F:\29120 Ind Law Rev 48-2\48 masthead.wpd Indiana Law Review Volume 48 2015 Number 2 ARTICLES “CAN I PROFIT FROM MY OWN NAME AND LIKENESS AS A COLLEGE ATHLETE?” THE PREDICTIVE LEGAL ANALYTICS OF A COLLEGE PLAYER’S PUBLICITY RIGHTS VS. FIRST AMENDMENT RIGHTS OF OTHERS ROGER M. GROVES* A seven-minute rap music video by Sir Mix-A-Lot called “Baby Got Back” was mixed with music by the Seattle Symphony.1 The atypical concoction went viral on YouTube. There were more than 1.5 million views in five days in June 2014.2 Imagine if a college athlete on scholarship created such a video, incorporating his musical talent with his own end-zone dance, or spin move on the basketball court. Would he be able to profit from that video and retain his eligibility in the sport that brought him fame? Would he be able to keep others from profiting from his video without his permission? Those are the unresolved issues explored in this Article. INTRODUCTION In 2013, the law experienced something that had not happened in all of prior legal history. In that year, two federal cases provided a blueprint for whether collegiate athletes have rights in their name, image, and likeness (“NIL” also termed “publicity rights”) that is superior to the First Amendment rights of those who use the NIL without their permission.3 The cases established legal tests for determining if a celebrity, athlete, or entertainer can make claims under his or her publicity rights to enjoin an unauthorized producer of a work and claim damages for profits derived from that work.4 * Roger M. Groves is a Professor of Law and Director of the Business Law Certificate Program at Florida Coastal School of Law. He is a former tax judge and equity partner in Howard & Howard Attorneys, P.C. and weekly contributor to Forbes’ SportsMoney. 1. See Michael Cooper, A Symphony’s Viral Video, Debating Seattle Orchestra’s Foray with Sir Mix-A-Lot, N.Y. TIMES, June 11, 2014, at C1. 2. See id. 3. See infra Part III (discussing a more detailed definition of “publicity rights”). 4. See generally Keller v. Electr. Arts, Inc., 724 F.3d 1268 (9th Cir. 2013); Hart v. Electr. http://dx.doi.org/10.18060/4806.0001 370 INDIANA LAW REVIEW [Vol. 48:369 But those cases still leave many untested applications to new facts—facts that the courts have not faced. Particularly intriguing is how twenty-first century technology will apply to this area in future litigation. No publicity right, case, or article to date has explored the application of predictive analytics, computer programs, algorithms, and the discovery issues of electronically stored information.5 This Article does just that—explores the substance of the tests, the implications of those tests for future cases, and how algorithms may become an integral and important part of those controversies. This is one of the future battlegrounds for this rather unique and emerging body of law. The above referenced cases are Hart v. Electronic Arts, Inc., and Keller v. Electronic Arts, Inc.6 Both cases involve quarterbacks for universities whose football teams and players are regulated in large part by the National Collegiate Athletic Association (“NCAA”).7 A third case with larger public name recognition is O’Bannon v. NCAA.8 The trial in this case concluded on June 27, 2014.9 O’Bannon also involves the NIL of former college athletes, and current college athletes.10 Due to the pretrial Arts, Inc., 717 F.3d 141 (3d Cir. 2013). 5. Predictive Analytics has been defined as follows: “The use of statistics and modeling to determine future performance based on current and historical data. Predictive analytics look at patterns in data to determine if those patterns are likely to emerge again, which allows businesses and investors to adjust where they use their resources in order to take advantage of possible future events.” Predictive Analytics, INVESTOPEDIA.COM, http://www.investopedia.com/terms/p/ predictive-analytics.asp (last visited June 1, 2014), archived at http://perma.cc/HAN4-GCAH. 6. See Keller, 724 F.3d at 1268; Hart, 717 F.3d at 141. Both cases had not reached final rulings in these reported cases but ultimately settled. 7. Ryan Hart was a quarterback for Rutgers University for the 2002 through 2005 seasons. Hart, 717 F.3d at 145-146. Sam Keller was a quarterback for Arizona State University and Nebraska. Keller, 724 F.3d at 1271. The NCAA is a private and voluntary yet regulatory association of colleges and universities. The schools have agreed to abide by certain rules that are uniform within each of three divisions. MATTHEW MITTEN ET AL., SPORTS LAW AND REGULATION 99 (Wolters Kluwer, 3d ed. 2013). 8. The first published opinion in this five-year litany was O’Bannon v. NCAA, No. C 09- 3329 CW, 2009 U.S. Dist. LEXIS 122205 (N.D. Cal. Dec. 11, 2009). 9. The trial lasted fifteen days after five years of preliminary sparing among the parties. See Steve Berkowitz, Closing Briefs Are In; O’Bannon Case in Hands of Judge, USA TODAY (July 11, 2014, 10:21 AM), http://www.usatoday.com/story/sports/college/2014/07/10/ed-obannon-antitrust- case-against-ncaa-closing-judge-claudia-wilken/12510271/, archived at http://perma.cc/7SSC- BNEE. 10. See Steve Berkowitz, Judge will Allow Player to Join O’Bannon Suit, USA TODAY (July 5, 2013, 6:24 PM), http://www.usatoday.com/story/sports/college/2013/07/05/ed-obannon-ncaa- likeness-lawsuit/2492981/, archived at http://perma.cc/T3FB-BMQG; see also Roger Groves, Little Known Federal Court Ruling Hints At NCAA Showdown With Current Student Athletes, FORBES (July 10, 2013, 10:04 PM), http://www.forbes.com/sites/rogergroves/2013/07/10/little-known- federal-court-ruling-hints-at-ncaa-showdown-with-current-student-athletes/, archived at http://perma.cc/8ZKK-JJT2 (noting the inclusion of current players in lawsuit). 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 371 settlement of publicity rights issues, the only issue was whether the defendants violated antitrust laws.11 On August 8, 2014, Judge Claudia Wilken issued a ninety-nine page opinion holding that the NCAA cannot form agreements with its member institutions to prohibit players from receiving any money from their NILs while playing for the school.12 In antitrust terms, this is an unreasonable restraint of trade.13 Judge Wilken held that NCAA rules that prevent college players from sharing at all in the revenue unlawfully restrains price competition among FBS football and Division 1 basketball schools.14 The publicity rights portion of O’Bannon was resolved through a negotiated settlement of $20 million to the plaintiffs.15 Hart and Keller had been consolidated with O’Bannon in anticipation of trial, but a settlement was reached with the non-NCAA defendants, video game manufacturer Electronic Arts, Inc., and the NCAA licensing partner, CLC.16 The NCAA was the last defendant to settle on the publicity rights claims.17 This Article is focused entirely on the battle between athlete’s NIL/publicity rights and the First Amendment rights of those who use those rights in their own works or products. The aforementioned publicity rights settlements predictably do not provide any admission of liability or wrongdoing by the NCAA or any other defendants. Therefore, the primary source of legal authority and precedent on the publicity rights issues reside in the Hart and Keller opinions discussed in this Article. In all three cases, the athletes abided by the NCAA rules that required that they refrain from taking advantage of their relative fame through licensing their NILs or otherwise being paid in any form, directly or indirectly from the sport.18 In all three cases, the NCAA and its licensee partners gained substantial revenue 11. The plaintiffs claim the NCAA and its venture partners in video games engaged in price fixing, inter alia, in violation of Section 1 of the Sherman Act. O’Bannon, 2009 U.S. Dist. LEXIS 122205, at *3. 12. O’Bannon v. Nat’l Collegiate Athletic Assoc., No. C 09-3329 CW (N.D. Cal. Aug. 8, 2014), available at http://i.usatoday.net/sports/!Invesitgations-and-enterprise/OBANNONRULING. pdf, archived at http://perma.cc/HA63-PKYE. 13. See O’Bannon, 2009 U.S. Dist. LEXIS 122205, at *3-4. 14. See id. 15. The settlement may have been reached in dramatic fashion reminiscent of old television scripts at the steps of the courthouse the day of trial. In fact, the settlement was not announced until the opening minutes of the first day of trial: June 9, 2014. See Sharon Terlep, NCAA to Pay Ex- Athletes $20 Million To Settle Suit, WALL ST. J., June 11, 2014, at B-1. 16. See Sharon Terlep, NCAA reaches $20 Million settlement with Ex-Players Over videogames, WALL ST. J. (June 9, 2014), http://online.wsj.com/articles/ncaa-unveils-20-million- settlement-with-ex-players-over-videogames-1402330931, archived at http://perma.cc/2TZ5- QDUL. 17. Id. 18. Keller v. Electr. Arts, Inc., 724 F.3d 1268 (9th Cir. 2013); Hart v. Electr. Arts, Inc., 717 F.3d 141 (3d Cir. 2013). 372 INDIANA LAW REVIEW [Vol. 48:369 using the NILs of the athletes.19 Furthermore, in all three cases the athletes did not receive proceeds from the licensing or use of their NILs from the NCAA or its partners before or after the collegiate eligibility expired.20 O’Bannon is particularly relevant to this Article because it is the landmark holding of the fundamental point that the NCAA cannot form agreements to deny collegiate athletes (current and former) all proceeds from the players’ own NILs.21 That quite clearly means that these athletes have a legal right to share the revenue generated from the use of their NILs.22 Yet it is important to distinguish O’Bannon on other grounds. The antitrust issues faced in that case have separate legal standards designed to achieve a different legal result—to determine whether anticompetitive aspects of NCAA rules are an unlawful restraint of trade.23 Hart and Keller involve standard publicity right claims that are dependent on misappropriation of intellectual property, as discussed in detail below.24 So while O’Bannon authorizes the NCAA and its member schools to share licensed revenue from players’ NILs, form rules to cap amounts received, and provide only equal amounts to players from licensed NIL revenue, those allowances were only designed to meet antitrust standards.25 The court was not facing the misappropriation of NIL claims existing in Hart and Keller.26 In fact, individualized damages were no longer before Judge Wilken in O’Bannon.27 This Article will examine in detail the historic underpinnings for the relationship between athlete and NCAA (Part I), the legal precedents giving rise to publicity rights for athletes (Parts II-III), and the most recent standards used to decide whether athletes’ publicity rights trump the First Amendment rights of the users of their NILs (Parts IV-VI). The discussion then turns to the implications of the theories and practical application, including opportunities for predictive analytics to be part of the sophisticated handling of such cases (Section VIII). The question of whether antitrust laws prohibit the NCAA and related party use of the NILs is outside the scope of this Article.28 19. See Keller, 724 F.3d at 1268; Hart, 717 F.3d at 141; O’Bannon, 2009 U.S. Dist. LEXIS 122205. 20. See Keller, 724 F.3d at 1268; Hart, 717 F.3d at 141; O’Bannon, 2009 U.S. Dist. LEXIS 122205. 21. See O’Bannon v. Nat’l Collegiate Athletic Assoc., No. C 09-3329 CW, 2010 WL 445190, at *98 (N.D. Cal. Aug. 8, 2014), available at http://i.usatoday.net/sports/!Invesitgations-and- enterprise/OBANNONRULING.pdf, archived at http://perma.cc/TF7A-3S57. 22. Id. 23. Id. 24. See Keller, 724 F.3d at 1268; Hart, 717 F.3d at 141. 25. See O’Bannon, 2010 WL 445190, at *95-97. 26. See Keller, 724 F.3d at 1268; Hart, 717 F.3d at 141. 27. Judge Wilken noted the plaintiffs were “only seeking to enjoin restrictions on the sharing of group licensing revenue,” not traditional damages to recoup profits from the unauthorized use of player NIL. See O’Bannon, 2010 WL 445190, at *97. 28. O’Bannon v. NCAA is the lead case on whether the NCAA violated antitrust law in 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 373 I. THE NCAA RELATIONSHIP WITH THE ATHLETE’S PUBLICITY RIGHTS The NCAA has a contractual relationship with colleges and universities, both termed “member institutions.”29 Both the NCAA and its member institutions secure a contractual obligation from the athlete to refuse the commercial exploitation of his own NIL while performing as an “amateur” at the institution.30 The most pertinent provisions of the NCAA bylaws state that an athlete loses his amateur status and scholarship opportunity to play college sports if the athlete either “uses his . . . athletics skill (directly or indirectly) for pay in any form in that sport,”31 or “accepts any remuneration or permits the use of his . . . name or picture to advertise, recommend or promote directly the sale or use of a commercial product or service of any kind.”32 The scholarship athlete performing in NCAA-sanctioned sports, like football or basketball, therefore cannot profit from, and more literally, receive any money from his own name or picture for commercial products or services.33 This means he cannot license his name, for example, to promote Gatorade or some other sports drink, or Nike shoes. Importantly, neither of those provisions expressly prohibits any other entity from using and deriving revenue from the athlete’s name or likeness.34 That ambiguity is part of what gives rise to the litigation in Hart and Keller.35 Those cases center not on the actual attempts by the quarterbacks to profit from their own names,36 but whether New Age Athletes can prohibit the NCAA, its licensing partners, and video game manufacturers from profiting from their NILs.37 denying the plaintiff athletes their use of their NILs. The parties are awaiting a final decision and order. 29. See Who We Are, NCAA, http://www.ncaa.org/about/who-we-are/membership (last visited July 14, 2014), archived at http://perma.cc/N9DA-RRH5. 30. Article 12 of the NCAA bylaws codifies its rules of amateurism that the athlete must agree to be so governed. NCAA, 2011-12 NCAA DIVISION 1 MANUAL § 12.01.1 (2011), available at http://saas.usc.edu/files/2012/08/NCAA-2011-12-Manual.pdf, archived at http://perma.cc/RM99- B6V7. 31. Id. § 12.1.2(a). 32. Id. § 12.5.2.1. 33. Id. 34. Id. §§ 12.1.2, 12.5.2.1. 35. See Keller v. Electr. Arts, Inc., 724 F.3d 1268 (9th Cir. 2013); Hart v. Electr. Arts, Inc., 717 F.3d 141 (3d Cir. 2013). 36. See Keller, 724 F.3d at 1268; Hart, 717 F.3d at 141. 37. This Author’s symbiotic term for these plaintiffs as “New Age Athletes” reflects the growing entrepreneurial enlightenment of professional athletes, who in this context seek to exploit their own name and likeness rather than acquiesce to exploitation by others. A broader definition, including social entrepreneurial acumen is contained in Roger M. Groves, New Age Athletes as Social Entrepreneurs: Proposing A Philanthropic Paradigm Shift And Creative Use of Limited 374 INDIANA LAW REVIEW [Vol. 48:369 II. THE ATHLETE’S PUBLICITY RIGHTS DEFINED While this Article is confined to publicity rights of college athletes, perhaps the most comprehensive and binding definition of publicity rights is found in professional sports—the National Football League (“NFL”). A collective bargaining agreement is executed between the NFL and the players through their collective bargaining agent, the NFL Players Association (“NFLPA”).38 The description of those rights is most succinctly stated in the following provision: Player hereby grants to Club and League . . . the right and authority to use . . . his name, nickname, initials, likeness, image, picture, photograph, animation, persona, autograph/signature (including facsimiles thereof), voice, biographical information and/or any and all other identifying characteristics (collectively, ‘Publicity Rights’) . . . .39 There are no apparent loopholes or gaps in what comprises publicity rights.40 In Hart, the defendants acknowledged that the athletes had those rights without attempting to carve out exceptions.41 Rather, the defendants argued that those rights were unenforceable because the defendants’ First Amendment rights override the athletes’ publicity rights.42 Celebrities, including athletes, have publicity rights as a matter of common law and statutory law in some states.43 Such rights however are subject to limitations.44 Courts have recognized that in the public interest, people may lawfully appropriate and publish the image of another so long as the publication reports factual information, newsworthy items, or subjects within the public’s interest.45 Such limitations affect collegiate and professional athlete’s rights to publicity as they are typically the subject of newsworthy items and factual information contained in the statistics of the player or team.46 This Article, however, involves far more than factual news items. In Keller, Liability Company Joint Ventures, 11 WAKE FOREST J. BUS. & INTELL. PROP. L. 213 (2011). 