OPINION

BYBEE, Circuit Judge:

Video games are entitled to the full protections of the First Amendment, because "[l]ike the protected books, plays, and movies that preceded them, video games communicate ideas — and even social messages — through many familiar literary devices (such as characters, dialogue, plot, and music) and through features distinctive to the medium (such as the player's interaction with the virtual world)." Brown v. Entm't Merchs. Ass'n, ___ U.S. ___, 131 S.Ct. 2729, 2733, 180 L.Ed.2d 708 (2011).1 Such rights are not absolute, and states may recognize the right of publicity to a degree consistent with the First Amendment. Zacchini v. Scripps-Howard Broad. Co., 433 U.S. 562, 574-75, 97 S.Ct. 2849, 53 L.Ed.2d 965 (1977). In this case, we must balance the right of publicity of a former college football player against the asserted First Amendment right of a video game developer to use his likeness in its expressive works.

The district court concluded that the game developer, Electronic Arts ("EA"), had no First Amendment defense against the right-of-publicity claims of the football player, Samuel Keller. We affirm. Under the "transformative use" test developed by the California Supreme Court, EA's use does not qualify for First Amendment protection as a matter of law because it literally recreates Keller in the very setting in which he has achieved renown. The other First Amendment defenses asserted by EA do not defeat Keller's claims either.

I

Samuel Keller was the starting quarterback for Arizona State University in 2005 before he transferred to the University of Nebraska, where he played during the 2007 season. EA is the producer of the NCAA Football series of video games, which allow users to control avatars representing college football players as those avatars participate in simulated games. In NCAA Football, EA seeks to replicate each school's entire team as accurately as possible. Every real football player on each team included in the game 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. 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.

EA's game differs from reality in that EA omits the players' names on their jerseys and assigns each player a home town that is different from the actual player's home town. However, users of the video game may upload rosters of names obtained from third parties so that the names do appear on the jerseys. In such cases, EA allows images from the game containing athletes' real names to be posted on its website by users. Users can further alter reality by entering "Dynasty" mode, where the user assumes a head coach's responsibilities for a college program for up to thirty seasons, including recruiting players from a randomly generated pool of high school athletes, or "Campus Legend" mode, where the user controls a virtual player from high school through college, making choices relating to practices, academics, and social life.

In the 2005 edition of the game, the virtual starting quarterback for Arizona State wears number 9, as did Keller, and has the same height, weight, skin tone, hair color, hair style, handedness, home state, play style (pocket passer), visor preference, facial features, and school year as Keller. In the 2008 edition, the virtual quarterback for Nebraska has these same characteristics, though the jersey number does not match, presumably because Keller changed his number right before the season started.

Objecting to this use of his likeness, Keller filed a putative class-action complaint in the Northern District of California asserting, as relevant on appeal, that EA violated his right of publicity under California Civil Code § 3344 and California common law.2 EA moved to strike the complaint as a strategic lawsuit against public participation ("SLAPP") under California's anti-SLAPP statute, Cal.Civ.Proc. Code § 425.16, and the district court denied the motion. We have jurisdiction over EA's appeal pursuant to 28 U.S.C. § 1291. See Batzel v. Smith, 333 F.3d 1018, 1024-26 (9th Cir.2003).3

II

California's anti-SLAPP statute is designed to discourage suits that "masquerade as ordinary lawsuits but are brought to deter common citizens from exercising their political or legal rights or to punish them for doing so." Batzel, 333 F.3d at 1024 (internal quotation marks omitted). The statute provides:

Cal.Civ.Proc.Code § 425.16(b)(1). We have determined that the anti-SLAPP statute is available in federal court. Thomas v. Fry's Elecs., Inc., 400 F.3d 1206 (9th Cir.2005) (per curiam).

We evaluate an anti-SLAPP motion in two steps. First, the defendant must "make a prima facie showing that the plaintiff's suit arises from an act by the defendant made in connection with a public issue in furtherance of the defendant's right to free speech under the United States or California Constitution." Batzel, 333 F.3d at 1024. Keller does not contest that EA has made this threshold showing. Indeed, there is no question that "video games qualify for First Amendment protection," Entm't Merchs. Ass'n, 131 S.Ct. at 2733, or that Keller's suit arises from EA's production and distribution of NCAA Football in furtherance of EA's protected right to express itself through video games.

Second, we must evaluate whether the plaintiff has "establish[ed] a reasonable probability that the plaintiff will prevail on his or her ... claim." Batzel, 333 F.3d at 1024. "The plaintiff must demonstrate that the complaint is legally sufficient and supported by a prima facie showing of facts to sustain a favorable judgment if the evidence submitted by plaintiff is credited." Metabolife Int'l, Inc. v. Wornick, 264 F.3d 832, 840 (9th Cir.2001) (internal quotation marks omitted). The statute "subjects to potential dismissal only those actions in which the plaintiff cannot state and substantiate a legally sufficient claim." Navellier v. Sletten, 29 Cal.4th 82, 124 Cal.Rptr.2d 530, 52 P.3d 703, 711 (2002) (internal quotation marks omitted). EA did not contest before the district court and does not contest here that Keller has stated a right-of-publicity claim under California common and statutory law.4 Instead, EA raises four affirmative defenses derived from the First Amendment: the "transformative use" test, the Rogers test, the "public interest" test, and the "public affairs" exemption. EA argues that, in light of these defenses, it is not reasonably probable that Keller will prevail on his right-of-publicity claim. This appeal therefore centers on the applicability of these defenses. We take each one in turn.5

