In sum, the Court declines to exclude Dr. Hauser's choice-based conjoint surveys, and Dr. Vellturo's reliance on the results of those surveys, on Daubert grounds. Samsung's motion to exclude Dr. Hauser's survey evidence is therefore DENIED.

2. Samsung's Motion to Exclude Dr. Vellturo's Damages Analysis for Reasons Other than His Reliance on Dr. Hauser's Conjoint Survey Analysis

Dr. Vellturo's damages opinion consists of three elements: (1) a lost-profit award based on a "blackout period" (i.e., a "but for" world wherein Samsung would have been unable to sell any of the accused products for four months, until it had built non-infringing alternatives to the asserted patents); (2) a lost-profit award based on "diminished demand" for Samsung devices following the blackout period (i.e., a "but for" world wherein Apple would have sold additional devices even after Samsung introduced its design arounds because of the diminished demand for Samsung's products that were rendered inferior by the necessity of using unproven alternatives to the patented technology), and (3) a reasonable royalty for sales that do not qualify for lost profits. Vellturo Rep. ¶¶ 11-18, 146-152.

In general, Dr. Vellturo's damages opinion follows the statutory measure for infringement damages, which allows a patentee to recover "damages adequate to compensate for the infringement, but in no event less than a reasonable royalty for the use made of the invention by the infringer." 35 U.S.C. § 284; see Mor-Flo Indus., 883 F.2d at 1577 ("[T]he [damage] award may be split between lost profits as actual damages to the extent they are proven and a reasonable royalty for the remainder."). One way a patentee can prove lost profits is by establishing the four factors set out in Panduit Corp. v. Stahlin Bros. Fibre Works, Inc., 575 F.2d 1152, 1156 (6th Cir. 1978): (1) demand for the patented product, (2) absence of acceptable noninfringing substitutes, (3) capacity to exploit the demand, and (4) the amount of profit Apple would have made but for Samsung's infringement. Under section 284, a patentee can recover a reasonable royalty for those sales that do not meet the Panduit conditions. See Mor-Flo, 883 F.2d at 1577.

In addition to Samsung's challenges to Dr. Vellturo's use of design-around dates, see supra pp. 2-3, and Dr. Vellturo's use of Dr. Hauser's survey evidence, see supra pp. 24-36, Samsung attacks Dr. Vellturo's damages opinion on two grounds: (a) that Dr. Vellturo has improperly used a lost profits analysis to calculate reasonable royalty damages and (b) that Dr. Vellturo relied on counter-factual assumptions and ignored relevant evidence. The Court addresses each ground in turn.

a. Dr. Vellturo's Use of Lost Profits to Calculate Reasonable Royalties

Samsung challenges Dr. Vellturo's method of calculating the "reasonable royalty" element of his damages opinion where lost profits are not warranted. Dr. Vellturo's reasonable royalty opinion relies on what he calls an "Edgeworth Box" analysis. Dr. Vellturo explains this analysis as follows:

The likely price at which an agreement can be struck between two parties to exchange a good or asset is bounded by the expected valuations each party anticipates if the transaction does not take place. These valuations are commonly referred to in economics as forming the boundaries of an "Edgeworth Box," a bargaining range that governs economic interactions between the parties. As the licensor, the patentee takes into consideration that it would generally not be in its economic interest to accept less in royalty fees than the expected financial cost of granting such a license. Correspondingly, as the licensee, the infringer would consider the license fee relative to the expected additional value it would have anticipated realizing as a direct result of having access to the technology taught by the asserted patent.
Vellturo Rep. ¶ 340. In economic parlance, Dr. Vellturo explains that the "bottom" of the Edgeworth Box is the lowest amount the licensor, Apple, would be "willing to accept" to license its patents, and the "top" of the Edgeworth Box is the highest amount the licensee, Samsung, would be "willing to pay" to obtain a license to Apple's patents. Id. ¶ 340, n. 550-51.

As relevant here, Dr. Vellturo calculated Apple's "willing to accept" amount by looking at the amount of profits Apple would expect to lose by allowing Samsung to practice Apple's patents. Id. ¶ 415. He also calculated Samsung's "willing to pay" amount, but concluded that that amount for each patent was less than Apple's "willing to accept" amount, resulting in what Dr. Vellturo calls an "'inverted' Edgeworth Box—one in which Apple's minimum acceptable rate exceeds the maximum acceptable rate to Samsung." Id. ¶ 430. But this topsy-turvy result does not cause a problem, Dr. Vellturo contends, because in the real world Samsung would have simply raised the prices of its devices to account for Apple's high "willing to accept" amount. Id. ¶ 431. Thus, Dr. Vellturo concludes, "the governing benchmark rate from the Edgeworth Box analysis is the minimum acceptable rate to the patentee—here, Apple." Id. As Samsung accurately describes, the result of Dr. Vellturo's reliance on Apple's "willing to accept" amount as the sole hypothetical negotiation benchmark is that "Dr. Vellturo's . . . reasonable royalty . . . is entirely based on what Apple would expect to lose in profits from licensing to Samsung." Samsung Mot. at 12 (internal quotation marks omitted).