38. NATIONAL FOOTBALL LEAGUE, COLLECTIVE BARGAINING AGREEMENT 5 (2011), available at https://nfllabor.files.wordpress.com/2010/01/collective-bargaining-agreement-2011- 2020.pdf, archived at http://perma.cc/K4KN-7EQV. 39. Id. at 256. 40. See id. 41. Hart v. Electr. Arts, Inc., 717 F.3d 141, 148 (3d Cir. 2013). 42. Id. at 148-49. The Court in Keller stated the publicity claims in both cases were “materially identical” and that Keller is merely “incarnated in California” as opposed to the Third Circuit battle of Hart in New Jersey. See Keller v. Electr. Arts, Inc., 724 F.3d 1268, 1277 (9th Cir. 2013). 43. See Hart, 717 F.3d at 150. 44. Id. 45. See Keller, 724 F.3d at 1283. 46. Id. 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 375 the court resoundingly rejected the newsworthy exception defense by the defendant users of the athlete’s NIL.47 Rather than a means of reporting “factual data” about a real event, the court characterized the work as merely a means of glorifying its own virtual games.48 The break with factual reporting of reality was most evident from the video game’s omission of Keller’s name.49 In the court’s words: EA can hardly be considered to be “reporting” on Keller’s career at Arizona State and Nebraska when it is not even using Keller’s name in connection with his avatar in the game. Put simply, EA’s interactive game is not a publication of facts about college football; it is a game, not a reference source. These state law defenses, therefore, do not apply.50 The foregoing leads this author to conclude that publicity rights as applied to collegiate athletes includes the broad based definition already used in the NFL. Similarly, the athletes’ rights are limited by the newsworthy carve out in the public interest. III. FOUNDATIONAL UNDERPINNING FOR ATHLETIC IP PROTECTIONS A. Right of Privacy The legal authority of one person to prevent others from using his or her NIL dates back to at least 1905.51 The issue concerned unauthorized advertisements.52 The legal theory was based on a right of privacy.53 The privacy right is intuitively difficult for athletes since they voluntarily seek publicity, not privacy, and profit from the very public celebrity status. So when the college football program of Texas Christian University, in 1939, included the picture of All-American quarterback Davey O’Brien, the Court was not sympathetic to his attempt to enjoin the defendant from using his NIL for profit.54 The Court noted that O’Brien had given prior authorizations of his picture to many other media and individuals.55 In the Court’s view, that left him with no independent legal right to the commercial value of his name and likeness.56 In particular, O’Brien 47. Id. at 1283-84. 48. See id. at 1283. 49. Id. 50. Id. 51. Pavesich v. New England Life Ins., 50 S.E. 68, 68 (Ga. 1905). 52. See id. at 68. 53. Id. at 69. 54. O’Brien v. Pabst Sales, 124 F.2d 167, 169 (1941). 55. Id. at 169-70. 56. Id. at 170; see also Hanna Mfg. v. Hillerich & Bradsby, 78 F.2d 763, 768 (1935) (holding a manufacturer’s patent invalid because it violated a competitor’s property rights by using the names of athletes who contracted with the competitor). 376 INDIANA LAW REVIEW [Vol. 48:369 objected to his picture alongside a glass of Pabst Blue Ribbon beer.57 O’Brien is instructive in understanding the detrimental legal effect of having the publicity right grounded in the right of privacy rather than a property right. If the publicity rights emanate from “privacy,” that right is more easily defeated by a voluntary exposure to others. Obviously, the privacy is eliminated on the celebrity’s own volition. That was the predictable result in O’Brien. But a right based on property rights should not hinge on exposure to the public any more than letting the public know you are a homeowner. The exposure to the public of that property does not authorize the public or anyone else to use it without your permission.58 Thus, the publicity rights asserted by the athletes in Hart and Keller have grounded their claims in their property interests.59 The economic value of those rights include preventing the unauthorized use and appropriation of those NIL rights. In theory, one has the exclusive right to the fruits of his image for they are the result of his own efforts.60 Tort law seeks to protect one’s economic interest in his image by recognizing that one has the right in his own publicity, which provides a remedy for misappropriation of one’s image.61 The plaintiff athletes therefore have a stronger basis to prevail over First Amendment advocates with the property rights basis than the privacy basis noted in prior iterations of publicity rights. B. Publicity Rights as a Property Interest Several cases have now established that a right of publicity is available to celebrities, including athletes, quite apart from a right of privacy.62 The court in Hart relied upon the Restatement (Second) of Torts, particularly the comment that “the right created by the rule [liability for appropriating the name or likeness of another] is in the nature of a property right.”63 In doing so, the court reached back to a 1907 case, where a New Jersey court enjoined a company from using the name and likeness of Thomas Edison to promote its products.64 The court rhetorically asked why the “peculiar cast of one’s features is not one’s property . . . rather than to the person seeking to make an unauthorized use of it.”65 The court in Hart took that implicitly to mean the Edison court recognized that “an individual enjoyed a property interest in his or her identity.”66 The distinction between basing publicity rights in property rather than 57. O’Brien, 124 F.2d at 168. 58. Hart v. Electr. Arts, Inc., 717 F.3d 141, 150 (3d Cir. 2013). 59. See id. at 150-51; see also Keller v. Electr. Arts, Inc., 724 F.3d 1268, 1283 (9th Cir. 2013). 60. Hart, 717 F.3d at 151. 61. Id. 62. See id. at 150 (providing a history of the right to privacy). 63. Id. at 151 (citing RESTATEMENT SECOND OF TORTS § 652C (1977)). 64. Edison v. Edison Polyform Mfg. Co., 73 N.J. Eq. 136, 140 (N.J. Ch. 1907). 65. Id. at 141. 66. Hart, 717 F.3d at 150. 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 377 privacy is important because as discussed above, a privacy right can easily give way to other types of protected interests when the privacy claimant thrusts himself or herself into the public eye. As stated in Hart, “a man in public life may not claim the same immunity from publicity that a private citizen may.”67 A property interest, in this author’s view, has a greater sense of permanence since it does not change when it is exposed to the public. For example, a house does not lose its character as property just because it is exposed to the public voluntarily by its owner. The court in Hart cited a seminal case by the highest court in New Jersey for this same proposition—that the right to one name and reputation has immutable characteristics, and those characteristics are not altered just because they are publicized.68 “It is unfair that one should be permitted to commercialize or exploit or capitalize upon another’s name, reputation or accomplishments merely because the owner’s accomplishments have been highly publicized.”69 The casting of the publicity right as a property interest is corroborated by the Hart court’s use of the term “misappropriation” of the right of publicity, much like misappropriation of other forms of real and personal property, or the misappropriation of corporate assets, which is comprised of both tangible and intangible property.70 The property right underpinning is also valuable because it is a more direct tie to economic rights. The United States Supreme Court highlighted that point in the case discussed immediately below. C. Case Law and Implications for Future Cases The cornerstone case for establishing the right of publicity was the United States Supreme Court decision in Zacchini v. Scripps-Howard Broadcasting.71 After nearly four decades, this case is still the only occasion where the nation’s highest court has directly addressed the First Amendment juxtaposed against the right of publicity.72 The purported owner of the publicity rights was not an athlete, but an entertainer.73 Hugo Zacchini was a “human cannonball” that was shot from a cannon into a net some 200 feet away at an Ohio state fair.74 A television reporter filmed the fifteen-second performance in its entirety.75 The 67. Id. (citing Edison, 73 N.J. Eq. at 140). 68. Id. at 151 (citing Palmer v. Schonhorn Enters., 232 A.2d 458, 462 (N.J. Super. Ct. Ch. Div. 1967)). 69. Palmer, 232 A.2d at 462. 70. Hart, 717 F.3d at 150. 71. See generally Zacchini v. Scripps-Howard Broad., 433 U.S. 562 (1977) (discussing the establishment of the right of publicity). 72. Id. at 567-68. 73. Id. at 563. 74. Id. 75. Id. at 564. 378 INDIANA LAW REVIEW [Vol. 48:369 show then aired on the local television station.76 The station claimed the airing was protected from publicity rights claims by its First Amendment rights.77 The Court held the entertainer’s right of publicity outweighed the broadcaster’s First Amendment rights.78 A key passage of the rationale is: “The broadcast of a film of petitioner’s entire act poses a substantial threat to the economic value of that performance . . . [T]his act is the product of [Zacchini’s] own talents and energy, the end result of much time, effort, and expense.”79 As will be discussed in greater detail below, federal courts have recently declared that college and former college athletes also have rights of publicity.80 They too have claimed that unauthorized users—the NCAA and video game manufacturers—have created near-identical usage of their performances.81 The other noteworthy component of Zacchini is the recognized importance of the publicity rights claimant’s “time, effort, and expense.”82 As the facts in Hart and Keller will reveal, the video game depiction of the athletes highlighted their skills, which like Zacchini, was a culmination of time and effort, and shared expense.83 The video games, like the television broadcast, are arguably a “substantial threat” to the “economic value” of the athletes’ performance, although not a complete airing of a game.84 The extent of the time, effort and expense has been recently exposed when Northwestern University football players asserted they were “employees” with a right to collectively bargain for their working conditions under the National Labor Relations Act.85 The players prevailed before the administrative law judge.86 The ALJ’s findings included the following facts: 1. The players spent 50-60 hours a week on their “football duties” during the one month training camp before the school year even started. 2. The players spent an additional 40-50 hours a week on football duties 76. Id. 77. Id. 78. Id. at 575. 79. Id. (emphasis added). 80. See Keller v. Electr. Arts, Inc., 724 F.3d 1268, 1273 (9th Cir. 2013); Hart v. Electr. Arts, Inc., 717 F.3d 141, 149 (3d Cir. 2013). 81. See Keller, 724 F.3d at 1272. 82. See Zacchini, 433 U.S. at 562 (discussing the establishment of the right of publicity). 83. Id. 84. Obviously the institution that provides a scholarship has an expense to help hone the skills of the athlete. There are no cases suggesting that the cost of a scholarship, however, nullifies the NIL of the athlete, or the fame that the fans bestow on that athlete. See id. (discussing that the broadcasting of a film poses a substantial threat to economic value). 85. Northwestern Univ. and Coll. Athletes Players Ass’n (CAPA), 13-RC-121359 N.L.R.B. 2 (2014), available at http://www.nlrb.gov/sites/default/files/attachments/basic-page/node- 1727/stylemanual.pdf, archived at http://perma.cc/VVW5-6F87. 86. The administrative law judge was Peter Sung Ohr, regional director of the National Labor Relations Board Region 13, in Chicago, Illinois. 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 379 during the 4-5 month football season. 3. The football hour commitments were “more hours than many undisputed full-time employees work at their jobs.” 4. The football duties involved “many more hours than the players spend on their studies.”87 During the training camp, for example, the student athletes did not attend classes.88 Even during the academic school year, players spent twenty hours a week in class, and over twice as much time engaging in football duties.89 Even incorporating study time for class, the administrative law judge was not convinced that academics were “primary.”90 The amount of time and training spent by college football players provides a compelling argument that they develop a skill level that brings great praise from the sports fan and fame to the athlete. Indeed, since the United States is the primary venue for this version of football, a reasonable contention is that these eighteen- to twenty-one-year-olds are the most skilled on the planet at their age. The uniqueness and scarcity of supply leads to the economic and rational conclusion that they are in high demand, with significant economic value associated with their evolved skill. This supports the notion that these college football and basketball players can reach celebrity status and valuable publicity rights on par or beyond those of the human cannonball in Zacchini. Conversely, a distinction can be made between an individual who is solely responsible for the entire performance, as was the human cannonball in Zacchini, and the college players in a team sport like football or basketball. A team sport involves a synergy among players. They all contribute to a product that is greater than the sum of the individual parts, much like the combined connectivity of electronics and cabling exceeds the value of those same items individually for a cable company. While there is compelling logic to that point, the distinction is more a matter of degree, not of kind. In both team sports and an individualized entertainer’s performance, the acts are still uniquely associated with talent and skill capable of being separated from the First Amendment rights of users of the celebrity. The litmus test really emanates from the public. Sports fans decide by their actions whether to single out someone for fanatical praise. They vote by their purchase of things such as jerseys with the name or number of a particular jersey. They vote by paying to attend or send their children to attend camps featuring that player. But for the NCAA rule, there could be many other occasions where the collegiate athlete uses his NIL for public consumption. The market for his celebrity then would decide individualized economic value from the publicity 87. Northwestern Univ. and Coll. Athletes Players Ass’n (CAPA), 13-RC-121359 N.L.R.B. 15-16 (2014), available at http://www.nlrb.gov/sites/default/files/attachments/basic-page/node- 1727/stylemanual.pdf, archived at http://perma.cc/VVW5-6F87. 88. Id. at 18. 