A

The California Supreme Court formulated the transformative use defense in Comedy III Productions, Inc. v. Gary Saderup, Inc., 25 Cal.4th 387, 106 Cal.Rptr.2d 126, 21 P.3d 797 (2001). The defense is "a balancing test between the First Amendment and the right of publicity based 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." Id. 106 Cal.Rptr.2d 126, 21 P.3d at 799. The California Supreme Court explained that "when a work contains significant transformative elements, it is not only especially worthy of First Amendment protection, but it is also less likely to interfere with the economic interest protected by the right of publicity." Id. 106 Cal.Rptr.2d 126, 21 P.3d at 808. The court rejected the wholesale importation of the copyright "fair use" defense into right-of-publicity claims, but recognized that some aspects of that defense are "particularly pertinent." Id.; see 17 U.S.C. § 107; see also SOFA Entm't, Inc. v. Dodger Prods., Inc., 709 F.3d 1273, 1277-78 (9th Cir.2013) (discussing the "fair use" defense codified in 17 U.S.C. § 107).

Comedy III gives us at least five factors to consider in determining whether a work is sufficiently transformative to obtain First Amendment protection. See J. Thomas McCarthy, The Rights of Publicity and Privacy § 8:72 (2d ed.2012). First, if "the celebrity likeness is one of the `raw materials' from which an original work is synthesized," it is more likely to be transformative than if "the depiction or imitation of the celebrity is the very sum and substance of the work in question." Comedy III, 106 Cal.Rptr.2d 126, 21 P.3d at 809. Second, the work is protected if it is "primarily the defendant's own expression" — as long as that expression is "something other than the likeness of the celebrity." Id. This factor requires an examination of whether a likely purchaser's primary motivation is to buy a reproduction of the celebrity, or to buy the expressive work of that artist. McCarthy, supra, § 8:72. Third, to avoid making judgments concerning "the quality of the artistic contribution," a court should conduct an inquiry "more quantitative than qualitative" and ask "whether the literal and imitative or the creative elements predominate in the work." Comedy III, 106 Cal.Rptr.2d 126, 21 P.3d at 809. Fourth, the California Supreme Court indicated that "a subsidiary inquiry" would be useful in close cases: whether "the marketability and economic value of the challenged work derive primarily from the fame of the celebrity depicted." Id. 106 Cal.Rptr.2d 126, 21 P.3d at 810. Lastly, the court indicated that "when an artist's skill and talent is manifestly subordinated to the overall goal of creating a conventional portrait of a celebrity so as to commercially exploit his or her fame," the work is not transformative. Id.

We have explained that "[o]nly if [a defendant] is entitled to the [transformative] defense as a matter of law can it prevail on its motion to strike," because the California Supreme Court "envisioned the application of the defense as a question of fact." Hilton, 599 F.3d at 910. As a result, EA "is only entitled to the defense as a matter of law if no trier of fact could reasonably conclude that the [game] [i]s not transformative." Id.

California courts have applied the transformative use test in relevant situations in four cases. First, in Comedy III itself, the California Supreme Court applied the test to T-shirts and lithographs bearing a likeness of The Three Stooges and concluded that it could "discern no significant transformative or creative contribution." Id. 106 Cal.Rptr.2d 126, 21 P.3d at 811. The court reasoned that the artist's "undeniable skill is manifestly subordinated to the overall goal of creating literal, conventional depictions of The Three Stooges so as to exploit their fame." Id. "[W]ere we to decide that [the artist's] depictions were protected by the First Amendment," the court continued, "we cannot perceive how the right of publicity would remain a viable right other than in cases of falsified celebrity endorsements." Id.

Second, in Winter v. DC Comics, the California Supreme Court applied the test to comic books containing characters Johnny and Edgar Autumn, "depicted as villainous half-worm, half-human offspring" but evoking two famous brothers, rockers Johnny and Edgar Winter. 30 Cal.4th 881, 134 Cal.Rptr.2d 634, 69 P.3d 473, 476 (2003). The court held that "the comic books are transformative and entitled to First Amendment protection." Id. 134 Cal.Rptr.2d 634, 69 P.3d at 480. It reasoned that the comic books "are not just conventional depictions of plaintiffs but contain significant expressive content other than plaintiffs' mere likenesses." Id. 134 Cal.Rptr.2d 634, 69 P.3d at 479. "To the extent the drawings of the Autumn brothers resemble plaintiffs at all, they are distorted for purposes of lampoon, parody, or caricature." Id. Importantly, the court relied on the fact that the brothers "are but cartoon characters ... in a larger story, which is itself quite expressive." Id.

Third, in Kirby v. Sega of America, Inc., the California Court of Appeal applied the transformative use test to a video game in which the user controls the dancing of "Ulala," a reporter from outer space allegedly based on singer Kierin Kirby, whose "`signature' lyrical expression ... is `ooh la la.'" 144 Cal.App.4th 47, 50 Cal.Rptr.3d 607, 609-10 (2006). The court held that "Ulala is more than a mere likeness or literal depiction of Kirby," pointing to Ulala's "extremely tall, slender computer-generated physique," her "hairstyle and primary costume," her dance moves, and her role as "a space-age reporter in the 25th century," all of which were "unlike any public depiction of Kirby." Id. at 616. "As in Winter, Ulala is a `fanciful, creative character' who exists in the context of a unique and expressive video game." Id. at 618.

