Le et al v. Zuffa, LLC, No. 2:2015cv01045 - Document 839 (D. Nev. 2023)

Court Description: ORDER Granting in part and Denying in part 518 Motion to Certify Class. IT IS FURTHER ORDERED that the Court will set a status conference for the parties to address Plaintiffs claims for injunctive relief pursuant to Federal Rule of Civil Proc edure 23(b)(3) and invite additional briefing if required.IT IS FURTHER ORDERED that the Court appoints named Plaintiffs as Class Representatives and appoints Interim Class Counsel as Class Counsel.Status Conference set for 8/21/2023 at 03:30 PM in LV Courtroom 7C before Judge Richard F. Boulware II. Signed by Judge Richard F. Boulware, II on 8/9/2023. (Copies have been distributed pursuant to the NEF - SLR)

Download PDF

Le et al v. Zuffa, LLC Doc. 839 Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 1 of 80 1 2 3 4 UNITED STATES DISTRICT COURT 5 DISTRICT OF NEVADA 6 *** 7 8 9 CUNG LE, NATHAN QUARRY, JON FITCH, BRANDON VERRA, LUIS JAVIER VASQUEZ, and KYLE KINGSBURY, on behalf of themselves and all others similarly situated, 10 13 ORDER Plaintiffs, 11 12 Case No. 2:15-cv-01045-RFB-BNW v. ZUFFA, LLC D/B/A ULTIMATE FIGHTING CHAMPIONSHIP AND UFC, Defendant. 14 15 16 17 18 19 20 21 I. INTRODUCTION Before the Court is Plaintiffs’ Motion to Certify Class. ECF No. 518. For the reasons stated below, the Court grants the motion in part and denies the motion in part. The Court grants the motion as to the Bout Class but denies the motion as to the Identity Class. The Court grants the motion with respect to Defendant’s unlawful use of its monopsony power in the Relevant Input Market of Fighter Services for live Mixed Martial Arts (“MMA”) bouts. 22 23 A. Parties 24 Plaintiffs and proposed class representatives Cung Le, Nathan Quarry, Jon Fitch, Brandon 25 Vera, Luis Javier Vazquez, and Kyle Kingsbury (collectively, “Plaintiffs”) are all current or former 26 fighters in the Ultimate Fighting Championship (“UFC”). ECF No. 208. Defendant Zuffa, LLC 27 (“Defendant” or “Zuffa”), doing business as the UFC, is the preeminent MMA event promoter in 28 the United States. It is headquartered in Las Vegas, Nevada. Dockets.Justia.com Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 2 of 80 1 B. 2 3 4 5 6 7 8 Overview of Alleged Antitrust Conduct Plaintiffs allege that Zuffa’s widespread success and dominance in MMA is due to anticompetitive behavior. Specifically, Plaintiffs allege that Zuffa did the following: a) used exclusive contracts with specific provisions to retain fighters within the UFC; b) used its market power in both the input and output markets to render its fighter contracts effectively perpetual, and c) acquired or drove out rival promoters. The alleged effect of Zuffa’s conduct (the alleged “Scheme”) was to establish such overwhelming market dominance that it could pay its fighters substantially less than what they would have been paid in a competitive market for their services. 9 C. 10 11 12 Class Definitions Plaintiffs seek to certify the following classes, for which all named Plaintiffs seek to be class representatives, except for Quarry for the Bout Class: 13 • Bout Class: All persons who competed in one or more live professional UFCpromoted MMA bouts taking place or broadcast in the United States from December 16, 2010 to June 30, 2017. The Bout Class excludes all persons who are not residents or citizens of the United States unless the UFC paid such persons for competing in a bout fought in the United States. • Identity Class: Each and every UFC Fighter whose Identity was expropriated or exploited by the UFC, including in UFC Licensed Merchandise and/or UFC Promotional Materials in the United States from December 16, 2010 to June 30, 2017. 14 15 16 17 18 19 20 The class period is from December 16, 2010 to June 30, 2017 (“Class Period”). 21 22 23 24 25 II. PROCEDURAL BACKGROUND On December 16, 2014, Plaintiffs filed their Complaint in the Northern District of California, asserting that Zuffa violated Section 2 of the Sherman Antitrust Act, 15 U.S.C. § 2. ECF No. 1. On June 2, 2015, the Northern District of California transferred the case to the District 26 of Nevada upon Zuffa’s motion. ECF No. 93. On June 11, 2015, this Court granted Plaintiffs’ 27 Motion to Consolidate Cases. ECF No. 101. On September 25, 2015, this Court denied Zuffa’s 28 -2- Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 3 of 80 1 Motion to Dismiss. ECF Nos. 186, 314. On December 18, 2015, Plaintiffs filed a consolidated 2 amended Class Action Complaint (“CAC”) with a jury demand. ECF No. 208. 3 On February 1, 2017, Zuffa filed motions for partial summary judgment. ECF Nos. 347, 4 348. The Court held hearings on the partial summary judgment motions on September 21, 2017. 5 ECF No. 492. The Court found that it was premature to rule on the motions for summary judgment 6 and denied them without prejudice. ECF No. 493. 7 On February 16, 2018, Plaintiffs filed the instant Motion to Certify the Class. ECF No. 8 518. In support of their motion for class certification, Plaintiffs retained Dr. Hal J. Singer,1 sports 9 economist Dr. Andrew Zimbalist,2 and labor economist Dr. Alan Manning3 to provide expert 10 reports.4 11 Defendant responded with its opposition to the Motion to Certify Class on April 6, 2018. 12 ECF No. 540. In support of its response, Defendant retained its own experts, Dr. Robert Topel,5 13 Dr. Paul Oyer,6 and Dr. Roger D. Blair,7 to generate competing reports. Plaintiffs replied on May 14 15 16 1 Dr. Hal J. Singer is a principal at Economists Incorporated, a Senior Fellow at George Washington University’s Institute for Public Policy and has served as an Adjunct Professor at Georgetown University’s McDonough School of Business. Prior to joining Economists Incorporated, he was Managing Partner at Navigant Economics, and before that, Chief Executive Officer of Empiris, a litigation and regulatory firm. 2 17 18 19 20 21 22 23 24 25 26 27 28 Dr. Andrew Zimbalist is a professor and chair of the Economics Department at Smith College in Northampton, Massachusetts. He has published twenty-six books and dozens of articles in the fields of sports economics, comparative economic systems, and development economics, and is a founding member and editorial board member of the Journal of Sports Economics. 3 Dr. Alan Manning is a professor of economics at the London School of Economics. He has published over 50 articles in peer reviewed journals. He was chair of his department from 2009-2012 and an editor of the Journal of Labor Economics. 4 Plaintiffs also retained Guy Davis, CPA who generated two reports for Plaintiffs about the capacity of Zuffa to have paid more in total compensation to the fighters during the Class Period than it actually did. Mr. Davis did not testify at the evidentiary hearings for the class certification motion. 5 Dr. Robert H. Topel is a professor of economics at the University of Chicago Booth School of Business. From 1993 to 2003 he served as the editor of the Journal of Political Economy, and from 1991 to 1993 he was a member of the Editorial Board of the American Economic Review. He is also a past founding editor of the Journal of Labor Economics. 6 Dr. Paul Oyer is a professor of economics at Stanford University’s Graduate School of Business. He is also the current editor-in-chief of the Journal of Labor Economics, and a Research Associate at the National Bureau of Economic Research. 7 Dr. Roger D. Blair is a Professor of Economics at the University of Florida in Gainesville, Florida. His major fields of academic concentration are antitrust economics, industrial organization, sports economics, and applied economics. He has authored, coauthored and edited over 200 books, book chapters, and articles in economics journal and law reviews, including chapters in the Hovenkamp & Areeda treatise Antitrust Law. -3- Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 4 of 80 1 30, 2018. ECF No. 554. 2 Zuffa also filed motions seeking to exclude the expert testimony of Hal Singer, Andrew 3 Zimbalist, Guy Davis, and Alan Manning on Daubert grounds. ECF Nos. 517, 522, 524. The Court 4 denied those motions without prejudice on September 27, 2018. ECF No. 600. At that time, the 5 Court stated that it would “consider the arguments in the Motions and Responses in its review of 6 the Motion to Certify Class.” ECF No. 601.8 7 The Court subsequently held evidentiary hearings regarding the motion for class 8 certification on August 26, 2019 through August 30, 2019, September 12 through September 13, 9 2019, and September 23, 2019. ECF Nos. 723, 725, 729, 735, 740, 746, 747. 10 The Court held a status conference on September 17, 2020 and another on April 9, 2021. 11 ECF Nos. 768, 814. The Court deferred ruling on the pending motions because of related antitrust 12 class action appellate litigation in the Ninth Circuit that had not been finalized. Then, on April 8, 13 2022, Plaintiffs filed a Motion for Leave to File Supplemental Authority in light of the Ninth 14 Circuit’s recent decision in Olean Wholesale Grocery Coop., Inc. v. Bumble Bee Foods LLC, 31 15 F.4th 651 (9th Cir. 2022). That motion was granted. This appellate decision became final upon 16 denial of certiorari by the United States Supreme court on November 14, 2022. The Court now 17 finds it appropriate to issue its final written ruling at this time. 18 19 20 21 22 23 24 25 III. FACTUAL BACKGROUND9 The Court makes the following factual findings based on its review of the parties’ submissions and evidence adduced at the hearings held in this case. The Court holds that such findings have been adequately established by the record in this case. To the extent the parties may have disputed a particular fact, the Court finds sufficient evidence has been adduced to allow it to resolve such disputes and make its findings here. 26 8 27 Today, the Court addresses the substance of these motions in this instant order. 9 28 In this factual finding section, the Court may use the present tense at times, but all findings should be understood to have occurred during the Class Period as this was and is the focus of the evidence and this Court’s findings. -4- Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 5 of 80 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 A. Mixed Martial Arts MMA is a combat sport that combines techniques from several disciplines, including boxing, wrestling, karate, muay thai, Brazilian jiu jitsu, and judo. Similar in many ways to boxing, modern MMA fights (or “bouts”) between competitors are timed rounds in which opponents use these fighting techniques to acquire points through the contact they make with the body of their opponent. A bout can also be won by either knocking out an opponent, or by forcing the opponent to submit (or “tap out”) through a wrestling hold or lock that causes extreme pain requiring submission. In the event that neither fighter capitulates by the end of the last round, the winner is determined by a scoring system employed by judges based on the various types of contact or blows by each fighter on the other. Fighters are divided into ten major weight classes. The top fighter of each division is awarded a championship belt. In a title fight, a challenger is matched against the defending champion. Live MMA events are produced by promoters such as the UFC, which plan and execute MMA bouts for commercial entertainment (“Live MMA Events”). Live MMA Events are multihour productions consisting of a series of bouts between pre-matched fighters. This series of bouts (the “fight card”) occurs sequentially from the bottom, with the most anticipated fight of the night generally occurring last (the “top of the card”). Unlike in boxing, MMA championships are promoter specific. This is to say that the UFC awards championship belts to fighters within its own organization. There are no independent sanctioning organizations that either declare champions or regulate the rules as to who is declared a champion. The UFC does not cross-promote Live MMA Events with other MMA promoters. There are third-party ranking systems, however, that rank all MMA fighters within weight classes and on a “pound for pound” basis. B. Zuffa 26 Zuffa is a Nevada limited liability company founded in 2000 and headquartered in Las 27 Vegas, Nevada. Lorenzo and Frank Fertitta purchased the UFC for $2 million in 2001. Zuffa was 28 then sold to William Morris Endeavor-International Management Group (“WME-IMG”) in -5- Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 6 of 80 1 August 2016 for approximately $4 billion.10 Prior to its sale to WME-IMG, Zuffa was privately 2 owned and operated by a small group of individuals, initially consisting of Lorenzo and Frank 3 Fertitta, who each held 45% equity in the venture, and Dana White, who owned the remaining 4 10%.11 Lorenzo Fertitta functioned as the UFC’s CEO and Mr. White as its President. According 5 to Lorenzo Fertitta, Zuffa generated over $600 million in revenue in 2015. 6 Zuffa is in the business of promoting live Elite Professional MMA bouts in the United 7 States and elsewhere, under the trade names of the Ultimate Fighting Championship® or UFC®. 8 Under the UFC trademark, which is wholly owned by Zuffa, Zuffa promotes professional MMA 9 events for live audiences as well as live television, Internet, and PPV broadcasts. Additionally, it 10 licenses, markets, sells, and distributes UFC Licensed Merchandise, and Promotional Materials, 11 including, but not limited to, tickets to bouts, live and taped television programming, broadcasts 12 over an Internet subscription service, sponsorships and other merchandise, including video games, 13 action figures, gyms, fitness products, athletic equipment, apparel, footwear, hats, photographs, 14 toys, collectibles, trading cards, and digital media products. 15 During the Class Period, Zuffa earned its revenue primarily from four different sources: 16 (1) selling access to live events and taped content to a variety of media outlets, such as residential 17 and commercial pay-per-view (“PPV”), cable and broadcast television, video-on-demand, and the 18 UFC Fight Pass online subscription service, (2) ticket sales generated from Live MMA Event 19 attendance, (3) income generated from sponsorships, branding, and the advertising of products 20 during its events, and (4) revenues and royalties generated from UFC branded merchandise such 21 as video games. 22 23 24 25 26 27 28 10 This fact, along with the non-redacted versions of the instant motion for class certification, the accompanying briefs, and the numerous exhibits and reports attached to these moving papers, were filed under seal. “Unless a particular court record is one traditionally kept secret, a strong presumption in favor of access is the starting point. In keeping with the strong public policy favoring access to court records, most judicial records may be sealed only if the court finds compelling reasons.” Oliner v. Kontrabecki, 745 F.3d 1024, 1025 (9th Cir. 2014) (citation omitted). This Order relies on many of the sealed records cited in the moving papers to make its findings and explain its reasoning. Given the public’s significant interest in this Order’s substance, the Court now finds that the strong public policy favoring access to portions of these court records outweighs any prior compelling reason to seal any motions, briefs, and exhibits related to Plaintiff’s request for class certification. See Oliner, 745 F.3d at 1025 (“[C]ourt records often provide important, sometimes the only, bases or explanations for a court’s decision.”). For consistency, this order refers to Defendant entity as “Zuffa” for its presale and postsale (to WMEIMG) conduct. 11 -6- Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 7 of 80 1 C. 2 3 4 5 6 7 8 9 10 11 Fighters are treated as independent contractors not employees. Generally, fighters are only compensated upon their participation in a bout; and are not paid when they are not fighting. They receive no salary or benefits. Importantly, Zuffa does not pay for a fighter’s training, or any other expenses related to a fighter’s development or maintenance of their respective fighting skills. A fighter’s compensation terms are defined in a Promotional and Ancillary Rights (“PAR”) Agreement that Zuffa requires all its fighters to execute. Fighters do not receive compensation from Zuffa without signing such a contract. Each contract generally governs a fighter’s relationship with Zuffa from the moment of execution. Prior to any bout, a UFC fighter also signs a “Bout Agreement” which affirms event-specific compensation details. If one does not sign these contracts with Zuffa, one will not be able to fight in a Zuffa bout in any capacity. 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 Fighter Contracts and Compensation In the UFC, fighter pay is dependent on, among other things, a fighter’s bout participation. Typically, fighters receive a predetermined “show” amount to participate and perform in a bout, with additional compensation in the form of a “win” payment if the fighter is declared the victor. Show and win payments are the same amount; winning a bout typically doubles a fighter’s base pay.12 Zuffa utilized a tiered system with respect to pay to maintain consistency for compensating fighters. Compensation is typically structured such that if a fighter wins the first fight in the contract, they move up to the next pay tier in their subsequent fight. If a fighter loses the fight, however, they will maintain the original level of pay. Zuffa had specific employees, such as Joe Silva, who were tasked with the responsibility of maintaining consistency within Zuffa’s tiered compensation structure for its fighters. This is to say that Zuffa sought to and did maintain a consistent structured payment system for its fighters in the Class Period. UFC fighters may also, in very limited cases, receive discretionary, performance-based bonuses. For example, certain top-tier fighters, such as defending champions, may receive a share of the total revenues generated by the event, usually through a share of PPV revenue. This only 27 28 The base pay is the amount of the “show” payment: the minimum amount of money a fighter receives for participation in a bout. 12 -7- Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 8 of 80 1 occurs for a very small number of fighters and is determined on a bout-by-bout basis. 2 Fighters typically sign contracts with Zuffa either for a minimum period of time or for a 3 specific number of bouts. Importantly, and pursuant to the contracts, the timing of bouts and 4 matching of opponents was determined unilaterally by Zuffa. Consequently, this means that the 5 PAR Agreements generally do not guarantee, for instance, a minimum number of fights per year 6 and hence do not guarantee a minimum salary for each fighter. 7 8 9 Zuffa’s contracts contain several different clauses. The following clauses are the most salient ones at issue in this lawsuit: • Exclusion clause: requires a fighter to fight exclusively for Zuffa and prohibits a fighter from working with other MMA promoters. 10 11 • Cut clause: allows Zuffa to cut a fighter that loses a bout. 12 • Right-to-Match clause: allows Zuffa, up to a year after its contract(s) with a fighter 13 has “expired,” to match a competing promoter’s offer.13 Zuffa can elect to retain a 14 fighter by matching the competitor’s terms. These clauses typically prohibit the 15 competitor from countering, as a fighter is contractually obligated to accept Zuffa’s 16 matching terms. 17 • Champion’s clause: if a fighter becomes a title-holding champion at any point 18 during the term of the contract, the champion’s clause allows Zuffa to automatically 19 extend the contract for one year or three bouts. 20 • Retirement clause: permits Zuffa to suspend a fighter’s contract indefinitely. This 21 prevents fighters from coming out of retirement to fight for competitors, even after 22 many years. 23 • a fighter refuses an opponent. 24 25 Promotion clause: allows Zuffa to extend a fighter’s contract term in the event that • Tolling clause(s): allows Zuffa to extend the Term of an agreement for the period 26 27 28 13 Zuffa’s right-to-match clause grants Zuffa the right-to-match promoters’ offers for a limited period of time after the expiration of a fighter’s agreement. A fighter’s contract typically expires “on the earlier of (i) Fifteen (15) months after the first bout promoted by Zuffa involving Fighter under this Agreement; or (ii) the date on which Fighter has participated in at least three (3) Bouts promoted by ZUFFA under this Agreement (“Termination Date”), unless terminated sooner or extended further pursuant to the provisions of this Agreement.” -8- Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 9 of 80 1 of time that a fighter is unable or unwilling to compete in a UFC bout to cover 2 situations such as injuries or if a fighter believes that an opponent who is being 3 offered is unacceptable or inappropriate. • 4 Exclusive Negotiation clause: requires a fighter to negotiate exclusively with Zuffa for renewal of their contracts for a period of 30 to 90 days.14 5 6 During the Class Period, four individuals were exclusively responsible for negotiating 7 contracts with fighters: Mr. Silva,15 Sean Shelby, Mr. White, and Lorenzo Fertitta. Of these four 8 individuals, Mr. White and Mr. Fertitta only dealt with a select number of the highest ranked 9 fighters. Most contracts were negotiated by Mr. Silva and Mr. Shelby, who were known by the 10 UFC as “matchmakers.” In this capacity, Mr. Silva and Mr. Shelby were responsible for 11 contracting with fighters for different weight classes. For instance, Mr. Shelby handled contracting 12 for all fighters in the 145, 135, and 125-pound men and women’s divisions. Mr. Silva and Mr. 13 Shelby were also responsible for pairing fighters up in bouts and determining when fighters would 14 participate in a bout. 15 When signing new fighters to contracts, Mr. Silva and Mr. Shelby considered objective 16 factors, such as the ranking, popularity, relative need (based on geographic location and weight 17 class staffing) to determine the structure of their show/win payments over the multiple bouts in the 18 initial contract. While Mr. Silva and Mr. Shelby also took into account one or two subjective 19 factors, such as “potential” when signing fighters, the objective factors provided the primary basis 20 for setting fighter compensation. 21 D. Rival Promoters 22 23 24 25 Five years after it acquired the UFC, Zuffa purchased the World Fighting Alliance (“WFA”) and World Extreme Cagefighting (“WEC”) in 2006. Both purchases were announced in December of that year. The following year, Zuffa bought Pride Fighting Championships (“Pride”). 26 27 14 28 15 This clause was not used in Zuffa contracts from 2011-2014. It was reinstated for use in 2014. Mr. Silva provided testimony at the evidentiary hearing on September 23, 2019 as a former employee of Zuffa. Mr. Silva worked for the UFC for a total of 22 years and worked for Zuffa for 15 of those years. -9- Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 10 of 80 1 As a U.S.-based MMA clothing company, Affliction Clothing was one of the primary 2 clothing sponsors of Zuffa fighters. Backed by prominent investors, Affliction put on two 3 successful Live MMA Events. Zuffa terminated its sponsorship relationship with Affliction in 4 early 2008 after it received information that the company was planning on entering the MMA 5 promoter business. Further, Zuffa barred Affliction from sponsoring any of its events or any of its 6 fighters. In turn, Affliction exited the MMA promotion business, based in large part on its loss of 7 business with Zuffa. Zuffa then purchased Affliction’s assets and fighters in 2009. Finally, in March 2011, Zuffa purchased Strikeforce, the second largest MMA promoter at 8 9 the time. 10 After the acquisition of Strikeforce, Zuffa’s primary domestic competitor during the Class 11 Period, was Bellator, a promoter with a deal with Viacom to broadcast on Spike TV. Zuffa also 12 faced international competition from pan-Asian promoter, ONE Championship and Russian 13 promoter, Absolute Championship Berkut. 