89. Id. 90. Id. 380 INDIANA LAW REVIEW [Vol. 48:369 rights of that particular athlete. This would be the case regardless of whether the athlete had performed in the sport alone or with other athletes. Additionally, other difficult allocations are made in the professional world. It is now commonplace that highly paid executives and board members of major corporations develop formulas to determine bonuses for individuals based upon a rather nebulous, tenuous, and unscientific attribution of that individual’s value based on the synergistic contribution of the entire company workforce.91 There is more commonality than dissimilarity between a company workforce working together and a college team working together. They both depend on each other to reach a common goal unachievable alone. If allocations can be made for executives who are removed from the actual on-the-ground activity that generates the revenue, then allocations can be made for football and basketball athletes who are literally on the ground making the plays that generate the revenue. In any event, whether there is one actor or several to be parsed or allocated, there is sufficient sophistication to develop an appropriate model. The failure to have such a method currently in place is not a valid basis for completely eliminating a right of publicity, or subrogating it to the First Amendment rights of the user. Therefore, Zacchini has continuing applicability to current collegiate athlete’s rights, laying the foundation for the current cases discussed below. IV. CURRENT RIGHT OF PUBLICITY CASES INVOLVING COLLEGE ATHLETES The most recent articulation of this right of publicity test is stated in Keller.92 The stated elements are noted below: 1. the defendant’s use of the plaintiff’s identity; 2. the appropriation of plaintiff’s name or likeness to defendant’s advantage, commercially or otherwise; 3. lack of consent; and 4. a resultant injury.93 While there are now several federal cases and a few state statutes delineating publicity rights, there are several open questions regarding the application of this test. Each element in fact has issues that each attorney for opposing parties 91. Bonus or stock options often provide millions of dollars to executives based on stock options granted even though the executive was not directly responsible for generating the revenue. In many such cases, the sums were guaranteed even if there was no correlation with increased profitability. Indeed, stock options are defined as the receipt of stock “regardless of shifts in market value” of the stock price. BLACK’S LAW DICTIONARY 1459 (8th ed. 2004). 92. Keller v. Electr. Arts, Inc., 724 F.3d 1268, 1273 n.4 (9th Cir. 2013) (citing Stewart v. Rolling Stone LLC, 105 Cal. Rptr. 3d 98, 111 (Ct. App. 2010), as modified on denial of reh’g (Feb. 24, 2010). Such an important rule was probably relegated to a footnote because Hart was decided months before on the same issue, already provided the preliminary groundwork, and was decided and reviewed prior to publishing the opinion in Keller. See id. As noted infra, footnote 20, the majority in Keller viewed the publicity rights claims to be materially identical. See id. at 1273. 93. Stewart, 105 Cal. Rptr. 3d at 111. 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 381 should view as potentially contentious and still unresolved from the case law to date. Those issues include the following in the order of element presentation: 1. Use. Is there a “use” of the plaintiffs NIL when a trait of the celebrity is a relatively insignificant part of a larger collage, montage, movie, comic strip or other media presentation? As discussed below, the judges in both Hart and Keller were divided on the fundamental issue of whether other elements of creativity outside of the celebrity’s depiction should matter.94 The dissent in both cases viewed the other items of the work as a whole to be relevant.95 Those judges would have held that First Amendment rights of expression of the authors of the video games prevailed over the publicity rights of the athletes.96 That issue is a question not of fact but as a matter of law.97 Future claimants will undoubtedly renew that debate. 2. Commercial Exploitation or ‘Otherwise”. While in most cases it should be clear whether the creator of a work gained a commercial advantage, there is no case to date that explores the meaning of “or otherwise.” Could the creator of the work still be subject to an injunction against further production even if the project was used for nonprofit purposes? That too is an open question. 3. Consent. As of the writing of this Article, the O’Bannon opinion rejects the notion that the plaintiff college athletes impliedly consented to allowing NCAA member institutions to use their NILs.98 The O’Bannon complaint alleges that the NCAA, its licensing agent and video manufacturer Electronic Arts, Inc. have conspired to profit from the image of athletes without the consent of those athletes. 99 Assuming the NCAA appeals the ruling, the NCAA is playing defense in more ways than one. Scholars on this precise issue have cautioned, “The scope of such consent is necessarily determined on a case-by-case basis and requires careful contract drafting to delineate which rights are conferred.”100 4. Injury. Reported cases to date have not provided a plethora of guidance on damages, the methodology or calculations, or the parameters for the types of injury on which those calculations are made. To the extent misappropriation of a publicity right is found, defendant’s counsel 94. See Keller, 724 F.3d at 1285; Hart v. Electr. Arts, Inc., 717 F.3d 141, 171 (3d Cir. 2013). 95. Keller, 724 F.3d at 1285; Hart, 717 F.3d at 171. 96. Keller, 724 F.3d at 1289; Hart, 717 F.3d at 176. 97. Keller, 724 F.3d at 1274; Hart, 717 F.3d at 170. 98. O’Bannon v. Nat’l Collegiate Athletic Ass’n, C 09-1967 CW, 2010 WL 445190, at *2 (N.D. Cal. Feb. 8, 2010). 99. See id. at *4. 100. See MATTHEW J. MITTEN ET AL, SPORTS LAW AND REGULATION, CASES, MATERIALS, AND PROBLEMS 1036 (Wolters Kluwer, 3d ed. 2013); see also O’Brien v. Pabst Sales, 124 F.2d 167, 167 (1941). 382 INDIANA LAW REVIEW [Vol. 48:369 will no doubt explore the typical tort-based damage issues, e.g., whether the plaintiff’s quantification is speculative, missing important components, or lacking in causal connection to the work. A related open question is whether the courts will create a de minimis rule, where an injury below that is of a certain monetary threshold will be subrogated to the holder of First Amendment rights in a work using the NIL. Those unresolved questions aside, there is significant guidance “as a matter of law” from the above-noted 2013 cases of Hart and Keller. A. The Hart-Keller Factual Background In Hart, quarterback Ryan Hart became publicly recognizable for leading the Rutgers football team to rare levels of success, including its first Bowl game in seventeen years.101 He was also widely considered the most prolific on-field performer in school history holding many career records at Rutgers.102 Hart abided by his contractual obligation not to profit or take any direct or indirect remuneration from his own name.103 Yet Electronic Arts, Inc. (“EA”) produced a profitable video game entitled, NCAA Football.104 The video has annual editions.105 The 2006 edition included Hart as the high profile Rutgers quarterback in that rare Bowl game ending the prior season.106 There were several elements of the video game that included the NIL of Hart.107 The similarities between the video game Rutgers quarterback and the real Ryan Hart are noted below: 1. The video used the same jersey number (13) that Hart actually used at Rutgers. 2. The video player had the same height and weight as Hart (6’ 2”, 197 pounds). 3. Hart’s home state and town, and class year were identical on the video. 4. Hart had a characteristic left wristband and helmet visor. Both were used in the video, though a game user (“gamer”) could change those items. 5. The video included Hart throwing a pass in the bowl game that was from actual game film footage.108 101. Hart, 717 F.3d at 145-46. 102. Id. Hart led the Rutgers Scarlet Knights to the Insight Bowl. At the time the case was written, Hart held the records for career passing attempts and completions. Id. at 146 n.4. 103. Id. at 145. 104. Id. at 146. EA was noted by the court to be “one of the world’s leading interactive entertainment software companies.” Id. 105. Id. 106. Id. at 147. 107. Id. at 146. 108. Id. 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 383 The court described some of the above as “immutable” characteristics of Hart, e.g., physical characteristics, hometown, and class year.109 In Keller, plaintiff Samuel Keller was also a quarterback, albeit from Arizona State University and then Nebraska.110 He too was included in EA’s video game, NCAA Football.111 The factual similarities between the video game and the actual athlete parallel Hart.112 In that regard, the court made the following factual observation about the NCAA Football video game: EA seeks to replicate each school’s entire team as accurately as possible. Every real football player on each team . . . has a corresponding avatar in the game with the player’s actual jersey number and virtually identical height, weight, build, skin tone, hair color, and home state.113 As in Hart, EA’s video game allows the gamers to manipulate various aspects of the athletes in the game.114 But the court’s point of emphasis was about context.115 The focus was on the near identical replication in the video games of the way in which the athletes gained fame.116 The court in Keller summarized those points as follows: EA attempts to match any unique, highly identifiable playing behaviors by sending detailed questionnaires to team equipment managers. Additionally, EA creates realistic virtual versions of actual stadiums; populates them with the virtual athletes, coaches, cheerleaders, and fans realistically rendered by EA’s graphic artists; and incorporates realistic sounds such as the crunch of the players’ pads and the roar of the crowd.117 To be more precise, the court noted that the game’s 2005 version included Keller’s same “jersey number, height, weight, skin tone, hair color, hair style, handedness, home state, play style (pocket passer) visor preference, facial features, and school year.”118 Thus, both Hart and Keller involved EA video games that practically replicated both the plaintiff quarterbacks and the environment that brought them a relative level of notoriety.119 Having noted the factual similarities of the two 109. Id. 110. Keller v. Electr. Arts, Inc., 724 F.3d 1268, 1271 (9th Cir. 2013). 111. Id. at 1272. 112. Id. at 1278 (citing Hart, 717 F.3d at 163 n.28). 113. Id. at 1271. 114. Id. 115. Id. at 1275 (citing No Doubt v. Activision Publ’g, Inc., 122 Cal. Rptr. 3d 397, 411 (Cal. Ct. App. 2011)). 116. Id. at 1271. 117. Id. 118. Id. at 1272. 119. Id. at 1271-72; Hart v. Electr. Arts. Inc., 717 F.3d 141, 146, 166 (3d Cir. 2013). 384 INDIANA LAW REVIEW [Vol. 48:369 cases and the focus on how the video games in both cases replicated the actual athletes, the question then turns to what legal tests are appropriate to determine whether the athletes’ publicity rights trump the First Amendment rights of those who used the athletes’ NILs.120 The courts in both cases waged that war of principalities and discussed appropriate tests.121 This author analyzes those tests below. B. The Battle of Tests to Determine the Superiority of Publicity Rights vs. First Amendment Rights The Hart-Keller cases examined several tests to determine whether the publicity rights of athletes prevailed over the First Amendment rights of expression by users of the athletes’ names and likenesses.122 In Hart, the court followed prior case law and articulated three tests to be potentially used to decide the case: (1) the Predominant Use Test, (2) the Rogers Test, and (3) the Transformative Use Test (“TUT”).123 In Keller, the court mentioned four tests.124 But both courts found only two tests to be most fundamentally at odds and most relevant for decision making purposes—the Rogers Test and the TUT.125 The courts in both instances decided to use the TUT.126 While both courts rejected the Rogers Test,127 its consideration is still important. These are only two cases from the very recent lineage of the still burgeoning publicity rights jurisprudence as applied to collegiate athletes. There will most certainly be another examination of these tests. On review, the test selection may change. This Article will therefore examine the Rogers Test for substantive differences and the implications and impact if it is indeed selected instead of the TUT. 1. The Rogers Test.—Among the secondary tests (i.e., those other than the 120. Keller, 724 F.3d at 1273; Hart, 717 F.3d at 149. 121. Keller, 724 F.3d at 1273; Hart, 717 F.3d at 149. 122. In Keller, these tests were offered by the defendant user of the athletes’ likeness. Keller, 724 F.3d at 1273. The defendant's burden was to establish that the athlete would not have a reasonable probability of success on the merits of his publicity rights claim. Id. The probability of success was an element of proof under California's anti-SLAPP statute, which the defendant used as the legal basis for its motion to strike the complaint. Id. at 1272. Ironically, the Strategic Lawsuit Against Public Participation (“SLAPP”) statute is designed to protect the smaller, often public interest, entities against oppressive tactics by large companies. Id. In Keller, the anti- SLAPP statute was used by the Goliath entity against the individual of meager means, a former college quarterback, Samuel Keller. Id. 123. Hart, 717 F.3d at 153. 124. The tests as articulated in Keller were (1) the transformative use, (2) Rogers Test, (3) the “public interest” test, and (4) the “public affairs” exemption. Keller, 724 F.3d at 1273. 125. See id.; Hart, 717 F.3d at 153. 126. Hart, 717 F.3d at 164. 127. Keller, 724 F.3d at 1282; Hart, 717 F.3d at 155. 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 385 TUT), the Rogers Test was most discussed in both Hart and Keller.128 Between the two cases, Hart provided the more detailed analysis of this test.129 The reason is likely due to the fact that Hart was decided first.130 The Keller court made more of a summary reference to its secondary tests since it too chose the TUT.131 Keller therefore could afford the luxury of dismissing redundant discussion of the secondary tests, and focus or build upon the primary test, TUT.132 The Rogers Test primarily examines the relationship between the athlete’s NIL and the work “as a whole.”133 As will be discussed below, the court in Hart rejected that approach because it was overbroad.134 It would “immunize” the defendant users of an athlete’s NIL from a “broad swath of tortious activity.”135 The tortious activity is the misappropriation of an athlete’s NIL by those who use the NIL without the athlete’s authorization.136 The test is named after the case Rogers v. Grimaldi, where famous actress Ginger Rogers sued the producers of a film.137 The film was titled, Ginger and Fred.138 The plaintiff claimed the film’s title was an obvious reference and attempt to profit from the names and images of the famous performances of her on-screen dancing with Fred Astaire.139 The Rogers court’s rule was essentially that the celebrity can only prevail if he or she establishes that the use of the NIL was “wholly unrelated” to the movie or was “simply a disguised commercial advertisement for the sale of goods or services.”140 The court’s application of the rule resulted in a dismissal of the publicity rights claim.141 The court found the title was clearly related to the content of the movie and was not a veiled commercial advertisement for the sale of goods or services apart from the movie content.142 In Hart, the court observed that the defendant sought to expand the Rogers Test beyond the use of the NIL in its title.143 The defendant sought to compare the athlete’s NIL to the entire movie.144 In other words, Hart would have to 128. See Keller, 724 F.3d at 1279-82; Hart, 717 F.3d at 154-58. 129. See Hart, 717 F.3d at 154-58. 130. Both opinions were filed in 2013. The Hart opinion was filed in May 2013, while the Keller opinion was filed in July 2013. See Keller, 724 F.3d 1268; Hart, 717 F.3d 141. 131. See Keller, 724 F.3d at 1279-83. 132. Id. at 1274-76. 133. Hart, 717 F.3d at 154. 134. Id. at 155. 135. Id. 136. Id. at 157. 