Finally, in No Doubt v. Activision Publishing, Inc., the California Court of Appeal addressed Activision's Band Hero video game. 192 Cal.App.4th 1018, 122 Cal.Rptr.3d 397, 400 (2011), petition for review denied, 2011 Cal. LEXIS 6100 (Cal. June 8, 2011) (No. B223996). In Band Hero, users simulate performing in a rock band in time with popular songs. Id. at 401. Users choose from a number of avatars, some of which represent actual rock stars, including the members of the rock band No Doubt. Id. at 401. Activision licensed No Doubt's likeness, but allegedly exceeded the scope of the license by permitting users to manipulate the No Doubt avatars to play any song in the game, solo or with members of other bands, and even to alter the avatars' voices. Id. at 402. The court held that No Doubt's right of publicity prevailed despite Activision's First Amendment defense because the game was not "transformative" under the Comedy III test. It reasoned that the video game characters were "literal recreations of the band members," doing "the same activity by which the band achieved and maintains its fame." Id. at 411. According to the court, the fact "that the avatars appear in the context of a videogame that contains many other creative elements[] does not transform the avatars into anything other than exact depictions of No Doubt's members doing exactly what they do as celebrities." Id. The court concluded that "the expressive elements of the game remain manifestly subordinated to the overall goal of creating a conventional portrait of No Doubt so as to commercially exploit its fame." Id. (internal quotation marks omitted).

We have also had occasion to apply the transformative use test. In Hilton v. Hallmark Cards, we applied the test to a birthday card depicting Paris Hilton in a manner reminiscent of an episode of Hilton's reality show The Simple Life. 599 F.3d at 899. We observed some differences between the episode and the card, but noted that "the basic setting is the same: we see Paris Hilton, born to privilege, working as a waitress." Id. at 911. We reasoned that "[w]hen we compare Hallmark's card to the video game in Kirby, which transported a 1990s singer (catchphrases and all) into the 25th century and transmogrified her into a space-age reporter, ... the card falls far short of the level of new expression added in the video game." Id. As a result, we concluded that "there is enough doubt as to whether Hallmark's card is transformative under our case law that we cannot say Hallmark is entitled to the defense as a matter of law." Id.6

With these cases in mind as guidance, we conclude that EA's use of Keller's likeness does not contain significant transformative elements such that EA is entitled to the defense as a matter of law. The facts of No Doubt are very similar to those here. EA is alleged to have replicated Keller's physical characteristics in NCAA Football, just as the members of No Doubt are realistically portrayed in Band Hero. Here, as in Band Hero, 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 Band Hero. The context in which the activity occurs is also similarly realistic — real venues in Band Hero and realistic depictions of actual football stadiums in NCAA Football. 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." Keller v. Elec. Arts, Inc., No. C 09-1967 CW, 2010 WL 530108, at *5 (N.D.Cal. Feb. 8, 2010).

EA argues that the district court erred in focusing primarily on Keller's likeness and ignoring the transformative elements of the game as a whole. Judge Thomas, our dissenting colleague, suggests the same. See Dissent at 1285. We are unable to say that there was any error, particularly in light of No Doubt, which reasoned much the same as the district court in this case: "that the avatars appear in the context of a videogame that contains many other creative elements[ ] does not transform the avatars into anything other than exact depictions of No Doubt's members doing exactly what they do as celebrities." No Doubt, 122 Cal.Rptr.3d at 411.7 EA suggests that the fact that NCAA Football users can alter the characteristics of the avatars in the game is significant. Again, our dissenting colleague agrees. See Dissent at 1286-87. In No Doubt, the California Court of Appeal noted that Band Hero "d[id] not permit players to alter the No Doubt avatars in any respect." Id. at 410. The court went on to say that the No Doubt avatars "remain at all times immutable images of the real celebrity musicians, in stark contrast to the `fanciful, creative characters' in Winter and Kirby." Id. The court explained further:

Id. at 410-11. Judge Thomas says that "[t]he Court of Appeal cited character immutability as a chief factor distinguishing [No Doubt] from Winter and Kirby." Dissent at 1287. Though No Doubt certainly mentioned the immutability of the avatars, we do not read the California Court of Appeal's decision as turning on the inability of users to alter the avatars. The key contrast with Winter and Kirby was that in those games the public figures were transformed into "fanciful, creative characters" or "portrayed as ... entirely new character[s]." No Doubt, 122 Cal. Rptr.3d at 410. On this front, our case is clearly aligned with No Doubt, not with Winter and Kirby. We believe No Doubt offers a persuasive precedent that cannot be materially distinguished from Keller's case.8,9

The Third Circuit came to the same conclusion in Hart v. Electronic Arts, Inc., 717 F.3d 141 (3d Cir.2013). In Hart, EA faced a materially identical challenge under New Jersey right-of-publicity law, brought by former Rutgers quarterback Ryan Hart. See id. at 163 n. 28 ("Keller is simply [Hart] incarnated in California."). Though the Third Circuit was tasked with interpreting New Jersey law, the court looked to the transformative use test developed in California. See id. at 158 n. 23 (noting that the right-of-publicity laws are "strikingly similar ... and protect similar interests" in New Jersey and California, and that "consequently [there is] no issue in applying balancing tests developed in California to New Jersey"); see also id. at 165 (holding that "the Transformative Use Test is the proper analytical framework to apply to cases such as the one at bar"). Applying the test, the court held that "the NCAA Football ... games at issue ... do not sufficiently transform [Hart]'s identity to escape the right of publicity claim," reversing the district court's grant of summary judgment to EA. Id. at 170.

As we have, the Third Circuit considered the potentially transformative nature of the game as a whole, id. at 166, 169, and the user's ability to alter avatar characteristics, id. at 166-68. Asserting that "the lack of transformative context is even more pronounced here than in No Doubt," id. at 166, and that "the ability to modify the avatar counts for little where the appeal of the game lies in users' ability to play as, or alongside[,] their preferred players or team," id. at 168 (internal quotation marks omitted), the Third Circuit agreed with us that these changes do not render the NCAA Football games sufficiently transformative to defeat a right-ofpublicity claim.