14 There are several additional promoters who act as so-called “minor leagues” to the UFC, 15 which essentially develop fighters to eventually farm out to the UFC. It is common that contracts 16 with these “minor league” promoters, as well as many international promoters, have “UFC-out” 17 clauses, which allow fighters to exit their contract if they receive an offer from the UFC. 18 19 IV. LEGAL STANDARD 20 21 22 23 24 25 26 27 28 The party seeking to certify a class under Federal Rule of Civil Procedure 23 bears the burden of showing that they have satisfied each of the four requirements of Rule 23(a) and at least one of the three requirements of Rule 23(b). Comcast Corp. v. Behrend, 569 U.S. 27, 33 (2013). Rule 23(a) provides four requirements that must be met in any class action: (1) “the class is so numerous that joinder of all members is impracticable”; (2) “there are questions of law or fact common to the class”; (3) “the claims or defenses of the representative parties are typical of the claims or defenses of the class”; and (4) “the representative parties will fairly and adequately protect the interests of the class.” Fed. R. Civ. P. 23(a). As to Rule 23(b), a plaintiff need only show that any one of the three provisions is satisfied. - 10 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 11 of 80 1 See Fed. R. Civ. P. 23(b)(1)-(3). Here, Plaintiffs seek certification of the proposed classes pursuant 2 to Rule 23(b)(3) and therefore must demonstrate: first, that “questions of law or fact common to 3 class members predominate over any questions affecting only individual members,” and second, 4 that “a class action is superior to other available methods for fairly and efficiently adjudicating the 5 controversy.” Fed. R. Civ. P. 23(b)(3). 6 “Rule 23 does not set forth a mere pleading standard.” Wal-Mart Stores, Inc. v. Dukes, 564 7 U.S. 338, 350 (2011). Rather, “[a] party seeking class certification must affirmatively demonstrate 8 his compliance with the Rule—that is, he must be prepared to prove that there are in fact 9 sufficiently numerous parties, common questions of law or fact, etc.” Id. The Court must engage 10 in a “rigorous analysis,” often requiring some evaluation of the “merits of the plaintiff’s underlying 11 claim,” before finding that the prerequisites for certification have been satisfied. Id.; see also 12 Comcast, 569 U.S. at 33-34. “Although some inquiry into the substance of a case may be 13 necessary[,]” the court should not advance a decision on the merits at the class certification stage. 14 Staton v. Boeing Co., 327 F.3d 938, 954 (9th Cir. 2003) (citations and internal quotation marks 15 omitted). “Merits questions may be considered to the extent—but only to the extent—that they are 16 relevant to determining whether Rule 23 prerequisites for class certification are satisfied.” Amgen 17 Inc. v. Conn. Ret. Plans & Trust Funds, 568 U.S. 455, 466 (2013). Plaintiffs must demonstrate that they meet the requirements of Rule 23 by a preponderance 18 19 of the evidence. Olean, 31 F.4th at 664. 20 21 22 23 24 25 V. DISCUSSION The Court will first address Defendant’s motion to exclude Plaintiffs’ experts’ testimony. Second, the Court will examine the Bout and Identity classes for purposes of class certification. Finally, the Court will evaluate the appointment of class counsel and class representatives. 26 A. Expert Evidence at the Class Certification Stage 27 In a class proceeding, the defendants may challenge the reliability of an expert’s evidence 28 under Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579 (1993), and Rule 702 of the Federal - 11 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 12 of 80 1 Rules of Evidence. See Tyson Foods, Inc. v. Bouaphakeo, 577 U.S. 442, 459 (2016); see also 2 Comcast, 569 U.S. at 32 n.4., Olean, 31 F.4th at 665, Sali v. Corona Reg’l Med. Ctr., 909 F.3d 3 996, 1006 (9th Cir. 2018). Under Daubert, “the district court judge must ensure that all admitted 4 expert testimony is both relevant and reliable.” Wendell v. GlaxoSmithKline LLC, 858 F.3d 1227, 5 1232 (9th Cir. 2017). “Scientific evidence is reliable if the principles and methodology used by an 6 expert are grounded in the methods of science.” Id. “The focus of the district court’s analysis must 7 be solely on principles and methodology, not on the conclusions that they generate,” and the 8 court’s task “is to analyze not what the experts say, but what basis they have for saying it.” Id. 9 (emphasis added). In conducting this analysis, the district court may consider “whether the theory 10 or technique employed by the expert is generally accepted in the scientific community; whether 11 it’s been subjected to peer review and publication; whether it can be and has been tested; and 12 whether the known or potential rate of error is acceptable.” Id.; see also Grodzitsky v. Am. Honda 13 Motor Co., 957 F.3d 979, 984-85 (9th Cir. 2020). 14 Importantly, at the class certification stage, the determination of whether expert evidence 15 is capable of resolving a class-wide question may include “weighing conflicting expert testimony” 16 and “resolving expert disputes.” Olean, 61 F.4th at 666 (internal citations omitted). This is to say 17 that district courts deciding class certification should resolve any disputes between the parties’ 18 experts whose resolution is necessary to the certification of the respective class. Id. 19 On the same day that Plaintiffs filed their motion for class certification, Defendant moved 20 to exclude the expert testimony of two of Plaintiffs’ experts, Dr. Zimbalist and Dr. Singer, on the 21 grounds that they did not meet the requirements of Daubert and Federal Rule of Evidence 702. See 22 ECF Nos. 522, 524. On September 27, 2018, the Court denied these motions without prejudice as 23 premature, writing that it would “consider the arguments in the Motions and Responses in its 24 review of the Motion to Certify Class.” ECF No. 600. 25 For the reasons discussed at length below and which are based upon the submissions of the 26 parties and the expert testimony presented at the evidentiary hearings, the Court finds that the 27 opinions of Dr. Singer and Dr. Zimbalist satisfy the Daubert and Federal Rule of Evidence 702 28 standards. Their reports, evidence, and testimony are based on accepted and reliable scientific - 12 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 13 of 80 1 methods and principles. In fact, their methodology, especially the use of multivariate regression 2 analysis, has long been accepted as reliable in econometric analysis. It has also been subject to 3 extensive peer review. The Court thus finds, based on the reports and testimony of all the experts, 4 that it is undisputed that the multivariate regression analysis used by Plaintiffs is a reliable, 5 scientific methodology used by social scientists, including econometricians who study monopolies 6 and monopsonies. Moreover, the Court finds that the mere fact that Plaintiffs’ experts’ analysis is 7 being conducted regarding a set of facts in a new market or industry does not alter or undermine 8 the generally accepted nature of their methodological techniques. As noted, Defendant’s experts 9 undisputedly rely on the same scientific methods (multivariate regression modeling). 10 Furthermore, the Court finds, based upon the record, including Defendant’s experts’ 11 testimony, that such methods are accepted in the field of econometrics when applied to industries 12 for which there may not have been previous significant regression analysis. In fact, all the experts’ 13 testimony taken together supports the conclusion that evaluating these methodologies for 14 reliability or validity is not topic specific. Rather, in this context, the reliability of these 15 methodologies can be ascertained from various testing as to the internal accuracy or confidence of 16 the respective model and consideration of the relative explanatory power or depth of the model. 17 Indeed, Defendant appears to contend that regression models may not be deployed in new 18 or emerging industries, or with new or differently defined variables. The Court disagrees. If 19 Defendant was correct, econometrics and regression analysis could never have developed into the 20 reliable disciplines and methodologies they are today, and certainly would not be used in antitrust 21 cases. See Olean, 31 F.4th at 677 (“In antitrust cases, regression models have been widely accepted 22 as a generally reliable econometric technique to control for the effects of the differences among 23 class members and isolate the impact of the alleged antitrust violations on . . . class members.”). 24 Defendant’s argument misstates the fundamental principles upon which regression analysis is 25 based. To reiterate, this methodology is not topic specific; it is applied in different contexts. The 26 explanatory power of a given regression model is evaluated on a case-by-case basis, to ensure that 27 a relationship is not found when it does not exist and a relationship between variables is not missed 28 when it does exist. See, e.g., Phillip Johnson, Edward Leamer, Jeffrey Leitzinger, Statistical - 13 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 14 of 80 1 Significance and Statistical Error in Antitrust Analysis, 81 Antitrust L. J. 641 (2017). Both sets of 2 experts acknowledged the fact that the reliability of a regression model cannot be judged by its 3 topic. 4 Thus, the Court does not find that the statistical methodologies employed by Plaintiffs’ 5 experts were unreliable or not established. The Court’s rigorous Rule 23 analysis of Plaintiffs’ 6 theory and methodology below provides further detailed reasoning for this finding. 7 8 9 10 B. The Bout Class Plaintiffs seek to certify a class identified as the “bout class,” pursuant to Federal Rule of Civil Procedure 23(a), 23(b)(2), and 23(b)(3). The Bout Class is defined by Plaintiffs as containing: 13 Bout Class: All persons who competed in one or more live professional UFC-promoted MMA bouts taking place or broadcast in the United States from December 16, 2010 to June 30, 2017. The Bout Class excludes all persons who are not residents or citizens of the United States unless the UFC paid such persons for competing in a bout fought in the United States. 14 The Court begins by discussing whether the Bout Class meets the requirements set forth by Rules 15 23(a) and 23(b)(3) and then turns to the issue of Rule 23(b)(2)’s requirements. 11 12 16 1. Rule 23(a) Requirements 17 18 a. Numerosity 19 Rule 23(a)(1) requires the party seeking certification to show the “class is so numerous that 20 joinder of all members is impracticable.” Fed. R. Civ. P. 23(a)(1). “There is no specific number of 21 class members required.” Johnson v. City of Grants Pass, 50 F.4th 787, 803 (9th Cir. 2022); 22 “However, proposed classes of less than fifteen are too small while classes of more than sixty are 23 sufficiently large.” Id.; see also Rannis v. Recchia, 380 F. App’x 646, 651 (9th Cir. 2010) (While 24 “[t]he numerosity requirement is not tied to any fixed numerical threshold,” courts generally “find 25 the numerosity requirement satisfied when a class includes at least 40 members.”); 5 James Wm. 26 Moore et al., Moore’s Federal Practice, § 23.22[1] (3d ed. 2023). 27 The Bout Class contains over 1,200 members. Additionally, class members reside 28 throughout the United States. This large number of members spread across the country would - 14 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 15 of 80 1 make joinder impracticable, and thus numerosity is satisfied. See Harik v. Cal. Teachers Ass’n, 2 326 F.3d 1042, 1051-52 (9th Cir. 2003) (finding numerosity satisfied where the “members exceed 3 sixty, and the defendants never presented any reason to the district court why they did not meet 4 the numerosity requirement”). 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 b. Commonality The commonality requirement of Rule 23(a)(2) requires that “class members’ claims ‘depend upon a common contention’ such that ‘determination of its truth or falsity will resolve an issue that is central to the validity of each [claim] in one stroke.’” Mazza v. Am. Honda Motor Co., 666 F.3d 581, 588 (9th Cir. 2012) (quoting Dukes, 564 U.S. at 350), overruled on other ground by Olean, 31 F.4th. at 682, n.32. The common issues do not need to be legally and factually identical. Rather, the “common questions may center on ‘shared legal issues with divergent factual predicates [or] a common core of salient facts coupled with disparate legal remedies.’” Jimenez v. Allstate Ins. Co., 765 F.3d 1161, 1165 (9th Cir. 2014) (quoting Hanlon v. Chrysler Corp., 150 F.3d 1011, 1019 (9th Cir. 1998)); see also Dukes, 564 U.S. at 359 (“[e]ven a single [common] question will do”). In other words, the inquiry is whether resolution of a common issue will “drive the resolution of the litigation.” Jimenez, 765 F.3d at 1165. The Rule 23(a)(2) commonality and Rule 23(b)(3) predominance inquiries often overlap, creating a “fuzzy line” separating the two determinations. In re Capacitors Antitrust Litig., No. 17md-2801-JD, 2018 U.S. Dist. LEXIS 195310, at *47 (N.D. Cal. Nov. 14, 2018). To conduct the necessary “rigorous analysis,” many courts consider both together, folding the two determinations into one. See, e.g., id. at *45-46. Here, Defendants focus their arguments on predominance, rather than commonality. Because the predominance inquiry is “more demanding than Rule 23(a),” Comcast, 569 U.S. at 34, the Court focuses its analysis on the predominance determination. For the same reasons the Court finds below that predominance has been established, the Court concludes that commonality has also been established. 28 - 15 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 16 of 80 1 c. Typicality 2 Under Rule 23(a)(3), a party seeking class certification must show “the claims or defenses 3 of the representative parties are typical of the claims or defenses of the class.” “The test of 4 typicality ‘is whether other members have the same or similar injury, whether the action is based 5 on conduct which is not unique to the named plaintiffs, and whether other class members have 6 been injured by the same course of conduct.’” Ellis v. Costco Wholesale Corp., 657 F.3d 970, 984 7 (9th Cir. 2011) (quoting Hanon v. Dataproducts Corp., 976 F.2d 497, 508 (9th Cir. 1992)). 8 Defendant challenges typicality on the ground that fighters with greater popularity and 9 prominence had negotiating leverage that the named Plaintiffs did not. Defendant also argues that, 10 pursuant to Plaintiffs’ expert’s model, fourteen fighters with the most bargaining power actually 11 benefitted from the alleged Scheme, rather than experience any adverse impact from it. This 12 argument is unavailing. Claims need only be “reasonably coextensive” with absent class members, 13 not “substantially identical.” Just Film, Inc. v. Buono, 847 F.3d 1108, 1116 (9th Cir. 2017). A 14 small percentage of class members who are unharmed or who are helped by the alleged unlawful 15 conduct does not defeat typicality. Here, the Court finds that there is typicality between the name 16 Plaintiffs and class members in terms of alleged harm by the Scheme, with only a de minimis 17 number of potential class members not being impacted by the alleged Scheme. Indeed, according 18 to Plaintiffs’ expert testimony, which the Court credits, the unharmed class members make up only 19 between 0.4 and 0.9% of the class. This de minimis number of class members, who may have been 20 unharmed, does not defeat typicality here. 21 Lastly, Defendant challenges typicality on the grounds that Defendant has individualized 22 defenses to claims brought by Mr. Quarry, Mr. Vera, Mr. Le, Mr. Vazquez, Mr. Fitch, and Mr. 23 Kingsbury. But Defendant fails to show that these defenses “threaten to become the focus of the 24 litigation.” Hanon, 976 F.2d at 508 (collecting cases). Defendant raises a variety of individualized 25 defenses to the named Plaintiffs’ claims related to the way these individuals participated in or, in 26 some cases, ultimately exited their contracts from the UFC. Defendant has not established that the 27 individual defenses undermine the typicality of the nature of the injury alleged in the CAC: the 28 suppression of compensation due to monopsonistic, anticompetitive behavior. Plaintiffs have - 16 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 17 of 80 1 established that, regardless of how many bouts an individual fought in the UFC or how they 2 ultimately were able to exit that contract, their wages were still suppressed. This is particularly the 3 case in the context of Plaintiffs’ Sherman Act § 2 claims, where the focus of the analysis is on the 4 firm’s behavior, not Plaintiffs’ individual behavior. Because of this, Defendant’s individualized 5 defenses do not undermine the typicality of the experience of the named Plaintiffs. 6 d. Adequacy 7 Rule 23(a)(4) requires that “the representative parties will fairly and adequately protect the 8 interests of the class.” The adequacy inquiry turns on (1) whether the “named plaintiffs and their 9 counsel have any conflicts of interest with other class members” and (2) whether “the named 10 plaintiffs and their counsel [will] prosecute the action vigorously on behalf of the class.” Hanlon, 11 150 F.3d at 1020. “Adequate representation depends on, among other factors, an absence of 12 antagonism between representatives and absentees, and a sharing of interest between 13 representatives and absentees.” Ellis, 657 F.3d at 985 (citing Molski v. Gleich, 318 F.3d 937, 955 14 (9th Cir. 2003), overruled on other grounds by Dukes v. Wal-Mart Stores, Inc., 603 F.3d 571, 617 15 (9th Cir. 2010)). 16 Defendant argues that, because some of the named Plaintiffs are former fighters with the 17 UFC, their interests will conflict with the interests of current UFC fighters. Former fighters, for 18 instance, will tend to prefer to maximize monetary damages over injunctive relief. The Court finds 19 reliance on this argument is misplaced for two reasons. First, all class members, whether they are 20 former or current UFC fighters, have the shared goal of being appropriately compensated for their 21 services during the period in which Defendant allegedly suppressed their wages. Cf. Glass v. UBS 22 Fin. Servs. Inc., 331 F. App’x 452, 455 (9th Cir. 2009) (affirming class of former and current 23 employees who had shared interest in compensation for allegedly unlawful practices); Hisps. 24 United of DuPage Cnty. v. Vill. of Addison, 160 F.R.D. 681, 690 (N.D. Ill. 1995) (finding fact that 25 some plaintiffs may be entitled to monetary damages does not create conflict between those 26 plaintiffs and others who would be limited to injunctive relief). Any potential speculative conflict 27 is overridden by this shared goal. See Hispanics United, 160 F.RD. at 690 (“In any event, the 28 purported conflict, if there be any, is substantially outweighed by the class members’ overriding - 17 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 18 of 80 1 common interest.”). Defendant offers no evidence that pursuit of monetary damages by former 2 fighters would harm class members who are current fighters.16 3 Further, the Court finds any potential conflict in this case is speculative at this stage. See 4 Cummings v. Connell, 316 F.3d 886, 896 (9th Cir. 2003) (noting that “this circuit does not favor 5 denial of class certification on the basis of speculative conflicts”); see also In re Online DVD- 6 Rental Antitrust Litig., 779 F.3d 934, 942 (9th Cir. 2015). Defendant’s argument that Plaintiffs 7 who are not presently employed by Zuffa have incentive to hurt Zuffa is conjectural. The bottom 8 line is that conflicts among class members that may arise at the damages stage of the litigation 9 does not demand denial of class certification. Blackie v. Barrack, 524 F.2d 891, 905 (9th Cir. 10 1975). Accordingly, the Court finds that the proposed class representatives can meet the adequacy 11 requirement.17 12 In conclusion, the Court holds that the Rule 23(a) requirements of numerosity, 13 commonality, typicality, and adequacy are all met with respect to the Bout Class. The Court now 14 addresses the Rule 23(b) requisites. 15 2. Rule 23(b)(3) “Predominance” Requirements 16 17 First, the Court turns to the question of whether the putative class meets the requirements of Rule 23(b)(3). A class may be certified under Rule 23(b)(3) only if the Court determines that 18 19 20 questions common to the class “predominate” over individualized questions and that using the class action device is “superior” to the individual pursuit of claims. Fed. R. Civ. P. 23(b)(3). To reiterate, Plaintiffs must demonstrate that they satisfy these requirements by a preponderance of 21 the evidence. Olean, 31 F4th at 665. 22 23 24 25 26 27 28 16 The Court further finds that due to the length of this litigation and its delay arising from the COVID-19 pandemic, Defendant’s concern does not appear applicable at this time. We are now five years removed from the Class Period, and Defendant has not established that their objections or arguments are still germane. 17 This Order defers addressing the appropriateness of injunctive relief or, more specifically, the named Plaintiffs’ adequacy for representing a class for the purposes of injunctive relief. As the parties have acknowledged, evaluation of injunctive relief is much more time-sensitive, especially since the onset of the COVID-19 pandemic. It would thus be inappropriate to make a determination as to injunctive relief based solely on the earlier pleadings of the parties. As such, the Court’s analysis focuses primarily on the issue of damages, while noting common violations and impact for class members. - 18 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 19 of 80 1 a. Predominance 2 “The predominance inquiry asks whether the common, aggregation-enabling, issues in the 3 case are more prevalent or important than the non-common, aggregation-defeating, individual 4 issues.” Tyson Foods, 577 U.S. at 453. Consideration of “whether questions of law or fact common 5 to class members predominate begins, of course, with the elements of the underlying cause of 6 action.” Erica P. John Fund, Inc. v. Halliburton Co., 563 U.S. 804, 809 (2011). This is because the 7 plaintiffs must show that the common questions relate to a central issue in the plaintiffs’ claim. 8 See Wal-Mart, 564 U.S. at 349-50; Olean, 31 F.4th at 666-67. A court is required to conduct a 9 “rigorous assessment” before it may conclude that the predominance inquiry is satisfied by a 10 preponderance of the evidence. Olean, 31 F4th at 665-66. 11 “In determining whether the common question prerequisite is met, a district court is limited 12 to resolving whether the evidence establishes that a common question is capable of class-wide 13 resolution, not whether the evidence in fact establishes that plaintiffs would win at trial.” Id. at 14 666-67. At the class certification stage, the district court must also resolve factual disputes between 15 the parties “if necessary to determine whether plaintiffs’ evidence is capable of resolving a 16 common issue central to the plaintiffs’ claims.” Id. at 667. While such an analysis may “entail 17 some overlap with the merits of the plaintiff’s underlying claim,” Wal-Mart, 564 U.S. at 351, 18 merits questions may only be considered if they are relevant to determining whether the class 19 certification prerequisites under Rule 23 are satisfied, Amgen, 568 U.S. at 466; see also Ellis, 657 20 F.3d at 983 n.8. “Rule 23 grants courts no license to engage in free-ranging merits inquiries at the 21 certification stage.” Amgen, 568 U.S. at 466; see also Olean, 31 F.4th at 667. 