137. Rogers v. Grimaldi, 875 F.2d 994, 1004 (2d Cir. 1989). 138. Id. at 996. 139. Id. at 997. 140. Id. at 1004. 141. Id. at 996. 142. Id. at 1001. 143. Hart v. Electr. Arts, Inc., 717 F.3d 141, 155 (3d Cir. 2013). 144. Id. at 156. 386 INDIANA LAW REVIEW [Vol. 48:369 establish that his NIL was unrelated to any part of the work, not just the title.145 The defendant asserted then that since the video game was about college football it was related to the use of Hart’s NIL.146 Upon showing that relationship, the First Amendment rights prevail.147 The Hart court considered this rule extension a counterintuitive “blunt instrument” that was “unfit for widespread application.”148 In the court’s words, “It cannot be that the very activity by which Appellant achieved his renown now prevents him from protecting his hard-won celebrity. We decline to endorse such a conclusion and therefore reject the Rogers test as inapplicable.”149 Thus, Hart rejects a rule that only requires some level of relatedness between the NIL and any part of the work that appropriates the NIL.150 The Hart court stated that adopting that rule “threatens to turn the right of publicity on its head.”151 This author’s translation is that the rule would have the opposite of its intended effect. Instead of providing a basis for a celebrity to protect his or her own NIL, it would be used as a weapon (“blunt instrument”) to destroy the publicity rights sought to be protected.152 The defendant, as an unauthorized user of the NIL, would have too little to prove—i.e., only some relatedness within any part of the work.153 Therefore, the Rogers Test would do more to protect the defendant, the unintended beneficiary of publicity rights, than the plaintiff, the intended beneficiary of those rights.154 In this author’s view, the adoption of this rule would likely have a profound effect on the outcome of future cases. If a court need only find some level of relatedness between the celebrity’s NIL and the work, then the creator of the work would likely prevail in almost every instance. A maker of a video game, for example, would prevail whenever the game depicted the same sport as played by the athlete.155 Any actor would fail in court as did the graceful Ginger Rogers if the movie contained dancing scenes reminiscent of, or even imitations of Ms. Rogers.156 This author agrees with the Hart majority not only because of the over- breadth of the Rogers Test, but also because of the established notion that the battle between the publicity rights and First Amendment rights is inherently a “balancing test.”157 As stated in Hart, the court’s duty is to “balance the interests 145. Id. 146. Id. at 157. 147. Id. at 158. 148. Id. at 157. 149. Id. at 158. 150. See id. at 157-58. 151. Id. at 157. 152. Id. 153. Id. 154. Id. 155. See generally id. 156. See id. at 155. 157. See id. at 149, 152-53; see also Zacchini v. Scripps-Howard Broad. Co., 433 U.S. 562, 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 387 in protecting the relevant property right against the interest in free expression.”158 This was an issue of first impression in Hart, so the Court was well aware of the magnitude of its ruling.159 Simply put, the Rogers Test is too heavily weighted in favor of First Amendment claimants to be capable of a fair balancing of the interests,160 and defendants can too easily immunize themselves from publicity rights claimants.161 Accordingly, Hart correctly rejected the test.162 In Keller, the defendants recast the argument but it was fundamentally the same.163 2. The Work-as-a-Whole Defense.—Importantly, the common contention of the defendants in both cases is that even if the TUT is used, the work as a whole should be examined to determine the extent of creative predominance by the author of the work.164 In Keller, for example, the defendants and dissenting Judge Thomas claimed that the majority wrongly cast the issue.165 In the view of Judge Thomas, the court must view the work in a holistic sense—the entire work—not just the more narrow view of how the particular celebrity was depicted.166 In the words of Judge Thomas, The majority confines its inquiry to how a single athlete’s likeness is represented in the video game, rather than examining the transformative and creative elements in the video game as a whole . . . The salient question is whether the entire work is transformative, and whether the transformative elements predominate, rather than whether an individual persona or image has been altered.167 This author interprets the above passage to essentially mean this: The defendant’s work would be sufficiently transformative by simply creating other actions or depictions surrounding the athlete. A work would therefore be protected against publicity claims even if the exact likeness of the athlete was used. Judge Thomas emphatically noted that the video game in question had “thousands of virtual actors,” not just Sam Keller the quarterback.168 The work also included the “lifelike roar of the crowd” and the “crunch” of shoulder pads.169 Judge Thomas and the defendant game maker viewed these features as creative elements that should be taken into account in determining whether the 574-75 (1977). 158. Hart, 717 F.3d at 155. 159. See id. at 151-52. 160. See id. at 157-58. 161. See id. 162. Id. 163. See generally Keller v. Elec. Arts, Inc., 724 F.3d 1268 (9th Cir. 2013) (Thomas, J., dissenting). 164. See id. at 1276; Hart, 717 F.3d at 156-57. 165. See Keller, 724 F.3d at 1285. 166. See id. 167. Id. (emphasis added). 168. Id. at 1287. 169. Id. 388 INDIANA LAW REVIEW [Vol. 48:369 work was transformative.170 The majority ignored such elements as irrelevant to the central inquiry of whether there was a literal depiction of the athlete, regardless of the surrounding creative elements.171 A graphic illustration of this critical difference is depicted below. The Keller majority was only concerned with the narrow question of whether there was a literal depiction of the athlete.172 The dissent and the defendants claim the court should focus on the other creative elements within the entire work.173 Once described in this fashion, it is readily apparent that the framing of the issue is likely to foretell the legal conclusion. The work is far more likely to be transformative if the applicable rule allows various aspects of a video game to be included in the determination. A game maker could easily add scenery, actors, and scripts that go beyond the athlete’s name, image and likeness. In this author’s view, the difficulty of the dissent in Keller is no different than the majority discussed in Hart—the standard is so broad that a game maker would almost always prevail without a meritorious balancing of interests. Every video game or comic book can contain pictures, words, and sounds that have little to do with the athlete. If that were the only requirement for a transformative work, every work would qualify and the now time-honored right of publicity would become nearly a nullity. None of the recent publicity rights cases were willing to be responsible for the implicit death of such a right. This author considers Keller to be rightly decided. The issue was correctly cast as the narrow examination of the depiction of the athlete, not the ancillary elements of the work. Yet there is another element of the First Amendment claims that is only noticed on close examination of the cases. That element involves interactivity, discussed below. 170. Id. at 1285-87. 171. See id. at 1268 (Thomas, J., dissenting). 172. Id. 173. See id. at 1284-90. 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 389 3. The Related Element of Interactivity.—The dissent in Keller highlights the fact that the disputed video game allows consumers (gamers) the ability to manipulate the game.174 The descriptive phrase used by Circuit Judge Thomas is that the game is a “work of interactive historical fiction.”175 Gamers therefore can create their own game situations from a common base.176 The gamer can create a “virtual image” of himself and role play himself as a high school athlete trying out for a team, perhaps making the team, and participating under changing circumstances of his choosing for a college team in a position of his or her choosing.177 The gamer can also select to role-play as a coach of the college team.178 Judge Thomas then applied those facts to what he called the “proper holistic context.”179 That context is the very crux of the controversy between the majority and the dissent.180 The dissent is referencing what this author has termed the “As- A-Whole rule.” Judge Thomas viewed the game as containing many creative and transformative elements.”181 One comes to that conclusion, however, only if the creative and transformative elements are not focused on the plaintiff, but on the entire work. According to the majority, the only relevant comparison is between the name and likeness of the real athlete and the way he is depicted in the game, not how anyone or anything else is depicted.182 The fallacy of the gamer involvement justification for defeating publicity rights can also be exposed from a different perspective. Assuming all the above interactive opportunities, the question can be posed: How marketable would the work be without a strong athlete depiction? If sales are unaffected by the lack of the athlete, then the work’s creative elements were transformative because the market did not recognize the athlete’s NIL as the dominant cause for purchasing the work. If, on the other hand, sales would be virtually nonexistent without the athlete, then athlete’s NIL is dominant in the work. The work would not be transformative because the NIL was retained as primary. There would be no evidence of the work transforming into something else from a value standpoint. The question of whether a transformation occurred is a precursor to the discussion of the TUT, which follows in the next section of this Article. The question may then arise: How can we know whether a purchase of a video game was motivated primarily by the athlete depicted, or rather the creative elements of the video? This author’s response is that we can certainly tap into the substantial social science use of focus groups and surveys to provide empirical 174. See id. 175. Id. at 1285. 176. Id. 177. Id. 178. Id. at 1286. 179. Id. 180. Id. 181. Id. at 1284-87. 182. See id. at 1276. 390 INDIANA LAW REVIEW [Vol. 48:369 support for why consumers purchase a product.183 Predictive analytics is also an emerging area giving rise to increasingly admissible evidence in court proceedings.184 One possibility, then, is for a group of consumers to first view the work with the athlete as sold at retail. The same group can be shown a simulation of the same video but without the athlete or any of his or her NIL. Any variance would then be measured and quantified. Future courts may have to determine the admissibility of such proffered evidence which makes for an interesting component of publicity rights litigation going forward. 4. The Transformative Use Test.—The TUT was first articulated by the Supreme Court of California in Comedy III Productions, Inc. v. Gary Saderup, Inc. (“Comedy III”).185 The oft-quoted definition is below: . . . the central purpose of the inquiry . . . is to see . . . whether the new work merely ‘supercede[s] the objects’ of the original creation, or instead adds something new, with a further purpose or different meaning, or message; it asks, in other words, whether and to what extent the new work is ‘transformative.’”186 The rationale for the overall superiority of TUT was most cogently stated in Hart: In our view, the [TUT] appears to strike the best balance because it provides a flexible—yet uniformly applicable—analytical framework . . . Unlike the Rogers Test, the [TUT] maintains a singular focus on whether the work sufficiently transforms the celebrity’s identity or likeness, thereby allowing courts to account for the fact that misappropriation can occur in any market segment, including those related to the celebrity.187 This passage articulates a strong basis for the selection of the TUT in this author’s view. The court quite obviously has a legitimate interest in selecting a rule that can be flexible enough to apply to varying fact patterns, while still having practical workability as a test. There have been various contexts where the court did not select a test because it did not see how it could practically be applied despite some theoretical favorability.188 183. Focus groups typically involve carefully selected groups of between four to fifteen consumers or experts that interactively examine a topic through a moderator, typically either for market research or problem solving. The results are scientifically coalesced into findings that help analyze that topic. See Focus Group, BUSINESSDICTIONARY.COM, http://www.businessdictionary. com/definition/focus-group.html (last visited July 12, 2014), archived at http://perma.cc/GW9K- DXGW. 184. Predictive Analytics will be discussed infra Part VII. 185. Comedy III Prod., Inc. v. Gary Saderup, Inc., 21 P.3d 797 (Cal. 2001). 186. Id. at 808 (quoting Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 579 (1994)) (internal quotation marks omitted). 187. Hart v. Electr. Arts, Inc., 717 F.3d 141, 163 (3d Cir. 2013). 188. One such example is where courts have faced various tests for allocating business income among taxing jurisdictions for companies that have multistate business activity. The tests that were 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 391 In both Hart and Keller, the Court adopted the TUT to determine whether the publicity rights of an athlete are prioritized over the First Amendment rights of the user of the athlete’s NIL.189 This test is simply focused on whether the work in question adds significant creative elements so as to be transformed into something more than a mere celebrity likeness or imitation.190 Both cases held there was no transformative use and the First Amendment of the works’ creators cannot shield the defendants from the privacy rights claims as a matter of law.191 The rationale for this conclusion was multi-layered, yet it relies upon a fundamental principle: The more transformative the creative elements of the user, the less likely there will be an interference with the athlete’s publicity rights.192 The lack of a bright line difference is why this author views the TUT as a continuum. That discussion follows. V. ELEMENTS AND CONTINUUM OF FACTORS IN THE TUT As a point of emphasis, the TUT is a balancing act, juxtaposing the athlete’s publicity rights and the creator’s rights in a “work” (e.g., video game, comic book or movie) that uses the NIL of the athlete.193 Accordingly, the gravamen of the court’s analysis in Keller and Hart involves a comparison of those interests.194 The TUT has several layers of complexity because of its various articulations of elements. We start anew with the fundamental question of whether the work adds significant creative elements so as to transform that work into something more than a mere recreation or imitation of the celebrity’s likeness.195 If a work, like the video games in Hart and Keller, was found to have those “significant creative elements,” then it would be sufficiently transformative and the author could successfully assert the First Amendment defense.196 If, however, there were insufficient creative elements so that there was nothing more than a “mere” likeness or imitation of the celebrity, then the celebrity would prevail, and the user would be liable for damages or an injunction too difficult to apply were rejected. For example, a single factor formula using sales to determine a state tax liability gave way to the use of an apportionment formula that incorporated property, payroll, and sales as a more workable means of fairly reflecting the extent of income attributable to a state. See Complete Auto Transit v. Brady. 430 U.S. 274 (1977). 