Silicon Knights, Inc. v. Epic Games, Inc.
ORDER: The court GRANTS Epic's motion to exclude the reports and testimony of Terry Lloyd [D.E. 620], and excludes Lloyd's opinions on damages. The court DENIES in part and GRANTS in part Epic's motion to seal the supporting memorandum and exhibits [D.E. 623]. The court also DENIES in part and GRANTS in part Silicon Knight's motion to seal the response in opposition and accompanying exhibits [D.E. 644]. Signed by Chief Judge James C. Dever III on 12/22/2011. (Sawyer, D.)
IN THE UNITED STATES DISTRICT COURT
FOR THE EASTERN DISTRICT OF NORTH CAROLINA
SILICON KNIGHTS, INC.,
EPIC GAMES, INC.,
On July 8, 2011, Epic Games, Inc., ("Epic" or "defendant") filed a motion to exclude the
reports and testimony ofTerry Lloyd, an expert for Silicon Knights, Inc. ("SK" or "plaintiff"), and
a motion to seal Epic's supporting memorandum and exhibits [D.E. 620, 623]. On July 15,2011,
SK responded in opposition and filed a motion to seal the response [D.E. 643, 644]. On July 22,
2011, Epic replied [D.E. 662]. On November 30,2011, the court held a hearing on the motion to
exclude [D.E. 691]. On December 6, 2011, SK filed a submission in response to the court's
inquiries on November 30,2011 [D.E.695]. As explained below, the court grants Epic's motion to
exclude the reports and testimony of Terry Lloyd. As for the motions to seal, the court grants the
motions to seal in part and denies the motions to seal in part.
SK and Epic are both in the business of developing video games. On May to, 2005, SK
entered into a license agreement to use Epic's video game engine, the Unreal Engine 3 ("UE3"), in
SK's development of Too Human, a game that SK produced under a publishing agreement with
Microsoft. SK contends that problems with UE3 eventually forced SK to develop its own game
engine ("SKE"), delaying the release ofToo Human by nearly two years. SK claims that this delay
and additional costs caused by the delay caused decreased sales of Too Human, caused Microsoft
to end negotiations to develop two sequels to Too Human, damaged SK's reputation, and impaired
SK's ability to secure future development projects. In 2007, SK filed suit against Epic. SK alleges
that, during the parties' license-agreement negotiations, Epic made false representations concerning
the license agreement and the functionality of UE3. SK asserted multiple claims against Epic,
including breach ofcontract, fraud, and other torts. In response, Epic filed counterclaims asserting
breach of contract, copyright infringement, and misappropriation of trade secrets, and seeking
imposition of a constructive trust. The parties engaged in substantial discovery. At the end of
discovery, the court granted in part and denied in part cross motions for summary judgment. SK's
remaining claims are fraudulent inducement, negligent misrepresentation, Unfair and Deceptive
Trade Practices Act ("UDTPA") violations, and common law unfair competition. Epic's remaining
counterclaims are copyright infringement, breach of license agreement, misappropriation of trade
secrets, and its request for imposition of a constructive trust.
On July 8, 2011, Epic filed amotion to exclude SK's damages expert, Terry Lloyd, and his
proposed testimony regarding various categories of SK's alleged lost profits and damages.
Thereafter, SK responded in opposition and Epic replied. On November 30, 2011, the court held a
hearing on Epic's motion to exclude.
Terry Lloyd ("Lloyd") is a Certified Public Accountant and Chartered Financial Analyst.
SK's Opp'n Mot. Exclude [D.E. 643] 5. SK retained Lloyd to render an opinion regarding SK's
alleged damages. Id. Lloyd has identified six categories of damages: (1) lost royalties due to
decreased sales of Too Human caused by its delayed release (approximately $6,200,000); (2) lost
income from publisher bonus payments based on deceased sales of Too Human (approximately
$750,000); (3) lost income from royalties on Too Human "ancillary" sales (approximately
$810,000); (4) lost profits from two Too Human sequels (approximately $16,142,000 for Too
Human II and approximately $14,252,000 for Too HumanID) and The Ritualyst, another video game
(approximately $8,748,000); (5) the additional cost for SK to develop a replacement game engine
(approximately $2,308,000); and (6) economic harm to SK's reputation, which limited future game
development opportunities (approximately $8,951,000). Revised Report ofTerry Lloyd [D.E. 564-1]
6-7 (May 4,2010) (hereinafter "Lloyd Report"); see SK's Opp'n Mot. Exclude 5-7. Alternatively,
Lloyd has calculated SK's damages as a portion of Epic's profits based on a theory of unjust
enrichment. See Lloyd Report 7; SK's Opp'n Mot. Exclude 7. Lloyd also has prepared a rebuttal
report in response to Epic's expert witness, Phillip Beutel ("Beutel"), regarding the amount of
damages Epic seeks in its counterclaims. See Report of Terry Lloyd [D.E. 621-2] (May 7, 2010)
(hereinafter "Lloyd Rebuttal").
SK originally planned to release Too Human in November 2006, but due to various delays,
SK did not release Too Human until 2008. See Lloyd Decl. [D.E. 643-2] , 10; Lloyd Report 9.
Lloyd estimated that about half of the two-year delay was attributable to ''normal slippage" in the
industry. See Lloyd Decl. , 10; Lloyd Report 23 & n.39. Therefore, in order to calculate SK's lost
profits allegedly caused by Epic, Lloyd attempted to predict what Too Human's sales would have
been if it had been released in August 2007. See Lloyd DecL , 10. Lloyd referred to this projection
as a "but-for world", where SK's development of Too Human would not be affected by Epic's
alleged wrongful conduct. See Lloyd Report 22-24; Lloyd Dep. [D.E. 564-4] 67-68. Lloyd
employed a methodology known as a ''yardstick'' or "comparables" approach, which measured SK's
damages by comparing Too Human to similar games in order to project the sales SK would have
generated from a 2007 release ofToo Human. Lloyd DecL , 2. Lloyd contended that he conducted
a comparables analysis by looking at games with similar characteristics to Too Human that were
released during the same general time period. Id.; see Lloyd Dep. 131-32. Lloyd described the
process ofdetermining which variables he used to fmd comparable games as "very much an inexact
science," and admitted that there was "a fair amount of subjectivity" and "uncertainty" involved.
See Lloyd Dep. 14-16, 178.
Lloyd's first step in finding comparable games was to sort video games by genre. Lloyd
DecL ~ 4. Lloyd initially excluded any game that was not of a genre similar to Too Human's. Id. l
Lloyd explained that the chosen genres were a "judgment call based on what we understood about
the game." Lloyd Dep. 131. Second, Lloyd eliminated any video game published before 2006
because the Xbox 360 was released in November 2005. Lloyd Decl. ~ 5. Third, Lloyd eliminated
any game with a Metacritic2 rating lower than Too Human's rating, because Lloyd believed that Too
Human's rating would have been at least as high ifit had been released a year earlier. Id. ~ 6. Lloyd
also excluded all games that did not sell at least 640,000 units, the number of units Too Human
actually sold. 3 Lloyd Dep. 151-52. After applying the above criteria, Lloyd added or removed
games from the list based on Lloyd's "judgment" and discussions with his colleagues at his ftnancial
consulting company and with Denis Dyack ("Dyack"), SK's president. See id. at 134-37, 140-43,
167-71. Lloyd explained that he "added and subtracted as [he] saw ftt" until he created a group of
games that he felt was comparable. Id. 136. For example, Lloyd removed Halo "because [it] just
seemed like such an outlier, and it was different enough, in [his] judgment, that it wasn't really a
comp[arable game]." Id.135. Lloyd added Grand Theft Auto IV based on the advice of Jeremy
Sherlock ("Sherlock"), an employee ofSega Id. 142-43. Lloyd explained that he "judgmentally
included" Grand Theft Auto IV, even though it did not meet his identifted criteria. Lloyd Decl. ~
1 Because Lloyd decided that Too Human had elements ofmultiple genres, he included games
from six different genres: (1) first person, action, shooter; (2) action, adventure; (3) first person,
shooter; (4) ftrst person, action, role-playing games ("RPG"), shooter; (5) action; (6) action, RPG.
2 Metacritic operates a video game review website. Epic's Mem. Supp. Mot. Exclude [D.E.
621] 17 n.21. In determining a game's Metacritic rating, "Metacritic combines a number ofscores
from individual critics using a proprietary algorithm and assigns a single [Metacritic rating]." Id.