22 A district court cannot decline certification merely because it considers the plaintiffs’ 23 evidence relating to the common question to be unpersuasive and unlikely to succeed in carrying 24 the plaintiffs’ burden of proof on that issue. See Olean, 31 F.4th at 667; Amgen, 568 U.S. at 459- 25 60. “Nor can a district court decline to certify a class that will require determination of some 26 individualized questions at trial, so long as such questions do not predominate over the common 27 questions.” Olean, 31 F4th at 668 (citing Fed. R. Civ. P. 23(b)(3)). Rather, “Tyson Foods 28 established the rule that if ‘each class member could have relied on [the plaintiffs’ evidence] to - 19 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 20 of 80 1 establish liability if he or she had brought an individual action,’ and the evidence ‘could have 2 sustained a reasonable jury finding’ on the merits of a common question, then a district court may 3 conclude that the plaintiffs have carried their burden of satisfying the Rule 23(b)(3) requirements 4 as to that common question of law or fact.” Olean, 31 F.4th at 667 (quoting Tyson Foods, 577 U.S. 5 at 455). 6 Here, the class action claims are alleged violations of the Sherman Antitrust Act, 15 U.S.C. 7 § 2.18 A Section 2 violation focuses on the actions a single firm has taken to maintain their 8 dominance or consolidated power in a marketplace.19 Plaintiffs allege that Defendant possesses 9 monopsony power in the Relevant Input Market.20 “[T]o prove there is a common question of law 10 or fact that relates to a central issue in an antitrust class action, plaintiffs must establish that 11 essential elements of the cause of action, such as the existence of an antitrust violation or antitrust 12 impact, are capable of being established through a common body of evidence, applicable to the 13 whole class.” Olean, 31 F.4th at 666. 14 The elements of a claim for such antitrust action are (i) the existence of an antitrust 15 violation; (ii) “antitrust injury” or “impact” flowing from that violation (i.e., the conspiracy or 16 scheme); and (iii) measurable damages. See Big Bear Lodging Ass’n v. Snow Summit, Inc., 182 17 F.3d 1096, 1101-02 (9th Cir. 1999); In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 311 18 (3d Cir. 2008), as amended (Jan. 16, 2009) (citing 15 U.S.C. § 15). “Antitrust injury” is “injury of 19 the type the antitrust laws were intended to prevent and that flows from that which makes 20 defendants’ acts unlawful.” Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 489 21 18 22 23 The Supreme Court has long recognized that class actions serve a valuable role in the enforcement of antitrust laws. See Reiter v. Sonotone Corp., 442 U.S. 330, 344 (1979); Hawaii v. Standard Oil Co. of Cal., 405 U.S. 251, 262 (1972). Section 2 penalizes “[e]very person who shall monopolize, or attempt to monopolize, or combine or conspire with any other person or persons, to monopolize any part of the trade or commerce among the several States, or with foreign nations.” 15 U.S.C. § 2. By contrast, Section 1 of the Sherman Antitrust Act prohibits agreements between two or more individuals or independent entities that unreasonably restrain trade. See 15 U.S.C § 1. 19 24 25 26 27 28 20 The Court finds that, in the course of briefing and arguing their motion for class certification, Plaintiffs have abandoned their theory raised in the CAC that liability flows from Defendant’s monopoly power in the output market. Plaintiffs do not focus on market concentration in the output market in a manner extensive enough to pass muster under the “predominance” standard for certification under Rule 23(b)(3). Thus, the Court only considers arguments about Defendant’s dominance in the output market insofar as it bolsters its exercise of monopsony power in the input market. - 20 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 21 of 80 1 (1977). Damages are measured only after each plaintiff has demonstrated that the defendant’s 2 conduct caused the plaintiff to suffer an antitrust injury. See In re Hydrogen Peroxide, 552 F.3d at 3 311. 4 In short, to determine predominance in this antitrust action, the Court must evaluate 5 whether Plaintiffs can demonstrate the impact of the alleged anticompetitive behavior using 6 common evidence involving common questions of law or fact. More specifically, to determine 7 whether Plaintiffs meet the predominance threshold for class certification, the Court must find that 8 Plaintiffs have established, by a preponderance, the following four factors: (1) the existence of a 9 scheme or conduct which violates antitrust law, (2) that that scheme was, or is, commonly applied 10 to the class, such that there are common questions of law or fact, (3) that the common application 11 of that scheme led to common or similar injury amongst the class, and (4) that there are measurable 12 monetary damages proven by common evidence which arise from the anticompetitive conduct and 13 not procompetitive conduct.21 To make these determinations, the Court has considered evidence 14 including quantitative statistical analysis and qualitative documentary evidence from the record 15 and testimony provided during the seven-day evidentiary hearing. 16 17 For the reasons discussed below, the Court holds that Plaintiffs have demonstrated that common questions predominate over individualized issues for the purpose of class certification. 18 19 20 21 22 23 24 25 b. Overview of Predominance Analysis Applied to Bout Class The Court’s application of the predominance analysis for the Bout Class will proceed as follows. First, it will discuss the existence of the Scheme. Plaintiffs argue, Defendant engaged in various conduct to unlawfully maintain dominance in the market for fighter services in order to suppress fighter wages for cost-cutting purposes all while increasing Defendant’s own profits. The Court first defines the market for fighter services. Next, the Court examines evidence of the (1) 26 21 27 28 The Court reiterates that its findings in this order are not meant to be dispositive as to any factual dispute beyond class certification. When the Court indicates here that Plaintiffs or Defendant have ‘established’ a particular fact or position, this simply means that they have satisfied the necessary legal threshold in the context of determining class certification as set forth in Olean. 31 F4th at 667-68. The Court’s findings here are also intended to and do resolve any factual dispute between the parties necessary to resolution of the issue of class certification. Id. at 666-67. - 21 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 22 of 80 1 existence of the scheme to maintain market dominance, including efforts to buy or curtail the 2 success of rival promoters, (2) the exclusionary contracts that kept fighters “locked up” and 3 therefore unavailable to rival promoters for a period of time that approached the average duration 4 of their career, and (3) the coercive conduct that extended these exclusionary contracts to make 5 them virtually perpetual. 6 Second, the analysis will evaluate how the Scheme was commonly applied to the Bout 7 Class and involves common questions of law or fact. In doing so, it will focus on evidence of 8 common or broad application of the Scheme to the class, including the systemic use of the 9 exclusionary contracts and common coercive tactics which limited fighters’ options in 10 renegotiating a contract with Zuffa or, in the alternative, entering into a contract with a rival 11 promoter. 12 Third, the analysis will examine how common injury and damages flow from the common 13 application of the alleged Scheme. For this analysis, the Court engages in two distinct inquiries. 14 To begin with, the Court examines to what extent the class was injured. Evidence of this injury is 15 provided by Plaintiffs’ statistical regression model(s) which asserts a causal relationship between 16 Defendant’s anticompetitive conduct and the suppression of fighters’ compensation. The model 17 demonstrates that, as Zuffa increasingly dominated the market by means of anticompetitive 18 conduct, it suppressed fighters’ wages substantially below the level that they would have been in 19 a competitive market. Importantly, the Court examines whether the model meets the “rigorous 20 analysis” standard required for class certification and evaluates whether any of Defendant’s many 21 objections fundamentally undermine Plaintiffs’ demonstration of the Scheme’s impact. Thereafter, 22 the analysis looks at whether this impact was common to the class, specifically whether, and to 23 what extent, common issues relating to the injury predominate over the specific situations of 24 individual class members. In doing so, the Court assesses evidence of a common, or consistent, 25 compensation structure and the extent to which most putative class members were uninjured by 26 the Scheme. 27 Finally, the analysis will address whether monetary damages can be calculated by common 28 evidence—that is, whether there are measurable damages for the class members that can be traced - 22 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 23 of 80 1 directly to the alleged antitrust violation or conduct. Briefly, Plaintiffs compare the real-world 2 situation in which the Scheme was affected with a comparable situation in which the Scheme was 3 not carried out by Defendants, to calculate the damages amount. This enables Plaintiffs to show 4 what compensation for the fighters would have been “but for” the Scheme conduct. To do this, 5 Plaintiffs present calculations based on their model as well as evidence comparing MMA to other 6 sports, such as boxing. 7 8 9 c. An Unlawful Antitrust Scheme Plaintiffs assert violations of Section 2 of the Sherman Act based on Defendant’s alleged monopsony power. 10 11 A monopsony is a market structure where a firm has concentrated market power on the buyer side of the market. See Weyerhaeuser Co. v. Ross-Simmons Hardwood Lumber Co., Inc., 12 549 U.S. 312, 320 (2007), United States v. Syufy Enterprises, 903 F.2d 659, 663 n.4 (9th Cir. 13 14 1990) (“[m]onopsony is a market situation in which there is a single buyer . . . monopsony and monopsony power are the general equivalent on the buying side of monopoly and monopoly power 15 on the selling side.”); see also Roger D. Blair & Jeffrey L. Harrison, Antitrust Policy and 16 17 Monopsony, 76 Cornell L. Rev. 297 (1991). Because of the symmetrical relationship between monopsony and monopoly power, they are subject to similar legal standards. See Weyerhaeuser 18 19 20 Co., 549 U.S. at 322 (“The kinship between monopoly and monopsony suggests that similar legal standards should apply to claims of monopolization and to claims of monopsonization.”); see also California ex rel. Harris v. Safeway, Inc., 651 F.3d 1118, 1161 (9th Cir. 2011) (Reinhardt, J., 21 22 23 concurring in part) (“The Supreme Court has made clear . . . that because antitrust law operates to correct all distortions of competition, it condemns market actors who distort competition, whether on the buyer side or seller side Accordingly, the Court has long understood the Sherman Act to 24 25 26 condemn buyer side cartels.” (citation omitted)). Therefore, proof of violation of the antitrust law for a monopsony claim does not differ from that for a monopoly claim. As such, it requires that the plaintiff prove that the defendant: (1) 27 28 “possessed monopsony power in the relevant markets,” (2) “willfully acquired or maintained its monopsony power through exclusionary conduct,” and (3) “caused antitrust injury” through such - 23 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 24 of 80 1 conduct. Am. Pro. Testing Serv., Inc. v. Harcourt Brace Jovanovich Legal & Pro. Publications, 2 Inc., 108 F.3d 1147, 1151 (9th Cir. 1997). The Court now applies these factors to Defendant’s 3 alleged monopsony claim. 4 1. 5 Relevant Markets Zuffa Possessed Monopsony/Market Power in 6 The Court finds that Plaintiffs have provided sufficient evidence that Defendant maintained 7 8 market power in the relevant input market. “Market power may be demonstrated through either of two types of proof. One type of proof is direct evidence of the injurious exercise of market power. 9 If the plaintiff puts forth evidence of restricted output and supracompetitive prices, that is direct 10 11 12 13 14 proof of the injury to competition which a competitor with market power may inflict, and thus, of the actual exercise of market power.” Rebel Oil Co. v. Atl. Richfield Co., 51 F.3d 1421, 1434 (9th Cir. 1995). That said, “[t]he more common type of proof is circumstantial evidence pertaining to the structure of the market. To demonstrate market power circumstantially, a plaintiff must: (1) define the relevant market, (2) show that the defendant owns a dominant share of that market, and 15 (3) show that there are significant barriers to entry and show that existing competitors lack the 16 17 18 capacity to increase their output in the short run.” Id. Plaintiffs have supplied sufficient evidence of Defendant’s market power through circumstantial evidence.22 To explain below why, the Court will first define the relevant market. It will then discuss evidence Plaintiffs provide of how Zuffa 19 20 maintained dominance in that market. Finally, the Court will discuss the natural barriers to entry in this market, which discourage entry and allow for greater market dominance with 21 anticompetitive conduct. 22 23 a. Defining the Relevant Market 24 25 First, the Court finds that Plaintiffs have adequately defined the relevant markets, 26 22 27 28 The Court also finds, as discussed later in this Order, that Plaintiffs have provided, through the statistical modeling of their expert, Dr. Singer, sufficient evidence to establish direct evidence of market power. As discussed below, Plaintiffs have established that Defendant exercised sufficient market power to suppress wages. Accordingly, Plaintiffs have established antitrust injury in the form of suppressed market wages due to Defendant’s conduct. The Court concludes that its holding extends to finding direct evidence of market power. - 24 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 25 of 80 1 upholding its previous finding in its order on Zuffa’s Motion to Dismiss. See Le v. Zuffa, LLC, 2 216 F. Supp. 3d 1154, 1166 (D. Nev. 2016). Antitrust law requires the definition of both a product 3 market and a geographic market. See Newcal Indus., Inc., v. Ikon Off. Sol., 513 F.3d 1038, 1045 4 n.4 (9th Cir. 2008). Plaintiffs define two relevant MMA product markets: the input market (fighter 5 services) and the output market (bout events). 6 The relevant input market is “the market for Elite Professional MMA Fighter services.” At 7 a very basic level, the input market is the supply side of the equation. It is the market in which 8 Zuffa and its competitors purchase fighter services in order to be able to create their product of 9 MMA bout events.23 In order to define the relevant input market, Plaintiffs use industry accepted 10 databases measured in two, alternative ways: “Tracked” and “Ranked.” Under the “Tracked” 11 measure, Dr. Singer used any fighter who fought for an MMA promoter that was included in the 12 third party “FightMetric” database, and all fighters associated with MMA promoters acquired by 13 Zuffa. For the “Ranked” measure, Dr. Singer expanded the relevant input market to include the 14 “Tracked” fighters but also any fighters ranked in the FightMatrix database. Dr. Singer also had a 15 relevant input submarket consisting of “Headliners,” defined as the top fifteen fighters in any of 16 the ten major MMA weight classes tracked in the USA Today/MMA Junkie rankings, according 17 to the FightMatrix ranking database. The relevant market is the whole market for elite fighter 18 services24 and it contains fighters from Zuffa and non-Zuffa MMA promoters.25 These measures 19 23 20 21 22 23 24 25 26 27 It is important to note here that the substantial overlap and strong connection between the input and output markets is somewhat unique in MMA. This overlap results from the fact that it is common in this industry for the buyers on the input side of the market (fighter services) to generally also be sellers on the output side of the market (bout events). This is because the ranking and champion status of a fighter is also determined by combined sellers/buyers, such as Zuffa, and not by an independent third party, such as a boxing commission. There is generally no cross-promotion. Further, the Relevant Input Market here is for “elite” fighter services: those fighters who the UFC and its competitors may realistically view as eligible to compete for their promotions. While there are many non-Zuffa MMA promoters, these promoters, except for Bellator and Strikeforce, are small regional outfits that do not compete with the UFC for talent. Fighters with these promoters are typically not “ranked” or “tracked” by the industry databases. Moreover, these non-Zuffa regional promoters consider themselves to be a minor league to the major league of Zuffa (and its few competitors). Fighters with these minor league promoters are excluded from the relevant input market definition because they are not reasonable substitutes for the market of individuals from which Zuffa and its competitors draw. 24 25 28 The Court further finds that combat-oriented sports outside of MMA, such as boxing, kickboxing, sumo wrestling, and professional wrestling, do not constitute sufficiently viable employment alternatives for MMA Fighters to defeat an exercise of monopsony power by a hypothetical monopsonist in the Relevant Input Market because - 25 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 26 of 80 1 are used to define the total size of the market for fighter services. The Court finds that Plaintiffs 2 have sufficiently defined the relevant input market. As discussed above, the Court finds that 3 Plaintiffs have only maintained their claims regarding Defendant’s monopsony power. As a result, 4 the precise definition of the input market is the relevant market to evaluate throughout this order. 5 The output market is relevant insofar as Zuffa’s market power on the seller side reinforces its 6 ability to maintain a monopsony in the input market. As a single firm dominates more and more 7 of the product market, especially by purchasing competitors, the labor that creates that product is 8 necessarily more concentrated as well, allowing for opportunities for wage suppression.26 As a 9 result, a clear definition of the output market is crucial to understanding the totality of the 10 circumstances related to market dominance by Zuffa. 11 Here, the relevant output market is the “promotion of live Elite Professional MMA bouts 12 between Professional MMA fighters who compete in one-on-one fights known as bouts.” Plainly 13 stated, the output market consists of the product of MMA live bout events and the associated 14 products that are sold to customers, including viewers, cable networks, broadcast networks, and 15 sponsors. This market definition excludes other combat sports such as boxing because of low 16 crossover between audiences and differences in the play and rules of the sport. The output market 17 definition also excludes non-combat sports such as football and baseball, because record evidence 18 of cross promotion indicates that these sports leagues may be complements but are not substitutes. 19 As with the relevant input market, the Court finds that Plaintiffs have sufficiently defined the 20 relevant output market. 21 Finally, the relevant geographic markets for both the input and output markets are “the 22 United States, and in the alternative, North America.” Logically, this definition excludes promoters 23 such as ONE Championship, a pan-Asian promoter, and Absolute Championship 24 Akhmat(formerly known as Absolute Championship Berkut), an MMA promoter based in Russia. 25 Record evidence indicates that Zuffa, and its North American competitors, do not meaningfully 26 compete with these promoters for customers. The Court finds this evidence is sufficient to establish 27 training to become an MMA fighter tends to be specific to MMA and not generally transferrable to and from other combat sports. 28 26 See T. Hyclak, G. Johnes & R. Thonrton, Fundamentals of Labor Economics 211 (2d ed. 2012). - 26 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 27 of 80 1 that the relevant geographic market for both the input and output markets is the United States. 2 In summary, the Court finds that Plaintiffs have sufficiently defined the input and output 3 markets in the relevant geographic region for the purpose of class certification. The Court next 4 examines the evidence provided by Plaintiffs that Defendant dominated these markets. 5 b. 6 Dominant Market Share 7 Next, the Court finds that Plaintiffs have adequately demonstrated market dominance. To 8 satisfy the second element of the inquiry regarding monopsony market power, Plaintiffs must show 9 that Defendant controls, or controlled, a dominant share of the relevant market.27 10 To show control over a dominant share of the relevant input market, Plaintiffs’ expert Dr. 11 Singer divided the number of MMA fighters contracted with Zuffa by the total number of MMA 12 fighters in the Relevant Input Market and Submarket, weighted to reflect differences in quality of 13 fighters across different MMA promoters.28 Dr. Singer’s analysis showed that from December 14 2010 through June 2017, Zuffa’s share of the Relevant Input Market fluctuated between 94 and 15 99% (using the Tracked measure) and between 71 and 91% (using the Ranked measure). 16 Furthermore, from December 2010 through June 2017, Zuffa’s share of the Headliner29 market 17 18 19 20 21 22 23 27 The threshold percentage of market domination that a firm must meet to satisfy the standard for an antitrust violation is an evolving area of the law. The Ninth Circuit has held that a “market share of 44 percent is sufficient as a matter of law to support a finding of market power, if entry barriers are high and competitors are unable to expand their output in response to supracompetitive pricing.” Rebel Oil, 51 F.3d at 1438. Antitrust legal scholars have posited that 20 to 30% market dominance is sufficient to infer anticompetitive effects. See Herbert Hovenkamp, XI Antitrust Law ¶1821, at 152, 160, 164-65 (1998); see also Einer Elhauge, Tying, Bundled Discounts, and the Death of the Single Monopoly Profit Theory, 123 Harv. L. Rev. 469 (2009) (“it makes sense (when effects are not directly proven) to require the same 20-30% foreclosure share threshold that is required to infer anticompetitive effects from exclusive dealing”). For purposes of this analysis, the Court uses these sources as guidelines for determining when and whether the market dominance creates anticompetitive effects. Dr. Singer’s weighting of fighters, which the Court finds credible, with regards to this calculation is done in the following manner: Dr. Singer constructed a database of PPV revenue and gate revenue by promoter. Zuffa fighters were weighted by the average revenue per Zuffa fighter. Non-Zuffa fighters were weighted by the average revenue per Fighter for non-Zuffa promoters. However, revenue data is only available for larger non-Zuffa MMA promoters such as Bellator and Strikeforce (pre-acquisition), so fighters from promotions smaller than those were weighted as if they were affiliated with a larger non-Zuffa promoter. 28 24 25 26 27 28 29 MMA promoters rely on top-ranked, fighters fighting at the top of the card to draw audiences and viewers to Live MMA events. Accordingly, Dr. Singer created a “Headliner” metric to evaluate how much of this market was captured by the various promoters, including Defendant. The Headliner measure includes each of the top fifteen fighters in any of the ten major MMA weight classes tracked in the USA Today/MMA Junkie rankings, according to - 27 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 28 of 80 1 fluctuated between 95 and 99% (if fighters are weighted by revenue). Id. The Court finds that 2 Plaintiffs have presented sufficient evidence of Defendant’s market dominance. 3 The Court rejects Defendant’s arguments to the contrary. Defendant and its experts argue 4 that this analysis is flawed because (1) the Tracked measure pulls from metrics databases that do 5 not include everyone in the sport, and (2) the Headliner measure fails to consider the ways 6 promoters help fighters develop into headliners. 