189. Keller v. Electr. Arts, Inc., 724 F.3d 1268, 1280 (9th Cir. 2013); Hart, 717 F.3d at 163- 65. A companion article will separately examine the legal basis for determining the value of those publicity rights. That is to say, once it is firmly determined that the publicity rights can prevail, the next question is what value is ascribed to those rights. The next article is devoted to that yet under- explored cutting edge issue of publicity rights valuation. 190. See Hart, 717 F.3d at 159. 191. See generally id.; Keller, 724 F.3d at 1268. 192. See generally Keller, 724 F.3d at 1268; Hart, 717 F.3d at 141. 193. See Hart, 717 F.3d at 159. 194. See generally id.; Keller, 724 F.3d at 1268. 195. See Hart, 717 F.3d at 159. 196. See generally id. 392 INDIANA LAW REVIEW [Vol. 48:369 preventing unauthorized profits or exploitation of the name and likeness of the celebrity.197 As applied in Hart and Keller, the celebrity is an athlete whose NIL is used by the NCAA and its partner, EA, in producing video games that include depictions of those athletes.198 That basic formulation of the TUT also reveals what could be considered the polar ends of a continuum. In this author’s view, the above rule contemplates a continuum because a work may have a widely varying range of creative elements. Only when those elements reach a certain and seemingly indefinable level is there a transformation into “something more than the mere celebrity likeness or imitation.”199 One end of the continuum is clear—the celebrity prevails when the user creates a work that is a “mere” imitation or literal replica of the celebrity. The circumstances under which the work’s creator can prevail are far less clear. The TUT is satisfied when “something” more than the imitation is created by the user of the celebrity’s name and likeness.200 No case to date has explained the beginning and end points. If the analysis incorporates only the rule statement, the continuum is depicted below: The common sense underlying the continuum is that the more transformative the creative elements, the more something else is created and the less likely to be interference with the economic interests of the celebrity athlete.201 In the attempt to make the TUT a truly workable test, the court in Keller incorporated the five- factor analysis endorsed by the California Supreme Court in Comedy III.202 With that context and purpose, this author summarizes the five factors as follows: 197. See generally Keller, 724 F.3d at 1268; Hart, 717 F.3d at141. 198. See generally Keller, 724 F.3d at 1268; Hart, 717 F.3d at 141. 199. See generally Hart, 717 F.3d at 159; Keller, 724 F.3d at 1268. 200. See Hart, 717 F.3d at 159. 201. See generally id.; Keller, 724 F.3d at 1268. 202. Keller, 724 F.3d at 1274 (citing Comedy III Prod., Inc. v. Gary Saderup, Inc., 21 P.3d 797 (Cal. 2001)). 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 393 1. “Raw Materials” vs. “Sum and Substance”: If the work uses the celebrity’s name and likeness only as the raw materials of the work which is “synthesized” into something else, the work is more likely transformative and protected from the celebrity’s publicity claims. If, conversely, the name and likeness is the sum and substance of the work, the transformation is not present, and the celebrity would likely prevail.203 2. “Defendant’s Own Expression”: The work is transformative and protected from the celebrity’s publicity rights claims if it is primarily the defendant user’s own expression and “something other than the likeness of the celebrity.”204 3. Quantitative Dominance: The work is transformative and protected from the celebrity’s publicity rights claims if the creator’s own creative efforts quantitatively predominate in the work.205 The court cautioned that a qualitative judgment is inappropriate.206 Presumably then, the Court would not analyze whether a movie, music CD or video game depicted the celebrity athlete as the most memorable even if the appearance time was significantly less than other characters. A purely quantitative review would measure dominance based on actual time spent depicting the celebrity. 4. Marketability and Economic Value: The work is not transformative and not protected if the marketability and economic value of the work is predominately from the celebrity.207 5. Subrogation of User’s Skills: The work is not transformative and not protected if the “skill and talent” of the work’s author is subordinated to the “overall goal of creating a conventional portrait of a celebrity so as to commercially exploit his or her name.”208 It is readily apparent that these factors do not create a bright line test. Not surprisingly, the court did not expressly state that any one factor or factors are worthy of greater weight than other factors.209 The recurring question with most of these factors is whether some facts and circumstances “predominate” over competing facts.210 The underlying inquiry is also clear: Are the immutable aspects of the celebrity the primary source for the work? The factors view the facts quantitatively, or in marketability or in the related economic value for the work.211 And to the same conclusion, if the characteristics of the celebrity are 203. Id. 204. Id. 205. Id. 206. Id. 207. Id. 208. Id. 209. See generally id. 210. See id. 211. See id. 394 INDIANA LAW REVIEW [Vol. 48:369 subrogated to the creative efforts of the creator of the work, a transformation away from the name and likeness of the celebrity is likely to have occurred. A. Unanswered Questions of Fact Finding, Intent and Burden The intent of the work’s author is also worth examination, albeit implicitly. Factor 5 requires analysis of the overall goal of the author of the work. If the goal was to prioritize the celebrity’s NIL, to his or her subrogation of skill and talent, there is no resultant transformation. The celebrity’s publicity rights claim would prevail. Yet if traditional rules of civil litigation apply, the burden to establish the user’s intent is on the plaintiff, the athlete or celebrity that claims infringement.212 Hart and Keller have not addressed intent and the related burden as a primary issue.213 Thus the question is unresolved from these most recent cases. Curiously, those same cases favor the athlete/celebrity.214 So if the burden was on the athlete, the burden was apparently met. Nonetheless, intent is difficult to prove. The plaintiff is in a more difficult battle if he or she must establish that the user of his NIL intended to make a “conventional portrait” of him, and that the intent included a desire to “commercially exploit” his or her name.215 The athlete’s quite plausible rejoinder could be that intent is not required. Rather, one can ascertain whether the work was a conventional portrait from an objective analysis of the facts. That is fundamentally the same inquiry into whether the work was designed to commercially exploit the NIL. Indeed, the Keller court’s analysis of factors indicates they had no problem making factual findings on their own without interposing an intermediate burden on the athlete to establish the intent of the defendant. Courts should also be more comfortable with a standard that relies more on objective factual analysis than the inherently amorphous task of ascertaining one’s intent. We must leave this question for future litigants to argue and future courts to decide. This author views greater judicial certainty from Hart’s rejection of an earlier and oft-overlooked case. In Matthews v. Wozencraft, the court accepted the First Amendment defense of the author of a fictional book that incorporated events of a narcotics officer.216 The court held for the author because there was no actual malice.217 Hart barely discussed the case but rejected it implicitly through its non-adoption.218 If the plaintiff had to establish actual malice by the First Amendment claimant, the publicity rights claimant would have a significantly heavy burden. The lack of that element of proof may implicitly be used to state that no malice is required—a comforting argument for the publicity rights 212. See generally Hart v. Electr. Arts, Inc., 717 F.3d 141 (3rd Cir. 2013). 213. See Keller, 724 F.3d at 1268; see also Hart, 717 F.3d at 141. 214. Keller, 724 F.3d at 1268; Hart, 717 F.3d at 141. 215. Comedy III Prod., Inc. v. Gary Saderup, Inc., 21 P.3d 797, 810 (Cal. 2001). 216. Matthews v. Wozencraft, 15 F.3d 432, 439 (5th Cir. 1994). 217. Id. at 440. 218. Hart, 717 F.3d at 173. 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 395 claimant. B. Application of the Five-Factor Test The court in Keller adeptly and concisely reviewed four cases where California courts applied the TUT.219 As noted above, none of the factors were explicitly declared to be primary over any of the other factors.220 It appears, rather, that in both Hart and Keller the facts are more critical to the holdings than the law of publicity rights or First Amendment rights.221 Under the TUT, factual findings as to the extent to which the work mimics the celebrity or is a literal depiction of the celebrity resolves the matter.222 Comedy III is particularly relevant to the Keller court’s analysis of Factor Five.223 This author entitled that factor: Subrogation of User’s Skills. The “work” at issue in Comedy III was comprised of T-shirts and lithographs of the comedic Three Stooges.224 The court opined that the author’s skill in creating the work was subordinated to the overall goal of creating “literal, conventional depictions” of the Three Stooges to “exploit their fame.”225 Whether a work is a literal or conventional depiction is a factual finding.226 Whether an artist’s skill is “subrogated” to those depictions is even more difficult to determine.227 The Quantitative Dominance factor cautions against using qualitative judgments about the skill or talent of the author (i.e., how effective or convincing or aesthetically pleasing or powerful the work).228 So the court must have the discipline to examine only the amount, the quantity, of creativity infused into the work compared to the extent of literal depictions of the celebrity.229 That too, this author opines, is a factual finding. If so, then the whole case can turn on the facts of a particular case. In Comedy III, the court must have been convinced that the creators of the T-shirts and lithographs added very little that could be considered artistic (regardless of whether it was “good” art).230 The work, therefore, was not transformed into anything beyond the literal 219. See Keller, 724 F.3d at 1268 (reviewing No Doubt v. Activision Publ’g, Inc., 192 Cal. App. 4th 1018 (Cal. Ct. App. 2011); Kirby v. Sega of America, Inc., 144 Cal. App. 4th 47 (Cal. Ct. App. 2006); Winter v. DC Comics, 69 P.3d 473 (Cal. 2003); Comedy III Prods. v. Gary Saderup, Inc., 21 P.3d 797 (Cal. 2001)). 220. Id. at 1274. 221. See id. at 1268; see also Hart, 717 F.3d at 141. 222. Comedy III Prod., 21 P.3d at 808 (citing Zacchini v. Scripps-Howard Broad. Co., 433 U.S. 562 (1977)). 223. Id. at 811. 224. Id. at 800-01. 225. Id. at 811. 226. Id. 227. Keller v. Electr. Arts, Inc., 724 F.3d 1268, 1274 (9th Cir. 2013). 228. Id. 229. Id. 230. Comedy III Prod., 21 P.3d at 810-11. 396 INDIANA LAW REVIEW [Vol. 48:369 depiction of the Three Stooges.231 The other cases analyzed in Keller also evidence fact-driven analysis and holdings.232 Two of the cases held for the author of works that used the name and/or likeness of the celebrity.233 Those opinions too are fact driven and logically and legally consistent.234 More importantly, additional clarity can be gleaned from the cases about which facts lead to the holding. The two cases are analyzed below. The first case is Winter v. DC Comics where the court found comic books to be transformative and thus protected from publicity rights claims.235 The court determined that the renditions of the celebrity rock stars, Johnny and Edgar Winter, were distortions of the celebrities, intending to “lampoon, parody, or caricature” those artists.236 The Keller court characterized the work as one made “for the purpose of parody.”237 Again, whether an item is a parody or caricature is a conclusion not based on a pure rule of law, but a factual finding or no less than a mixed question of law and fact. A declaration of the artist’s intent from the court is similarly a finding of the court. Future litigants will likely have extensive discovery on the question of intent. This seems to favor the user of the name and likeness. If the work appears to greatly exaggerate the features of the celebrity, the artist can easily corroborate the obvious depiction with a statement of intent to create a caricature and/or a parody of the artist. The legal symmetry is that the more a depiction is exaggerated and differs from the literal depiction, the more likely the artist has independent creative input to transform the work into something else. The celebrity has fewer teeth to the claim of an appropriation of his likeness if the work is less like the celebrity. Accordingly, these two cases are legally consistent even if the holdings differ. In Winter, the Court ruled in favor of the artists even though they used the same first names and only changed the last name “Winter” to “Autumn.”238 Appropriately, the court looked beyond the simple question of whether the name was used.239 So no athlete or celebrity should feel assured of prevailing by simply showing that someone else used his or her name without authority to do so. The facts mattered most again in Kirby v. Sega of America, Inc.240 The court held that the video game was transformative, though it used a singer’s “signature” 231. Id. at 811. 232. See Keller, 724 F.3d at 1268 (reviewing No Doubt v. Activision Publ’g, Inc., 192 Cal. App. 4th 1018 (Cal. Ct. App. 2011); Kirby v. Sega of Am., Inc., 144 Cal. App. 4th 47 (Cal. Ct. App. 2006); Winter v. DC Comics, 69 P.3d 473 (Cal. 2003); Comedy III Prods., 21 P.3d at 797)). 233. See Comedy III Prod., 21 P.3d at 797; see also Winter, 69 P.3d at 473. 234. Id. 235. Winter, 69 P.3d at 473. 236. Id. at 479. 237. Keller, 724 F.3d at 1275 (citing Winter, 69 P.3d at 473). 238. Winter, 69 P.3d at 476. 239. Id. at 479. 240. Kirby v. Sega of Am., Inc., 144 Cal. App. 4th 47, 50-53 (Cal. Ct. App. 2006). 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 397 lyrical expression “ooh la la.”241 The court found that the hairstyle, costume, dance moves and the principal character’s role as a “space-age reporter” were not literal depictions of the singer.242 In fact, the court stated those creative elements were “unlike any public depiction” of the actual celebrity.243 The court characterized the primary character as “fanciful.”244 The question arises then, how is a court to determine whether a disputed character is sufficiently “fanciful” to be beyond the literal depiction of the athlete or celebrity?245 If Kirby is guidance, then the court should perform a comparative analysis. The elements of comparison include not only hairstyle, as noted in the case. All other elements of appearance should be equally fair game for analysis. The creator of the work who is attempting to avoid publicity claims should create significant variances in clothing, shoes, skin tones, nails, ears, eyelashes, and the like. Variances in movements, running style, and walking patterns also arguably evidence the requisite creativity. C. The “Same Activity and Context” Test The Court in Keller found one case to be most factually aligned with case at issue—No Doubt v. Activision Publishing, Inc.246 True to the fact-sensitive analysis of the other cases, the court found that case most persuasive in holding that Keller’s publicity rights were infringed.247 In No Doubt, the work in question was a video game that allowed customers to select avatars, including one entitled, Band Hero, depicting vocals by the rock band No Doubt.248 The Court held the use of the band was no more than “literal recreations of the band members” within the same activity for which they became famous.249 While there were various legal aspects to the Keller decision, the court appears most influenced by certain factual similarities between the instant case and No Doubt.250 The following passage best codifies the Court’s comparative analysis: “Here, as in [the video game], users manipulate the characters in the performance of the same activity for which they are known in real life—playing football in this case, and performing in a rock band in [the rock band video 241. Id. at 59. 