3 Lloyd acknowledged the possibility that Too Human could have sold fewer units ifSK had
released it in 2007, but explained that his assumptions and criteria did not account for such an
outcome. See Lloyd Dep. 195-97.
7.4 Lloyd also used his judgmentto add Gears ofWar. See Lloyd Dep. 135-40. Lloyd testified that
he included some games because he believed "they would have been blockbusters." Lloyd Dep. 168,
181. 5 Lloyd did not use any ofSK's previous games as comparables. See Lloyd Dep. 113-15.6
Thus, Lloyd described the process ofidentifying comparable games as "evolutionary" and noted that
it was not a simple two-step process. Id. 141. In fact, Lloyd could not recall "[what games] came
in or out specifically after that first filter or first sort." Id. 170.
Lloyd's "evolutionary" process yielded a "comparables" data set ofthirty-one games. Lloyd
Decl. ~ 7. Lloyd had confidence that a wider range of games would lead to a reasonable projection
of Too Human's expected sales from a 2007 release. Lloyd Dep. 131-33. Next, Lloyd decided to
give extra weight to one game, Mass Effect, based on his "qualitative judgment" that Mass Effect's
characteristics were so close to Too Human's that it deserved more weight Lloyd Dec!.
Although Lloyd believed that Mass Effect was a "particularly good" comparison, he admitted that
this belief was "based on what are admittedly somewhat SUbjective factors." Lloyd Dep. 170,
220-21. Lloyd explained that Mass Effect's similarities were "good enough that we should give it
fifty percent ofthe weighting ...." Id.221. Lloyd then calculated Too Human's projected unit sales
Lloyd said that it ''wouldn't have occurred to [him] to include Grand Theft Auto IV [as a
comparable game], but [Sherlock] thought it was. He works in the industry. He stated [it was a
comparable game]. We said we'll accept his judgment on that." Lloyd Dep. 142.
s As of May, 2010, Grand Theft Auto IV had sold 7.7 million units and Gears of War had
sold 6.0 million units. Lloyd Report, Revised App. A, [D.E. 621-1] Schedule A-I.
6 Lloyd says that he indirectly considered SK's historical sales as a subjective factor in his
analysis, but that he did not directly consider SK's prior games because Too Human was "a different
product," offered on "a different platform," at a "different date and time." Lloyd Dep. 113-15. SK's
prior games resulted in total sales as follows: Blood Omen, Legacy ofKain, 320,082 units; Eternal
Darkness, 403,757 units; Metal Gear Solid, Twin Snakes, 350,539 units. See Lloyd Report 22 n.34.
At oral argument, SK's counsel asserted that Blood Omen, Legacy ofKain sold over 2 million units.
SK's counsel, however, cited nothing in the record to support this assertion.
by giving equal weight to Mass Effect's total unit sales7 and the average unit sales ofthe thirty-one
comparable games. See Lloyd Decl.' 9; Lloyd Report, Revised App. A, Schedule A-1.R
Lloyd concluded that if Too Human had been released in August 2007, SK would have sold
an additional 1.9 million units of the game, for a total of 2.5 million units. Lloyd Report 23-24 &
Revised App. A, Schedule A-1.9 Based on this sales projection, Lloyd further concluded that SK
would have received $750,000 in bonus payments from Microsoft, and Microsoft would have sold
other "ancillary" products, such as downloadable content lO ("DLC") for Too Human. resulting in
$810,000 additional income for SK. Id. at 26-28 & Revised App. A, Schedule A. Lloyd's
methodology does not have any statistical margin of error. Lloyd Dep. 130. Lloyd is not aware of
any other cases where he (or any other expert witness) has applied this particular methodology in the
way that he did in this case. See id. at 121-24.
Working from his conclusion that Too Human would have sold 2.5 million units if it had
been released in August 2007, Lloyd then calculated the projected sales for SK's undeveloped
sequels, Too Human II and Too Human m.ll Lloyd Dep. 223-24, 248. For each sequel, Lloyd used
Mass Effect's total unit sales were 2.2 million. Lloyd Report, Revised App. A, Schedule
8 The average total unit sales ofthe thirty-one selected games were
2,542,830. Lloyd Report,
Revised App. A, Schedule A-I.
9SK actually released Too Human in 2008, resulting in sales of640,076 units (through June
2009). Lloyd Report 22-23.
10 "Downloadable content" includes extra game features or keys that users can purchase for
small amounts and then download over the internet. Lloyd Report 27. Lloyd stated that these
products "can be highly profitable given the ease of distribution and direct access between the
consumer and the publisher." Id.
II SK's publishing agreement with Microsoft gave Microsoft the option to purchase the rights
to the sequels, see SK-Microsoft Agreement [D.E. 533-17] § 7.3; however Microsoft never exercised
its option and SK never developed either game. Epic's Mem. Supp. Mot. Exclude 6. According to
Epic, SK began "predevelopment work" for Too Human II, but did no work on Too Human m.
Epic's Mem. Supp. Mot. Exclude 6.
a small subset of the thirty-one comparable games identified for Too Human. Lloyd Decl. , 11.
Specifically, Lloyd assumed that SK would have released Too Human II in 2008, and therefore
considered only games that were: (1) released in 2008; and (2) released as part ofa franchise. Id.
Lloyd identified three such comparable games, Fable II, Left ~ Dead 2, and Far ~ 2. See id.; Lloyd
Report, Revised App. A, Schedule A-I. Based on the average sales ofthe three comparable games,
Lloyd projected that Too Human II would have sold 2.5 million units. Lloyd Report, Revised App.
A, Schedule A-I. As for Too Human ill, Lloyd assumed that SK would have released the game in
2009 or 2010, and therefore only considered games that were (1) released in 2009 or 2010,12 and (2)
released as part of a franchise. Lloyd Decl. , 12. Lloyd identified only two comparable games,
Assassin's Creed II and Bioshock 2. See id.; Lloyd Report, Revised App. A, Schedule A-I. Based
on the average sales ofthe two comparable games, Lloyd projected that Too Human ill would have
sold 2.4 million units. See Lloyd Report, Revised App. A, Schedule A-l.13
Lloyd also calculated the projected sales for another undeveloped game, The Ritualyst. SK
had a publishing agreement with Sega to produce The Ritualyst; however, Sega assigned its rights
to another company, THQ. See Epic's Mem. Supp. Mot. Exclude 7. THQ cancelled the project in
2009 when THQ ran out ofmoney (for reasons unrelated to The Ritualyst). Id. The game was never
finished. Id. Lloyd initially calculated the projected sales for The Ritualyst by simply averaging the
sales from two games, Grand Theft Auto IV and Assassin's Creed II. See Lloyd Report, Revised
12 Because Lloyd's calculations relied on data compiled in May 2010, Lloyd excluded games
released or sold after that date. Lloyd Decl. , 12.
13 SK never had an agreement to publish these sequels. However, Lloyd's calculations
assume that SK would have negotiated publishing agreements similar to the agreement for Too
Human. See Lloyd Dep. 246 ("Our assumption is that in the but-for world, the subsequent contracts
would have looked remarkably like the original contract, including [the units sold bonus provision].
That Microsoft would have been happy enough with [Too Human], that the volumes would have
been high enough to say-that they would have said let's do it again. Same royalty rates, same
milestones, same bonus payments, things like that.").
App. A, Schedule A-I; Lloyd Dep. 226-28. However, the two comparables had an average ofS.7
million units sold, which "struck [Lloyd] as high." Lloyd Dep. 228-29. Lloyd decided that he "felt
more comfortable with a sales volume in line with the original Too Human forecast and applied that
to The Ritualyst." Lloyd Dec1. ~ 13; see Lloyd Dep. 228-29. Thus, Lloyd returned to the average
units sold from the Too Human comparable game set and concluded that 2.8 million units sold was
a reasonable estimate for The Ritualyst. See Lloyd Report, Revised App. A, Schedule A-I; Lloyd
Decl. ~ 13.
Next, Lloyd calculated the additional costs SK incurred to develop its own engine to replace
Lloyd Report 28-29 & Revised App. A, Schedule E.
Lloyd relied on SK's
representation of the amount of time it took SK to develop its own engine, and calculated an
additional cost of approximately $3.1 million. See id.