7 The Court finds these arguments are unavailing in view of the relevant standard for the 8 purpose of class certification. Defendant’s arguments are factual and merits-based and cannot 9 therefore defeat certification here. See Amgen, 568 U.S. at 466; Olean, 31 F.4th at 667. The Court 10 need only find that Plaintiffs’ have presented, by a preponderance of evidence, that Zuffa had 11 market dominance. Plaintiffs have credibly done so. See Olean, 31 F.4th at 664. 12 Second, the Court finds Defendant’s arguments unpersuasive factually. The Court finds 13 Defendant’s assertion that every possible or potential fighter, including fighters who would never 14 have fought for Zuffa and who may never fight or fight in the relevant market, should be included 15 in the market is a thinly veiled attempt to dilute its market dominance. Moreover, with regard to 16 the Tracked measure, FightMetric—the database which contains the information from which Dr. 17 Singer culled his analysis—is a business that provides information to third parties about those 18 competing with major MMA promoters. The relevant market for the purposes of this analysis is 19 fighters who have commercial value for promoters. Inclusion in a database that third parties pay 20 for is an appropriate proxy for a fighter’s inclusion in the market. Inversely, exclusion from that 21 database is an appropriate proxy for a fighter’s absence in the market because the market will be 22 unwilling to pay to track the careers of fighters who lack a threshold level of commercial value. 30 23 Similarly, Defendant’s contention that the Headliner metric does not adequately capture 24 “promoter acumen” lacks merit. Defendants fail to establish that there is in fact a separate credible 25 factor of “promoter acumen,” apart from what may be captured in the Headliner metrics. Indeed, 26 the FightMatrix ranking database. 27 Fighters who fight for promoters in the so-called “minor leagues” are sometimes not included in these databases. These fighters are not in the relevant input market because their services are not sought after by Zuffa or its market peers. The fact that these “minor league” contracts typically contain “UFC-out” clauses is further proof of their proper exclusion from the relevant input market. 28 30 - 28 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 29 of 80 1 the Court finds here, and elsewhere, that Defendant’s claims of some nebulous factor of “promoter 2 acumen” or “special sauce of promotion” are unsupported by the record, including by the statistical 3 modeling and expert testimony. Rather than present a substantively, more persuasive critique or 4 alternative to the metric, which might change the market power outcome, Defendant merely offers 5 a less cogent alternative explanation for how that market power was obtained. The Court is 6 unpersuaded that this critique undermines Plaintiffs’ evidence and position. 7 Separately, Defendant argues that Plaintiffs’ revenue-weighted share calculation (or 8 inverse rank for the Headliner submarket) does not reflect differences in fighters’ ability or 9 marketability. The Court finds, however, that this method of accounting is reasonable and 10 appropriate based upon Plaintiffs’ modeling. Fighters are not homogenous; they vary in their 11 ability to attract viewers. Plaintiffs’ revenue and inverse ranking weights allow for the expression 12 of this difference within the market in a model that applies to all members of the Bout Class. 13 Indeed, if Plaintiffs had not sought to capture some variability in the common model, Defendants 14 would be referring to such variable’s absence as a fundamental flaw in the model. 15 There is ample evidence that Zuffa also dominated the output market financially. Dr. Singer 16 estimated that Zuffa’s market share has been as high as approximately 90% by revenue since at 17 least 2008. Dr. Singer also analyzed financial data obtained from the three largest non-Zuffa 18 promoters in the Relevant Output market: Strikeforce, Bellator, and World Series of Fighting. 19 Since 2009, the combined revenue of these three competitors never represented more than 9% of 20 the total revenues. Zuffa’s level of market share and dominance in the input (and output) market 21 is more than double the percentage recognized by scholars and the Ninth Circuit as establishing 22 monopsony market power.31 23 Further, there is myriad record evidence that market participants and observers viewed and 24 treated Zuffa as the dominant power in the industry and changed their behavior as a result. A 2013 25 Deutsche Bank report dismissed the competitive significance of other MMA promoters generally, 26 and Bellator in particular, writing “[i]n recent years in MMA, promotions that have tried to position 27 themselves as the UFC’s primary competitor, such as Bellator, have always failed. Bellator’s most 28 31 See supra note 27. - 29 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 30 of 80 1 recent event at UC Irvine’s Bren Center attracted only 4,000 people. . . . none of the Bellator 2 fighters are household names.” This market dominance appears to have been part of what 3 motivated WME-IMG to purchase Zuffa. Additionally, Plaintiffs present record evidence that 4 promoters, such as Alliance MMA, disavowed any attempts to compete directly with Zuffa. In a 5 prospectus to its investors, Alliance MMA stated that it “is our intention to serve as a 6 developmental organization for the UFC and other premier national MMA promotions in the same 7 fashion as college athletic programs serve as ‘feeders’ to professional sports leagues] … [s]hould 8 the UFC or another premier national MMA promotion view us as a threat they could use their 9 significantly greater resources to frustrate our business and growth strategy and materially and 10 adversely affect our business.” 11 These findings, which the Court credits, demonstrate Defendant’s dominance in the input 12 (and output) market. It further establishes how Zuffa’s dominance in the output market enhanced 13 its ability to suppress compensation in the input market. Now that the Court has found that 14 Plaintiffs have established Defendant’s dominance in the markets, the Court moves on to analyzing 15 evidence of competitors’ natural and artificial barriers to entry in the market. 16 17 c. Barriers to Entry 18 To satisfy the third element of the monopsony market power inquiry, Plaintiffs must show 19 that there are significant barriers to entry such that existing competitors lack the capacity to 20 increase their output in the short run. Rebel Oil, 51 F.3d at 1439. “Entry barriers are additional 21 long-run costs that were not incurred by incumbent firms but must be incurred by new entrants, or 22 factors in the market that deter entry while permitting incumbent firms to earn monopoly returns.” 23 Id. “The main sources of entry barriers are: (1) legal license requirements; (2) control of an 24 essential or superior resource; (3) entrenched buyer preferences for established brands; (4) capital 25 market evaluations imposing higher capital costs on new entrants; and, in some situations, (5) 26 economies of scale.” Id. (footnote omitted). “To justify a finding that a defendant has the power to 27 control prices, entry barriers must be significant—they must be capable of constraining the normal 28 operation of the market to the extent that the problem is unlikely to be self-correcting.” Id. On the - 30 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 31 of 80 1 other hand, “[b]arriers to entry are insignificant when natural market forces will likely cure the 2 problem.” Id. 3 The Court finds that there were significant barriers to entry for new participants in the 4 relevant input market. These barriers included significant fixed costs to event promotion, 5 advertising, marketing, production, and distribution. Additionally, for the reasons discussed at 6 length below, exclusionary contracts limit the ability for new entrants to obtain the essential 7 input—fighter services—for the business. 8 The Court now examines evidence related to artificial barriers32 to entry. The Court finds 9 that Plaintiffs have established evidence of three main barriers to entry for competitors: (1) 10 coercive contracts that restrict the overall supply of fighters that competitors can access, (2) lack 11 of cross-promotion, and (3) dominance in the Headliner market. These barriers to entry made it 12 prohibitively expensive and difficult for new entrants in the input (and output) market to 13 effectively compete with Zuffa. Given the artificial and intentional nature of these barriers, they 14 are not self-correcting through natural market forces. 15 This Order analyzes the nature and effect of Zuffa’s coercive contracts at length below. 16 However, for purposes of the instant discussion of barriers to entry in this section, the Court notes 17 that in addition to allegedly harming the fighters by suppressing wages, these coercive contracts 18 damage the overall market environment by artificially restricting competitors’ access to strong 19 fighter talent which could be used to grow their business. While harm to competitors is not the 20 focus of liability in this matter, competitors’ inability to effectively compete with Zuffa in the 21 market for talent has a direct and enhancing effect on Zuffa’s ability to artificially, and improperly, 22 limit fighter compensation. 23 Second, Zuffa has made the choice to eschew matchups between a fighter under contract 24 32 25 26 27 28 The Court notes the distinction between a natural barrier and an artificial barrier to entry. A natural barrier to entry is the type that would-be entrants seeking to compete would face even in the absence of anticompetitive behavior. Natural barriers are relevant in so far as they distinguish an organic market situation from conduct by Defendant. The Court finds that several natural barriers to entry exist including sport regulation and the costs associated with planning, producing, marketing, and distributing a large-scale event. The Court, however, finds that Plaintiffs have adequately established that it was artificial barriers to entry, deriving from Zuffa’s alleged anticompetitive conduct, that prevented new entries into the relevant input (and output) markets. See generally, Ronald A Cass and Keith Hylton, Antitrust Intent, 74 S. Cal. L. Rev. 657, 726-27 (2001)(comparing natural and artificial barriers to entry). - 31 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 32 of 80 1 with Zuffa and a fighter under contract with another promoter, a concept known as cross- 2 promotion.33 The evidence establishes that cross-promotion is much more common in other 3 combat sports such as boxing. By contrast, UFC fighters may only ever fight other UFC fighters. 4 This means that if a fighter is signed with the UFC, they are cut off from fighting non-UFC talent. 5 Moreover, as Zuffa has acquired competitors and signed more and more fighters, competitors are 6 cut off from increasingly large portions of the talent pool. The difficulty of accessing a talent pool 7 capable of attracting enough commercial interest to turn a profit is another artificial barrier to entry 8 for competitors. 9 Finally, and relatedly, Zuffa’s dominance in the Headliner market specifically creates a 10 barrier to entry for competitors. Control of fighters who are Headliners is essential within MMA 11 for revenue purposes. Sales estimates from MMA events revealed that marquee names are of 12 particular importance for driving PPV sales for live fights. Zuffa’s top matchmaker, Mr. Silva, 13 confirmed the importance of Zuffa’s dominance in the Headliner market. In a colloquy discussing 14 a February 2011 email sent by Mr. Silva, titled “We Own MMA,” Mr. Silva explained that his 15 statement that Zuffa “own[ed]” MMA” was based in part on the fact that Zuffa had most of the 16 top fighters in most of the top weight classes. The purpose of the email was to disclose to Mr. 17 Silva’s bosses, Mr. White and Mr. Lorenzo Fertitta, that Zuffa had contracts with most of the top 18 fighters in the most popular commercial divisions. Mr. Silva further confirmed that control over 19 the Headliner market increased even more after the acquisition of Strikeforce. Clearly, the 20 Headliner market was important. And Zuffa’s dominance over it created an artificial barrier to 21 entry for competitors. 22 Based on this evidence, the Court finds that there were significant artificial barriers to entry 23 for Zuffa’s competitors in the input and output markets. As discussed above, while harm to 24 competitors is not the focus of liability in this matter, competitors’ inability to effectively compete 25 with Zuffa because of barriers to entry in the market significantly augments Zuffa’s ability to 26 suppress the compensation of its fighters. A market situation with high artificial barriers to entry 27 33 28 There has been one notable exception to this general rule: a co-promoted event featuring UFC fighter Conor McGregor and boxer Floyd Mayweather on August 26, 2017. Notably, however, the two fighters were boxing, not engaging in the sport of MMA. The UFC has not cross-promoted an MMA event with any other promotion. - 32 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 33 of 80 1 is an environment in which anticompetitive conduct can more easily thrive. This analysis informs 2 the context of this specific market, which allows the Court to evaluate the appropriateness of 3 certifying a class under a totality of the circumstances common to that of the class, as required by 4 law. The extent to which these artificial barriers are willfully maintained by Defendant for the 5 purpose of creating conditions to suppress compensation is also a relevant consideration and will 6 be discussed in the next section. 7 Accordingly, the Court finds that Plaintiffs have provided sufficient evidence that 8 Defendant “possessed monopsony power in the relevant [input] markets.” Am. Pro. Testing Serv., 9 108 F.3d at 1151. 10 11 12 2. Zuffa Used Anticompetitive Conduct to Support Its Monopsony 13 Next, the Court finds that Defendant maintained its dominant position in the fighter input 14 market through anticompetitive conduct. As noted above, the second factor in determining whether 15 there has been a violation of the antitrust law focuses on whether the defendant “willfully acquired 16 or maintained [market] power by engaging in predatory or anticompetitive conduct designed to 17 destroy competition.” Movie 1 & 2 v. United Artists Commc’ns, Inc., 909 F.2d 1245, 1255 (9th 18 Cir. 1990). “Under the best of circumstances, applying the requirements of § 2 can be difficult 19 because the means of illicit exclusion, like the means of legitimate competition, are myriad.” 20 Verizon Commc’ns Inc. v. L. Offs. of Curtis V. Trinko, LLP, 540 U.S. 398, 414 (2004). For 21 example, “[t]hey may involve a course of dealing that, even if profitable, indicates a willingness 22 to forsake short-term profits to achieve an anticompetitive end.” Pac. Bell Tel. Co. v. linkLine 23 Commc’ns, Inc., 555 U.S. 438, 458 (2009). 24 The Court finds that Plaintiffs have established that Defendant willfully engaged in 25 anticompetitive conduct to maintain or increase their market power. This was mainly accomplished 26 in three ways: first, through the enforcement of exclusionary contracts, which kept fighters “locked 27 up” in an anticompetitive manner; second, through extracontractual methods to make fighter 28 contracts effectively perpetual; and third, through acquisition and shutting down of rivals. The - 33 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 34 of 80 1 Court finds that Plaintiffs have established that these forms of anticompetitive conduct, both 2 individually and collectively, along with dominant output market power, allowed Defendant to 3 exercise substantial monopsony power over putative class members in the input market for fighter 4 services. Ultimately, the Court finds, for the purposes of class certification, that Plaintiffs have set 5 forth sufficient evidence that this Scheme amounted to a violation of the antitrust law. 6 7 a. Exclusionary Contracts 8 Defendant utilized contracts containing exclusionary provisions that effectively negated a 9 fighter’s mobility to competitors and their ability to meaningfully negotiate with Zuffa rivals. As 10 discussed above, before a fighter can participate in a Zuffa event, they are required to sign a PAR 11 Agreement that governs the terms of a fighter’s compensation over the course of multiple bouts. 12 Prior to any bout, a UFC fighter also signs a “Bout Agreement” which affirms event-specific 13 compensation details. Under these agreements, fighters are not paid except for their participation 14 in bouts (i.e., they are not paid a salary to train, or develop or maintain their fighting skills). 15 Virtually every PAR Agreement contains an “exclusion clause” that requires fighters to 16 fight exclusively with Defendant for the duration of the contract and prohibits fighters from 17 working with other MMA promoters. These exclusion clauses do not obligate Defendant to 18 maintain their relationship with the fighter for any period of time. The PAR Agreements also 19 contain tolling provisions that can sua sponte extend the contract term for a variety of reasons 20 without the consent of the fighter. For instance, the contract can be tolled for injury, or for refusing 21 to accept an opponent that Defendant selected, a selection over which Defendant has complete 22 control. The PAR Agreements additionally have a retirement clause permitting Defendant to 23 suspend a fighter’s contract indefinitely, so that even if a fighter were to announce their retirement 24 and then come out of retirement several years later, they would still be required to fight a requisite 25 number of fights for Defendant. At the same time, however, a fighter’s contract can be terminated 26 early after one or two consecutive losses. Zuffa’s PAR Agreements also typically contain a 27 champion’s clause that automatically extends a contract for one-year or three bouts for a fighter 28 who becomes a title-holding champion at the end of the initial term of the contract. - 34 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 35 of 80 1 In addition, Defendant’s contracts commonly contain a right-to-match clause which grants 2 Defendant the option to match a competing promoter’s offer for a limited period of time, after the 3 expiration of a fighter’s agreement. In almost all PAR Agreements containing this clause, the right- 4 to-match period lasts for one year. If another MMA promoter attempts to sign a fighter during this 5 period, the fighter must share the terms of that offer with Defendant, and Defendant can elect to 6 retain the fighter by matching the terms offered by the other MMA promoter. Fighters are 7 contractually obligated to accept Defendant’s matching terms. Further, many of Defendant’s PAR 8 Agreements include an exclusive-negotiation clause, which requires fighters to negotiate 9 exclusively with Defendant for renewal of their contracts for a period of 30 to 90 days. Importantly, 10 this meant that, during the crucial period of renegotiation of a new contract, fighters cannot even 11 talk to rival promoters to meaningfully understand or know their value outside of what Zuffa is 12 offering them. 13 Here, the Court finds that the combined effect of the contracts’ restrictive clauses created 14 a situation where Zuffa had the sole power to control a fighter’s ability to make money for the 15 majority of the average fighter’s career. Through these restrictive contracts, Zuffa controlled how 16 17 18 19 20 21 22 23 24 25 much, where, and when fighters could earn compensation by participating in a bout. Additionally, the contracts could be extended for a variety of reasons including injury, which is exceedingly common in MMA. Mr. Silva confirmed that it was common for fighters to have multiple injuries over the course of their career given the nature of the sport. On the face of the contracts alone, these fighters were effectively “locked up” by these contracts. These contracts’ clauses, both individually and collectively, operated to lock up fighters. The Court further finds that these contracts, with their various restrictive clauses, existed throughout the class period and operated to lock up fighters with Zuffa. These contracts therefore restricted fighter mobility and allowed Zuffa to build concentrated power in the market for fighter services. b. Coercive Tactics Derived From Contract 26 27 Clauses 28 - 35 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 36 of 80 1 In addition to the ways in which the provisions of the exclusionary contracts permitted 2 Zuffa to increase its market dominance, Plaintiffs have also established that Defendants used a 3 variety of ruthless coercive techniques to prevent fighters from becoming free agents—rendering 4 these contracts effectively perpetual. To start, Defendant maintained significant control over a 5 fighter’s career by means of controlling the details of bouts. More specifically, for example, 6 Defendant controlled the timing of bouts, the pairing of a fighter with opponents, the placement of 7 a fight on a fight card, whether a bout would be PPV or televised, whether a fighter could get 8 discretionary bonuses, and the overall promotion of a fighter’s career. Because Zuffa fighters did 9 not get paid unless they fought, this enabled Zuffa to use various strategies related to the timing, 10 placement, number and opponents of a fighter’s bouts to coerce fighters into renewing their 11 contracts early or extending their contracts in order to earn a paycheck. 12 These coercive tactics in conjunction with the restrictive provisions of the contracts 13 themselves had a devastating effect on fighters’ ability to control their careers and compensation. 14 For example, Plaintiffs have established that Zuffa leveraged the right-to-match clauses in Zuffa’s 15 contracts. These clauses prevented fighters from becoming free agents until 12 to 15 months after 16 their last bout to incentivize fighters to stay with Zuffa. The 12 to 15 month waiting period would 17 be one in which the fighter would not be earning any money from fighting and constituted a 18 significant portion of a fighter’s career. Significantly, Zuffa did not pay for any of a fighters’ 19 expenses associated with maintaining skills to be competitive during this 12 to 15 month period. 20 Yet, a fighter needed to train and live to maintain their skills so they could maximize their revenue 21 from a bout (when it came) and improve in ranking. The costs associated with training facilities, 22 receiving coaching, setting up sparring matches, and the basics of living (housing, food, health 23 benefits, etc.) were therefore all borne exclusively by the fighter. This meant that competitive 24 fighters could not practically endure a 12 to 15 month period of no bouts and no revenue if they 25 wished to both live and remain competitive. Thus, the devastating economic effect of having to 26 wait 12 to 15 months to earn again would effectively and practically force fighters to re-sign with 27 Defendant if they wanted to continue to be competitive MMA fighters. 28 - 36 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 37 of 80 1 Defendant also leveraged its bargaining power arising from the contracts to pressure 2 fighters to renew their contracts before the last fight of their existing contract. Defendant typically 3 opened negotiations for a contract renewal before a fighter’s existing contract expired. And, the 4 record establishes that, as part of the alleged Scheme, Zuffa used its contractual authority prior to 5 the expiration of a PAR Agreement, to limit a fighter’s options. This general strategy was 6 confirmed by Zuffa officials themselves. Mr. Silva testified that it was his practice to present a 7 new contract to a fighter before the last fight in the fighter’s current contract: “I always renegotiate 8 before the last fight.” Additionally, Michael Merch, Zuffa’s Assistant General Counsel and Vice 9 President of Business, Legal and Government Affairs, explained that “if a fighter is successful 10 under a 4 fight deal, we typically negotiate a new agreement after the 3rd fight so he never will 11 see the end of his contract and, assuming the fighter is successful, or at least competitive, that is 12 the process that will continue thereafter.” 