Judge Ambro dissented in Hart, concluding that "the creative components of NCAA Football contain sufficient expressive transformation to merit First Amendment protection." Id. at 175 (Ambro, J., dissenting). But in critiquing the majority opinion, Judge Ambro disregarded No Doubt and Kirby because "they were not decided by the architect of the Transformative Use Test, the Supreme Court of California." Id. at 172 n. 4. He thus "d[id] not attempt to explain or distinguish the[se cases'] holdings except to note that [he] believe[s] No Doubt, which focused on individual depictions rather than the work in its entirety, was wrongly decided in light of the prior precedent in Comedy III and Winter." Id. We recognize that we are bound only by the decisions of a state's highest court and not by decisions of the state's intermediate appellate court when considering statelaw issues sitting in diversity jurisdiction. See In re Kirkland, 915 F.2d 1236, 1238-39 (9th Cir. 1990). Nonetheless, where there is no binding precedent from the state's highest court, we "must predict how the highest state court would decide the issue using intermediate appellate court decisions, decisions from other jurisdictions, statutes, treatises, and restatements as guidance." Id. at 1239 (emphasis added). As stated above, we believe No Doubt in particular provides persuasive guidance. We do not believe No Doubt to be inconsistent with the California Supreme Court's relevant decisions, and we will not disregard a well-reasoned decision from a state's intermediate appellate court in this context. Like the majority in Hart, we rely substantially on No Doubt, and believe we are correct to do so.

Given that NCAA Football realistically portrays college football players in the context of college football games, the district court was correct in concluding that EA cannot prevail as a matter of law based on the transformative use defense at the anti-SLAPP stage. Cf. Hilton, 599 F.3d at 910-11.10

B

EA urges us to adopt for right-of-publicity claims the broader First Amendment defense that we have previously adopted in the context of false endorsement claims under the Lanham Act: the Rogers test.11 See Brown v. Elec. Arts, 724 F.3d at 1239-41, 2013 WL 3927736, at *1-2 (applying the Rogers test to a Lanham Act claim brought by former NFL player Jim Brown relating to the use of his likeness in EA's Madden NFL video games).

Rogers v. Grimaldi is a landmark Second Circuit case balancing First Amendment rights against claims under the Lanham Act. 875 F.2d 994 (2d Cir.1989). The case involved a suit brought by the famous performer Ginger Rogers against the producers and distributors of Ginger and Fred, a movie about two fictional Italian cabaret performers who imitated Rogers and her frequent performing partner Fred Astaire. Id. at 996-97. Rogers alleged both a violation of the Lanham Act for creating the false impression that she endorsed the film and infringement of her common law right of publicity. Id. at 997.

The Rogers court recognized that "[m]ovies, plays, books, and songs are all indisputably works of artistic expression and deserve protection," but that "[t]he purchaser of a book, like the purchaser of a can of peas, has a right not to be misled as to the source of the product." Id. "Consumers of artistic works thus have a dual interest: They have an interest in not being misled and they also have an interest in enjoying the results of the author's freedom of expression." Id. at 998. The Rogers court determined that titles of artistic or literary works were less likely to be misleading than "the names of ordinary commercial products," and thus that Lanham Act protections applied with less rigor when considering titles of artistic or literary works than when considering ordinary products. Id. at 999-1000. The court concluded that "in general the Act should be construed to apply to artistic works only where the public interest in avoiding consumer confusion outweighs the public interest in free expression." Id. at 999. The court therefore held:

Id.

We first endorsed the Rogers test for Lanham Act claims involving artistic or expressive works in Mattel, Inc. v. MCA Records, Inc., 296 F.3d 894, 902 (9th Cir. 2002). We agreed that, in the context of artistic and literary titles, "[c]onsumers expect a title to communicate a message about the book or movie, but they do not expect it to identify the publisher or producer," and "adopt[ed] the Rogers standard as our own." Id. Then, in E.S.S. Entertainment 2000, Inc. v. Rock Star Videos, Inc., we considered a claim by a strip club owner that video game maker Rock Star incorporated its club logo into the game's virtual depiction of East Los Angeles, violating the club's trademark right to that logo. 547 F.3d 1095, 1096-98 (9th Cir.2008). We held that Rock Star's use of the logo and trade dress was protected by the First Amendment and that it therefore could not be held liable under the Lanham Act. Id. at 1099-1101. In so doing, we extended the Rogers test slightly, noting that "[a]lthough this test traditionally applies to uses of a trademark in the title of an artistic work, there is no principled reason why it ought not also apply to the use of a trademark in the body of the work." Id. at 1099.

In this case, EA argues that we should extend this test, created to evaluate Lanham Act claims, to apply to right-of-publicity claims because it is "less prone to misinterpretation" and "more protective of free expression" than the transformative use defense. Although we acknowledge that there is some overlap between the transformative use test formulated by the California Supreme Court and the Rogers test, we disagree that the Rogers test should be imported wholesale for right-of-publicity claims. Our conclusion on this point is consistent with the Third Circuit's rejection of EA's identical argument in Hart. See Hart, 717 F.3d at 154-58. As the history and development of the Rogers test makes clear, it was designed to protect consumers from the risk of consumer confusion — the hallmark element of a Lanham Act claim. See Cairns v. Franklin Mint Co., 292 F.3d 1139, 1149 (9th Cir. 2002). The right of publicity, on the other hand, does not primarily seek to prevent consumer confusion. See Hart, 717 F.3d at 158 ("[T]he right of publicity does not implicate the potential for consumer confusion...."). Rather, it primarily "protects a form of intellectual property [in one's person] that society deems to have some social utility." Comedy III, 106 Cal.Rptr.2d 126, 21 P.3d at 804. As the California Supreme Court has explained:

Id. 106 Cal.Rptr.2d 126, 21 P.3d at 804-05 (internal quotation marks and citations omitted).