Last, Lloyd calculated the economic harm SK suffered due to reputational harm. See Lloyd
Report 29-31 & Revised App. A, Schedule F; SK's Opp'n Mot. Exclude 2S-26. SK claims it
suffered harm to its reputation and goodwill in the form of lost development opportunities with
publishers Vivendi, THQ, Capcom, and NAMCO. See SK's Opp'n Mot. Exclude 2S-26. Lloyd
concluded that SK's lost profits from these undeveloped games were an accurate measure ofSK's
reputational harm. Lloyd Report 29-31; Lloyd Dep. 264. SK was negotiating development plans
for two ofthe projects, King's Quest and Sandmim, and had developed a prototype for King's Quest.
See SK's Opp'n Mot. Exclude 2S. However, there were no specific games associated with the other
two projects, only "discussions" with the publishers. Lloyd Dep. 277-78. In calculating the lost
profits for these projects, Lloyd did not attempt to define the genres ofthe undeveloped projects (as
he did for Too Human). See id. 264-66. Instead, Lloyd simply calculated damages by using the
expected units sold from his analysis of Too Human. Id. 26S. As a result, Lloyd based his
calculations on the assumption that each project would sell 2.S million units. See Lloyd Report,
Revised App. A, Schedule F. Lloyd also applied a "probability factor" to each project to account
for the possibility that a project might not be developed. See Lloyd Report 30 & Revised App. A,
Schedule F. To determine the probability factor for each project, Lloyd said that he made "judgment
call[s]," estimating each project's likelihood ofdevelopment based on what he understood about the
project and what he thought was reasonable. Lloyd Dep. 267-78. Lloyd acknowledged that he chose
the probability factors from ranges of possible percentages, and that he could have used other
numbers. Id. 14 Lloyd did not calculate a confidence interval for the assigned percentages. Id. 274.
Adding together all ofhis damages calculations, Lloyd concluded that SK incurred $63.6 million in
damages as a result of Epic's alleged conduct. See Lloyd Report, Revised App. A, Schedule A;
Lloyd Dep. 293.
As an alternative measure of damages, Lloyd calculated SK's damages as a portion of
Epic's profits. See Lloyd Report 31-37,38-40. Lloyd noted that Epic's profits may be relevant to
a punitive damages award or as compensatory damages under a theory of unjust enrichment. See
id. & Revised App. A, Schedule A. Lloyd calculated Epic's profits from 2005 to 2009 as between
$8.1 million and $49.6 million. IS See id. He determined that SK is entitled to between $8.7 million
and $52.9 million. See id. 16
Specifically, Lloyd assigned King's Quest a 90 percent probability, but stated that he
thought the range of reason was somewhere between SO to 100 percent. See Lloyd Dep. 268-73.
As for Sandman, Lloyd assigned a 75 percent probability, but stated that the range of reason was
approximately 40 to 90 percent. See id. 274-77. For the other two unnamed projects, Lloyd
assigned a twenty-five percent probability, but said that the range of reason was between zero and
fifty percent. See id. 277-78.
Lloyd calculated Epic's profits in two ways. Lloyd Report 33. His first calculation
considered only ''the historical, actual profits realized from the release of Unreal Tournament ~
Gears of War, and Gears of War II," and the actual value of UE3. Id. The second calculation
considered the ongoing value to Epic of the UE3-related products, calculated through reference to
Epic's cash flow from such products. Id.
Lloyd also calculated SK's entitlement to Epic's profits using alternative share rates. He
reached his first share rate of 15.3 percent by comparing Epic's costs in developing UE3 with the
amount that SK paid Epic under the licensing agreement, which Lloyd stated reflected SK's
investment in UE3. Lloyd Report 36. He reached his second share rate of2.S percent by considering
As for Lloyd's rebuttal opinions, Lloyd argued that Epic's expert, Beutel, overstated the
amount ofdamages SK would owe if Epic prevailed on its counterclaims. See Lloyd Rebuttal 3-7.
Beutel opined that Epic's damages are $4.45 million for unpaid fees under the license agreement and
over $55 million for misappropriation oftrade secrets or copyright infringement. See Beutel Report
[D.E. 636-7] 6-13. Lloyd posited that SK could have renegotiated its license agreement with Epic
or used a free version of UE3 to reduce the amount of licensing fees that SK would have owed to
$3.85 million. See Lloyd Rebuttal at 7-10. Lloyd also challenged the accuracy of Beutel's
calculation of Epic's damages for SK's alleged misappropriation. See id. 10-11.
Epic argues that Lloyd's opinions should be excluded for three reasons. First, Lloyd is not
qualified to make the assumptions regarding the video game industry necessary to support his
calculations. See Epic's Mem. Supp. Mot. Exclude 10-12. Second, Lloyd's opinions do not fit the
facts ofthe case. See id. 12-13. Third, Lloyd's methodologies are unreliable and speculative. See
id. 13-21. SK argues that Epic's objections go to the weight that should be given to Lloyd's
opinions, not their admissibility, and are matters best left for cross-examination. See SK's Opp'n
Mot. Exclude 1-3, 12, 18,21.
Rule 702 ofthe Federal Rules of Evidence governs the admission ofexpert testimony. Fed.
R. Evid. 702.
The proponent of the expert testimony bears the burden of establishing its
admissibility by a preponderance ofevidence. Cooper v. Smith & Nephew. Inc., 259 F.3d 194, 199
(4th Cir. 2001). A district court has broad latitude in making its detennination on the admissibility
of proposed expert testimony. United States v. Gastiaburo, 16 F.3d 582,589 (4th Cir. 1994).
the amount that Epic received in UE3 licensing fees between 2005 and 2006 from the various
licensees, and comparing that to the amount that Epic received from SK to license UE3 during this
period. Id. & Revised App. A, Schedule J.
Rule 702 provides that expert testimony is appropriate when it "will assist the trier of fact
to understand the evidence or to determine a fact in issue." Fed. R. Evid. 702. Rule 702 further
provides that a witness qualified as an expert may be permitted to testify where "(1) the testimony
is based upon sufficient facts or data, (2) the testimony is the product of reliable principles and
methods, and (3) the witness has applied the principles and methods reliably to the facts ofthe case."
Id. Courts have distilled Rule 702's requirements into two crucial inquiries: whether the proposed
expert's testimony is relevant and whether it is reliable. Kumho Tire Co. v. Carmichael, 526 U.S.
137, 141 (1999); Daubert v. Merrell Dow Pharm.. Inc., 509 U.S. 579, 589 (1993); United States v.
Forrest, 429 F.3d 73, 80 (4th Cir. 2005). The trial court must carry out the special gatekeeping
obligation ofensuring that expert testimony meets both requirements. Kumho Tire, 526 U.S. at 147.
To be relevant, the prop()sed expert testimony must be helpful to the trier of fact. See
Daubert, 509 U.S. at 591-92. "Testimony from an expert is presumed to be helpful unless it
concerns matters within the everyday knowledge and experience ofa lay juror." Kopfv. Skyrm,993
F.2d 374, 377 (4th Cir. 1993); accord Koger v. Norfolk S. Ry. Co., No. 1:08-0909, 2010 WL
692842, at'" 1 (S.D. W. Va. Feb. 23, 2010) (unpublished).
"[T]he test ofreliability is flexible and the law grants a district court the same broad latitude
when it decides how to determine reliability as it enjoys in respect to its ultimate reliability
determination." United States v. Wilson, 484 F.3d 267, 274 (4th Cir. 2007) (quotation omitted); see
Kumho Tire, 526 U.S. at 141-42. A witness may qualify to render expert opinions in anyone ofthe
five ways listed in Rule 702: knowledge, skill, experience, training, or education. See Kumho Tire,
526 U.S. at 147. When a party challenges an expert's qualifications, "'the test for exclusion is a
strict one, and the purported expert must have neither satisfactory knowledge, skill, experience,
training nor education on the issue for which the opinion is proffered.'" Kopf, 993 F.2d at 377
(quoting Thomas 1. Kline. Inc. v. Lorillard. Inc., 878 F.2d 791, 799 (4th Cir. 1989».
Other factors may also bear on the reliability of the expert's testimony. They include: (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) whether a technique has a high known or potential
rate of error and whether there are standards controlling its application; and (4) whether the theory
or technique enjoys general acceptance within the relevant community. Kumho Tire, 526 U.S. at
149-50; Daubert, 509 U.S. at 593-94; United Statesv. Crisp, 324 F.3d261, 265--66 (4thCir. 2003).