13 Moreover, Defendant could and did impose, or credibly threaten to impose, delays between 14 bouts, particularly before the final bout of a fighter’s contract. During this period, fighters could 15 not earn MMA income from any source. The record establishes that Zuffa regularly delayed 16 scheduling a fighter for a final bout in the contract in order to coerce them into renegotiating prior 17 to the end of their contract. As named Plaintiff Vera, explained: “Every time a fighter has -- is 18 coming up on his renegotiation period, it was common knowledge in our industry that if you didn’t 19 sign the new agreement, that you were going to get frozen out or put on a dark show so that nobody 20 would ever see your last fight. That’s common knowledge across the board.” Plaintiff Fitch 21 explained how such tactics were applied to him in the context of his fight career and a contract that 22 ultimately extended into the Class Period: 23 24 25 26 27 28 [JON FITCH]: . . . . I hadn’t fought since September, so money’s running low, I need money, I need to fight. This agreement is sent to me in February; right? So -- oh, wait a minute, that’s January. So that’s two months before that fight. They’re holding my bout agreement hostage. They’re holding my next fight hostage until I sign this. They do that often. I think if you look at the other one, you’ll probably see that too. Signed 4/6, yes. Signed 5/6, which is May, yeah, so I signed that in May. I fought in June. [QUESTION]: So you’re saying that at that time they were withholding a fight from you until you resigned? - 37 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 38 of 80 1 2 [JON FITCH]: I’m saying they do that to everybody. We’re going to hold your bout agreement until you sign your extension. We won’t allow you to become a free agent. 3 4 One of the most coercive tactics used by Zuffa related to its exclusive control of a fighter’s 5 opponent in bouts. Defendant could, and did, credibly threaten to match a fighter with a suboptimal 6 opponent to coerce acquiescence to an exclusionary contract. The opponent could be suboptimal 7 in two ways. First, the opponent could hurt the ranking of the fighter. Second, the opponent could 8 be more skilled or brutal and therefore likely to hurt or seriously injure the fighter. Plaintiff 9 Kingsbury provided an illustration of this coercive tactic: 10 11 12 13 14 15 16 17 18 19 20 21 22 Joe Silva wanted me to fight [Glover Teixeira, a training partner of Chuck Liddell] on the undercard in Glover’s first fight in the UFC. And that is exactly the kind of thing that hurts fighters. Nobody knew who he was. He was extremely talented. It was a lose-lose. If I beat him, I’m supposed to beat him. If I don’t beat him, I just got my ass kicked by a guy that nobody knows. I had another unknown opponent in his first fight in the UFC. I had to fly all the way to England for it. It was on the undercard, and it was against a guy who at the time was 11 and 0, no losses, and not a single opponent had made it out of the first round with him. So he was highly decorated and not known. Again, as stated, a lose-lose for me. And then I had received the damage that I did in that fight. And with my second orbital blowout on the same side, I was really concerned for my health fracturing the eyebrow as well and was just considering is it worth it? I can’t get out of my contract. There’s nowhere else to go at this point. Strikeforce didn’t exist anymore. Pride didn’t exist anymore. And I had no way of saying no. Indeed, Zuffa matchmakers admitted that they used such brutal coercive tactics. In April 2010, Mr. Silva admitted, regarding negotiations with UFC fighter, Nick Diaz, “I lowballed them on purpose the first offer knowing they would turn it down. How bout I come back with 29+29 32+32 35+35 38+38. If they turn it down I put him in a prelim against a really tough guy for his 23 last fight.” Mr. Silva was known to have told fighters, including Nate Quarry, “if you don’t like 24 the first fight I offer you, you’re sure as shit not going to like the second one.” 25 Alternatively, Defendant might place a fight in a lower spot on the fight card, which would 26 stymie a fighter’s career trajectory and value, regardless of their skill level. As one fighter 27 explained “[t]hrough threats and direct punishment, [Mr. White] really controlled how our careers 28 went. And he could stick you on the undercard in your next fight against some unfavorable - 38 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 39 of 80 1 opponent. If you’re on the undercard, you would make far less. If anything in sponsorship because 2 you weren’t on TV.” The variety of coercion tactics available to Defendant, in conjunction with 3 the restrictive contractual provisions, ensured that Defendant could maintain almost perpetual 4 control over fighters. 5 Additionally, the Court finds that the coercive nature of these tactics and contracts were 6 particularly effective because of the short duration of an average MMA fighter’s career. As 7 Plaintiffs established, a fighter’s average career length was between 31 and 41 months or about 8 five to six bouts. Plaintiffs’ expert also found that the average duration of exclusivity of the 9 contracts during the class period was 36.7 months,34 -- an almost complete overlap. Thus, the 10 ineluctable effect of these contract provisions combined with Zuffa’s coercive tactics was to 11 prevent fighters from even understanding their true market value and engaging in negotiations with 12 any other entity besides Zuffa. Due to this anticompetitive, coercive conduct, fighters were trapped 13 by Zuffa’s exclusionary contracts and their restrictive terms, creating a situation in which Zuffa 14 had unfettered power and opportunity to suppress fighters’ compensation. 15 In conclusion, the Court finds that Plaintiffs have established that Defendant’s tactics were 16 anticompetitive. Defendant evinced a clear intent to acquire and maintain monopsony power. 17 Defendant has not presented sufficient evidence that these exclusionary contracts and coercive 18 tactics were procompetitive or contributed to the overall development of the sport or the injury 19 such that the Court would decline to certify the class. The evidence indicates the opposite: these 20 tactics harmed competition and the development of the sport. 21 c. 22 Horizontal Acquisition 23 In addition to preventing mobility of fighters through their contracts and coercive 24 techniques, Defendant’s horizontal acquisition of competitors limited the availability of 25 alternatives to the UFC. Plaintiffs credibly establish that Zuffa systematically acquired, or forced 26 27 34 28 This calculation includes the initial contract term, the options period, the exclusive negotiations period, and the right-to-match period, but does not take into consideration contracts that followed on from the original due to tolling provisions, champion’s clauses, or the frequency of pre-expiration renegotiation. - 39 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 40 of 80 1 the shutdown of, rival promoters. Between December 2006 and May 2007, Zuffa acquired its three 2 most significant rivals at the time. In March 2011, it acquired Strikeforce, which at the time of 3 purchase, was the second, most prominent MMA organization in the world, including the most 4 prominent women’s MMA division. Defendant’s horizonal acquisitions reinforced their 5 dominance in the input market.35 These acquisitions consciously eliminated would-be rivals to 6 which fighters might otherwise have switched. This conduct substantially foreclosed competition 7 and had market-wide anticompetitive effects without any recognizable procompetitive benefits. 8 The Court agrees with Plaintiffs that Defendant’s willful anticompetitive conduct maintained, or 9 increased, Defendant’s monopsony power. The Court finds that Defendant fails to present any 10 meaningful evidence that these acquisitions were procompetitive or contributed to the 11 development of the industry. 12 The record establishes that undermining competitors was important to Zuffa leadership and 13 an essential part of its strategy to establish market dominance. This is evidenced, for example, in 14 a February 2014 text message discussion between Mr. White and Mr. Lorenzo Fertitta regarding 15 re-signing fighter Gilbert Melendez. Melendez’s contract with Zuffa was ending, and he had come 16 to an agreement with competitor, Bellator. Regarding that contract, Mr. Lorenzo Fertitta texted 17 Mr. White: “We gotta keep taking these fuckers oxygen till they tap out. We have sacrificed too 18 much to let anyone get traction now[.]” At his deposition, Mr. Lorenzo Fertitta confirmed that the 19 term “fuckers” referred to Bellator, while the “oxygen” referred to fighters Gilbert Melendez, and 20 another fighter, Eddie Alvarez, both of which Bellator was attempting to recruit. 21 Evidence also indicates that Zuffa’s leadership attitude motivated the acquisition of several 22 of Zuffa’s rivals. Specifically, the purchase of these rivals was undertaken in an effort to swell the 23 ranks of Zuffa’s fighters and remove potential competition. Mr. Silva explained that the acquisition 24 of rivals was generally motivated by the desire to obtain rivals’ fighter contracts and increase 25 control over popular fighters. Because the horizontal acquisitions caused non-Zuffa promoters to 26 lose access to the top (non-Zuffa) fighters that generated the most revenue, non-Zuffa promoters 27 28 35 Plaintiffs also offer credible evidence that, in addition to horizontal acquisitions, Defendants used a variety of other tactics to undermine competition. This includes engaging in counter-programming (scheduling free conflicting events at the same time as rival promoter events) in attempts to suppress demand for events from other MMA promoters and using litigation, or the threat thereof, against rivals. - 40 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 41 of 80 1 were further precluded from offering competitive compensation and promising career paths to 2 fighters. The Court finds that the record demonstrates that Defendant’s horizontal acquisition of 3 its rivals was an intentional effort to further increase its concentration in the relevant input market 4 in service of maintaining a favorable market situation for itself and to suppress fighters’ 5 compensation. 6 Beyond the qualitative evidence establishing this anticompetitive conduct, quantitative 7 evidence indicates that this strategy was successful in discouraging competition and maintaining 8 market power. The Herfindahl-Hirschman Index (“HHI”) is a common measure of market 9 concentration and is used to determine market competitiveness pre-merger and post-merger. U.S. 10 Dep’t of Just. and Fed. Trade Comm’n, Horizontal Merger Guidelines § 5.3 (2010). The HHI for 11 the Relevant Input Market and Submarket therefore exceeds 8,100 under the Tracked measure 12 exceeds 6,400, under the Headliner Submarket, and exceeds 4,900 under the Ranked measure. 13 Each of these HHI values are far above 2,500, the threshold for a market to be classified as “highly 14 concentrated” according to the antitrust agencies’ Horizontal Merger Guidelines. Id. 15 Defendant contends that there are procompetitive rationales for Defendant’s contracting 16 practices that rebut Plaintiffs’ argument that Defendant willfully acquired, and worked to maintain, 17 monopsony power in violation of the antitrust law. Plaintiffs’ do not, however, have to establish 18 that their position and arguments, based upon the record, are “better” than those of Defendant and 19 thus establish liability. Olean, 61 F.4th at 666-67. Rather, Plaintiffs need only establish the class- 20 wide viability or reasonableness of their position by a preponderance of the evidence. Id. The 21 Court considers Defendant’s arguments to the extent that they might undermine the soundness or 22 viability of Plaintiffs’ arguments. They have not done so. While Defendant may present evidence 23 at trial as to the potential procompetitive effects of its conduct, Plaintiffs have sufficiently met 24 their burden here. 25 To the extent required at this certification stage, the Court also explicitly rejects 26 Defendant’s expert opinion on this issue. Defendant’s experts argue that the challenged contract 27 provisions are efficient responses to externalities and transaction costs that would otherwise hinder 28 the development of MMA as a sport. There is, however, a lack of evidence in the record indicating - 41 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 42 of 80 1 that these contractual provisions eliminated transaction costs. There is also a dearth of evidence 2 that these contract provisions contributed to the development of MMA as a sport by, for instance, 3 attracting additional capital investment to the sport or in any other way. 4 Defendant’s reliance on the existence of some of these same contractual provisions existing 5 in its competitors’ contracts is also misplaced. The mere fact that a competitor may be attempting 6 to imitate, or respond to, Zuffa’s anticompetitive conduct with anticompetitive conduct of its own 7 does not render such conduct procompetitive or efficient. Moreover, the effect of conduct 8 undertaken by a firm with significant market power can be harmful to competition in a way that 9 that the same conduct undertaken by a non-dominant firm is not. 10 To be sure, the central inquiry for the Court is the overall effect of the various contract 11 provisions and Defendant’s concomitant coercive conduct, rather than the consideration of these 12 different factors in isolation. Indeed, this is the approach the Ninth Circuit requires of courts in 13 this circuit. See In re Nat’l Football League’s Sunday Ticket Antitrust Litig., 933 F.3d 1136, 1152- 14 53 (9th Cir. 2019) (finding that courts are “required to take a holistic look at how the interlocking 15 agreements actually impact competition”). The essential inquiry is thus “whether or not the 16 challenged restraint enhances competition under “the totality of the nature or character of the 17 contracts.” Id. at 1153. The Court has engaged in exactly this inquiry here to support its findings. 18 In conclusion, the Court finds that the record establishes the anticompetitive nature of the 19 contracts, the use of extracontractual coercion techniques, and the intentional horizontal 20 acquisition of prospective market competitors. The totality of these tactics demonstrates 21 Defendant’s willful acquisition, and maintenance, of market dominance through anticompetitive 22 conduct in the relevant input market. 23 Accordingly, having conducted the above rigorous analysis, the Court finds that Plaintiffs 24 have shown, by a preponderance of the common evidence and for purposes of class certification, 25 that Defendant’s anticompetitive Scheme violated the antitrust law. 26 The Court further finds that the resolution of certain common questions regarding Zuffa’s 27 antitrust conduct predominates over individual ones as it relates to this antitrust scheme, including 28 a.) was the scheme applied broadly and consistently across the class, b.) did Zuffa’s exclusionary - 42 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 43 of 80 1 contracts lock up fighters in the relevant market, c.) did Zuffa’s anticompetitive conduct prevent 2 competitors from accessing fighters and fighters from accessing its competitors, d.) did Zuffa 3 intentionally and successfully eliminate meaningful competition, and e.) did Zuffa’s market 4 dominance lead to the systemic suppression of fighter compensation. These common questions of 5 fact and law predominate over the individual questions as to this antitrust scheme. Olean, 31 F.4th 6 at 669. As further elaborated below in this order, the Court does not find that individualized 7 questions about the unlawful conduct alleged here will “overwhelm” these “common ones.” Id. 8 9 10 3. Common Application of the Scheme to Class Members 11 The Court finds next that Plaintiffs have established by a preponderance through common 12 evidence that this anticompetitive Scheme was commonly applied to the class. The Court finds 13 that the nature of this violation necessarily entails questions of law and fact that are common to 14 class members, which predominate over individualized questions. 15 First, the Court finds that the fighter contracts involved all class members. Fighters were 16 required to sign them prior to being brought into the pool of UFC fighters. The key contractual 17 provisions described above were also common. Dr. Singer’s data set examined 2,136 available 18 valid PAR Agreements signed by both Zuffa and the fighter from May 2001 to November 2015. 19 Of these, for example, 100% of the contracts had an exclusion clause, a right-to-match period, and 20 tolling provisions for injury and retirement. 21 Further, as previously noted, the Court finds based upon the record that the coercive tactics 22 used by Zuffa to induce fighters into re-signing contracts or risk punishment were commonly 23 applied to all class members. Record evidence indicates both that these tactics were intentionally 24 and consistently used by management to maintain contractual control of fighters and to send a 25 message to fighters that they were essentially stuck with UFC for the life of their careers. The 26 record further demonstrates that Zuffa’s practice and policy was widely known by its fighters. The 27 structure of these deals, particularly the fact that the fighter was only paid when they fought, meant 28 that these tactics were a credible threat to every fighter under contract with Zuffa. Because these - 43 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 44 of 80 1 contractual features are common to the class, the coercive tactics were common to the class, and 2 because the combination of the two dominated fighters’ negotiations for compensation and 3 alternatives to fighting for Zuffa, the resolution of the parties’ factual and legal disputes regarding 4 these issues predominate over individualized questions regarding their use. 5 Finally, the impact of the horizontal acquisitions is also common to the class. The 6 concentration of input (and output) market in Zuffa constrained exit options to competitors for all 7 fighters. As one fighter explained: “There’s nowhere else to go at this point. Strikeforce didn’t 8 exist anymore. Pride didn’t exist anymore. And I had no way of saying no.” Additionally, as WEC, 9 Pride, Strikeforce, and others were subsumed under Zuffa, so too were fighters’ ability to sign with 10 these competitors, further constraining all fighters in the class. 11 12 4. Common Impact/Injury 13 Next, the Court examines whether Plaintiffs have established common injury flowing from 14 the common application of the alleged Scheme. See Big Bear Lodging Ass’n v. Snow Summit, 15 Inc., 182 F.3d 1096, 1101-02 (9th Cir. 1999); In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 16 305, 311 (3d Cir. 2008), as amended (Jan. 16, 2009) (citing 15 U.S.C. § 15). To show antitrust 17 impact, Plaintiffs must establish that the antitrust violation is “capable of being established through 18 a common body of evidence, applicable to the whole class.” Olean, 31 F.4th at 666; see also In re 19 High-Tech Emp. Antitrust Litig., 289 F.R.D. 555, 567 (N.D. Cal. 2013) (finding that plaintiffs 20 “must establish, predominantly with generalized evidence, that all (or nearly all) members of the 21 class suffered damage as a result of Defendants' alleged anti-competitive conduct.”) 22 Here, the Court engages with the record evidence in two distinct inquiries: first, the Court 23 analyzes whether Plaintiffs have met their burden in establishing whether the violation of the 24 antitrust law has resulted in an impact to the putative class members. In other words, the Court 25 analyzes whether and to what extent the class was injured. Second, the Court evaluates whether 26 this impact was common to the class, specifically whether and to what extent common impact 27 relating to the injury predominate over the specific situations or injury of individual class members. 28 To establish this common injury, Plaintiffs rely on Dr. Singer’s expert report, which used - 44 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 45 of 80 1 regression analysis to indicate that the wage share36 of bout class members is negatively and 2 significantly correlated with Defendant’s “foreclosure share.”37 Dr. Singer also described common 3 evidence, based upon qualitative and quantitative data, of an internal compensation structure 4 within Zuffa, including through significant documentary and record evidence, in addition to 5 econometric evidence. All this evidence, Plaintiffs argue, supports the proposition that Zuffa’s 6 monopsony power that artificially suppressed the fighters’ wages was applied across the Bout 7 Class. 8 Regression analysis has been established by economists and courts as a reliable scientific 9 and mathematical method of establishing common injury.38 “In antitrust cases, regression models 10 have been widely accepted as a generally reliable econometric technique to control for the effects 11 of the differences among class members and isolate the impact of the alleged antitrust violations 12 on [class members].” Olean, 31 F4th at 677. Regression analysis is a statistical process for 13 estimating the relationship between a dependent variable and one or more independent variables. 14 It quantifies a correlation between these variables. In a simple linear regression, there is only one 15 dependent and one independent variable. In most situations, as is the case here, however, multiple 16 factors may influence the outcome of an event. For these situations, social scientists use a related 17 technique called multivariate regression to determine the relationship between multiple 18 independent variables and a dependent variable. This modeling determines whether the 19 independent variables have a statistically significant, i.e., a nonrandom, effect on the dependent 20 variable. Once this correlation is established, the modeling permits a further evaluation of the 21 extent of the effect on the dependent variable by the independent variables. See City of Oakland 22 v. Wells Fargo Bank & Co., 972 F.3d 1112, 1119 n.7 (9th Cir. 2020). Rule 23(b)’s predominance 23 analysis requires a district court to determine whether plaintiffs’ modeling is sound and reliable. 24 In re Rail Freight Fuel Surcharge Antitrust Litig., 725 F.3d 244, 255, 406 U.S. App. D.C. 371 25 Wage share is defined as the share of Defendant’s event-specific revenue paid to a fighter that participated in a given event at a given point in time. 36 26 27 28 37 Foreclosure share is the percentage of the overall input market that Defendant has made unavailable to competitors through exclusionary conduct in violation of antitrust law. 38 See generally J. Woolridge, Introductory Econometrics: A Modern Approach (Thompson 4th ed. 2009). - 45 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 46 of 80 1 (D.C. Cir. 2013) (“It is now clear . . . that Rule 23 not only authorizes a hard look at the soundness 2 of statistical models that purport to show predominance—the rule commands it.”). 3 The Court finds that Plaintiffs’ expert Dr. Singer has presented a sound and reliable model 4 that satisfies Plaintiffs’ burden for class certification for the Bout Class.39 To measure class-wide 5 impact on a common basis, Dr. Singer used a multivariate regression model. The model’s 6 dependent variable is revenue share, or the percent of event revenue provided to a fighter who 7 participates in a given event. The model’s key independent variable is Zuffa’s foreclosure share, 8 or the percentage of the MMA market subject to Zuffa’s Exclusive Contracts. Here, Dr. Singer 9 uses his multivariate regression models to estimate the impact that Zuffa’s relative control and 10 dominance of the input market for fighter services, i.e., foreclosure of the market to competitors, 11 had on fighter compensation The model also controls for, or removes, the possible effect of many 12 other variables that may impact compensation. 