Samsung contends that Dr. Vellturo's use of a lost profits analysis for certain infringing sales and an expected lost profits analysis for the remaining sales is flawed. According to Samsung, the very reason Apple must resort to a reasonable royalty analysis for certain sales is because it cannot justify its entitlement to lost profits under Panduit. Allowing Dr. Vellturo to present a reasonable royalty number based solely on Apple's expected lost profits, Samsung continues, would allow Apple to receive some measure of lost profits damages even though it cannot prove the Panduit requirements for lost profits. "As a matter of law," Samsung contends, "this is improper." Samsung Mot. at 12.

Samsung's argument is foreclosed by Federal Circuit precedent. In Rite-Hite Corp. v. Kelley Co., 56 F.3d 1536 (Fed. Cir. 1995) (en banc), the Federal Circuit expressly upheld a claim for reasonable royalties based on the profits the patentee would have expected to lose as a result of a license. See id. at 1554-55. The patentee (Rite-Hite) successfully premised its claim for lost profits by tracing back Rite-Hite and the infringer's (Kelley) competition on "specific transactions." Rite-Hite Corp. v. Kelley Co., 774 F. Supp. 1514, 1525 (E.D. Wisc. 1991). For a subset of those transactions, however, Rite-Hite "had not proved that it contacted the Kelley customer prior to the infringing Kelley sale," and, accordingly, was not entitled to lost profits on those particular sales. 56 F.3d at 1554 (citing 774 F. Supp. at 1534). Nonetheless, the Federal Circuit affirmed an award of reasonable royalties to Rite-Hite for those sales "equal to approximately fifty percent of Rite-Hite's estimated lost profits per unit sold to retailers." Id. (citing 774 F. Supp. at 1535). The Federal Circuit, sitting en banc, rejected the contention that Rite-Hite could not rely on estimated lost profits to support its reasonable royalty award, holding that "the fact that the award was based on and was a significant portion of the patentee's profits also does not make the award unreasonable." Id. at 1555.

Samsung relies on Judge Nies's partial dissent in Rite-Hite, where she emphasized her disagreement with the majority on the point at issue here. See id. at 1576 (Nies, J., dissenting in part) ("[W]here a patentee is not entitled to lost profit damages, lost profits may not, in effect, be awarded by merely labeling the basis of the award a reasonable royalty.") (citing SmithKline Diagnostics, Inc. v. Helena Labs. Corp., 926 F.2d 1161, 1165 (Fed. Cir. 1991)). At least as to the fact pattern presented in Rite-Hite, however, Judge Nies's position did not carry the day.

In SmithKline Diagnostics, the patentee used a reasonable royalty as a "fallback" position to its lost profits analysis, but the reasonable royalty was "the same purported lost profit figure translated into a percent of [the infringer's] sales." 926 F.2d at 1165. The district court rejected this argument, and the Federal Circuit affirmed, holding that the patentee's "evidence to support a 48% royalty figure was not credible." Id. at 1168.
--------

Rite-Hite precludes Samsung's contention that a reasonable royalty analysis based on lost profits is legally impermissible. That is not to say that relying on a lost profits analysis to support a reasonable royalty award will be proper in every case. In particular, there may be situations where an expert's contention that a patentee could anticipate lost profits at the time of the hypothetical negotiation is plainly unreasonable. See Georgia-Pacific Corp, 318 F. Supp. at 1121 (listing as a factor to be considered in the reasonable royalty analysis "the anticipated amount of profits that the prospective licensor reasonably thinks he would lose as a result of licensing the patent as compared to the anticipated royalty income.") (emphasis added). Here, however, Samsung did not move to exclude Dr. Vellturo's analysis on the basis that it is unreasonable in light of the circumstances of this case. Accordingly, Samsung's motion to exclude Dr. Vellturo's reasonable royalty opinion is denied.

b. Dr. Vellturo's Factual Assumptions and Evidence

Samsung also contends that Dr. Vellturo improperly relied on certain counter-factual assumptions and "turn[ed] a blind eye to contradictory relevant evidence." Samsung Opp. at 14. For the reasons stated below, the Court finds that Samsung's complaints go to the weight, not the admissibility, of Dr. Vellturo's opinions.

First, Samsung contends that Dr. Vellturo failed to consider Apple's prior offer to Samsung or the HTC Agreement. See Samsung Mot. at 7, 14-15. These arguments are without merit. As the Court ruled above, these data points are unreliable for purposes of calculating damages. See supra Part III.A.1.b.i-ii. Dr. Vellturo cannot be faulted for declining to incorporate them into his analysis.