"In making its initial determination of whether proffered testimony is sufficiently reliable,
the court has broad latitude to consider whatever factors bearing on validity that the court finds to
be useful; the particular factors will depend upon the unique circumstances ofthe expert testimony
involved." Westbeny v. Gislaved Gummi AB, 178 F.3d 257,261 (4th Cir. 1999) (citing Kumho
Tire, 526 U.S. at 150--52). When proposed expert testimony pertains to damages, the testimony
should be excluded when it "consists of an array of figures conveying a delusive impression of
exactness in an area where ajury's common sense is less available than usual to protect it." Tyger
Constr. Co. v. Pensacola Constr. Co., 29 F.3d 137, 145 (4th Cir. 1994) (quotations and citations
omitted); see Boucher v. U.S. Suzuki Motor Com., 73 F.3d 18, 21 (2d Cir. 1996) (per curiam)
("Where lost future earnings are at issue, an expert's testimony should be excluded ... if it is based
on unrealistic assumptions regarding plaintiffs future ... prospects.").
SK argues that Lloyd's use of the comparables methodology is reliable and cites Simo v.
Mitsubishi Motors N. Am., Inc., 245 F. App'x 295, 300--01 (4th Cir. 2007) (unpublished), in support
ofLloyd's methodology. See SK's Opp'n Mot. Exclude 15-21. However, Epic does not challenge
the reliability of the yardstick or comparables methodology. See Epic's Reply 4--6. Instead, Epic
contends that Lloyd "did not use a valid comparables methodology" but rather "made up" a
methodology, and that his calculations are not rationally related to the damages he purportedly
sought to quantify. Id. Epic also describes Lloyd's expert assumptions as "uninformed guesses."
Id. at 1-2, 7-8.
In addition to requiring that an expert's methodology be generally reliable, Ru1e 702 requires
an expert to reliably apply his methodology to the facts. See Fed. R. Evid. 702; Wilson, 484 F.3d
at 276; Cooper, 259 F.3d at 200. The court must exclude an expert's testimony when it "is
connected to existing data only by the ipse dixit ofthe expert." Gen. Elec. v. Joiner, 522 U.S. 136,
146 (1997). In such cases, "[a] court may conclude that there is simply too great an analytical gap
between the data and the opinion proffered." Id.
As explained in Celebrity Cruises. Inc. v. EssefCOl:p., 434 F. Supp. 2d 169, 179 (S.D.N.Y.
2006), "the comparable companies method is reliable only to the extent that the companies chosen
are truly comparable.,,17 While a minor flaw in an expert's reasoning or methodology does not
render the expert's opinion inadmissible, the data and methodology must be adequate to support the
conclusions reached or the testimony must be excluded. See id. at 176 (citing Amorgianos v. Nat'l
R.R. Passenger Com., 303 F.3d 256, 266-67 (2d Cir. 2002)). Moreover, "[a] reliable expert opinion
must be based on scientific, technical, or other specialized knowledge and not on belief or
specu1ation, and inferences must be derived using ... valid methods." Oglesby v. Gen. Motors
Com., 190 F.3d 244, 250 (4th Cir. 1999).
As for Lloyd's lost profit calcu1ations for Too Human, Lloyd has failed to reliably apply the
methodology to the facts and his opinion is inadmissible. First, Lloyd's process for determining Too
17 The court further explained that
the use of comparable companies has the advantageous feature of "controlling" in a
rough way for market factors. That is, factors that impact the market generally are
assumed to affect the target company . . . and the comparators alike. Thus, a
downturn in [the target company's] profits is not attributed to the incident if the
comparators suffer similar losses over the same period.
ld. at 180.
Human's comparable games reveals a series ofad hoc decisions based on subjective considerations,
rather than identifiable (or principled) criteria. Lloyd identified six factors that he believed were
relevant in predicting how many units a video game will sell. Lloyd Dep. 11-12. These factors were
the game's publisher, the game's developer, the game's platform, the game's time of release, the
marketing of the game, the game's franchise, and the condition of the economy at the time of the
game's release. Id. However, Lloyd did not use any of those factors in selecting Too Human's
comparable games. See Lloyd Decl.
4-7; Lloyd Dep. 131-32; Lloyd Report, Revised App. A,
Schedule A-1. Significantly, despite asserting that a game's marketing budget was the key factor
in predicting a game's total sales, Lloyd did not choose games with marketing budgets similar to Too
Human's. See Lloyd Report 12 n.15, 13 & n.18; Lloyd Dep. 211-12. Moreover, Lloyd could not
even recall the marketing budgets for Too Human or the comparable games that he had identified.
Lloyd Dep. 70-73, 161--62, 182. 18
Second, the factors that Lloyd did use were ineffective in identifying games comparable to
Too Human. For example, Lloyd excluded all games that sold fewer units than Too Human. Lloyd
Dep. 151-53; Lloyd Report, Revised App. A, Schedule A-I. Imposing this limitation precluded the
possibility that Too Human would have sold fewer units had it been released in 2007 instead of
2008. However, Lloyd admitted that it was possible that Too Human would have sold fewer units
if it had been released in 2007. Lloyd Dep. 195-97. Indeed, Lloyd concedes that there were more
18 Despite not using a game's marketing budget as a specific criterion in selecting comparable
games, Lloyd stated that when he made ad hoc revisions to his list (after having imposed his six
criteria), he "included other games that were ... expected to have big marketing budgets, big pushes,
big rollouts, that sort of thing, that would get the benefit of that big marketing push." Lloyd Dep.
182. Lloyd stated that this strategy justified his inclusion of Grand Theft Auto IV, an admitted
"blockbuster," on his comparables list. Id. 181-82. However, he stated that at no point did he know
the marketing budget for Grand Theft Auto IV or any other game on the list. Id. 182. Additionally,
Lloyd's only basis for assuming that Too Human's marketing budget would have been comparable
to the marketing budget that he assumed for Grand Theft Auto IV and similar "blockbusters" was
Dyack's assurance that had Too Human been released in 2007, it ''would have had a big marketing
budget, would have been a big seller." Id. 181.
Xbox 360s in homes in 2008 than in 2007, Lloyd Dep. 157, and videogame sales were 23 percent
higher in 2008 than in 2007. See Lloyd Report 23. Therefore, failing to consider less successful
games meant that Lloyd's lost-profits figure failed to take into account all possible outcomes in the
"but-for world" he sought to construct.
Lloyd committed a similar error in excluding all games with Metacritic ratings lower than
Too Human's. Lloyd's only basis for not considering games with lower Metacritic ratings was that
"but for the wrongful conduct, things would have been better .... We would have had the operable
engine. We would have had a better rating. All of those sorts of things." Lloyd Dep. 188. This
conclusory statement provides no support for the argument that SK would have released a game with
better content, thereby earning a better Metacritic rating, in 2007 than it did in 2008. In fact, it is
as likely that Too Human's rating would have been lower had SK released it in 2007, because SK
probably made some improvements to the game's content during the two-year delay.19
Third, Lloyd stated that genre was an important factor in predicting a game's total sales.
Lloyd Dep. 11-13. However, when selecting comparables, he chose games from six different genres
that he considered related to Too Human. Lloyd Dec!. ~ 4. Lloyd did so because he ''understood that
Too Human has elements of different types (or genres) ofvideogames." Id. However, Lloyd failed
to identify Too Human's specific aspects that justified placing it in all six genres. Lloyd did not even
provide limiting definitions of the genres that he identified. See id. Lloyd effectively rendered his
own genre limitation moot and allowed himselfto choose as comparables the most successful games
from a wide spectrum of the video game market.
19 SK suggests that instead ofgiving it time to make improvements to Too Human, the delay
was analogous to a marathon runner suffering "a hit to the knee at Mile 15" of a 26.2 mile race.
SK's Opp'n Mot. Exclude 16 n.5. However, SK provides no evidence to support the proposition that
the delay caused it to release a game that was inferior to the game that it would have produced in
2007. See id.
Fourth, Lloyd sought to choose games as comparables that had the same target audience as
Too Human. Lloyd Dep. 198. However, again Lloyd implemented this criterion in a way that
rendered it meaningless: he personally determined Too Human's target audience, despite having
no expertise with video games. Lloyd Dep. 198.20 In addition, Lloyd circularly described Too
Human's target audience as video game purchasers who purchased games similar to Too Human.
ld. 199-201.21 In effect, Lloyd never defined Too Human's target audience, which gave Lloyd the
freedom to retain hugely successful video games as he narrowed his comparables list.