The right of publicity protects the celebrity, not the consumer. Keller's publicity claim is not founded on an allegation that consumers are being illegally misled into believing that he is endorsing EA or its products. Indeed, he would be hard-pressed to support such an allegation absent evidence that EA explicitly misled consumers into holding such a belief. See Brown v. Elec. Arts, 724 F.3d at 1242-43, 2013 WL 3927736, at *4 (holding under the Rogers test that, since "Brown's likeness is artistically relevant to the [Madden NFL] games and there are no alleged facts to support the claim that EA explicitly misled consumers as to Brown's involvement with the games," "the public interest in free expression outweighs the public interest in avoiding consumer confusion"). Instead, Keller's claim is that EA has appropriated, without permission and without providing compensation, his talent and years of hard work on the football field. The reasoning of the Rogers and Mattel courts — that artistic and literary works should be protected unless they explicitly mislead consumers — is simply not responsive to Keller's asserted interests here. Cf. Hart, 717 F.3d at 157 ("Effectively, [EA] argues that [Hart] should be unable to assert a claim for appropriating his likeness as a football player precisely because his likeness was used for a game about football. Adopting this line of reasoning threatens to turn the right of publicity on its head.").

We recognize that Rogers also dealt with a right-of-publicity claim-one under Oregon law — and applied a modified version of its Lanham Act test in order to adapt to that particular context:

875 F.2d at 1004. However, the Rogers court was faced with a situation in which the "Oregon Courts ... [had] not determined the scope of the common law right of publicity in that state." Id. at 1002. In the absence of clear state-law precedent, the Rogers court was "obliged to engage in the uncertain task of predicting what the New York courts would predict the Oregon courts would rule as to the contours of a right of publicity under Oregon law." Id. In light of Comedy III and its progeny, we are faced with no such uncertain task.

Lastly, we note that the only circuit court to import the Rogers test into the publicity arena, the Sixth Circuit, has done so inconsistently. In Parks v. LaFace Records, the Sixth Circuit indicated that the Rogers test was appropriate for right-of-publicity claims, noting that the Restatement (Third) of Unfair Competition had endorsed use of the test in that context. 329 F.3d 437, 461 (6th Cir.2003) (citing Restatement (Third) of Unfair Competition § 47 cmt. c). Subsequently, in ETW Corp. v. Jireh Publishing, Inc., the court acknowledged the Parks decision but did not apply the Rogers test to the Ohio right-of-publicity claim in question. 332 F.3d at 915, 936 & n. 17 (6th Cir.2003). Instead, the court applied a balancing test from comment d in the Restatement (analyzing "the substantiality and market effect of the use of the celebrity's image ... in light of the informational and creative content"), as well as the transformative use test from Comedy III. Id. at 937-38; see Hart, 717 F.3d at 157 ("We find Parks to be less than persuasive [as to the applicability of the Rogers test to right-of-publicity cases] given that just over a month later another panel of the Sixth Circuit decided [ETW], a right of publicity case where the Circuit applied the Transformative Use Test."). Similarly, the Tenth Circuit in Cardtoons, L.C. v. Major League Baseball Players Ass'n, 95 F.3d 959 (10th Cir.1996), and the Eighth Circuit in C.B.C. Distribution and Marketing, Inc. v. Major League Baseball Advanced Media, L.P., 505 F.3d 818 (8th Cir.2007), rejected the Rogers test in favor of a flexible case-by-case approach that takes into account the celebrity's interest in retaining his or her publicity and the public's interest in free expression. Therefore, we decline EA's invitation to extend the Rogers test to right-of-publicity claims.

C

California has developed two additional defenses aimed at protecting the reporting of factual information under state law. One of these defenses only applies to common law right-of-publicity claims while the other only applies to statutory right-of-publicity claims. Montana v. San Jose Mercury News, Inc., 34 Cal.App.4th 790, 40 Cal.Rptr.2d 639, 640 (1995). Liability will not lie for common law right-of-publicity claims for the "publication of matters in the public interest." Id. at 640-41. Similarly, liability will not lie for statutory right-of-publicity claims for the "use of a name, voice, signature, photograph, or likeness in connection with any news, public affairs, or sports broadcast or account, or any political campaign." Cal. Civ.Code § 3344(d). Although these defenses are based on First Amendment concerns, Gill v. Hearst Publ'g Co., 40 Cal.2d 224, 253 P.2d 441, 443-44 (1953), they are not coextensive with the Federal Constitution, New Kids on the Block v. News Am. Publ'g, Inc., 971 F.2d 302, 310 n. 10 (9th Cir.1992), and their application is thus a matter of state law.

EA argues that these defenses give it the right to "incorporate athletes' names, statistics, and other biographical information" into its expressive works, as the defenses were "designed to create `extra breathing space' for the use of a person's name in connection with matters of public interest." Keller responds that the right of publicity yields to free use of a public figure's likeness only to the extent reasonably required to report information to the public or publish factual data, and that the defenses apply only to broadcasts or accounts of public affairs, not to EA's NCAA Football games, which do not contain or constitute such reporting about Keller.

California courts have generally analyzed the common law defense and the statutory defense separately, but it is clear that both defenses protect only the act of publishing or reporting. By its terms, § 3344(d) is limited to a "broadcast or account," and we have confirmed that the common law defense is about a publication or reporting of newsworthy items. Hilton, 599 F.3d at 912. However, most of the discussion by California courts pertains to whether the subject matter of the communication is of "public interest" or related to "news" or "public affairs," leaving little guidance as to when the communication constitutes a publication or reporting.