13 Dr. Singer further defines the relevant input market of fighters in two different ways. As 14 discussed previously in this order, the “tracked” definition of the input market includes fighters 15 who fought for an MMA promotor included in the “FightMetric” database and all fighters 16 associated with MMA promoters that Zuffa acquired; the “ranked” definition of the input market 17 expands the “tracked” definition to also include fighters ranked in the “FightMetric” database. 18 The regression analysis shows that an increase in Zuffa’s foreclosure share in the input market— 19 using either the “Tracked” or the “Ranked” definition—is associated with a decrease in fighter 20 revenue share at a statistically significant level. This further speaks to the model’s ability to capture 21 class-wide impact. 22 Although the parties disagree as to the definition and the appropriateness of the variables 23 used by Dr. Singer, the parties do not dispute that Singer’s results are statistically significant at 24 accepted levels. While statistical significance is not required to establish impact, it is relevant to 25 the analysis. “It is for the [reviewing] judge to say, on the basis of the evidence of a trained 26 27 28 39 To the extent the Court has had to “weigh” the competing expert opinions of the parties as required by Olean for class certification, the Court, as elaborated in this order, finds that the evidence strongly and clearly weighs in favor of Plaintiffs’ expert opinions regarding the certification of the Bout Class. 61 F.4th at 666. The Court reaches the opposite conclusion as to the Identity Class. - 46 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 47 of 80 1 statistician, whether a particular significance level, in the context of a particular study in a 2 particular case, is too low to make the study worth the consideration of judge or jury.” Kadas v. 3 MCI Systemhouse Corp., 255 F.3d 359, 363 (7th Cir. 2001). In various contexts, the United States 4 Supreme Court has declined to set a legal standard for “statistical significance” to establish a 5 relationship between illegal conduct and alleged harm. See, e.g., Watson v. Fort Worth Bank and 6 Trust, 487 U.S. 977, n. 3 (1988) (“We have emphasized the useful role that statistical methods can 7 have in Title VII cases, but we have not suggested that any particular number of "standard 8 deviations" can determine whether a plaintiff has made out a prima facie case in the complex area 9 of employment discrimination.”); Matrixx Initiatives, Inc. v. Siracusano, 563 U.S. 27, 43 (2011) 10 (noting that in the securities fraud context, the reviewing court must conduct a “fact-specific” 11 inquiry into adverse event reports and that “[t]his is not to say that statistical significance (or the 12 lack thereof) is irrelevant -- only that it is not dispositive of every case”). 13 At this class certification stage, the Court has considered all the evidence before it, and 14 concludes that Dr. Singer’s reports and testimony are reliable sources of proof as to antitrust 15 impact. While Defendant makes many critiques of Plaintiffs’ model, these critiques do not 16 undermine the explanatory power of the model for class certification purposes. Instead, Defendant 17 asserts small and merits-based quibbles regarding the definition of variables and the correct data 18 to input into the model—questions reserved for the jury when the case proceeds to trial. 19 Alternatively, Defendant seeks to create a straw man: augmenting the fundamental assumptions of 20 the model and running the regression again to show that if the model is rerun using different 21 assumptions, it fails. Defendant’s alternative model fails to actually grapple with the underlying 22 assumptions of Plaintiffs’ model in any way that would undermine Plaintiffs’ assumptions or 23 render Dr. Singer’s model incapable of proving impact or damages on a class-wide basis. 24 Accordingly, Defendant’s arguments here are unavailing. See, e.g., In re NFL’s Sunday Ticket, 25 2023 U.S. Dist LEXIS 21752 at *28 (“Defendants' arguments against [Plaintiffs’ experts]—that 26 this-or-that figure and assumption should not have been used—go to weight.”). 27 For the reasons stated below, Plaintiffs’ model(s) illustrates the common impact of 28 Defendant’s antitrust violation. The Court finds that, based upon the totality of evidence before it, - 47 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 48 of 80 1 Plaintiffs have established based on common evidence that common questions predominate as to 2 the impact of Zuffa’s foreclosure of the MMA market on the Bout class, and that members of the 3 Bout class suffered economic injury as a result of Defendant’s anti-competitive conduct. 4 The Court’s analysis of common impact occurs in three stages. First, the Court reviews the 5 key features of Plaintiffs’ model. Second, the Court reviews Defendant’s objections and responds 6 to those objections. Third, the Court explains why Plaintiffs’ regression analysis establishes 7 common impact for the Bout Class. 8 a. Evidence of Antitrust Class Injury 9 10 One of Plaintiffs’ main experts, Dr. Singer, set forth the quantitative basis for common 11 injury or impact. Regarding impact, Dr. Singer’s report sets out to accomplish two tasks: (1) to 12 determine whether the alleged antitrust violation caused members of the Bout Class to be 13 compensated below levels that would have prevailed in the absence of the violation and (2) to 14 determine whether injury in the form of under-compensation generated a “common impact” on the 15 Bout Class. In preparing his report, Dr. Singer considered many litigation sources to conduct his 16 analysis. 17 The Court now reviews Dr. Singer’s model(s). The Court finds that the models and his 18 findings provide sound and reliable evidence of Plaintiffs’ argument and class certification, and 19 that Plaintiffs’ expert opinions effectively rebut Defendant’s expert opinions. 20 Dependent Variable: The dependent variable in Dr. Singer’s regression is the portion of 21 Defendant’s revenue from MMA events that went to fighters; this is called “revenue share” or 22 “wage share.”40 23 For any given bout in any given event, the total compensation paid to each fighter in the 24 Bout Class is the aggregation of up to four different streams: (1) show and win purses, (2) 25 discretionary/performance pay, (3) PPV royalties, and (4) letters of agreement.41 The sum of these 26 27 28 40 These terms are used interchangeably by the experts. A small number of top-tier fighters negotiate individual “Letters of Agreement” (“LOA”) with Zuffa prior to a Live MMA Event. These are generally lump sum payments to fighters following a Live MMA Event, with the 41 - 48 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 49 of 80 1 categories represents a fighter’s compensation for a specific bout. The dependent variable in the 2 regression model is equal to the share of Defendant’s event-specific revenue that was paid to a 3 fighter that participated in that event. Defendant’s event revenues were calculated using 4 Defendant’s profit and loss statements and include revenues from ticket sales, PPV and 5 broadcasting fees, web buys, event-specific merchandise, event specific sponsorships, and other 6 event-specific revenue streams. The dependent variable, wage share, is calculated by dividing the 7 fighter’s compensation by Zuffa’s event-specific revenue, resulting in a percentage. 8 Wage or revenue share is used instead of compensation as a proxy for the putative 9 competitive wage. Dr. Singer explains that the reason for this is because in the context of a sport 10 like MMA, where the fighter is the product, revenue share is a better approximation of marginal 11 revenue product (“MRP”)42 than wage level. 12 Key Independent Variable: The key or primary independent variable in understanding 13 antitrust injury or suppression of competitive wages in Dr. Singer’s regression model is 14 “foreclosure share.” 15 Foreclosure share is intended to capture the percentage of the overall input market, i.e., the 16 market for fighter services, that Defendant has made unavailable to competitors through 17 exclusionary conduct in violation of antitrust laws under the Sherman Act. Dr. Singer ran his 18 regression with two alternative definitions of foreclosure. 19 Under the first definition, Dr. Singer classified all Zuffa fighters as foreclosed. Under this 20 definition, foreclosure share converges to market share, fluctuating between 94 and 99% (using 21 the Tracked measure) and between 71 and 91% (using the Ranked measure) in the relevant input 22 markets. Foreclosure share of the “Headliner” input submarket fluctuated between 95 and 99% 23 (revenue weighted) and 52 and 91% (inverse rank weighted). 24 Under the second definition, Dr. Singer classified Zuffa fighters as foreclosed if their 25 amount often contingent on the outcome of the fight. Some LOAs specify bonuses based on a specific outcome (such as a knockout) and may also incorporate sponsorship or endorsement agreements or both. 26 27 28 42 According to basic economic principles, labor in a competitive market receives compensation equal to its MRP, defined as the additional dollar value generated by adding a single extra unit of production resource (e.g., a worker or, in this case, a fighter). In a monopsonist market, the exercise of market power by a firm drives the wage paid to labor below the MRP. Therefore, the gulf between the actual wage paid to a fighter and their MRP is the degree of monopsony power the firm has. - 49 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 50 of 80 1 individual contract was characterized by all three of the following qualities: (1) exclusivity to the 2 UFC, (2) a champion’s clause allowing the automatic renewal of the contract upon the fighter 3 becoming a “champion” and (3) a length of at least 30 months. Pursuant to these calculations, 4 Defendant’s foreclosure share of the relevant input market fluctuated between 91 and 98% (using 5 the Tracked measure) and between 68 and 90% (using the Ranked measure) during the class 6 period. Foreclosure share of the “Headliner” input submarket fluctuated between 91 and 99% 7 (revenue weighted) and 50 and 91% (inverse rank weighted) during the same period. The second 8 definition is the one used in the subsequent analysis in this Order. 9 Control and Fixed Effect Variables: Other variables in the regression control for other 10 factors that may influence fighter compensation. Dr. Singer’s regression model used over 200 11 control variables, in addition to 1,396 fighter fixed effects—effects that control for all a fighter’s 12 individual-specific attributes that remain constant over time. 13 The Court will not review in this order the staggering number of control variables or fixed 14 effects, but the Court does note a few significant ones. The regression model includes a term which 15 controls for individual fighter-level attributes that remain constant over time. There is a ‘win’ 16 indicator to determine whether the fighter won the bout, a variable to control for the fighter’s rank 17 at the time of the bout, a control variable for a fighter’s win/loss record, fixed effects for gender, 18 promoter, weight class, year, country, and venue, as well as a time trend. There is also an indicator 19 variable for how the fight ended (e.g., submission, technical knockout, disqualification, etc.). The 20 use of these variables, and the sheer number of them enable the model to reliably and robustly 21 specify the correlation between foreclosure share and revenue share. 22 Regression Results: The multivariate regression performed by Dr. Singer indicates that 23 the explanatory variables in the model collectively explain a large proportion (more than 65%) of 24 the variation in the dependent variable. As expert opinion confirmed, this is an extremely 25 significant level of explanatory power for a regression model. 26 Regardless of how the relevant input market is defined (i.e., whether by the Tracked or 27 Ranked measure), foreclosure share has a negative coefficient and is statistically significant. 28 Therefore, an increase in the foreclosure share (key independent variable) is associated with a - 50 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 51 of 80 1 decrease in the revenue share (dependent variable). 2 Having reviewed the expert reports in this case and heard the testimony of the experts, the 3 Court is persuaded by more than a preponderance that Dr. Singer’s model represents a 4 scientifically reliable method for establishing antitrust injury to the class based upon the alleged 5 antitrust conduct of Defendant. The Court finds that Plaintiffs’ multivariate regression indicates 6 that foreclosing a substantial share of the relevant input market had the impact of suppressing the 7 amount of revenue share paid to fighters. That is, Defendant’s dominance in the input market for 8 fighter services had the effect of reducing fighters’ compensation below what would have been the 9 competitive compensation for their services. 10 b. Defendant’s Expert Opposition to Dr. Singer’s 11 Analysis 12 Defendant contends that Dr. Singer’s model suffers from multiple deficiencies, rendering 13 the model and thus Plaintiffs’ common evidence incapable of proving common impact to the Class. 14 In response to Dr. Singer’s empirical work, Defendant employed their own experts, 15 including Dr. Robert Topel, to analyze Dr. Singer’s model. After reviewing Dr. Singer’s report, 16 Dr. Topel asserts that Dr. Singer’s analysis contains several flaws which critically undermine the 17 model’s capability of establishing class-wide impact. The Court discusses these objections below. 18 At the outset, the Court notes that the appropriate concerns at this stage are not about the 19 categorization or definitions of data Dr. Singer used or whether he included every conceivable or 20 potential variable in his model. “Challenges along these lines…[go to] matters of weight and 21 probative value for a jury to evaluate.” In re Capacitors Antitrust Litig., 2018 U.S. Dist. LEXIS 22 195310, at *55 (citing In re Urethane Antitrust Litig., 768 F.3d 1245, 1263 (10th Cir. 2014)). In 23 Comcast, the Supreme Court indicated that while a court’s determination about what a statistical 24 regression model may prove or is capable of proving is not a question of fact, disputes raised by 25 “the data contained within an econometric model” may be disputed issues of fact appropriate for 26 a jury. Comcast Corp., 569 U.S. at 36, n.5, see also Olean, 31 F.4th at 662-663. 27 Many of Defendant’s arguments miss this salient point by criticizing how Dr. Singer 28 defined specific variables or what data he did or did not account for in his analysis. While these - 51 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 52 of 80 1 critiques may be raised at the trial to argue against the weight of this statistical evidence, the 2 relevant inquiry here is whether the Court finds that that Plaintiffs’ models and analysis are 3 “capable of showing class-wide impact” in order to satisfy one of the prerequisites of Rule 4 23(b)(3). Id.; see also Olean, 31 F.4th at 667 (“The district court was not required to resolve factual 5 disputes regarding ultimate issues on the merits…therefore, a district court cannot decline 6 certification merely because it considers plaintiffs' evidence relating to the common question to be 7 unpersuasive and unlikely to succeed in carrying the plaintiffs' burden of proof on that issue.”). 8 Yet, to the extent necessary for the determination of certification and the scientific reliability of 9 Plaintiffs’ modeling, the Court does find the record weighs in favor of Plaintiffs’ expert opinions 10 and that Defendant’s substantive critiques do not undermine Plaintiffs’ expert opinions. Id. at 666 11 (noting the need at times to weigh expert opinion for class certification). c. 12 Revenue Share Is An Appropriate Measure 13 The most strenuous criticism of Plaintiffs’ expert testimony is the use of wage or revenue 14 share by Plaintiffs’ expert(s). Dr. Topel contends that this is a fundamentally incorrect dependent 15 variable to demonstrate that Defendant’s conduct increased its monopsony power. As discussed 16 above, the dependent variable in Dr. Singer’s regression is the fighter’s share of event revenue, 17 called wage or revenue share. Dr. Topel asserts that the correct metric is compensation or wage 18 level43 (used interchangeably). 19 Both Plaintiffs’ and Defendant’s experts agree that the economically accurate way to 20 determine compensation level is by examining the fighters’ marginal revenue product or MRP. 21 This is because the MRP allows a labor economist to directly identify and measure the specific 22 contribution a fighter makes to Defendant’s overall revenue. Being compensated at this level is the 23 most economically efficient outcome in a perfectly competitive market. MRP is, however, a 24 notoriously difficult metric to precisely quantify.44 Indeed, Defendant’s expert, Dr. Topel, noted 25 this challenge at the hearing in this case: “there’s no reliable data on marginal revenue product. 26 27 28 43 In his re-running of Dr. Singer’s regression using his dependent variable, Dr. Topel uses the logarithm of compensation from an event in dollars. Using the log of an athletes’ compensation makes the compensation regression informative about changes in athletes’ compensation in percentage terms. 44 See e.g., G. Scully, The Business of Major League Baseball (1989). - 52 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 53 of 80 1 People have tried to go and estimate things. It’s a really complicated process. So, it’s not as if we 2 can just go out and collect data on marginal revenue product.” Both sets of experts acknowledged 3 that econometricians are thus forced to develop reasonable assumptions when defining MRP in 4 their models. 5 The parties’ experts disagree as to whether wage or revenue share can serve as an 6 appropriate proxy for MRP or competitive wage in a competitive market. Defendant and its experts 7 argue that wage share is an inappropriate proxy for marginal revenue product because (1) it is not 8 commonly used and (2) when examining negative impact due to harmful employer activity, 9 workers are most concerned with the dollar value of their compensation, not whether their share 10 of revenue is increasing or decreasing. To determine impact, Defendant’s experts suggest 11 modeling a scenario where the foreclosure share is zero and comparing the logarithm of wage 12 compensation in that scenario to the log of compensation with the real-world foreclosure share. 13 In response to Dr. Singer’s analysis, Defendant’s expert Dr. Topel runs a separate 14 regression using a different model. In Dr. Topel’s model, the dependent variable is the logarithm 15 of wage compensation instead of revenue share.45 Instead of substantively challenging the 16 underlying assumptions of Plaintiffs’ model, Dr. Topel, however, attempts to sidestep Plaintiff’s 17 analysis by introducing a model with different underlying assumptions. Using his model, Dr. Topel 18 found that there was no statistical significance between foreclosure share and his chosen dependent 19 variable (actual compensation). The Court finds Dr. Topel’s model to be unpersuasive in terms of 20 its ability challenge the fundamental soundness of Plaintiffs’ model. Dr. Topel’s model fails to 21 account for the fact that wages may rise in a monopsony but just not to what would be their 22 competitive level in a competitive market. Although he admitted that it is possible for absolute 23 wages to increase even where monopsony power was suppressing the MRP (and therefore causing 24 harm to a putative class of fighters), he could not explain how his alternative model took this 25 possibility into account. His model thus failed to test the proposition at the heart of Plaintiffs’ 26 model. 27 28 See generally D. Baldus & J. Cole, Statistical Proof of Discrimination, 76 (1980) (explaining that the “[u]se of exponential or logarithmic terms . . . may compensate for nonlinearity.”). 45 - 53 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 54 of 80 1 Defendant also argues that wage share has not been used in regression analysis in the 2 manner that Plaintiffs have used it in this case. The Court is persuaded that wage share is an 3 appropriate proxy for MRP. First, this industry is unique in several respects. In this market, the 4 fighters, themselves, are the product. The expert testimony and record establish that this industry 5 is different from the standard employer-worker relationship. For most workers, the connection 6 between their productivity and firm revenues is attenuated. But that is not the case for professional 7 athletes. Expert opinion in this case confirmed that in sports, particularly in combat sports, such 8 as MMA or boxing, there is a clear connection between revenues generated and individual fighters’ 9 MRP. See also, Lawrence M. Kahn, The Sports Business as a Labor Market Laboratory, 143 J. 10 Econ. Persps., 75-94 (2000) (noting that there “is no research setting other than sports where we 11 know the name, face, and life history of every production worker and supervisor in the industry”). 12 Additionally, the Court notes that Defendant itself (or its agents) used wage share to 13 evaluate its own business; this further bolsters the Court’s finding that wage or revenue share is 14 appropriate in this industry. For example, a presentation prepared by WME-IMG, Defendant’s 15 parent company, explicitly analyzes fighter “costs” (i.e., Zuffa’s payments to fighters) in terms of 16 fighters’ share of company revenue. Specifically, WME-IMG noted that fighter costs fluctuated 17 between 16 and 19% of revenue between 2012 and 2015. The presentation states that the risk of 18 “Fighter Compensation Inflation” was “the most asked question by financing sources and is a 19 critical cost that we must actively manage.” In response, Zuffa management intended to “contain” 20 fighter costs at no more than 20% of revenue. Additionally, in preparation for Mr. Lorenzo 21 Fertitta’s interview with the New York Times, Zuffa prepared talking points comparing UFC 22 fighter compensation to “Other League Payroll Comps,” including MLS, NHL, NBA, and MLB. 23 Defendant fails to demonstrate that Plaintiffs’ use of wage share is statistically unsound or 24 fatally flawed. While Defendant challenges some of the assumptions of Dr. Singer’s model which, 25 at best, go to the weight of this evidence, the defense experts do not sufficiently establish any 26 internal errors or problems with the statistical model itself that undermine its reliability and 27 validity. Indeed, they used similar econometric models, albeit with different assumptions and 28 variables. Thus, Defendant’s arguments are relevant for the ultimate factfinder at trial but have not - 54 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 55 of 80 1 established that Plaintiffs’ model is fundamentally flawed or fundamentally unreasonable in its 2 assumptions. 3 The Court also finds that the reliability of Plaintiffs’ modeling is further augmented by the 4 extent and detail of the data available to and used by Plaintiffs. As Plaintiffs’ expert Dr. Manning 5 points out, the data available in this case is far more extensive than is required or is found in other 6 contexts. Here, there is specific, event-level data that can directly link the contributions of 7 particular fighters to event revenue. As he explained: “[w]e have an event, a discrete flow of 8 revenue for a particular piece of work, and we know exactly which worker were associated with 9 that.” This amount of data and its detail allow for the measurement of wage share in a robust and 10 detailed manner. Many other studies have been able to measure and hence use wage share as a 11 variable in regression analysis, because this level of detailed data is not usually available. It is a 12 basic principle of econometrics that a more complete data set will lead to more accurate results 13 within regression analysis. See generally William H. Greene, Econometric Analysis, 277 (1990) 14 (discussing the limited accuracy and predicative power of incomplete data sets). 