Fifth, Lloyd selected comparable games from the pool of video games released between
2006 and 2010. Lloyd Decl. W5, 12; see Lloyd Report, Revised App. A, Schedule A-I. In doing
so, he admitted that it made sense to eliminate games released before November 2005, suggesting
that data from two years prior to 2007 would not be relevant in predicting Too Human's 2007 total
sales. See Lloyd Decl ~ 5. However, Lloyd did not explain why he included games released after
2007, or even after Too Human's 2008 actual release. See id. Moreover, Lloyd admitted that he
could not ensure that he had captured the appropriate range of years, stating that in choosing the
range there was "a fair amount of art or subjectivity." Lloyd Dep. 178. However, this subjectivity
allowed him to select high-selling games that post-dated Too Human's actual release, like Call of
Dull': Modem Warfare 2 (10.7 million total units sold) and Assassin's Creed II (3.75 million total
20 Lloyd stated that he determined Too Human's target audience after consulting with "people
familiar with the industry," and after reviewing marketing documents, reviews, and "other
information." Lloyd Dep. 198,200.
21 Lloyd described Too Human's target audience as "[p]eople who buy role playing games.
People who buy action/adventure games. Possibly some people who buy shooter games. Possibly
more than the average number of females, given the role playing elements, given the enhanced
characteristics of the story lines and the characters." Id. 200-01.
SK argues that Lloyd included these games because he intended his list to serve as a set
ofcomparable games for not only Too Human, but also planned sequels to Too Human that "Lloyd
After compiling a list ofgames based on flawed criteria, Lloyd made arbitrary modifications
to his list. See Lloyd Dec!.
mr 7-8; Lloyd Dep. 135-37, 168--69. Lloyd added Grand Theft Auto IV
even though that game did not meet Lloyd's criteria and Lloyd initially did not think that Grand
Theft Auto IV was comparable to Too Human. Lloyd Dec!. ~ 7; Lloyd Dep. 142-43. Lloyd stated
that his only basis for adding Grand Theft Auto IV was Sherlock's recommendation. Id. However,
he admitted that Sherlock might have actually suggested that Grand Theft Auto IV was comparable
to The Ritualist, not Too Human. Id. Lloyd added Gears of War based on his own judgment,
because it "had similar features or aspects" to Too Human's, including its having been built by Epic
on the version ofUE3 that SK claims Too Human should have used. Lloyd Dep. 135-36.23 Lloyd
considered Gears ofWar' s "large advertising budget" an additional reason for the game's inclusion,
although Lloyd did not know what Gears of War's marketing budget had been. Id. 136, 168. Lloyd
also decided to give disproportionate weight to Mass Effect, because he believed that its
"characteristics ... were so close to Too Human that it deserved more weight in the calculation than
simply as one of the 31 other games." Lloyd Dec!.
8. However, Lloyd did not state how he
determined how much weight to give to Mass Effect or why he decided to give increased weight to
Mass Effect based on its strong similarity to Too Human, but did not assign varying weights to all
understood SK would have released" between 2006 and 2010. SK's Opp'n Mot. Exclude 17. This
argument is inconsistent with Lloyd's statement that he added games to his list of comparables
"based on what [he] understood about the game. The game being Too Human." Lloyd Dep. 131;
see also Lloyd Dec!. 3-4 ("To identify those games that would provide a basis for measuring lost
sales ofToo Human .... To identify those games that were most comparable to Too Human ....").
Lloyd's statement suggests that he included Gears ofWar only because Epic had produced
the game using the engine that SK felt the license agreement entitled SK to use. See id. This
reasoning suggests that Lloyd considered any game produced on a fully operable version of UE3
comparable to Too Human. Ifso, this statement further demonstrates the arbitrary nature ofthe list.
Lloyd provided no explanation for how adding some games to his list based on their qualitative
similarities to Too Human and other games based on their having been produced in a manner that
SK believes that Too Human should have been produced leads to a set of games that collectively
provide an accurate prediction of Too Human's expected sales.
or some ofthe other games on the list based on how similar he perceived them to be to Too Human.
SK characterizes these modifications as part of a specific, multi-step process, whereby Lloyd used
''the proxy games
Effect] as a measure of the but-for sales of [Too Human], and us[ed] his
comp set to confinn that his proxy's [sic] were not unreasonable" to "arrive at a reasonable
estimate ofSK's lost sales." SK's Opp'n Mot. Exclude 18. However, nothing in Lloyd's reports,
declaration, or deposition testimony suggests that he added games and modified the weighting ofthe
added games pursuant to such a pre-conceived, structured process. In fact, Lloyd admitted just the
opposite, stating that ''this was more ofan evolutionary list. You start here. You discuss it .... We
may have gone back and revisited the data." Lloyd Dep. 141; see also id. 136 ("We started with that
first cut, and then we added and subtracted as we saw fit based onjudgment.").
The failure to reliably apply a methodology is grounds for excluding an expert's testimony.
See Cooper, 259 F.3d at 200. Lloyd could not recall previously having used the results-oriented and
subjective methodology described to project lost profits, and there is no evidence ofa known error
rate for Lloyd's methodology. Lloyd Dep. 121-22; cf. Fed. R. Evid. 702 advisory committee note
(2000 amendment); Celebrity Cruises, 434 F. Supp. 2d at 179. Lloyd's use of his methodology
produced a damages figure based largely on his own subjective conclusions about an industry in
which he had no prior knowledge or experience. His proposed testimony regarding SK's lost profits
is connected to the video game industry data by little more than Lloyd's ipse dixit. See Gen. Elec.,
552 U.S. at 146. Therefore, SK has failed to establish that Lloyd's opinion meets the requirements
of Rule 702. Lloyd's projected lost profits for Too Human's sequels, The Ritualys!, and the four
undeveloped projects are all based on Lloyd's Too Human lost profits prediction, making them
likewise inadmissible. However, there are additional reasons for excluding these projections.
As for Lloyd's projected sales for Too Human II and Too Human ill, Lloyd relied on a much
smaller group ofcomparable games. Lloyd opined that the large sample size for Too Human (thirty
one games) gave him confidence in his calculation. Lloyd Decl. ~ 9. However, Lloyd's projections
for Too Human II used only three games for the comparables analysis, and his projections for Too
Human ill used only two games. See Lloyd Report, Revised App. A, Schedule A-I. Additionally,
Lloyd's projections regarding these sequels are highly speculative because they rely on the following
assumptions: (1) that Too Human would have sold the additional 1.9 million units; (2) that, based
on this success, Microsoft would have exercised its option to publish the sequels; (3) that Microsoft
and SK would have negotiated publishing agreements with the same terms as the publishing
agreement for Too Human; and (4) that the sequels also would have been "blockbuster" video
games. As for the assumption that Too Human and all of these other games would have been
"blockbusters," Lloyd ignores that SK has never actually produced a "blockbuster" video game. Cf.
Lloyd Report 22-26 & n.34.
Under North Carolina law, proof of damages, including lost profits, "must be made with
reasonable certainty." Olivetti Com. v. Ames Bus. Sys.. Inc., 319 N.C. 534, 546, 356 S.E.2d 578,
585 (1987); Weyerhaeuserv. Suru>ly Co.. Inc., 292 N.C. 557,560-61,234 S.E.2d 605, 607 (1977);
see BUs Day Spa. LLC v. Hartford Ins. Group, 427 F. SUpp. 2d 621,629 (W.D.N.C. 2006). As a
result, lost profit damages cannot be based upon "hypothetical or speculative forecasts." Blis Day
427 F. SUpp. 2d at 629 (quoting Iron Steamer. Ltd. v. Trinity Rest.. Inc., 110 N.C. App. 843,
847-48,431 S.E.2d 767, 770 (1993)); see Pharmanetics. Inc. v. Aventis Pharm.. Inc., No. 5:03-CV
817-FL(2), 2005 WL 6000369, at ·12 (E.D.N.C. May 4, 2005) (unpublished) ("Profit estimates
which are speculative or dependent on contingent circumstances not supported in the record are not
admissible to prove [Plaintiff's] damages." (quotations omitted)), aff'd, 182 F. App'x 267 (4th Cir.
2006) (per curiam) (unpublished). Claims for lost profits are more likely to be speculative when the
plaintiff lacks a predecessor product or profit history to use as a basis for recovery.
Pharmanetics, 2005 WL 6000369, at ·13; Olivetti, 319 N.C. at 546, 356 S.E.2d at 585-86; cf.
Scheduled Airlines Traffic Officers. Inc. v. Objective: Inc., 180 F.3d 583, 588 (4th Cir. 1999)
(applying Virginia law); Boucher, 73 F.3dat21-22 (excluding expert testimony regarding plaintiff's
future income based on unrealistic assumptions).