242. Id. 243. Id. (emphasis added). 244. Id. at 61. 245. Id. 246. Keller v. Electr. Arts, Inc., 724 F.3d 1268, 1276-1277 (9th Cir. 2013). 247. Id. at 1287-89. 248. No Doubt v. Activision Publ’g, Inc., 192 Cal. App. 4th 1018, 1023 (Cal. Ct. App. 2011); Kyle Kasper, Virtual Reality Headset, VIRTUALREALITYHEADSET.COM, available at http://www. virtualrealityheadset.com (last visited July 12, 2014), archived at http://perma.cc/SKZ7-L3S6 (Avatars are essentially virtual reality headsets and goggles used to show an individual user a simulation or virtual reality presentation). 249. No Doubt, 192 Cal. App. 4th at 1034. 250. Keller, 724 F.3d at 1018. 398 INDIANA LAW REVIEW [Vol. 48:369 game].”251 The Keller court then immediately added a companion factor: “The context in which the activity occurs is also similarly realistic—real venues in [the rock band video game] and realistic depictions of actual football stadiums in [the video game in Keller].”252 The court concluded: “As the district court found, Keller is represented as ‘what he was: the starting quarterback for Arizona State’ and Nebraska, and ‘the game’s setting is identical to where the public found [Keller] during his collegiate career: on the football field.’”253 This author views the above language as establishing a rule with duality. Both the same “activity” and the same “context” appear interrelated if not in fact required for the rule to have vitality and accomplish its intended purpose of clarifying when a literal depiction exists. It should now be clear, for example, that a video game manufacturer cannot claim protection under the First Amendment if it makes an unauthorized rendition of a college player that both replicates the actual person and also depicts him performing in the sport that provided his notoriety. VI. FUTURE APPLICATIONS Despite the existence of these various factors, unanswered questions remain. This rule would not seem to prevent the same video game manufacturer from using the likeness of an athlete performing some sport other than the one that he gained publicity—say trying his hand at baseball instead of football. What if the virtual game was performed in a baseball stadium? What if the venue was not a stadium at all, but the back yard of his parent’s home or a beach as part of a spring break fun-fest or on the planet Mars as selected by the video manufacturer? In each of the above scenarios the ultimate issue is the same—what is the extent of causal connection between the athlete’s name and likeness with the activity and context that the work depicting that athlete. The answer to those questions turn then on the dual analysis of what the author terms as the same activity and the same context. Those elements are discussed below. A. The Same Activity Prong The “same activity” prong is based on a fundamental but flexible notion.254 The more remote the work’s activity from the activity that made the athlete noteworthy, the more likely the court will find that work transformative. If an athlete gained his public persona from tennis, a video of him playing cards or some other activity not tied to athleticism may cause a court to find the game transformative. The opposite result would flow if the activity was racquetball, a sport more closely aligned with tennis. If the activity is not literally the same, but related, the court would have to decide whether the rule becomes the “same or substantially the same” activity or commonly identifiable with the activity or 251. Id. at 1276. 252. Id. 253. Id. 254. No Doubt, 192 Cal. App. 4th at 1018. 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 399 sport that gave rise to his or her notoriety under Keller.255 Future courts may also decide to define “activity” with greater particularity.256 A narrow definition would favor the creator of the work because that would establish many other activities as outside of the realm of sameness. If same literally meant the exact same activity, then in the above hypothetical racquetball would be considered different than tennis. A broad definition favors the celebrity/athlete because it would broaden the scope of what he can call his own. He could claim the activity of tennis and racquetball is using much of the same skill sets (e.g., hand eye coordination and arm movements) as tennis. Accordingly, an appropriation of his tennis activity would be equally an appropriation when the similar sport is the prop for using his NIL. B. The Same Context Prong As to the “same context” prong, isn’t a court more likely to find a work transformative if the athlete performs on a barren Mars surface than if the game was in a stadium?257 The author posits a “Yes” answer. The alternate planet venue was the creative element of the maker of the work. The extent of connection between the work and what brought the celebrity fame—a typical earthly stadium—would then be too far removed from the context for the game. In essence, the court is more likely to find that the game maker demonstrated a sufficient amount of creativity because the context of the work was transformative, not a literal depiction of the athlete’s persona in the same context that brought him his relative fame. If, however, the work uses a typical stadium, it is just a depiction of the preexisting context for the athlete’s performance. Even if the stadium was used for soccer instead of say football, they are substantially similar. There is little creativity demonstrated from the work’s author. Accordingly, context appears to be a potential factor in the court’s determination of whether there was a transformative work. The chart below depicts future scenarios where the context prong may become a defense by users of an athlete’s NIL against publicity rights infringement claims of athletes. The illustrations are from least to most likely settings favorable to the NIL users. 255. Keller, 724 F.3d at 1276-78. 256. No Doubt, 192 Cal. App. 4th at 1018. 257. Id. 400 INDIANA LAW REVIEW [Vol. 48:369 The corroborating logic to these scenarios in relation to the publicity rights rules to date is that the situations listed as “more likely” are also the situations that evidence more creativity by the NIL user, and thus more transformative. As stated in Keller, the truly transformative work is “‘also less likely to interfere with the economic interest protected by the right of publicity.’”258 The theory is that the athlete or celebrity has a lesser appropriation claim when the creativity of others is more dominant than his NIL.259 Upon comprehensive review of these factors and cases, we can now fill in the blanks of the above continuum. 258. Keller, 724 F.3d at 1285. 259. Id 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 401 C. Sourcing the Economic Value Another factor in deciding when the work has transformed the NIL to the level necessary for First Amendment protection is to determine the source of the economic value of that work. This is not separately identified as a factor in either Hart or Keller.260 But a careful reading of the cases reveals a subtle but important underpinning of how the courts make a finding about what is or is not a transformative work. Both Hart and Keller courts identify Comedy III as the first case to articulate the TUT.261 This was a precedent-setting case issued by the Supreme Court of California.262 As noted above, the court held that an artist’s T-shirts and prints of the Three Stooges were not transformative.263 The court defined transformative as requiring the artist to add something new, “‘with a further purpose or different character, altering the first with new expression, meaning, or message . . . .’”264 The important nuanced point here is that part of Comedy III’s rationale noted a need to scrutinize whether the work’s economic value was sourced primarily in the celebrity’s NIL, or conversely in the author of the work.265 The Court in Hart incorporated that element, explaining, “Going further, the court [in Comedy III] explained that works containing ‘significant transformative elements’ are less likely to interfere with the economic interests implicated by the right of publicity.”266 The Hart court then provided an example that reveals a further connection between economic interests and marketability of the work: “For example, ‘works of parody or other distortions of the celebrity figure are not, from the celebrity fan’s viewpoint, good substitutes for conventional depictions of the celebrity and therefore do not generally threaten markets for celebrity memorabilia that the right of publicity is designed to protect.’”267 The court in Comedy III also framed the issue to include both economic sourcing and marketability when it asked: “does the marketability and economic value of the challenged work derive primarily from the fame of the celebrity depicted?”268 If therefore a work sources its economic value primarily from the 260. See id.; see also Hart v. Electr. Arts, Inc., 717 F.3d 141 (3d Cir. 2013). 261. See Hart, 717 F.3d at 158; see also Keller, 724 F.3d at 1273. 262. Keller, 724 F.3d at 1273. 263. Comedy III Prod. v. Gary Saderup, Inc., 21 P.3d 797, 811 (Cal. 2001). 264. See Hart, 717 F.3d at 159 (citing Comedy III Prod., which cited the above quoted language from Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 579 (1994)). 265. Comedy IIl Prod., 21 P.3d at 810. 266. Hart, 717 F.3d at 159. (emphasis added). 267. Id. (citing Campbell, 510 U.S. at 579). 268. Comedy III Prod., 21 P.3d at 810 (emphasis added). The court in Hart described this as an “ancillary” question. This author examines the hypothetical as it becomes apparent additional rule refinements are necessary to address facts not present in either of those cases. Therefore, the author views the element of proximity as an appropriate additional factor. 402 INDIANA LAW REVIEW [Vol. 48:369 celebrity’s NIL, then the work is not sufficiently transformative to be protected from publicity rights claims of the celebrity. We would know that from analyzing whether the celebrity’s “market” (i.e., where the celebrity’s NIL is sold) is diminished. The example rings true to reality in that a parody is by definition a work so distorted from the real celebrity that the general public readily recognizes that the work is not an actual depiction of the celebrity. The example is designed to show that there is an apparent and obvious separation so that someone buying a product of the celebrity would not be duped into believing this is the genuine depiction of the celebrity. There is no bright line articulated by any court for when the markets diverge or the economic interests of the celebrity are not interfered with by the work. An example may illustrate the difficulty in application. Consider a circumstance where an animated cartoon-like film is produced of a Heisman Trophy winner. The film makes a parody of the player’s end zone dance, showing the dance in super-slow motion and accelerated motion at strategic times during the dance. The film also creates superhero actions where the athlete flies from the end zone to the press box, then into the stands to drink a Coke, and then back to the field. And assume the cartoon also grossly distorts his features with an extra-long nose, and elongated arms and legs. The film is then distributed and sold commercially in retail stores across the country. The broadly stated issue is whether the video is so transformative that it is protected from the athlete’s right of publicity claim. Applying the Comedy III economic sourcing rule, a court may reason that the source of revenue for the cartoon is transformative because of the significant creative elements of the animation and exaggerated features.269 If, however, the cartoon was an exact replica of the end zone dance, the court may find that the video is primarily sourced in the NIL of the athlete, even with exaggerated features and animation. The Court would also necessarily apply a case-by-case approach to the parody issue because the extent of exaggeration or distortion matters in the analysis. In my above hypothetical, a court would likely examine the extent of the elongation of limbs and exaggeration of features. That is because the greater the exaggeration the more likely a transformation would occur since the parody or distortions are judged by whether those items are something other than “good substitutes for conventional depictions“ of the celebrity.270 The court necessarily then will have to examine the cartoon to compare the distortions with the actual celebrity. That is why, in the author’s view, the test must be a continuum. No one description of the rule applies to all facts. And no one fact dismisses all other facts in making such a determination. A second prong of the Comedy III economic analysis is therefore required—market interference. There would be an interference with the economic rights of the celebrity if the work generally threatens markets for the celebrity.271 This author asserts there is a need for a refining factor to carefully 269. Id. 270. Id. at 808. 271. See Hart, 717 F.3d at 159-60 (citing Comedy III Prod., 21 P.3d at 808, which cited 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 403 analyze whether the work threatens a celebrity’s market. The author terms that factor the Marketing Proximity Factor, discussed below. D. Marketing Proximity Factor Neither Hart nor Keller directly faced the question of whether the economic sourcing analysis should include proximity, i.e., whether the athlete and the work are placed in settings close enough to each other to affect the market for either.272 In my above hypothetical, if the same actual Heisman winner made appearances on location where the cartoon was shown, akin to a book signing or promotional tour, would the physical proximity of the athlete lead the court to conclude that the cartoon derived its economic value from the NIL of the athlete? 273 Probably yes according to this author. Would the promotional activity of the athlete also lead the court to conclude that the cartoon interfered with the athlete’s market for his NIL? If so, then the cartoon was not transformative to something other than the NIL of the athlete and the cartoon’s creator would have no protection from publicity rights claims. If, however, a cartoon was being marketed in Arizona and the Heisman winner only made appearances in the state of Connecticut, the more likely finding is that the cartoon did not interfere with the market of the athlete. The critical fact in making that distinction was the lack of physical proximity of the athlete with the film. A second type of proximity is more directly economic in nature. Assume the Heisman hypothetical has this added fact: Although the athlete was not sharing physical space, the work was nonetheless marketed through the same retail outlets that were already under contract with the athlete. This author suspects that a court would consider that fact of decisional significance. The economic sourcing rule includes the implicit prohibition against the work threatening the market of the celebrity.274 That notion is endorsed by Campbell, then Comedy III, and then Hart.275 The Heisman athlete that has an existing retail distributor would likely prevail if the work is then sold through the same distributor. That would clearly be interference as contemplated in this context. Thus, apart from occupying the same physical space, the use of the same retail chain is a type of market commonality that would predictably favor a publicity rights claimant. The work is not transformative apart from the athlete’s NIL if the source of economic return for the work is tied so directly to the celebrity. The result would be the same regardless of whether the proximity is from physical proximity or economic proximity. Campbell, 510 U.S. at 569). 272. See id. at 141; see also Keller v. Electr. Arts, Inc., 724 F.3d 1268 (9th Cir. 2013). 