For instance, in Dora v. Frontline Video, Inc., a wellknown surfer sued the producer of a documentary on surfing entitled "The Legends of Malibu," claiming misappropriation of his name and likeness. 15 Cal.App.4th 536, 18 Cal.Rptr.2d 790, 791 (1993). The court held that the documentary was protected because it was "a fair comment on real life events which have caught the popular imagination." Id. at 792 (internal quotation marks omitted). The court explained that surfing "has created a lifestyle that influences speech, behavior, dress, and entertainment," has had "an economic impact," and "has also had a significant influence on the popular culture," such that "[i]t would be difficult to conclude that a surfing documentary does not fall within the category of public affairs." Id. at 794-95. Similarly, in Gionfriddo v. Major League Baseball, retired professional baseball players alleged that Major League Baseball violated their right of publicity by displaying "factual data concerning the players, their performance statistics, and verbal descriptions and video depictions of their play" in game programs and on its website. 94 Cal.App.4th 400, 114 Cal.Rptr.2d 307, 314 (2001). The court reasoned that "[t]he recitation and discussion of factual data concerning the athletic performance of these plaintiffs command a substantial public interest, and, therefore, is a form of expression due substantial constitutional protection." Id. at 315. And in Montana v. San Jose Mercury News, Inc., former NFL quarterback Joe Montana brought a right-of-publicity action against a newspaper for selling posters containing previously published pages from the newspaper depicting the many Super Bowl victories by Montana and the San Francisco 49ers. Montana, 40 Cal.Rptr.2d at 639-40. The court found that "[p]osters portraying the 49'ers' [sic] victories are ... a form of public interest presentation to which protection must be extended." Id. at 641 (internal quotation marks omitted).

We think that, unlike in Gionfriddo, Montana, and Dora, EA is not publishing or reporting factual data. EA's video game is a means by which users can play their own virtual football games, not a means for obtaining information about real-world football games. Although EA has incorporated certain actual player information into the game (height, weight, etc.), its case is considerably weakened by its decision not to include the athletes' names along with their likenesses and statistical data. 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.12

III

Under California's transformative use defense, EA's use of the likenesses of college athletes like Samuel Keller in its video games is not, as a matter of law, protected by the First Amendment. We reject EA's suggestion to import the Rogers test into the right-of-publicity arena, and conclude that statelaw defenses for the reporting of information do not protect EA's use.

AFFIRMED.

THOMAS, Circuit Judge, dissenting:

Because the creative and transformative elements of Electronic Arts' NCAA Football video game series predominate over the commercial use of the athletes' likenesses, the First Amendment protects EA from liability. Therefore, I respectfully dissent.

I

As expressive works, video games are entitled to First Amendment protection. Brown v. Entm't Merchs. Ass'n, ___ U.S. ___, 131 S.Ct. 2729, 2733, 180 L.Ed.2d 708 (2011). The First Amendment affords additional protection to NCAA Football because it involves a subject of substantial public interest: collegiate football. Moore v. Univ. of Notre Dame, 968 F.Supp. 1330, 1337 (N.D.Ind.1997). Because football is a matter of public interest, the use of the images of athletes is entitled to constitutional protection, even if profits are involved. Montana v. San Jose Mercury News, Inc., 34 Cal.App.4th 790, 40 Cal.Rptr.2d 639, 643 n. 2 (1995); see also Cal. Civ.Code § 3344(d) (exempting from liability the "use of a name ... or likeness in connection with any ... public affairs, or sports broadcast or account").

Where it is recognized, the tort of appropriation is a creature of common law or statute, depending on the jurisdiction. However, the right to compensation for the misappropriation for commercial use of one's image or celebrity is far from absolute. In every jurisdiction, any right of publicity must be balanced against the constitutional protection afforded by the First Amendment. Courts have employed a variety of methods in balancing the rights. See, e.g., Doe v. TCI Cablevision, 110 S.W.3d 363, 374 (Mo.2003) (en banc). The California Supreme Court applies a "transformative use" test it formulated in Comedy III Productions, Inc. v. Gary Saderup, Inc., 25 Cal.4th 387, 106 Cal.Rptr.2d 126, 21 P.3d 797 (2001).1

As the majority properly notes, the transformative use defense is "a balancing test between the First Amendment and the right of publicity based 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." Comedy III, 106 Cal.Rptr.2d 126, 21 P.3d at 799. The rationale for the test, as the majority notes, is that "when a work contains significant transformative elements, it is not only especially worthy of First Amendment protection, but it is also less likely to interfere with the economic interest protected by the right of publicity." Id. 106 Cal.Rptr.2d 126, 21 P.3d at 808.

The five considerations articulated in Comedy III, and cited by the majority, are whether: (1) the celebrity likeness is one of the raw materials from which an original work is synthesized; (2) the work is primarily the defendant's own expression if the expression is something other than the likeness of the celebrity; (3) the literal and imitative or creative elements predominate in the work; (4) the marketability and economic value of the challenged work derives primarily from the fame of the celebrity depicted; and (5) an artist's skill and talent has been manifestly subordinated to the overall goal of creating a conventional portrait of a celebrity so as to commercially exploit the celebrity's fame. Id. 106 Cal.Rptr.2d 126, 21 P.3d at 809-10.

Although these considerations are often distilled as analytical factors, Justice Mosk was careful in Comedy III not to label them as such. Indeed, the focus of Comedy III is a more holistic examination of whether the transformative and creative elements of a particular work predominate over commercially based literal or imitative depictions. The distinction is critical, because excessive deconstruction of Comedy III can lead to misapplication of the test. And it is at this juncture that I must respectfully part ways with my colleagues in the majority.