Second, Samsung contends that Dr. Vellturo improperly failed to consider Apple's public statements that it had product shortages in concluding that Apple would have had the production capacity to sell more devices but for Samsung's infringing sales. See Samsung Mot. at 15. Dr. Vellturo provided a factual basis for his opinion that Apple had excess capacity. See Vellturo Rep. ¶¶ 296-301. Samsung's contention that Dr. Vellturo did not address certain public statements or that he failed to have a cogent basis for ignoring those statements when pressed at deposition goes to the weight, not admissibility, of his opinion.

c. Conclusion Regarding Dr. Vellturo

For the foregoing reasons, the Court DENIES in part Samsung's motion to exclude Dr. Vellturo's testimony. The Court GRANTS in part Samsung's motion, to the extent Dr. Vellturo relies on notice dates rather than dates of first infringement to evaluate the design around periods in his lost profits calculation.

3. Samsung's Motion to Exclude Dr. Mowry's Inconsistent Opinions

Samsung contends that Apple's technical expert Dr. Mowry improperly provided inconsistent claim interpretations across his infringement and invalidity opinions and therefore that his testimony is unhelpful to the trier of fact. The Court disagrees that Dr. Mowry's testimony is unhelpful. Although Samsung presents instances where Dr. Mowry arguably did provide inconsistent testimony, Samsung can present these instances to the jury to impeach Dr. Mowry's credibility. Dr. Mowry's alleged transgressions do not rise to the level of the expert in DataQuill Ltd. v. Handspring, Inc., No. 01-4635, 2003 WL 737785, at *4 (N.D. Ill. Feb. 28, 2003), who did not follow the required two-step infringement analysis in reaching his infringement opinion, made conclusory assumptions that the accused products met certain claim limitations, could not explain at his deposition why features satisfied particular claim limitations, and did not write his own report. Similarly, the expert in Oxford Gene Tech. Ltd. v. Mergen Ltd., 345 F. Supp. 2d 431, 435-41 (D. Del. 2004), fundamentally failed to follow the two-step, element-by-element approach that the Federal Circuit requires to evaluate anticipation and obviousness. In both DataQuill and Oxford Gene, the problems went to the heart of the expert's methodology. Samsung's best challenge to Dr. Mowry, in contrast, is based on a strong suggestion that the expert's opinion as to one limitation is unbelievable. This is matter for cross examination. See Advanced Fiber Techs. Trust v. J & L Fiber Svcs., Inc., No. 07-1191, 2010 WL 1930569, at *6 (N.D.N.Y. May 11, 2010) (concluding that where an expert's qualifications are sufficient and report will otherwise "be helpful to a jury in making their factual determinations," then "[a]ny perceived inconsistencies should be addressed through cross-examination").

Accordingly, the Court DENIES Samsung's motion to exclude Dr. Mowry's testimony.

IV. CONCLUSION

For the foregoing reasons, the parties' Daubert motions are granted in part and denied in part. Apple's motion to exclude Dr. Chevalier's reliance on her statistical analysis of license agreements, the HTC Agreement, and the 2010 Apple Proposal to support her damages figure is GRANTED. Apple's motion to exclude Samsung's experts' reliance on Apple's experts' testimony from prior cases is DENIED. Samsung's motion to exclude Dr. Vellturo's reliance on design-around periods commencing on notice dates as opposed to the dates of first infringement is GRANTED. Apple shall serve on Samsung any supplementation of Dr. Vellturo's damages calculations to reflect the proper design-around periods by February 28, 2014. Samsung's remaining motions to exclude are DENIED.

IT IS SO ORDERED.

________________

LUCY H. KOH

United States District Judge


Summaries of

Apple Inc. v. Samsung Electronics Co., Ltd.

UNITED STATES DISTRICT COURT NORTHERN DISTRICT OF CALIFORNIA SAN JOSE DIVISION
Feb 25, 2014
Case No.: 12-CV-00630-LHK (N.D. Cal. Feb. 25, 2014)
Case details for

Apple Inc. v. Samsung Electronics Co., Ltd.

Case Details

Full title:APPLE, INC., a California corporation, Plaintiff and Counterdefendant, v…

Court:UNITED STATES DISTRICT COURT NORTHERN DISTRICT OF CALIFORNIA SAN JOSE DIVISION

Date published: Feb 25, 2014

Citations

Case No.: 12-CV-00630-LHK (N.D. Cal. Feb. 25, 2014)

Citing Cases

NetFuel, Inc. v. Cisco Sys.

Expert testimony based on license agreements "must provide a jury enough information to "evaluat[e] the…

Guido v. L'Oreal, U.S., Inc.

This concession parallels the conclusions of several district courts who have considered the admissibility of…