15 Moreover, the Court finds that Defendant’s experts, Dr. Topel and Dr. Oyer, through their 16 own work (and based upon their testimony), engaged in econometric analysis that has relied upon 17 revenue or wage share in the context of measuring or evaluating monopoly/monopsony power. 18 See, e.g., Roger Blair, Sports Economics 354 (Cambridge University Press 2012) (explaining that, 19 when professional athletes are paid less than their MRP, “[t]his is exploitation in the sense that the 20 player is paid less than his value to the club”); see also, David Berri, Michael Leeds, & Peter von 21 Allmen, Salary Determination in the Presence of Fixed Revenues, 10 International Journal of Sport 22 Finance 5-25 (2015). Dr. Topel’s own consulting work has implemented a detailed analysis of 23 wage share in professional football, undermining the credibility of his asserted opinion, in this 24 case, that revenue share is a “useless” measure. See Kevin Murphy & Robert Topel, Chicago 25 Partners, The Economics of NFL Team Ownership, (2009) (“Considering the changes in the 26 players' overall percentage over time, it appears that the players’ percentage did not keep pace with 27 the growth in NFL revenues after 2000. To the extent the 2006 CBA extension resulted in increases 28 - 55 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 56 of 80 1 in the salary cap, those increases have raised the players' percentage so that it is closer to the 2 historical average of approximately 59 percent of revenue” (emphasis added)) 3 Additionally, the Court finds that, boxing, the sport industry that most closely parallels 4 MMA also relies upon revenue share to understand player compensation. Boxing events are 5 organized by promoters who set up professional boxing contests pitting two individuals against 6 each other in hand-to-hand combat, governed by a set of rules established by state athletic 7 commissions. As with professional MMA fighters, boxers’ careers are relatively short, and they 8 face enormous risks each time they enter the ring. Promoters compensate boxers with “purses” in 9 exchange for appearing in bouts. While the market structure of boxing is somewhat different from 10 that of MMA,46 antitrust challenges have been mounted against boxing promoters.47 In the context 11 of these challenges, compensation was guaranteed as a percentage of event revenue, rather than as 12 an absolute compensation amount. In reviewing the record and hearing the expert testimony, the 13 Court finds that the comparison of aspects of boxing compensation to MMA for the purpose of 14 evaluating competitive compensation in MMA is reasonable. More specifically, this further 15 supports the Court’s finding that using revenue share as the dependent variable in Plaintiffs’ 16 regression modeling is appropriate. Dr. Singer’s model uses revenue or wage share as the 17 dependent variable and this does not depart from established methodological choices in sports 18 19 46 20 21 22 23 24 Unlike the limited competition in the input/output markets of MMA, boxing today has multiple prominent promoters who compete with each other to sign the best boxers, including titleholding championship boxers and contenders for titles, and who can promote PPV matches that attract large audiences. There are at least five different boxing promoters who are prominent enough to promote major championship fights each year, and probably more than ten. There are at least twenty boxing promoters in the United States today who have promoted nationally televised events, and of those, more than five have promoted at least one PPV event. These promoters compete with each other to sign boxing talent. It would appear based upon the record and experts that the MMA industry could evolve to such a competitive industry through the reduction of barriers to entry and restrictions on anticompetitive conduct within the industry. For example, in the 1950s the U.S. Department of Justice brought an antitrust suit, United States v. Int’l Boxing Club of NY, Inc., against two boxing promoters, James Norris and Arthur Wirtz, who had implemented a conspiracy to monopolize top-level prize fighting in 1949. Their scheme enabled them to gain almost complete control over the promotion of championship boxing matches, by among other strategies: (1) acquiring rival promoters; (2) signing the four championship-contender boxers to two-year exclusive contracts, with options to renew for additional two-year terms; and (3) requiring contenders for a title, as a condition of being afforded an opportunity to engage in a championship contest, to enter into “contingent” exclusive contracts pursuant to which the contender, if he won the contest and thereby became champion, was required to engage in title bouts exclusively under the promotion of defendants for a period of from three to five years. 47 25 26 27 28 - 56 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 57 of 80 1 economics; with this chosen dependent variable, the model is capable of determining class-wide 2 impact and injury. 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 d. Foreclosure Share Is An Appropriate Measure Defendant also criticizes Plaintiffs’ calculation and use of the key independent variable: “foreclosure share.” In Dr. Singer’s regression, foreclosure share is the percentage of fighters (1) in the input market that (2) have exclusive contracts with Zuffa, which are defined as (3) containing a champion’s clause and are over 30 months in duration. Defendant argues that the foreclosure share does not measure what effect the contracts had on competition or whether these fighters were actually foreclosed from the ability to compete in the market. Defendant also argues that other courts have concluded that contracts lasting more than 12 months are not anticompetitive. Additionally, Dr. Topel criticizes the data used in Dr. Singer’s foreclosure share calculation. The Court disagrees with Defendant and finds that Plaintiffs in fact link the mechanics of Defendant’s fighter contract to foreclosure effect. Dr. Singer credibly establishes Defendant’s contracts—which contain exclusionary clauses affecting renewals and extensions—effectively occupy almost the total duration of a fighter’s effective career. This, by definition, restricts athlete mobility: the fighter would effectively have only one contract for their viable professional career. Defendant thus suppresses compensation by reducing the availability of outside offers. This enhances its negotiating leverage and monopsony power, all the while transferring value from the fighters to itself. As Zuffa’s foreclosure share increases, its competitors’ access to the critical input of fighters decreases and their concomitant ability to increase or maintain revenue also declines. The rise in Zuffa’s foreclosure share thus prevents any meaningful competition in the market. Documentary evidence, as noted earlier, further indicates that Defendant understood their actions were having an anticompetitive effect on rivals and engaged in this conduct for that very reason. Emails by Zuffa personnel indicate that Zuffa understood its contracts with fighters prevented other MMA promoters from gaining “traction” by keeping fighters “locked up” or “tied up” and unavailable to other promoters. Further, Dr. Singer found that, during the Class Period, Defendant’s a foreclosure share ranged between 52 and 99%. The Court finds that based on the record before it, and for all the reasons outlined above, Plaintiff has shown the relationship between - 57 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 58 of 80 1 Defendant’s fighter contracts and foreclosure of competition in the MMA market. 2 For these reasons, the Court rejects Defendant’s reliance on other cases that have held that 3 contracts that last longer than 30 months were not anticompetitive. Defendant cites to two cases 4 outside of this Circuit to argue that “Singer’s opinions regarding foreclosure share” have been 5 rejected as “unreasonable” and that in antitrust cases, courts only “condemn” contracts exceeding 6 five years. ECF No. 742 at 3, 7 (citing Maxon Hyundai Mazda v. Carfax, Inc., No. 13-CV-2680 7 (AJN), 2016 WL 7231941 (S.D.N.Y. Dec. 9, 2016) and Spinelli v. NFL, 96 F. Supp. 3d 81 8 (S.D.N.Y. 2015)). Whether the duration of a contract is anticompetitive is a very fact intensive 9 inquiry, highly dependent upon the market and labor environment. The contracts at issue in the 10 Spinelli were licensing agreements between professional photographers and the NFL. Spinelli, 96 11 F. Supp.3d. Mazda dealt with exclusive dealing agreements between car manufacturers and used 12 car websites. Neither of these situations is close to approximating the contractual situation at issue 13 in this case, where Defendant’s contracts bound fighters for, essentially, their entire professional 14 careers. These decisions are therefore inapposite for determining the appropriate duration of a 15 contract in this situation. Furthermore, Dr. Singer successfully responds to Dr. Topel’s assumption 16 that the median career length for fighters is 8.9 years. Dr. Singer convincingly argues that Dr. 17 Topel’s alternative median career length calculation includes fighters outside the previously 18 defined relevant input market and includes time the fighter spent in the “minor leagues” where 19 they were not, realistically, competing for a contract with Zuffa or any of its competitors. 20 The Court finds, based on the full record before it, that foreclosure share, as defined in Dr. 21 Singer’s model, is a reasonable metric. Furthermore, Defendant's fighter contract terms reflect 22 exclusionary conduct and operate as a foreclosure mechanism in this very specific and unique 23 labor market. The Court finds that Defendant’s arguments do not fundamentally or substantially 24 undermine Plaintiffs’ argument and modeling. While Defendant may raise these various merit- 25 based or factual arguments at a trial, the Court finds that Plaintiffs have met their burden here. 26 Defendant makes other arguments regarding foreclosure share. For example, Defendant’s 27 experts challenge the appropriateness of using revenue weighting, inverse rank weighting, and 28 giving UFC fighters the same average weight. Defendant, however, continually applies the wrong - 58 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 59 of 80 1 standard to this class certification stage. As discussed above, so long as Plaintiffs have met their 2 burden in demonstrating that the model is capable of showing class-wide impact, Plaintiffs have 3 satisfied the requirements of this element of the legal inquiry. The Court does not find that any of 4 Defendant’s arguments fundamentally or substantively undermine the reliability and explanatory 5 effect of Plaintiffs’ modeling. Whether Plaintiffs’ modeling is sufficient to establish liability is a 6 jury question. See Comcast, 569 U.S. at 36 n.5. Defendants are free, at trial, to raise all of these 7 fact-intensive arguments regarding Plaintiffs’ modeling, but the Court is not persuaded that they 8 defeat class certification. 9 Furthermore, the Court finds that the assumptions embedded in Dr. Singer’s choice to 10 weight foreclosure share are reasonable and further support the Court’s finding that his model is 11 robust and reliable. Dr. Singer’s model applies two alternative weighting schemes: the first is 12 based on promoter PPV and gate revenue, and the second is based on fighters’ relative rankings. 13 Dr. Topel argues that Dr. Singer’s model is fatally flawed because it uses a weighted foreclosure 14 share as its key independent variable. To the contrary, Dr. Singer’s use of a weighted foreclosure 15 share supports a finding of robustness. As Dr. Singer explains: “Given that the point of assessing 16 the degree of foreclosure here is to determine the share of a critical input (namely, [f]ighters) Zuffa 17 has locked-up in exclusive arrangements, it would not make sense to ignore the relative differences 18 between [f]ighters and/or the promotions for which they fight.” Assigning the weight to fighters is 19 appropriate because the products here (individual fighters) do not each represent the same value 20 to the market. Dr. Topel’s quarrel is with the method of weighting by promoter, rather than 21 individually. This is, however, a red herring. The Court is persuaded by Dr. Singer’s argument that 22 revenue weights are agnostic to individual variation because of the way they are aggregated in the 23 foreclosure share calculation. Foreclosure share is a measurement of market dominance; what 24 matters is the relative weighting of Zuffa fighters vis-a-vis non-Zuffa fighters. 25 e. Use of Pre-Strikeforce Acquisition Data Is 26 Appropriate 27 Defendant also challenges the use of Strikeforce as a benchmark by Plaintiffs. Dr. Singer’s 28 regression data includes bouts from Defendant’s former rival promoter, Strikeforce. These bouts - 59 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 60 of 80 1 occurred before Zuffa acquired Strikeforce in March of 2011. At the time of purchase, Strikeforce 2 was “the second most prominent and recognized MMA organization in the world” and had the 3 most prominent women’s MMA division. Strikeforce produced 47 events across the United States 4 before being acquired. In some of his regressions, Dr. Singer uses pre-acquisition Strikeforce bouts 5 to serve as a benchmark to determine the compensation Zuffa athletes would have received absent 6 the anticompetitive conduct. 7 Dr. Topel argues that Dr. Singer’s use of pre-acquisition Strikeforce data is not an 8 appropriate benchmark for Zuffa. Dr. Topel submits that Dr. Singer does not perform the statistical 9 tests necessary to determine whether Strikeforce bouts observed prior to 2011 are a valid 10 benchmark for Zuffa bouts and that Dr. Singer did not adequately account for the factors that 11 underlie the differences in compensation between UFC and Strikeforce athletes. 12 Dr. Topel conducted a Chow test (or F-test) to determine the stability of coefficients across 13 subgroups or periods of time. And he indicates that Dr. Singer’s regression model fails all of the 14 Chow tests. From this finding, Dr. Topel concludes that pre-acquisition Strikeforce bouts are not 15 an effective benchmark for measuring what Zuffa athletes’ compensation would have been absent 16 Zuffa’s anticompetitive conduct. Dr. Topel also asserts that when he removes the Strikeforce 17 benchmark, Dr. Singer’s regression model predicts zero damages. 18 The Court finds, however, Dr. Singer’s rebuttal to be dispositive of this objection. As Dr. 19 Singer explains, the Chow test implemented by Dr. Topel was an improper and biased way to 20 challenge his model. This is because Dr. Topel tests the null hypothesis that all of the regression 21 coefficients for the Strikeforce pre-acquisition bouts are identical to the regression coefficients for 22 Zuffa. In other words, it tested whether all 63 coefficients on the independent variables on the 23 regression for the Zuffa bout data exactly match all 63 coefficients on the independent variables 24 for the Strikeforce data. In this instance, the null hypothesis could be rejected if even one of the 63 25 coefficients subjected to test were different. See generally, Woolridge, supra at 245 (“one 26 important limitation of the Chow test regardless of the method used to implement it, is that the null 27 hypothesis allows for no differences at all between the groups”). The Court thus agrees with Dr. 28 Singer that Dr. Topel’s application of the Chow test does not provide a basis for rejecting Dr. - 60 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 61 of 80 1 Singer’s model. It is not a reasonable test of the validity of Dr. Singer’s model given the number 2 of variables involved. See Olean, 31 F.4th at 675-76 (“Because there was a rational basis for 3 [Plaintiff’s expert’s] . . . model to demonstrate class-wide impact . . . the failure of the Chow test 4 did not require the [district] court to reject the model.”). Defendant is not precluded from raising 5 arguments about this and other statistical tests when the case proceeds to the merits. However, at 6 this juncture, inclusion of the pre-acquisition Strikeforce bout data does not undercut the reliability 7 and plausibility of Dr. Singer’s model. Plaintiffs’ 8 f. 9 Misspecified Control Variables Are Not 10 Defendant argues that multiple control variables in Plaintiffs’ modeling are misspecified, 11 because they do not have the expected statistically significant relationship with wage share. For 12 example, Defendant argues that the Win Flag, Rank, and PPV variables do not have the expected 13 relationship with the athlete’s fighter share. Defendant argues that these variables do have the 14 expected relationship when the dependent variable is the log of compensation, i.e., Dr. Topel’s 15 chosen dependent variable in his separate model. 16 The Court is again unpersuaded that this difference in the expected relationship of certain 17 variables with the fighter share renders Plaintiffs’ model unreliable for measuring common impact 18 and injury. As Dr. Singer explained, what Defendant notes is a statistical phenomenon called 19 multicollinearity.48 These results are likely due to multicollinearity between control variables in 20 the multivariate regression. More importantly, this multicollinearity, especially as it relates to 21 control variables, is not uncommon and it does not defeat the validity or predictive power of the 22 model.49 As Dr. Singer credibly explains: 23 24 25 26 27 28 48 See Wooldridge, supra at 147-48 (showing that jointly significant variables appear singly insignificant due to multicollinearity); see also In re: Elec. Books Antitrust Litig, 2014 U.S. Dist. LEXIS 42537 at * 79 n. 34 (S.D.N.Y Mar. 28, 2014) (“[A] model might estimate a smaller coefficient (influence) for one variable and a greater coefficient for other correlated variables; while the overall effect of these variables may be strongly statistically significant, each coefficient may not appear statistically significant on its own”). 49 See A. Gelman, J. Hill, and A. Vehtari, Regression And Other Stories 407 (Cambridge University Press 2021) (“Students learning about regression are often urged to avoid multicollinearity. However, since our focus is only on estimating the treatment effect, and the goal . . . is to create independence between the covariates and the treatment indicator, we are not particularly concerned about issues of multicollinearity among our covariates.”) - 61 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 62 of 80 1 2 3 4 5 6 7 8 9 10 11 12 13 Given the large number of control variables in my impact regression (a total of 270, in addition to 1,396 Fighter fixed effects), it is not at all surprising that the coefficients on some are individually statistically insignificant. When multiple variables move together (that is, they exhibit “collinearity”), a regression may find that they are individually statistically insignificant, but this does not imply that these control variables are not collectively significant. A standard F-test for the collective significance of Win Flag and other control variables—all of which are individually statistically insignificant at the five percent level—confirms that these variables are collectively highly statistically significant at the five percent level, indicating that these variables collectively contribute significantly to the explanatory power of the regression. Indeed, when Fighter Share is regressed on Win Flag individually (dropping all other explanatory variables from the model), the coefficient is positive and highly significant. This proves that the Win Flag variable is not “misbehaving” in my model, as Dr. Oyer wrongly asserts. Finally, the coefficient on Win Flag is positive and highly significant when we drop a single variable (such as the number of prior Wins) from my regression model. 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 The Court finds this analysis compelling and rejects Defendant’s critique on this point. The Court is persuaded by Dr. Singer’s explanation that the multicollinearity noted in these variables, and a few others, does not render the model invalid and arises out of the sheer number of control variables and detailed data available to the experts. Moreover, the purpose of the model is to determine the impact of Zuffa’s control or foreclosure of the input market on revenue share rather than assessing the impact of each of the various “control” variables. The modeling undisputedly demonstrates a relationship between Zuffa’s foreclosure share and revenue share at a statistically significant level. To conclude, the Court finds that Plaintiffs’ have established by a preponderance of the evidence that Defendant’s antitrust violation resulted in an impact to the putative class members. Evidence of this injury was provided, in part, by a statistical regression model. This regression describes the relationship between Zuffa’s actions to restrict fighter mobility in the market (foreclosure share) and fighter compensation as a portion of revenue (wage or revenue share). The model demonstrates that there was an inverse relationship between these variables: as Zuffa increasingly dominated the MMA market, fighters’ wages were suppressed. In its rigorous analysis - 62 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 63 of 80 1 of the model, the Court finds that the class certification standard is satisfied. The Court turns now 2 to evaluating whether this impact was common to the class, and to what extent common issues 3 relating to the injury predominate over the specific situations of individual class members. 4 g. 5 Extent of Class-Wide Injury 6 The Court now evaluates whether the impact described above was common to the class, 7 specifically whether and to what extent common issues relating to the injury predominate over the 8 specific situations of individual class members. The Court ultimately finds that Plaintiffs establish, 9 based on a preponderance of the evidence, that common questions predominate as to common 10 impact. 11 Plaintiffs use both statistical and documentary evidence to show that the effect 12 demonstrated by the multivariate regression is transmitted broadly across the Bout Class. Dr. 13 Singer captures this by running a separate regression where the dependent variable is the natural 14 log of a fighter’s event compensation and the independent variables include common and 15 objectively measured factors such as weight class, rank, gender, placement on the event card, year, 16 country, and venue, which affect fighter compensation. The regression results indicate that over 17 three-quarters of the observed variation in fighter compensation is explained by common factors. 18 Then, Dr. Singer performed a separate regression to determine whether gains (or losses) in 19 compensation were broadly shared across the Bout Class. In this regression, the dependent variable 20 was an individual fighter’s annual compensation per event and the independent variable was set 21 equal to either (1) the average compensation per event paid to all other fighters in that year or (2) 22 the average compensation per event paid to all other fighters in the prior year. The results of this 23 statistical test indicate that a fighter’s annual compensation per event is positively associated with 24 the per-event compensation paid to other fighters during that same year and the results were 25 statistically significant; in other words, fighter compensation moves together. These regressions 26 provide reasonable evidence of Plaintiffs’ argument that a rise or fall in compensation was and is 27 28 - 63 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 64 of 80 1 common across class members.50 2 Plaintiffs’ statistical evidence of common impact is buttressed by documentary evidence, 3 which separately and independently establishes class-wide impact. Internal Zuffa documents 4 demonstrate that fighters were generally compensated in identifiable tiers using objective factors 5 in order to ensure, according to matchmaker Mr. Shelby, “uniformity” in the payment structure. 