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. In my view, this approach contradicts the holistic analysis required by the transformative use test. See Hart v. Elec. Arts, Inc., 717 F.3d 141, 170-76 (3d Cir.2013) (Ambro, J., dissenting).2 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.

When EA's NCAA Football video game series is examined carefully, and put in proper context, I conclude that the creative and transformative elements of the games predominate over the commercial use of the likenesses of the athletes within the games.

A

The first step in conducting a balancing is to examine the creative work at issue. At its essence, EA's NCAA Football is a work of interactive historical fiction. Although the game changes from year to year, its most popular features predominately involve role-playing by the gamer. For example, a player can create a virtual image of himself as a potential college football player. The virtual player decides which position he would like to play, then participates in a series of "tryouts" or competes in an entire high school season to gauge his skill. Based on his performance, the virtual player is ranked and available to play at select colleges. The player chooses among the colleges, then assumes the role of a college football player. He also selects a major, the amount of time he wishes to spend on social activities, and practice — all of which may affect the virtual player's performance. He then plays his position on the college team. In some versions of the game, in another mode, the virtual player can engage in a competition for the Heisman Trophy. In another popular mode, the gamer becomes a virtual coach. The coach scouts, recruits, and develops entirely fictional players for his team. The coach can then promote the team's evolution over decades of seasons.

The college teams that are supplied in the game do replicate the actual college teams for that season, including virtual athletes who bear the statistical and physical dimensions of the actual college athletes. But, unlike their professional football counterparts in the Madden NFL series, the NCAA football players in these games are not identified.

The gamers can also change their abilities, appearances, and physical characteristics at will. Keller's impressive physical likeness can be morphed by the gamer into an overweight and slow virtual athlete, with anemic passing ability. And the gamer can create new virtual players out of whole cloth. Players can change teams. The gamer could pit Sam Keller against himself, or a stronger or weaker version of himself, on a different team. Or the gamer could play the game endlessly without ever encountering Keller's avatar. In the simulated games, the gamer controls not only the conduct of the game, but the weather, crowd noise, mascots, and other environmental factors. Of course, one may play the game leaving the players unaltered, pitting team against team. But, in this context as well, the work is one of historic fiction. The gamer controls the teams, players, and games.

Applying the Comedy III considerations to NCAA Football in proper holistic context, the considerations favor First Amendment protection. The athletic likenesses are but one of the raw materials from which the broader game is constructed. The work, considered as a whole, is primarily one of EA's own expression. The creative and transformative elements predominate over the commercial use of likenesses. The marketability and economic value of the game comes from the creative elements within, not from the pure commercial exploitation of a celebrity image. The game is not a conventional portrait of a celebrity, but a work consisting of many creative and transformative elements.

The video game at issue is much akin to the creations the California Supreme Court found protected in Winter v. DC Comics, 30 Cal.4th 881, 134 Cal.Rptr.2d 634, 69 P.3d 473, 476 (2003), where the two fabled guitarists Johnny and Edgar Winter were easily identifiable, but depicted as chimeras. It is also consistent with the California Court of Appeal's decision in Kirby v. Sega of America, Inc., 144 Cal.App.4th 47, 50 Cal.Rptr.3d 607, 609-10 (2006), where a character easily identified as singer Kierin Kirby, more popularly known as Lady Miss Kier, was transformed into a "`fanciful, creative character' who exists in the context of a unique and expressive video game." Id. at 618. So, too, are the virtual players who populate the world of the NCAA Football series.

No Doubt v. Activision Publishing, Inc., 192 Cal.App.4th 1018, 122 Cal.Rptr.3d 397 (2011), is not to the contrary. The literal representations in No Doubt were not, and could not be, transformed in any way. Indeed, in No Doubt, the bandmembers posed for motion-capture photography to allow reproduction of their likenesses, id. at 402, and the Court of Appeal underscored the fact that the video game did not "permit players to alter the No Doubt avatars in any respect" and the avatars remained "at all times immutable images of the real celebrity musicians," id. at 410. The Court of Appeal cited character immutability as a chief factor distinguishing that case from Winter and Kirby. Id. Unlike the avatars in No Doubt, the virtual players in NCAA Football are completely mutable and changeable at the whim of the gamer. The majority places great reliance on No Doubt as support for its proposition that the initial placement of realistic avatars in the game overcomes the First Amendment's protection, but the Court of Appeal in No Doubt rejected such a cramped construction, noting that "even literal reproductions of celebrities may be `transformed' into expressive works based on the context into which the celebrity image is placed." Id. at 410 (citing Comedy III, 106 Cal.Rptr.2d 126, 21 P.3d at 797).3

Unlike the majority, I would not punish EA for the realism of its games and for the skill of the artists who created realistic settings for the football games. Majority op. at 1279 n. 10. That the lifelike roar of the crowd and the crunch of pads contribute to the gamer's experience demonstrates how little of NCAA Football is driven by the particular likeness of Sam Keller, or any of the other plaintiffs, rather than by the game's artistic elements.

In short, considering the creative elements alone in this case satisfies the transformative use test in favor of First Amendment protection.

B

Although one could leave the analysis with an examination of the transformative and creative aspects of the game, a true balancing requires an inquiry as to the other side of the scales: the publicity right at stake. Here, as well, the NCAA Football video game series can be distinguished from the traditional right of publicity cases, both from a quantitative and a qualitative perspective.