Lloyd's projections for the Too Human sequels are based on unreliable and speculative
Blis Oay Sp~ 427 F. Supp. 2d at 629. Thus, SK's claims for lost profits relating
to the Too Human sequels are too speculative to be determined with reasonable certainty.
Lloyd's lost profit estimates for The RituaIyst and the four undeveloped projects are
inadmissible for the same reasons. Additionally, Lloyd did not even conduct comparables analyses
for each of these games. See Lloyd Report, Revised App. A. Instead, he simply substituted the
comparables analysis that he had completed for Too Human. See id. Lloyd did not make any effort
to ensure that the games that he determined were comparable to Too Human were also comparable
to The RituaIyst or any of the four undeveloped projects. Therefore, Lloyd did not have a proper
basis to perform a comparables analysis for these games. Accordingly, his conclusions are unreliable
Lloyd's estimates regarding SK's lost profits from royalties on ancillary sales and from
unpaid bonus payments from Microsoft are also unreliable and speculative. Lloyd estimated that had
Too Human been released in 2007 SK would have earned an additional $810,000 from the sale of
"ancillary products" and Microsoft would have paid SK $750,000 in bonus payments. Lloyd Report
26-28 & Revised App. A, Schedule A. Lloyd limited his estimate regarding ancillary products to
lost revenue from Microsoft's decision not to sell Too Human OLe. Lloyd Report 26-27; Lloyd
Oep.300. Lloyd "assumed that [Microsoft's] decision not to sell OLe [for Too Human] was due,
at least in part, to delays in the projected release date as a result ofproblems with the Engine." Lloyd
Report 27. However, Lloyd's only basis for this assumption was his belief that Too Human's
delayed release caused the game to lose momentum, which caused Microsoft to decide not to
promote Too Human to the extent that it had planned, which resulted in the market for Too Human
OLe evaporating. Lloyd Oep. 298. A series of such tenuous inferences does not provide adequate
support for a central premise underlying a damages calculation. Moreover, Lloyd admitted that
there was a possibility that even if SK had released Too Human in 2007, Microsoft still would not
have released DLC. Id.299. Lloyd stated that he factored this possibility into his calculation by
reducing the amount that he expected SK to earn from DLC by fifty percent. Id.; Lloyd Report 28
& Revised App. A, Schedule A. However, he admitted that his decision to discount his prediction
by fifty percent was based only on his own judgment and input from some unnamed parties. Lloyd
Dep. 299. Lloyd provided no evidence to discount the possibility that it was more likely than not
that Microsoft would not have released DLC, even if SK had released Too Human on time. See id.
299-300. In light ofhis admitted uncertainty regarding Microsoft's marketing plans, Lloyd's failure
to provide a factual basis for identifying the probability rate makes his resulting damages calculation
unreliable. Lloyd's DLC estimate is dependent upon a number of other factors for which he had no
information, including what DLC SK would have created, what the sale price of the DLC would
have been, what royalties SK would have received from DLC sales, and what SK's development
costs for DLC would have been.24 Finally, Lloyd determined his pre-discounted estimate ofSK's
profits from DLC sales as being five percent ofToo Human's estimated total sales, assuming a 2007
release. For the reasons previously stated, that estimate is unreliable, making the resulting DLC
estimate further unreliable.
Regarding the lost bonus payments, SK's contract with Microsoft entitled it to receive a
$500,000 bonus payment when Too Human sold more than 1 million units, and an additional
24 Lloyd appears to have attempted to account for these factors by estimating that SK's
revenue from DLC sales would have been five percent ofthe revenue SK earned from unit sales of
Too Human. Lloyd Report 27 & Revised App. A, Schedule A. However, Lloyd determined that a
game would generate DLC revenue equal to five percent of its unit sale revenue by looking at the
DLC revenue of just one game, Call of Duty, World at War. Lloyd Report, Revised App. A.,
Schedule A; Lloyd Dep. 301. Lloyd does not state that he considered Call of Duty, World at War
to be a comparable game to Too Human. His reliance on data pertaining to it in the context ofDLC
sales is undercut by his reliance on comparable games in other portions of the analysis.
Accordingly, his five-percent figure lacks an adequate factual basis and therefore is unreliable.
$250,000 bonus payment when Too Human sold more than 1.5 million units. See SK-Microsoft
Agreement § 6.1 (c); Lloyd Report 26. Lloyd determined that Microsoft would have paid SK a
$750,000 bonus, based on his conclusion that Too Human would have sold 2.5 million units had it
been released in 2007. Lloyd Report 26 & Revised App. A, Schedule A. As stated, Lloyd reached
his 2.5 million unit estimate using unreliable means. Therefore, his bonus estimate is unreliable.
Lloyd estimated SK's lost profits from four games that SK claims went undeveloped as a
result of Epic's misconduct. Lloyd Report 29-31 & Revised App. A, Schedule F. Lloyd intended
the lost profits from these games to serve as a proxy for measuring the reputational damage that SK
incurred. Id.; Lloyd Dep. 264. However, SK cites no North Carolina authority holding that a party
may recover reputational damages in breach of contract or UDTP A action. Even if reputational
damages are recoverable in such actions, Lloyd's calculation ofSK' s reputational harm is unreliable.
First, Lloyd determined SK's lost profits on each of these undeveloped games using an unreliable
methodology. He first determined that SK would have sold 2.5 million copies of each of the
undeveloped games, based on his assumption that SK would have sold 2.5 million copies of Too
Human if it had released Too Human in 2007. Lloyd Report, Revised App. A, Schedule F; Lloyd
Dep.269. Lloyd did not provide any facts to support his assumption that these four games would
all have sold 2.5 million copies. In making this unsupported assumption, he relied exclusively on
another unsupported assumption, that Too Human would have sold 2.5 million units but for the
delay. He also discounted SK's expected profits on each game by a percentage, in order to reflect
the probability that the game would not be produced. Lloyd Report 29-30. However, Lloyd
admitted that in assigning these probabilities, he made another "judgment call," and that there was
a wide range of reasonable probabilities that he could have chosen for each game. Lloyd Dep.
270-73. The wide range ofreasonable probabilities reflects the arbitrariness ofLloyd's calculation.
Second, even if Lloyd's calculations did produce a reliable estimate ofSK's lost profits from these
undeveloped games, neither Lloyd nor SK provide any basis for the proposition that lost profits from
four individual games are an adequate proxy for the overall harm inflicted on SK's corporate
In sum, the court excludes as unreliable and speculative Lloyd's opinions regarding SK's lost
profits relating to the Too Human trilogy, The Ritualyst, and SK's four undeveloped projects.
As for Lloyd's opinions regarding SK's costs to develop its own game engine, the parties
agree that Microsoft and Sega reimbursed SK for any costs that SK incurred in developing its own
SK-Microsoft Agreement, amend. 7. Under North Carolina law, a plaintiff
in a breach of contract or UDTPA action may not recover damages for an injury for which it has
already been compensated. See Chemimetals Processing. Inc. v. Schrimsher, 140 N.C. App. 135,
138,535 S.E.2d 594,596 (2000) (holding plaintiff-corporation seeking damages from its board of
directors in a breach of contract and UDTPA action could not recover from the board of directors
when it had already been fully compensated for its alleged injury by settling claims alleging the same
injuries against its corporate president); Duke Univ. v. St. Paul Mercwy Ins. Co., 95 N.C. App. 663,
681,384 S.E.2d 36, 47 (1989) ("Payment ofcompensation ... to plaintiffby a third party ... may
be shown in reduction of damages for breach of contract." (quoting 25 C.J.S. Damages § 97, at
1003-05 (1966)) (second alteration added)); seealsoPiedmontInst. ofPainMgmt. v. Staton Found.,
157 N.C. App. 577, 592-93, 581 S.E.2d 68, 78 (2003); Markham v. Nationwide Mut. Fire Ins. Co.,
125 N.C. App. 443, 455, 481 S.E.2d 349,357 (1997); cf. Sun Chern. Trading Com. v. SOS Control
Servs.. Inc., 159 F. App'x 459, 462 (4th Cir. 2005) (per curiam) (unpublished) ("It is well-settled that
North Carolina law precludes a plaintiff from recovering more than one-satisfaction for the same
injury."). Because SK cannot recover for the damages that it claims it incurred in designing and
building its own game engine, Lloyd's opinions on these damages are not relevant and therefore are
In opposition to this
SK states that had Microsoft not been required to
compensate SK for its additional costs in developing Too Human, Microsoft would have invested
an additional $4 million into the marketing of Too Human. SK's Opp'n Mot. Exclude 23-24.