273. This hypothetical cartoon does not implicitly include an assumption that athletes are parodied. 274. Comedy III Prod., 21 P.3d at 797. 275. See Campbell, 510 U.S. at 569; see also Comedy III Prod., 21 P.3d at 797; Hart, 717 F.3d at 141. 404 INDIANA LAW REVIEW [Vol. 48:369 The base logic of this economic sourcing factor is noteworthy. If a work is a mere substitute for the NIL of the celebrity, the work is not sufficiently transformative to have First Amendment protection against a right of publicity claimant. The logical consistency is that if the work interferes with the economic interests of the celebrity so that the celebrity’s existing markets are threatened, then an insufficient degree of transformation occurred because the economic value of the work is still too closely derived from the celebrity’s NIL. Conversely if the work is primarily the author’s own expression relative to the actual celebrity, there should be less economic overlap. The new independent expression by the creator of the work would have its own economic market and would not therefore interfere with the celebrity’s NIL. These rules therefore have interrelated symmetry and consistency in both theory and application, albeit in hypotheticals at this time. Yet, as the Heisman hypothetical illustrates, there are many unanswered questions when facts are introduced beyond those scrutinized in Hart and Keller.276 VII. PREDICTIVE ANALYTICS AND ESI A. The Economic Value Facture in TUT Determinations It is now well established that publicity rights are property rights that can give rise to damage claims for an unauthorized appropriation of an athlete’s NIL.277 The author asserts that the damages should be measured based on the economic value to the celebrity, not the value to the consumers of the product. In Keller, the court made the same celebrity-consumer distinction but in the broader context of publicity rights claims in total: “The right of publicity protects the celebrity, not the consumer.”278 The determination of damages is likely to not only look for prior unauthorized uses but also reasonably anticipated profits for the celebrity. In this section, the author asserts that predictive analytics may be a significant tool in that determination. The starting legal point, as a matter of law, is that the relevant economic value is the value to the celebrity.279 That point was made in Zacchini v. Scripps- Howard Broadcasting Co.280 The Court described the economic value of the publicity right of the entertainer who shot himself out of a cannon: Much of [the] economic value [of the performance] lies in the ‘right of exclusive control over the publicity given to his performance’; . . . [T]he broadcast of petitioner’s entire performance, unlike the unauthorized use of another’s name for purposes of trade . . . goes to the heart of 276. See id.; see also Keller, 724 F.3d at 1268 (These open questions were introduced in the initial discussion of the elements of the TUT). 277. Hart, 717 F.3d at 151. 278. Keller, 724 F.3d at 1281. 279. Zacchini v. Scripps-Howard Broad. Co., 433 U.S. 562, 575 (1977). 280. Id. 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 405 petitioner’s ability to earn a living as an entertainer . . . [and was] the appropriation of the very activity by which the entertainer acquired his reputation in the first place.281 At first blush, the above quote appears to cut against a college athlete’s right of publicity. Mr. Zacchini, the human cannonball, appears to be in a trade that freely allows him to profit from his performance.282 The collegiate athletes are not engaged in a for-profit performance since NCAA rules expressly prohibit profiteering by amateur athletes.283 This author does not view that fact as a basis to wholly distinguish Zacchini.284 The Zacchini court did not face a claim by college athletes that they have publicity rights.285 Nothing in the balancing of publicity rights and First Amendment rights hinges on whether the athlete or celebrity had a profit motive.286 What remains relevant from Zacchini is the underlying focus on the economic value to the performer and publicity rights as a means of protecting one’s economic value.287 Future cases are therefore likely to describe damages in economic terms—measuring the extent of economic gain by the unauthorized user and/or loss by the publicity rights claimant. The damages claim should, in turn, include a projection of a work’s future economic gains and losses which requires some predictions of future revenue from the use of the celebrity’s NIL. Hence, a discussion of predictive analytics is relevant and discussed below. B. Application of Predictive Analytics Predictive analytics involve creating large data sets that are sorted and analyzed by computer programs and algorithms to predict future behaviors.288 As applied to the subject of this Article, the same methods can be used to collect and analyze data associated with both parties to the litigation, an athlete’s NIL, and then the author or creator of the work. Algorithms are typically created through computer programs to quickly analyze large sets of data.289 This is part of data mining, which has been 281. Id. at 575-76. 282. Id. at 574. 283. Article 12 of the NCAA bylaws codifies its rules of amateurism that the athlete must agree to be so governed. NCAA, 2011-12 NCAA DIVISION I MANUAL § 12.01.1 (2011), available at http://saas.usc.edu/files/2012/08/NCAA-2011-12-Manual.pdf, archived at http://perma. cc/U7QD-4J6V. 284. Zacchini, 433 U.S. at 562. 285. See generally id. 286. See generally id. 287. Id. at 575. 288. See Julie E. Cohen, What is Privacy For, 126 HARV. L. REV. 1904, 1920-22 (2013) (where predictive analytics is described as a method of converting data flows into patterns, including the enhancement of climatologists’ understanding of weather patterns and improvement of epidemiologists’ insight regarding public health problems). 289. Roger M. Groves, The Implications of A Jeopardy! Computer Named Watson: Beating 406 INDIANA LAW REVIEW [Vol. 48:369 described in the complex field of antitrust litigation: This is precisely the value of data mining: with large quantities of diverse data, mathematical algorithms can be applied to identify useful but unintuitive correlations between inexpensively observed characteristics that can be exploited to make significantly more accurate predictions.290 These techniques are already being used to predict the highest interest rate a person is likely to accept when taking out a loan or the “pain point” for the upper limit of what a borrower would pay in a transaction.291 Such predictive data can be invaluable to a lender in both circumstances.292 The same techniques can have considerable utility in publicity rights cases. If a Court finds the work is not transformative, there is an appropriation of the athlete’s publicity rights. The question then is what is the economic value of those rights? Algorithms could be created to track any common elements of sale, marketing, purchasers, and selling brokers.293 Similarly, the court may use analytics when determining if the creative elements of the work’s author were primary. The analytics could assess motivations of the consumers who purchase the work that contained the athlete’s images. Indeed, this author envisions a time when a court will direct the defendant creator of a work to itemize all the “significant transformative elements” or “creative contributions” in all forms.294 Instead of a “pain point,” the question will likely be, “What is the ‘pleasure point’ for consumer/gamer usage or purchase?” A related question could be, “What was the source of that pleasure point: the athlete’s NIL or something else creatively supplied by the maker of the work?” Due to economic sourcing discussed above, it may not be sufficient to simply state each element of creativity as was done in Hart and Keller.295 In those cases, the work creators noted mechanisms that allowed game users to manipulate images, add themselves in tryouts with the team, and change the jersey colors or hair styles of the celebrity.296 Those items do not address the economic Corporate Boards of Directors at Fiduciary Duties?, 45 CREIGHTON L. REV. 377, 390 (2012). 290. Douglas M. Kochelek, Data Mining and Antitrust, 22 HARV. J. L. & TECH. 515, 519 (2009). 291. Id. at 518-19. 292. Id. at 519. 293. Algorithms are mathematical formulas that analyze large data sets, including data that can be arranged to make correlations to future behaviors. See id. at 519; see also Groves, supra note 289, at 415-16. 294. The California Supreme Court stated that the transformative elements and creative contributions can take many forms. Comedy III Prod. v. Gary Saderup, Inc., 21 P.3d 797, 809 (Cal. 2001). 295. See Hart v. Electr. Arts, Inc., 717 F.3d 141 (3rd Cir. 2013); see also Keller v. Electr. Arts, Inc., 724 F.3d 1268 (9th Cir. 2013). 296. See Hart, 717 F.3d at 167-69; see also Keller, 724 F.3d at 1276-77. 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 407 interference issues. The defendant should be prepared to offer proofs to demonstrate separation between the work’s economic base and the celebrity’s economic base. To accomplish that factual evidence, the defendant can have a computer program that isolates all sales, purchases, and marketing efforts to remove the work’s economic sourcing from the celebrity. The program can provide a comparative analysis showing the lack of economic intersection between the work and the athlete. C. Sentiment Analysis Sentiment analysis is an established field that examines opinions, sentiments, evaluations, attitudes, and emotions through written language.297 For example, one study was designed to analyze emotions and the way words are used to convey them.298 The components of emotions included “valence (the pleasantness of a stimulus), arousal (the intensity of emotion provoked by a stimulus), and dominance (the degree of control exerted by a stimulus).”299 Similarly, a study examined 7362 tobacco-related Twitter posts to ascertain sentiments about tobacco.300 If words and Twitter posts can be analyzed for emotion and be accepted by a court as relevant evidence, then written surveys can be evidence of consumers’ emotions about an athlete. The words may quite possibly be categorized or stratified to measure the relative positive and negative emotions associated with the athlete compared to other aspects of the work. If the court is convinced that there was a high level intensity of emotion for the athlete in purchasing the work, the court may more likely conclude that the work was not transformative since the athlete’s depiction was primary in the purchase. Another application involves the part of the TUT associated with the market. As discussed above, the TUT attempts to assess whether the “market” is predominantly using or buying the work because of the independent creativity of the work or, conversely, because of the NIL of the athlete.301 If there are sufficient transformative creative elements, the First Amendment rights of the creator of the work protect that creator against publicity rights attacks.302 As applied to calculating the plaintiff’s damages, a court may include a work sold in another venue (in the calculation). This is consistent with the marketing proximity factor previously discussed. This author’s view is that computer programs and algorithms will be 297. See LIU BING, SENTIMENT ANALYSIS AND OPINION MINING (Morgan & Claypool Publ’rs May 2012); see also Amy Beth Warriner, et al, Norms of Valence, Arousal, and Dominance for 13,915 English Lemmas, 45 BEHAV RES METHODS 1191-207. (2013). 298. See id. 299. Id. 300. See Mark Myslin et al, Using Twitter to Examine Smoking Behavior and Perceptions of Emerging Tobacco Products 15, J. MED. INTERNET RES. 1 (2013). 301. Comedy III Prod. v. Gary Saderup, Inc., 21 P.3d 797, 808 (Cal. 2001). 302. Id. http://dx.doi.org/10.2196/jmir.2534 408 INDIANA LAW REVIEW [Vol. 48:369 employed in future publicity rights cases to ascertain whether the public feels more strongly aligned with some aspects of a work compared to the NIL of the athlete. This is the very type of evidence a sentiment analysis is designed to produce. That is a likely future battleground in publicity rights cases, because the essential question is whether the public buys an item more because of the celebrity or, conversely, the work that used the NIL of that celebrity. Another future use of the sentiment analysis relates to the “trivial variation” rule.303 As noted above, a transformative use is not allowed for “merely trivial variation[s]” in a work.304 However, none of the recent publicity rights cases have attempted to define the term “trivial variations.” Thus, no case yet provides parameters as to cut the line for when a trivial variation ends and something termed a “significant” transformative element begins. There are mathematical constructs to measure the extent of variation and correlation between one item and another. The plaintiff publicity rights claimant is likely to assert that certain creative elements by the creator of the work are trivial. One potential method of proving triviality is surveying customers to determine if they even noticed the disputed elements of the work. In one of the above hypotheticals, the consumer would be asked if he noticed the use of racquetball instead of tennis or the extent of elongation in the athlete’s nose, and then, what level of intensity of like or dislike did the consumer feel. If the assessment was that items were barely noticed or lacked intensity for the work, the court is more likely to find the work’s elements trivial. Similarly, analytics may be employed to show that individual consumers were surveyed about what most influenced their buying decision—the other elements of the work or the actual celebrity. Market analytics are already being performed to also assess if people are more committed or influenced by some items over other items based on written or verbal responses.305 The evolution of these cases may then proffer evidence and expert testimony that attempts to digitize elements of creativity and weight the variances among those elements of the work. The variances would be weighted quantitatively, not qualitatively per the five-factors noted in Hart.306 This author envisions sophisticated formulas to measure those quantitative differences. For example, the extent of distortion of an athlete’s features can be measured quantitatively. A survey of consumers of the work can have their intensity of emotion or loyalty for the athlete versus the other aspects of the work measured quantitatively. Pitched battle is also anticipated regarding the admissibility of such programs and resulting data sets. The sourcing of economic value for a work is another future application of predictive analytics and sentiment analysis. Documents may be produced by defendants, for example, to establish that only 1.5% of the market sales of the work were sold in the markets already maintained or marketed by the celebrity. 303. See Winter v. DC Comics, 69 P.3d 473, 478-79 (2003). 304. Id. 305. See Groves, supra note 289, at 413. 306. Hart v. Electr. Arts, Inc., 717 F.3d 141, 163 (3d Cir. 2013). 