As a quantitative matter, NCAA Football is different from other right of publicity cases in the sheer number of virtual actors involved. Most right of publicity cases involve either one celebrity, or a finite and defined group of celebrities. Comedy III involved literal likenesses of the Three Stooges. Hilton v. Hallmark Cards, 599 F.3d 894, 909-12 (9th Cir.2009), involved the literal likeness of Paris Hilton. Winter involved the images of the rock star brother duo. Kirby involved the likeness of one singer. No Doubt focused on the likenesses of the members of a specific legendary band.

In contrast, NCAA Football includes not just Sam Keller, but thousands of virtual actors. This consideration is of particular significance when we examine, as instructed by Comedy III, whether the source of the product marketability comes from creative elements or from pure exploitation of a celebrity image. 106 Cal.Rptr.2d 126, 21 P.3d at 810. There is not, at this stage of the litigation, any evidence as to the personal marketing power of Sam Keller, as distinguished from the appeal of the creative aspects of the product. Regardless, the sheer number of athletes involved inevitably diminish the significance of the publicity right at issue. Comedy III involved literal depictions of the Three Stooges on lithographs and T-shirts. Winter involved characters depicted in a comic strip. Kirby and No Doubt involved pivotal characters in a video game. The commercial image of the celebrities in each case was central to the production, and its contact with the consumer was immediate and unavoidable. In contrast, one could play NCAA Football thousands of times without ever encountering a particular avatar. In context of the collective, an individual's publicity right is relatively insignificant. Put another way, if an anonymous virtual player is tackled in an imaginary video game and no one notices, is there any right of publicity infringed at all?

The sheer quantity of the virtual players in the game underscores the inappropriateness of analyzing the right of publicity through the lens of one likeness only. Only when the creative work is considered in complete context can a proper analysis be conducted.

As a qualitative matter, the essence of NCAA Football is founded on publicly available data, which is not protected by any individual publicity rights. It is true that EA solicits and receives information directly from colleges and universities. But the information is hardly proprietary. Personal vital statistics for players are found in college programs and media guides. Likewise, playing statistics are easily available. In this respect, the information used by EA is indistinguishable from the information used in fantasy athletic leagues, for which the First Amendment provides protection, C.B.C. Distribution & Mktg., Inc. v. Major League Baseball Advanced Media, L.P., 505 F.3d 818, 823-24 (8th Cir.2007), or much beloved statistical board games, such as Strat-O-Matic. An athlete's right of publicity simply does not encompass publicly available statistical data. See, e.g., IMS Health Inc. v. Sorrell, 630 F.3d 263, 271-72 (2d Cir.2010) ("The First Amendment protects `[e]ven dry information, devoid of advocacy, political relevance, or artistic expression.'" (quoting Universal City Studios, Inc. v. Corley, 273 F.3d 429, 446 (2d Cir.2001)) (alteration in original)).4

Further, the structure of the game is not founded on exploitation of an individual's publicity rights. The players are unidentified and anonymous. It is true that third-party software is available to quickly identify the players, but that is not part of the EA package. And the fact that the players can be identified by the knowledgeable user by their position, team, and statistics is somewhat beside the point. The issue is whether the marketability of the product is driven by an individual celebrity, or by the game itself. Comedy III, 106 Cal.Rptr.2d 126, 21 P.3d at 810. Player anonymity, while certainly not a complete defense, bears on the question of how we balance the right of publicity against the First Amendment. This feature of the game places it in stark contrast with No Doubt, where the whole point of the enterprise was the successful commercial exploitation of the specifically identified, world-famous musicians.

Finally, as a qualitative matter, the publicity rights of college athletes are remarkably restricted. This consideration is critical because the "right to exploit commercially one's celebrity is primarily an economic right." Gionfriddo v. Major League Baseball, 94 Cal.App.4th 400, 114 Cal.Rptr.2d 307, 318 (2001). NCAA rules prohibit athletes from benefitting economically from any success on the field. NCAA Bylaw 12.5 specifically prohibits commercial licensing of an NCAA athlete's name or picture. NCAA, 2012-13 NCAA Division I Manual § 12.5.2.1 (2012). Before being allowed to compete each year, all Division I NCAA athletes must sign a contract stating that they understand the prohibition on licensing and affirming that they have not violated any amateurism rules. In short, even if an athlete wished to license his image to EA, the athlete could not do so without destroying amateur status. Thus, an individual college athlete's right of publicity is extraordinarily circumscribed and, in practical reality, nonexistent.5

In sum, even apart from consideration of transformative elements, examination of the right of publicity in question also resolves the balance in favor of the First Amendment. The quantity of players involved dilutes the commercial impact of any particular player and the scope of the publicity right is significantly reduced by the fact that: (1) a player cannot own the individual, publicly available statistics on which the game is based; (2) the players are not identified in the game; and (3) NCAA college athletes do not have the right to license their names and likenesses, even if they chose to do so.6

II

Given the proper application of the transformative use test, Keller is unlikely to prevail. The balance of interests falls squarely on the side of the First Amendment. The stakes are not small. The logical consequence of the majority view is that all realistic depictions of actual persons, no matter how incidental, are protected by a state law right of publicity regardless of the creative context. This logic jeopardizes the creative use of historic figures in motion pictures, books, and sound recordings. Absent the use of actual footage, the motion picture Forrest Gump might as well be just a box of chocolates. Without its historical characters, Midnight in Paris would be reduced to a pedestrian domestic squabble. The majority's holding that creative use of realistic images and personas does not satisfy the transformative use test cannot be reconciled with the many cases affording such works First Amendment protection.7 I respectfully disagree with this potentially dangerous and out-of-context interpretation of the transformative use test.

For these reasons, I respectfully dissent.