Therefore~ SK appears to reason that the money that it received from Microsoft, while compensating
it for one injury~ caused another injury-decreased sales ofToo Human. However, SK provides no
factual support for its assumption that Microsoft took funds that it had previously allocated to market
Too Human to compensate SK for its increased development costs, other than the conclusory
statement that "Microsoft had the will to invest a certain amount of money in Too Human." Id.
Moreover, even assuming SK's factual proposition, SK does not cite and the court has not found any
authority to suggest that a plaintiff is not adequately compensated for his injury when such
compensation subsequently causes that plaintiff to incur a second, distinct injury.
The court also recognizes that SK' s publishing agreement with Microsoft entitled Microsoft
to receive all royalties otherwise payable to SK from the sales of Too Human until SK fully
compensated Microsoft for having covered SK's development costs. See SK-Microsoft Publishing
Agreement, amend. 7; Lloyd Report 20-21,23-25. SK argues that it did incur costs in developing
its own game engine because of Microsoft's collection of these royalties. Lloyd Report 24-26.
However, SK relies on Lloyd's lost-profits calculations to support SK's assumption that it would
have earned royalties sufficient to fully compensate Microsoft, meaning that SK would have repaid
Microsoft for the entire cost of developing the engine. Lloyd Report 25 & n.45. This argument
rests on the faulty assumption that Lloyd accurately calculated SK's lost profits attributable to the
delay. As discussed, Lloyd's calculation is flawed. Therefore, SK has no evidence that it would
have received sufficient royalties from total sales of Too Human to fully reimburse Microsoft.
Instead, the record shows that Microsoft compensated SK for its game engine development costs.
As for Lloyd's opinions regarding SK's damages based on Epic's unjust enrichment, Epic
argues that this testimony is not relevant because Epic's profits are not an adequate measure ofSK's
damages and because Lloyd used a flawed methodology in determining Epic's profits. See Epic's
Mem. Supp. Mot. Exclude 21-22. SK responds that Lloyd has accurately calculated Epic's wrongful
gain from Epic's alleged breach of the license agreement, and that SK is entitled to a portion of
Epic's unjust enrichment as compensatory damages for the breach. See SK's Mem. Opp'n Mot.
The court dismissed SK's unjust enrichment claim. See [D.E. 605]. In doing so, the court
adopted the magistrate judge's memorandum and recommendations. Id.; see id. Mem. & Rec. [D.E.
592]. The magistrate judge held that SK had "failed to come forward with any evidence sufficient
to raise a question as to the validity ofthe License Agreement," and thereby had failed "to adequately
support its unjust enrichment claim." Mem. & Rec. 33. Nevertheless, SK argues that the portion
of Epic's "unjust enrichment" to which SK is entitled serves as an alternative measure of SK's
breach ofcontract damages, even absent a substantive claim for unjust enrichment. See SK's Mem.
Opp'n Mot. Exclude 26. In support, SK cites Polo Fashions. Inc. v. Craftex. Inc., 816 F.2d 145, 147
(4th Cir. 1987). In Polo Fashions, plaintiff alleged trademark infringement in violation of the
Lanham Act and a violation of North Carolina's UDTPA, based on defendants' sales of clothing
bearing a mark substantially similar to plaintiff's iconic mark. 816 F.2d at 147. In affirming the
district court's award ofdamages for the UDTPA violation based on defendants' wrongful profits,
the Fourth Circuit held that "[t]he defendants' profits ... are a rough measure of the plaintiff's
damages. Indeed they are probably the best measure ofdamages available." Id. at 149.
Polo Fashions is distinguishable. First, in Polo Fashions, the court reasonably assumed that
defendants' profits from sales of clothing bearing the offending mark were a direct result of the
presence of the mark on the clothing. See id. Here, the link between Epic's allegedly wrongful
conduct and its profits from sales ofthree video games and licensing ofa video game engine is much
more tangential. Lloyd stated that Epic's unjust enrichment was an adequate measure of damages
because there was evidence that suggested that Epic "diverted its resources from the development
of [UE3] to development of its own portfolio of products," resulting in SK's having "indirectly
invested in the products Epic has developed with the resources not devoted to [UE3]." Lloyd Report
31-32. However, Lloyd does not provide any support for his assumption that Epic developed its
own video games and developed UE3 using funds diverted from its UE3 license agreements. It is
much easier for a court to assume that a defendant in a trademark infringement action sold clothing
because the clothing bore the infringing mark than it is for the court to assume that a business with
multiple revenue sources developed and profited from a line of products only because it received
revenue from but did not perform its obligations under a set ofcontracts. Second, the Polo Fashions
court looked to defendants' profits as a basis for damages because ofthe impossibility ofcalculating
plaintiff's damages in more exact terms. See 816 F.2d at 149 ("[I]nstead of having a fact finder
assess damages with little guidance, fairness to the infringers suggests strongly that the plaintiff's
damages should be limited to the defendants' profits."). SK does not contend that Epic's unjust
enrichment is the only available measure ofSK's damages. See SK's Mem. Opp'n Mot. Exclude
26-27. To the contrary, Lloyd asserted that the comparables analysis, which he described as a
"standard, widely-accepted in measuring damages," produced a reliable measure of SK's damages.
Lloyd Decl. ~ 2. SK provides no basis for holding that fairness requires the court to look to Epic's
profits in calculating SK's damages when SK has provided a theoretically reliable (but unreliably
executed) alternative means of calculating its damages. Therefore, Epic's profits cannot serve as a
basis for calculating SK's compensatory damages. See Watson v. Dixon, 352 N.C. 343, 347, 532
S.E.2d 175, 178 (2000) ("The objective of compensatory damages is to restore the plaintiff to his
original condition or to make the plaintiff whole."); United Labs. Inc. v. Kuykendall, 335 N.C. 183,
195,437 S.E.2d 374, 381 (1993). Instead, Lloyd's calculation ofSK's share of Epic's allegedly
Wljust enrichment is relevant only in measuring damages for a previously dismissed Wljust
enrichment claim. Accordingly, this testimony is not relevant and is excluded.
Alternatively, Lloyd's determination ofSK's entitlement to Epic's supposed wrongful gains
is unreliable. Lloyd calculated Epic's Wljust enrichment in two ways, both of which rested on
flawed assumptions. See Lloyd Report 33.
Lloyd's first measured Epic's Wljust enrichment based on ''the historical, actual profits
realized from the release ofUnreal Tournament;1, Gears of War, and Gears of War II, and [UE3]."
Id. & Revised App. A, Schedules G, K, L. However, his conclusion that SK is entitled to a share of
these profits rests on the unfoWlded assumption that ''that Epic's gains were at the detriment ofSK
and other UE3 licensees." Lloyd Report 33. As discussed, Lloyd cited no factual basis for this
assumption. Therefore, this determination of SK's claimed share in Epic's profits rested upon an
unfoWlded assumption, making it unreliable.
Lloyd's second approach attempted to predict the ongoing value of these four products to
Epic. Id. Lloyd calculated the products' ongoing value by attempting to calculate Epic's cash flows.
Id. 37 & Revised App. A, Schedule H. In determining Epic's cash flows, Lloyd considered:
Epic's historical and projected revenue; expected working capital for ongoing
operations; capital and debt considerations; revision of expected bonus payouts and
non..cash considerations (amortization and depreciation); a declining share rate of
UE3-Related Products to consider phase out ofUE3 Related Products from Epic's
product portfolio; and application of discoWlt rate to present value the future cash
Lloyd Report 35. Lloyd relied on a report prepared for Epic by a third-party auditor in gathering the
data for these factors. Id. 35 n.63, 39-40. However, he also conducted his own analysis of Epic's
fmancial data for the period between 2005 and 2008. Id. 39-40. In some instances Lloyd fOWld the
data ofthe third-party auditor unreliable. Id. In those cases, Lloyd replaced the auditor's data with
his own. Id. Lloyd stated that these substitutions were warranted because the auditor prepared its
report for purposes distinct from Lloyd's purposes and valuation studies like the auditor's report
are generally designed to benefit the company by undervaluing corporate assets. Id. 40. Epic
characterizes Lloyd's process as "discard[ing] the portions ofthe report that did not suit his desired
outcome and substitut[ing] assumptions that he preferred." Epic's Mem. Supp. Mot. Exclude 22.