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 409 The purpose of the offering would be to show the lack of market interference between the work and the athlete’s NIL. D. Illustration of Predictive Analytics—Market Penetration Analysis As noted above, part of the TUT endorsed in Hart requires an examination of whether the work is a substitute for the celebrity’s NIL that interferes with the celebrity’s “markets.”307 That question necessarily involves a finding of just what is the celebrity’s “market.”308 Once the market is defined, what is the penetration of the work in that market? There are cases that provide illustrations of the value of predictive analytics when faced with large data sets to determine economic value and market competition. The theory of substitution was also used in Hart.309 In FTC v. CCC Holdings Inc., the dispute involved the Federal Trade Commission’s attempt to enjoin a merger of two companies that provide valuation software used to estimate auto insurance company losses.310 As described by the court: [There are] twenty-five million automobile insurance claims each year and insurers, in turn, spend an estimated $100 billion annually to cover those claims. Most insurers and automotive repair shops use specialized computer software to estimate the cost of repair or the value of replacement in the event of a total loss. These software systems play a critical role in the automotive repair industry.311 Critical expert witness testimony and other evidence involved whether one party was competing in the market of another party.312 The court’s findings on that issue are noted below: [T]he real-world evidence shows that Books and TLV are not part of the same product market. See Arch Coal, 329 F. Supp. 2d at 116 (“[A]nti- trust theory and speculation cannot trump facts.”). The Book vendors do not consider themselves to be in competition with CCC, Mitchell, and Audatex. [citations omitted] . . . Most insurance companies do not view the Book as an adequate substitute for TLV products. See PX 664-005 (Mitchell); Brown Dep. at 90-90 (Erie).313 The court relied on deposition testimony of experts who collected large data sets.314 The experts then used predictive analytics to define the market and 307. Id. at 168. 308. Id. 309. Id. 310. FTC v. CCC Holdings, Inc., 605 F. Supp. 2d 26 (D.D.C. 2009). 311. Id. at 30. 312. Id. at 40-42. 313. Id. at 41 (emphasis added). 314. See id. at 70. 410 INDIANA LAW REVIEW [Vol. 48:369 determine if one party is providing a substitute within that market.315 The court relied on deposition testimony.316 Future litigation in publicity rights cases is likely to have the same dynamics: (1) a determination of a relevant market based on computer-generated analytics; and (2) a determination of whether there has been market penetration and competition from an analysis of large data sets. The parties should likewise prepare for depositions to stage a battle of experts. The court addressed predictive analytics in the following passage: Predictive analytics, an internal method of calculating future estimates based on an insurance company’s own empirical data, Marushka Dep. at 14-15 (CCC), may supplant the Estimatics software sold by CCC, Mitchell and Audatex one day, and the Defendants hint that it may be sooner rather than later when they suggest that the merger of two outdated platforms cannot hurt anyone.317 Thus, the court admitted testimony and evidence of future estimates based on current data.318 Therefore, may litigants in publicity rights cases project income and the market based on current sales, marketing, and sentiment analysis of consumers in those markets? The court rejected the particular estimates in CCC Holdings because: (1) the methods were too speculative since not used in the affected industry; (2) there were subjective aspects of a party’s predictive methodologies (also termed protocols) that caused the analysis to be unreliable; and (3) the pool of data sets used by that party omitted certain relevant information.319 Attorneys for defendants in publicity rights litigation should also consider similar claims of speculation, subjectivity, and material omissions that reveal a lack of relevance or probative value. In any event, we should expect increasing sophistication of both software programmers and the attorneys who use them when the issues involve the intersection of human behaviors and the economic value of the work they purchase. This Article presents issues and rule statements that indicate publicity rights cases are ripe for that type of analysis. And when millions of dollars are at stake, I suspect there is not going to be a law firm that is going to volunteer to be out-sophisticated, out-strategized, and defeated due to the lack of technology in court. The publicity rights cases to date involve plaintiffs that request class action status. A large pool of plaintiffs gives rise to claims in the hundreds of millions of dollars. The cases will necessarily involve millions of bits of data and revenue to be tracked nationally, if not internationally. Accordingly, this author opines that predictive analytics will be increasingly used 315. See id. at 41. 316. See generally id. 317. Id. at 59-60. 318. Id. 319. The court noted that predictive analytics was not currently a staple in the insurance industry. “It is acknowledged by CCC, however, that . . . no insurance company currently uses predictive analytics to calculate partial loss estimates and he does not know whether predictive analytics are allowed under insurance regulations.” Id. at 60. 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 411 as part of the evolution of publicity rights litigation. E. Electronically Stored Information (ESI) Based on the recent Hart, Keller, and O’Bannon cases filed to date, plaintiffs in publicity rights cases are likely to have millions of bits of emails, text messages, and other electronically stored information (ESI).320 That information is discoverable and analyzed as part of multi-million dollar damage claims, along with requests for permanent injunctions.321 There are now numerous cases, albeit with different substantive law issues, that nonetheless have similar ESI issues.322 Indeed an entire body of e-discovery law has developed concerning these mega- data cases.323 Publicity rights litigants can therefore expect the type of Case Management Order that managed the discovery in one such case that involved predictive analytics and high volume ESI.324 In the case of In re Actos (Pioglitazone) Products Liability Litigation, the court provides insight as to various aspects of e-discovery case management when predictive analytics is pursued.325 The court’s overall goal is for the parties to agree on the methodologies or protocol to be used in discovery.326 First, the parties were pushed to agree on the following: 1. The party that will provide the software program to search and review ESI, and develop predictive coding (or if they will jointly select such a program or programmer); 2. The party that will collect the ESI and create a “sample collection population”; 3. The joint execution of a Nondisclosure and Confidentiality Agreement and drafting of a Protective Order that binds all counsel, the parties, and their respective experts; 4. Training on the software used for predictive coding for both parties; 5. Privilege and redaction opportunities for affected parties; 6. Mandatory meetings to resolve any predictive coding disputes; 7. An agreed number of random samples from the sample collection population; 8. The party(s) that will be the custodian of the ESI and documents 320. See Hart v. Electr. Arts, Inc., 717 F.3d 141 (3rd Cir. 2013); see also Keller v. Electr. Arts, Inc., 724 F.3d 1268 (9th Cir. 2013); O’Bannon v. Nat’l Collegiate Athletic Ass’n, No. C 09-3329 CW, 2010 WL 445190 (N.D. Cal. Feb. 8, 2010). 321. ESI includes, but is not limited to, emails, text messages, voice mails, computer programs, and documents stored in electronic files. See Groves, supra note 289, at 377. 322. See generally id. 323. See generally id. 324. In re Actos (Pioglitazone) Prods. Liab. Litig., No. 6:11–md–2299, 2012 WL 7861249 (W.D. La. 2012). 325. Id. 326. Id. 412 INDIANA LAW REVIEW [Vol. 48:369 produced from the predictive coding.327 This Article is designed in part to sensitize parties with publicity rights disputes to the distinct possibility that the case will involve voluminous ESI associated with tracking revenue streams, sales at universities, sales from licensed merchandise, and particularly projection of future revenue from the use of NIL. There may also then be a case management order similar to In re Actos.328 One or both parties will look to examine the ESI and correlate past, current, and future revenue streams of the defendant users to the plaintiff athlete’s NIL. Skilled plaintiff attorneys should aggressively identify markets where defendants have profited from athletes’ NILs. Wise plaintiff counsel would therefore contemplate the various discovery challenges in advance of filing. The In re Actos order provides guidance in various areas.329 If the court is likely to request stipulations on the software program for collecting, sorting and predictive coding, a primary strategy is to interview top software programmers. The selected programmer should then be instructed to prepare a coding plan in advance. The party with the most prepared programmer at the early stages of litigation improves the odds of convincing the court to use that party’s expert. Such a strategy may be worth the expense. Understandably, plaintiff firms that take a case on a contingency basis are necessarily concerned with the costs, since they are essentially funding the litigation. But if class action status is achieved, the more prepared expert programmer may convince the court that plaintiff’s coding plan should be selected. The defendant, who is already playing defense, has the opportunity to vet plaintiff’s expert but the coding plan presumably favors the party who prepared it. The coding may result in millions of dollars of difference in damages. The potential of such awards can far exceed the expert fees, especially when the fee structure is carefully structured.330 The anticipation of such a discovery order should also allow both parties to contemplate in advance the ESI that is most advantageous in the sample collection population. It is obviously a greater benefit to the defendant creator of a video game to include a wealth of evidence showing penetration of sales that do not interfere with the celebrity’s markets. The plaintiff would rather weigh the collected ESI with evidence that the defendant’s sales were within his or her markets. Additionally the plaintiff would prefer the population to show that an increase in the defendant’s product sales caused a decrease in plaintiff’s sales from his use of his own NIL. This would be consistent with the economic sourcing rule, particularly the implicit prohibition of defendant creating a 327. Id. 328. Id. 329. Id. 330. Fee management includes structuring the fee schedule in stages, with itemization per task. This helps prevent billing without justification. The staging also forces the discipline and clarity that assures the witness is only paid for services to a certain point (e.g., preliminary analysis prior to discovery) followed a stage that responds to discovery requests. 2015] PUBLICITY RIGHTS OF COLLEGE PLAYERS 413 substitute for plaintiff’s NIL.331 Such a finding leads to a conclusion that the work is not transformative and, therefore, not protected from the publicity rights claims. Finally, there is an advantage of pre-thinking the discovery issues to aid in developing initial cost estimates. Those estimates should assist both parties in evaluating settlement options or in the plaintiff’s case, even whether to file the case. This author views these questions as a future legal battleground. The spokespersons will not be attorneys spouting theories. The key players will be the computer programs that have scientifically valid methods of measuring the degree of interrelatedness of a work’s creative elements with the NIL of the celebrity. CONCLUSION Hart and Keller have affirmed for college athletes what prior case law already established for professional athletes and entertainers—that they too have publicity rights that can trump First Amendment rights of those who use their names, images, and likenesses without authorization.332 Existing case precedent already provides illustrations of what will be considered an athlete’s NIL. A particular professional football player named Elroy Hirsch became known as “Crazy legs” because his running style was reminiscent of an egg-beater.333 A shaving gel for women under the same name was found to violate Mr. Hirsch’s publicity rights.334 Similarly, an established race car driver used a particular color scheme for his car.335 Cigarette maker, R.J. Reynolds Tobacco, created a commercial stating, “Winston tastes good, like a cigarette should.”336 The picture of the driver’s car was prominent in the commercial.337 Relief was provided to the driver on the unauthorized use of his color scheme even though the numbers on the car were changed in the commercial from eleven to seventy-one. 338 Assuming there is no judicially declared waiver by college athletes of their publicity rights, the types of NIL claims are easily imagined. Here are just a few examples of an athlete’s protectable intellectual property that could be protected under his or her NIL: (1) a player with a unique jump shot; (2) a signature spin move to elude defenders in basketball or football; (3) a unique dunk displayed in a dunk contest; (4) a newly effective free throw shooting technique; (5) a 331. Comedy III Prod. v. Gary Saderup, Inc., 21 P.3d 797, 810 (Cal. 2001). 332. See Hart v. Electr. Arts, Inc., 717 F.3d 141 (3d Cir. 2013); see also Keller v. Electr. Arts, Inc., 724 F.3d 1268 (9th Cir. 2013). 333. Hirsch v. S.C. Johnson, 280 N.W.2d 129, 141-42 (Wis. 1979). 334. Id. 335. Motschenbacker v. R.J. Reynolds Tobacco, 498 F.2d 821, 822 (9th Cir. 1974). 336. Id. 337. Id. 338. Id. at 827. 414 INDIANA LAW REVIEW [Vol. 48:369 particular hair style or beard; (6) an offensive lineman’s new blocking move; (7) a defensive end or linebacker with a signature hand techniques and body pivots that evade blockers to reach the quarterback; (8) a tattoo prominently displayed; (9) color schemes on a uniform, jersey, helmet, facemask, socks or shoes to the extent permitted under NCAA rules; and (10) apparel during camps or other non- regulated events to the extent permitted under NCAA rules. If a current college or former college player who is not performing in the professional ranks asserts those same rights, the legal environment appears more favorable now than in prior decades to allow those plaintiffs to prevail, notwithstanding their status as “amateurs” by the NCAA.339 This Article also attempts to highlight the evolving use of sophisticated technology, including, but not limited to, predictive analytics and sentiment analysis. Computer programs with algorithms designed to statistically reveal relationships between the work and NIL of the athlete may be offered as evidence. These techniques could favor either party in publicity rights litigation. A plaintiff’s purpose will be to establish that the work is not transformative, because it is either: (1) a mere trivial aspect to an otherwise identical depiction of the athlete; or (2) interfering with the economic value sourced in the athlete. Such a finding is likely to affirm a prioritization of publicity rights over the First Amendment defenses of the work’s creator. A defendant, on the other hand, may establish that the work transforms the NIL into a primary creative effort of the author of the work. In such a case, the First Amendment rights of the author would prevail. The defendant will also claim that consumers of the work have more intense sentiment for the work because of significant creative elements apart from the athlete, thus allowing the work to pass the TUT. Finally, this Article seeks to more directly expose the crux of the legal dispute. A careful analysis of Hart and Keller reveals that the most fundamental disagreement of the majority from the dissent and basis for the opposite conclusions involves the scope of the inquiry.340 The majority’s view that only the work’s depiction of the athlete is at issue differs sharply from the dissenters’ view that the court should examine the work as a whole to determine if there are sufficient creative elements to make the work transformative.341 The open questions presented in this Article should fuel future litigation, and all of the above issues may be in play, performed on a far different court than what the athletes are accustomed. 339. See Hart v. Electr. Arts, Inc., 717 F.3d 141 (3d Cir. 2013); see also Keller v. Electr. Arts, Inc., 724 F.3d 1268 (9th Cir. 2013). 340. See Hart, 717 F.3d at 141; see also Keller, 724 F.3d at 1268. 341. See Hart, 717 F.3d at 141; see also Keller, 724 F.3d at 1268.