Lloyd's substitutions did produce a cash flow calculation higher than the calculation would have
been had he relied exclusively on the auditor's report. See Lloyd Report, Revised App. A, Schedule
Assuming without deciding that the auditor's report undervalued Epic's total assets, Lloyd's
projection of Epic's cash flow is not reliable because Lloyd provided no factual support for the
alternate figures that he used. Cf. Lloyd Report 39-40. Notably, Lloyd determined that the bonus
payments that Epic paid to its corporate executives were "excessive," and he accordingly discounted
Epic's bonus payments by ninety percent. Id. & Revised App. A, Schedule H, M.2s Lloyd reached
this conclusion based on his colleague's review of "six or eight other videogame software
companies," the identities of which he could not recall in his testimony. Lloyd Dep. 279-80.26
Moreover, Lloyd did not state how he determined that ninety percent of Epic's executive bonus
payments were wrongly excluded from its cash flow, as opposed to some smaller percentage.
Therefore, Lloyd calculated Epic's cash flow based on his independent re-characterization of
significant business expenses, without providing an adequate factual basis for why the expenses
should be re-characterized or how he should re-characterize them. Lloyd's arbitrary use ofhis own
2S Lloyd concluded that Epic had falsely characterized as bonus payments what were actually
stock dividend payments to majority shareholders, thereby allowing the company to adequately
compensate its executives and simultaneously understate its cash flow. Lloyd Report 42. By the
time of his deposition, Lloyd stated that further analysis had led him to conclude that only five
percent ofEpic's payments to its executives were classifiable as employee bonus payments, and he
had accordingly adjusted his calculations. Lloyd Dep. 280.
26 Lloyd stated that he believed that the six companies that he considered were listed in his
rebuttal report. Lloyd Dep. 281. A review of that report does not indicate what companies Lloyd
considered. See Lloyd Rebuttal 1-12.
unsubstantiated judgment makes his calculation ofEpic' s cash flow unreliable. The unreliability of
Lloyd's calculations provide an alternative ground for excluding Lloyd's unjust enrichment
Lloyd also seeks to testify regarding Epic's cash flow in the context of SK's claim for
punitive damages. Lloyd Report 38-40. For the reasons already discussed, Lloyd's calculation of
Epic's cash flow is unreliable to prove punitive damages to the same extent that it is unreliable to
prove compensatory damages. Moreover, the court already has "reject[ed] SK's claim that Epic's
bonus payments, projected revenue, or projected value are relevant to punitive damages." See [D.E.
692] at 7-8; see N.C. Gen. Stat. § lD-15(c).
As for Lloyd's rebuttal report, Epic contends that Lloyd's rebuttal opinions are inconsistent
with the facts ofthe case. See Epic's Mem. Supp. Mot. Exclude 23. Indeed, Lloyd's rebuttal report
hypothesizes that SK could have renegotiated the terms ofthe UE3 license agreement or SK could
have used a free version of UE3 ("Unreal Development Kit" or "UDK"), which Epic made available
in November 2009. See Lloyd Rebuttal 8-10. According to Lloyd, if SK had taken these steps,
Epic's damages would be much lower. See id.
Lloyd's only factual support for his conclusion that Beutel erred in believing ''that SK could
not, or would not, have renegotiated their [sic] licensing agreement with Epic for [UE3]" was that
"Epic did renegotiate licensing fees with some UE3licensees." Lloyd Rebuttal 8. Epic'soccasional
renegotiations do not support Lloyd's conclusion. Between 2005 and 2008, Epic renegotiated eight
ofits UE3 license agreements. See Lloyd Rebuttal, App. A, Schedule A-I. During the same period,
Epic had license agreements with 175 licensees. See List ofLicensees [D.E. 67] .. Moreover, in only
four ofthe renegotiations that Lloyd cites did the renegotiation result in more favorable terms for the
licensee. See Lloyd Rebuttal, App. A, Schedule A-I. Nevertheless, Lloyd claims that it was
unreasonable for Beutel to assume that SK would not have favorably renegotiated its agreement with
Epic, despite the fact that Epic had agreed to such licensee-favorable modifications in only two
percent of its UE3 license agreements. Such modifications were hardly "firmly rooted in Epic's
practices," as SK would have the court believe. See SK' s Mem. Opp' n Mot. Exclude 29. Moreover,
Lloyd did not identify any incentive for Epic to renegotiate its agreement with SK. Epic instead
persuasively argues that "renegotiating the license agreement to lower SK's license fees only could
have harmed Epic's economic interests." Epic's Mem. Supp. Mot. Exclude 23.
In addition to being factually unsupported, Lloyd's argument concerning a hypothetical
contract renegotiation is legally unsound. SK does not cite, and the court has not located, a case
holding that a defendant in a breach of contract action can avoid liability for the alleged breach by
arguing that the plaintiffwould have favorably modified the contract had the defendant so requested.
Such a rule would produce the absurd result of essentially rendering contractual language
Lloyd also states that Beutel failed to consider the possibility that SK would have stopped
using UE3 in November 2009 and started using "a free version ofUE3," UDK. Lloyd Rebutta19
(emphasis in original). However, Epic states that SK would have had to pay a fee of$99 and 25%
ofall UDK-related income above $50,000 to use UDK. Epic's Mem. Supp. Mot. Exclude 24. Epic
argues that since SK's financial statements demonstrate that SK "recorded as revenue millions of
dollars in advance royalty payments from publishers between 2004 and 2008," the UDK license
would have cost SK a significant amount of money. [d.; see id. Ex. 9 (SK's financial statement).
Accordingly, in contrast to Lloyd's assumptions, switching to the UDK in 2009 would not have
been free to SK.
Rule 702 permits expert testimony based on, inter ali!!, "sufficient facts or data." Fed. R.
Evid. 702(b). "The language 'facts or data' is broad enough to allow an expert to rely on
hypothetical facts that are supported by the evidence."
[d. advisory committee note (2000
amendment) (citations omitted). Lloyd, however, relied on hypothetical facts that are not supported
by the evidence in forming his opinions regarding Beutel's report. Accordingly, this testimony is
not relevant and inadmissible. Fed. R. Evid. 702; see Daubert, 509 U.S. at 591; Newman v. Hy-Way
Heat Sys.. lnc., 789 F.2d 269, 270 (4th Cir. 1986) ("It is fixed law that an expert can give his opinion
on the basis of hypothetical facts, but those facts must be established by independent evidence
properly introduced." (quotation omitted»; Caseyv. Geek Squad Subsidiary Best Buy Stores. L.P.,
F. Supp. 2d --,2011 WL 5505376, at *6 (D. Md. Nov. 10,2011).
Epic also objects to Lloyd's reliance on information that SK provided and his conclusion that
SK's advance royalty payments are not considered revenue. See Epic's Mem. SUpp. Mot. Exclude
24. Although these disputes appear to go to the weight of Lloyd's testimony rather than its
admissibility, they are not material to the testimony, and the court concludes that Lloyd is not entitled
to testify on these immaterial issues when the remainder of his testimony is inadmissible.
Alternatively, Lloyd admitted that he assumed that Epic was liable for all ofSK's alleged
claims when preparing his report. Lloyd Dep. 20 ("Well the entire report is premised on the
assumption that Epic is guilty of the acts alleged."). However, since Lloyd prepared his report, the
court granted summary judgment as to SK's claims of intentional interference with contractual
relations, intentional interference with prospective economic advantage, unjust enrichment, and
rescission or reformation. [D.E.605]. Therefore, Lloyd's damages calculations are based on an
assumption regarding SK's success on all of its claims that is no longer viable. A damages model
premised on a false assumption is not reliable. See Pharmanetics, 2005 WL 6000369, at * 10 ("As
[the expert's] damages model is premised upon plaintiff's success on all of its causes ofaction, and
does not allow, even in the alternative, for an adjustment ofdamages to account for those claims no
longer remaining, it is not a reliable measure of damages."). Accordingly, Lloyd's opinions are
In sum, the court GRANTS Epic's motion to exclude the reports and testimony of Terry
Lloyd [D.E. 620], and excludes Lloyd's opinions on damages. The court DENIES in part and
GRANTS in part Epic's motion to seal the supporting memorandum and exhibits [D.E. 623].
Sections Ill, IV.C, IV.D, IV.E, and IV.F of Epic's memorandum and the accompanying exhibits
to the motion and memorandum are properly sealed. The balance ofthe memorandum is not. The
court also DENIES in part and GRANTS in part SK's motion to seal the response in opposition and
accompanying exhibits [D.E. 644]. Sections II, IV.C, IV.D, IV.E, and IV.F of SK's response are
properly sealed. The balance of the response and the accompanying exhibits are not.
SO ORDERED. This 1J. day of December 2011.