6 The record indicates that negotiations of fighter contracts centered around the compensation of 7 comparable fighters based on weight class, number of prior bouts, record, and championships. In 8 an email to a fighter’s manager, Chief UFC Matchmaker, Mr. Silva, explained, “[a]s I said, I have 9 a pay structure. I cannot mess it up for 1 fighter. I have to justify that to all the other managers… It’s 10 great if other people want to pay him more but that’s not how my business model works.” (emphasis 11 added). In his testimony at the evidentiary hearing, Mr. Silva elaborated that he determined the 12 parameters of fighter pay using objective factors such as expected performance and popularity. 13 Mr. Silva also testified that he wanted to have a comparable pay scale for fighters where he could 14 verbalize the specific factors that went into compensation. There is also documentary evidence of 15 a consistent compensation schedule: a contract will specify a show and win purse of a specific 16 amount for the first bout and increase that amount in a predetermined increment throughout the 17 bouts guaranteed by the contract (e.g., Bout #1 compensated at a show/win rate of $7,000/$7,000, 18 Bout #2 compensated at a show/win rate of $9000/$9000, Bout #3 compensated at a show/win rate 19 of $11,000/$11,000, etc.).51 20 Defendant argues that Plaintiffs fundamentally misstate how Zuffa determines 21 compensation. Indeed, its expert, Dr. Topel, asserts that compensation evolves over time, is 22 23 24 25 26 27 28 50 This methodology of showing class-wide impact has been used by economists in similar cases and accepted by courts as being capable of demonstrating class-wide impact. See In re High-Tech Emple. Antitrust Litig., 289 F.R.D. 555, 570 (N.D. Cal. 2013). 51 Internal documents exhibited by the Plaintiffs reveal that there was internal acknowledgement of this pay structure: In an internal email discussing fighter pay, UFC matchmaker Mr. Shelby wrote: “[G]oing forward please go up by 2’s most all of the time for the Strikeforce events, especially for a guy like Conor [McGregor]. For reference we start the majority of guys coming in at 6+6 in UFC and always go up by 2’s. To separate the challenger guys, top them out at 5 whenever possible. I get it if it ends up at 6K if signing a lot of fights for the Challengers. I say 5K to differentiate the two shows. I know it sounds like we are nickel and diming but it[’]s for the uniformity of it all and makes it so much easier for everyone the more boilerplate it is at the mid to lower end.” (emphasis in original). Mr. Shelby confirmed this approach in his deposition. - 64 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 65 of 80 1 specific to each athlete, and cannot be reduced to a formula. Dr. Topel further elaborates that Dr. 2 Singer’s regression analysis of athlete’s compensation is inherently flawed because the fact that a 3 fighter’s pay tends to move with the average pay does not imply the existence of a “pricing 4 structure.” In response, Dr. Singer contends that Dr. Topel misunderstands the nature of the inquiry 5 by Plaintiffs. By a “common structure,” Plaintiffs are referring to “a mechanism that would 6 transmit the artificially reduced compensation. . .broadly across the Class,” Dr. Singer thus 7 convincingly rebuts Dr. Topel’s claim that compensation dependent on individual fighter 8 characteristics such as “notoriety,” “representation” and “rival promoters’ interest in an athlete” 9 prevents the ability of the model to identify class-wide impact across the common compensation 10 structure. These individual factors do not disprove a common structure; they instead affect where 11 a fighter falls within Zuffa’s established compensation structure. Indeed, these individualized 12 factors are clearly contemplated by the inclusion of independent control variables based on various 13 effects. The fact that absolute fighter compensation varies in some ways by athlete does not imply 14 that there is no underlying commonality to the way compensation is distributed. The Court finds 15 that Dr. Topel’s alternative regressions do not present a convincing challenge to Dr. Singer’s 16 model or to the qualitative evidence of a systemic pay structure based upon objective individual 17 factors. Individual and collective consideration of the qualitative and quantitative data establishes 18 the existence of class-wide pay structure which distributed class-wide injury, and collective 19 questions predominate here. Further, Dr. Topel arrives at a different result than Dr. Singer in part 20 because of his rejection of all data prior to 2010, representing 19% of the observations used in Dr. 21 Singer’s analysis. As discussed at length throughout this order, Plaintiffs have presented a 22 reasonable and appropriate basis for including such data, and Defendant may still challenge this at 23 a trial. For these reasons, the Court is not persuaded, as to class certification, by Defendant’s 24 arguments that the regression does not determine a common compensation structure for the 25 purpose of determining whether the model is capable of showing class-wide impact. 26 Defendant also challenges predominance based upon its assertion that Plaintiffs’ statistical 27 analysis still shows a significant number of uninjured class members. When individualized 28 questions relate to the injury status of class members, Rule 23(b)(3) requires that the court - 65 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 66 of 80 1 determine whether individualized inquiries about such matters would predominate over common 2 questions. See Cordoba v. DIRECTV, LLC, 942 F.3d 1259, 1277 (11th Cir. 2019). The question 3 of what constitutes sufficient class-wide injury for class certification has been the subject of much 4 recent litigation. Circuits have held that the de minimis threshold is anywhere from 5 to 6%, In re 5 Rail Freight Fuel Surcharge Antitrust Litig., 934 F.3d 619, 624-25 (D.C. Cir. 2019) to “around 6 10%” In re Asacol Antitust Litig., 907 F.3d at 47, 51-58 (1st Cir. 2018). Recently, however, the 7 Ninth Circuit en banc overturned its previous elaboration of the de minimis rule, concluding that 8 9 10 11 12 13 14 15 16 a district court is not precluded from certifying a class even if plaintiffs may have to prove individualized damages at trial, a conclusion implicitly based on the determination that such individualized issues do not predominate over common ones…Therefore, we reject the dissent's argument that Rule 23 does not permit the certification of a class that potentially includes more than a de minimis number of uninjured class members. This position is inconsistent with Rule 23(b)(3), which requires only that the district court determine after rigorous analysis whether the common question predominates over any individual questions, including individualized questions about injury or entitlement to damages. Olean, 31 F.4th at 668-69. 17 Defendant makes various allegations to assert there is insufficient class-wide injury. 18 Defendant argues that, in Dr. Singer’s first report there were 1,214 members of the class, but Dr. 19 Singer’s second report only reports impact based on a class size of 1,056. In other words, it claims 20 that Dr. Singer has only been able to show impact to 87% of the class. The Court, however, finds 21 that Plaintiffs show that common questions predominate over individualized questions under either 22 standard: the de minimis rule or the more capacious rule established by the Ninth Circuit in Olean. 23 Defendant’s statement regarding the change from 1,214 class members to 1,056 class members is 24 misleading.52 Defendant claims that this change indicates that Plaintiffs do not have evidence that 25 26 27 28 52 Dr. Singer indicates in his first report that 1,214 fighters is his estimate as of June 30, 2017. Earlier in his report, Dr. Singer mentions that if the class definition were interpreted to be limited only to those class members who fought in a bout that physically took place within the United States during the Class Period (regardless of whether or not it was broadcast live into the United States), 12.7% of the Bout Class members would be excluded (154 out of 1,214 fighters). - 66 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 67 of 80 1 13% of the class is injured. The Court rejects this claim as a misstatement of the record and 2 Plaintiffs’ modeling. As Dr. Singer describes, at length, fighters are considered uninjured where 3 they may have benefitted from the antitrust conduct by earning relatively more than their peers. 4 Dr. Singer’s calculation of these individuals is consistent across his reports: initially he found that 5 the model did not show impact for 14 of the most prominent fighters at the UFC. In his second 6 report, he performed an analysis using the same data set used by Dr. Topel in doing a similar 7 calculation in an attempt to elucidate why Dr. Topel’s calculation of uninjured class members in 8 his report was incorrect. In doing this calculation, he again shows that the number of uninjured 9 class members is between 4 and 14 people out of a class of over a 1,000 fighters. The Court is 10 persuaded by this analysis. The denominator or size of the class may fluctuate based on the data 11 set used by the experts, but there is no evidence that this change has affected the number of 12 uninjured class members such that common issues will not predominate. In either case of class 13 size, the estimated percentage of uninjured fighters is below the de minimis threshold. Therefore, 14 common issues predominate individualized issues. 15 Plaintiffs have persuaded the Court by a preponderance that they can establish class-wide 16 impact and injury through quantitative and qualitative data. Dr. Singer’s regression model, 17 supplemented by documentary evidence, is compelling and capable of proving class-wide antitrust 18 impact, notwithstanding Defendant’s arguments. See Olean, 31 F.4th at 681. Based upon the 19 totality of evidence evaluated by the Court, Plaintiffs have presented sufficient evidence showing 20 common questions regarding impact predominate over individual ones. 21 22 h. Damages 23 To show that common questions regarding damages predominate, Plaintiffs must propose 24 damages calculations that rely on common evidence. See Meijer, Inc. v. Abbott Labs, 2008 U.S. 25 Dist. LEXIS 78219 (N.D. Cal. 2008); see also Stuart v. State Farm Fire & Cas. Co., 910 F.3d 371, 26 375 (8th Cir. 2018) (“A class may be certified based on common issues ‘even if other important 27 matters will have to be tried separately, such as damages or some affirmative defenses peculiar to 28 some individual class members.’”) (quoting Tyson Foods, 577 U.S. at 453)). Calculations need - 67 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 68 of 80 1 not be exact, see Story Parchment Co. v. Paterson Parchment Paper Co., 282 U.S. 555, 563 (1931), 2 but “at the class-certification stage (as at trial), any model supporting a plaintiffs’ damages case 3 must be consistent with its liability case, particularly with respect to the alleged anticompetitive 4 effect of the violation.” Comcast, 569 U.S. at 35. “At this stage, Plaintiffs need only show that 5 such damages can be determined without excessive difficulty and attributed to their theory of 6 liability.” Just Film, Inc., 847 F.3d at 1121; see also Bowerman v. Field Asset Servs., Inc., 39 F.4th 7 652, 663 (9th Cir. 2022). 8 The Court finds that Plaintiffs have established at least four different models for 9 determining damages derived from the alleged challenged conduct in the Scheme. Dr. Singer 10 presented two different benchmark models from former competitors Strikeforce and Bellator for 11 establishing the but-for compensation for Zuffa fighters during the Class Period. Strikeforce and 12 Bellator were both competitors of Zuffa; Strikeforce was ultimately acquired by Zuffa. The Court 13 finds that both entities represent, in the words of Defendant’s own expert, “natural candidates for 14 a competitive benchmark.” Beyond this concession by Defense experts, the Court finds that the 15 revenue share Bellator and Strikeforce paid to their fighters is a reasonable measure of calculating 16 the but-for damages of Zuffa’s fighters, even if Strikeforce and Bellator never acquired Zuffa’s 17 same market power.53 18 These two competitive benchmarks yield two different estimates for damages based upon 19 the revenue share Strikeforce and Bellator, respectively, paid to fighters. Based upon the record, 20 Strikeforce paid approximately 63% of its revenue to fighters between 2009 and 2010. Bellator 21 paid approximately 44.7% of its revenue to fighters over that timeframe between 2010 and 2015. 22 Zuffa itself paid approximately 19.5% of its Event Revenue to the Bout Class during the Class 23 53 24 25 26 27 28 The Court finds based upon the record that the use of these benchmark is consistent with elementary economics, which explains that competitive firms pay labor a share of revenue commensurate with labor’s productivity, based on the marginal product of labor. See, e.g., Roy Ruffin & Paul Gregory, Principles of Microeconomics, 331-336 (Harper Collins 5th ed. 1993); see also Michael Katz & Harvey Rosen, Microeconomics 264-65, 276-77 (Irwin McGraw-Hill 3rd ed. 1998). By contrast, firms that wield monopsony power pay a smaller share of revenue of labor, by restricting both the amount of labor hired and the compensation paid to labor. Id. Having reviewed the record and heard the testimony of the experts, the Court agrees with Plaintiffs’ expert that “although the Challenged Conduct, by reducing revenue opportunities for rival promoters, might have suppressed the level of compensation that Strikeforce (and other would-be rivals) could afford to pay to Fighters, the share of revenue paid to Fighters by Strikeforce remains an economically relevant benchmark for estimating but-for compensation.” Thus, the use of such benchmarks may actually underestimate rather overestimate the damages that Plaintiffs suffered. - 68 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 69 of 80 1 Period. Applying these benchmarks to Zuffa’s event revenue for the Class Period, Plaintiffs 2 estimate the damages to be between $811 million (Bellator benchmark) and $1.4 billion 3 (Strikeforce benchmark). 4 In finding these benchmarks to be reliable, the Court reiterates its finding that revenue or 5 fighter share is a reasonable and appropriate measure for calculating the but-for or market 6 compensation for fighters. The Court finds that Plaintiffs have adequately set forth a damages 7 calculation model based upon its impact regression model, which isolates the effect of Defendant’s 8 anticompetitive and violative conduct on fighter revenue share. The Court finds, for the reasons 9 previously stated, that Plaintiffs have proffered a damages model derived from the common impact 10 of Defendant’s antitrust conduct, based on common evidence, and which further aligns with its 11 theory of liability. 12 The Court also rejects Defendant’s arguments opposing these competitors as appropriate 13 benchmarks in the but-for compensation world. Defendant argues that these entities are not 14 adequate benchmarks, especially because Strikeforce and Bellator did not achieve total 15 profitability. The Court finds, based upon the record and expert testimony, that these entities are 16 appropriate benchmarks as data suggests that they achieved event level profitability. The Court 17 also agrees with Plaintiffs that any alleged lack of annual profitability (as opposed to event 18 profitability) relates to economies of scale based upon fixed costs, 54 which are distinct from 19 variable fighter costs. This does not prevent these entities from being comparators in the but-for 20 world. 21 Plaintiffs have also presented two adequate and reasonable damages models derived from 22 their impact regression analysis. First, the Plaintiffs estimated a damages model which removed 23 the Strikeforce data from the impact regression model. This model is based on the statistical 24 relationship between Zuffa’s foreclosure share and its own Fighter Shares over time. This model 25 mirrors the impact regression model discussed in detail above, except that all data points for 26 Strikeforce prior to its acquisition by Zuffa are discarded from the regression model.55 This model 27 54 Katz & Rosen, supra at 283-84. 55 Dr. Singer removed the Strikeforce data in anticipation of an objection from Defendant as to the use of this 28 - 69 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 70 of 80 1 conservatively used the Ranked measure of the Relevant Input Market to calculate damages. This 2 model estimates aggregate damages for the Class Period through June 30, 2017 as $894.3 million. 3 These estimates assume conservatively that Zuffa’s foreclosure share would fall to 30% in the but- 4 for world (a level sufficient to cause anticompetitive effects), instead of falling to zero. The Court 5 finds that this model provides a reasonable estimate of the damages to the class based upon the 6 alleged anticompetitive illegal conduct of Defendant. As the Court previously found, Plaintiffs’ 7 impact regression model controlled for the impact of other alleged factors, including any so-called 8 “special sauce” or business acumen of the Defendant, and establishes a statistically significant 9 relationship between Defendant’s alleged illegal conduct and Plaintiffs’ damages. 10 Second, Dr. Singer estimated Plaintiffs’ damages using the same impact regression analysis 11 with the addition of pre-acquisition Strikeforce data. Dr. Singer finds that this model estimates 12 class-wide damages at around $1.6 billion. For the reasons previously stated as to the regression 13 analysis, the Court finds that this damages model also adequately establishes Plaintiffs’ damages 14 that can be derived directly and solely from Defendant’s alleged antitrust violative conduct. Thus, 15 Plaintiffs’ impact regression models estimate damages between $811 million and $1.6 billion. 16 Defendant argues that Dr. Singer’s different damages model do not sufficiently isolate the 17 monetary effect of the alleged antitrust conduct from procompetitive conduct. The Court disagrees. 18 Defendant’s critiques of Dr. Singer’s analysis stem from their critiques of his regression models. 19 The Court has already found that Dr. Singer’s impact regression models adequately control for 20 various alleged neutral or procompetitive factors that might render his regression analysis 21 scientifically unreliable. For the reasons discussed at length above, the Court rejects the arguments 22 regarding Dr. Singer’s analysis. The Court reiterates its finding that Plaintiffs have established by 23 a preponderance of the evidence that they can identify, through the various regression models of 24 Dr. Singer, and the accompanying qualitative data in the record, damages directly derived and 25 arising from the alleged antitrust conduct alone. 26 The Court also rejects Defendant’s suggestion that the calculation of damages for the class 27 would be overwhelmed by individual factors. The fact that damages calculations for the class may 28 data in the damages model. The Court does not find that the removal of such data was necessary as the inclusion of additional relevant data, as noted, allows for more robust effect of control variables. - 70 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 71 of 80 1 involve some individual-level variation does not alter this Court’s finding in this case. See Olean 2 31 F.4th at 668 (explaining that a district court cannot “decline to certify a class” simply because 3 it “will require determination of some individualized questions at trial, so long as such questions 4 do not predominate over the common questions”); see also Fed. R. Civ. P. 23(b)(3). The Court is 5 satisfied that a common methodology based upon the regression models and the qualitative 6 evidence of a tiered and systemic compensation structure can resolve such questions. 7 Defendant’s critiques do not undermine Plaintiff’s argument that there is a coherent 8 methodology for establishing class-wide damages that predominates over any individual damages 9 analysis. Plaintiffs’ damages models align with their theory of liability, and monetary damages can 10 be calculated by common evidence in this case. Just Film, Inc., 847 F.3d at 1121. Plaintiffs’ 11 models are capable of proving impact and damages on a class-wide basis. 12 The Court finds that an antitrust violation, antitrust impact, and damages are common 13 questions capable of class-wide proof. The Court therefore concludes that Plaintiffs have 14 sufficiently established predominance. 15 16 17 18 19 20 21 3. Rule 23(b)(3) “Superiority” Requirements The final requirement for class certification is “that a class action [be] superior to other available methods for fairly and efficiently adjudicating the controversy.” Fed R. Civ. P. 23(b)(3). “In determining superiority, courts must consider the four factors of Rule 23(b)(3).” Zinser v. Accufix Research Inst., Inc., 253 F.3d 1180, 1190 (9th Cir. 2001). The Rule 23(b)(3) factors are: “(A) the class members’ interests in individually controlling the prosecution or defense of separate actions; (B) the extent and nature of any litigation concerning the controversy already begun by or 22 23 24 against class members; (C) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; and (D) the likely difficulties in managing a class action.” Fed. R. Civ. P. 23(b)(3). 25 26 The superiority inquiry focuses “on the efficiency and economy elements of the class action so that cases allowed under [Rule 23(b)(3)] are those that can be adjudicated most profitably on a 27 28 representative basis.” Zinser, 253 F.3d at 1190, as amended on denial of reh’g, 273 F.3d 1266 (9th Cir. 2001) (internal quotation marks omitted). A district court has “broad discretion” in - 71 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 72 of 80 1 determining whether class treatment is superior. Kamm v. Cal. City Dev. Co., 509 F.2d 205, 210 2 (9th Cir. 1975). 3 Plaintiffs maintain that class treatment is superior in this antitrust case. This is because 4 class treatment is superior where common issues, such as the violation of the antitrust law and 5 impact, predominate, Further, class treatment would be more manageable and efficient and 6 affordable for class members. The Court agrees. First, the Court finds that (1) the length of the 7 prosecution of this case in terms of years, (2) the tremendous resources that have been poured into 8 this litigation, (3) the complexity of the data collection and analysis regarding bouts, (4) the 9 extensive document review, the extended discovery period and discovery dispute litigation, and 10 (5) the need for multiple evidentiary hearings as to the legal issues in this case all underscore why 11 a class action is a superior vehicle for addressing the common issue facing class members. See 12 Fed. R. Civ. P. 23(b)(3)(A)-(D). Additionally, it would be simply too unwieldy and cost- 13 prohibitive for the alleged antitrust conduct, injury, and damages to be litigated on an individual 14 or small group basis. The sheer size of this case’s docket illustrates this point. Finally, while the 15 pandemic interrupted the litigation of this case (and almost all cases), this case demonstrates that 16 this type of litigation can be effectively managed and supervised through a class action. 17 The Court thus finds that Plaintiffs have demonstrated that a class action would be superior 18 to other available methods for fairly and efficiently adjudicating this case. Therefore, it concludes 19 that Plaintiffs have satisfactorily established superiority. 20 21 22 23 4. Injunctive Relief for Bout Class Plaintiffs seek injunctive relief on behalf of the Bout Class for Defendant’s violation of the antitrust law. In turn, the Court addresses whether Plaintiffs have standing to seek injunctive relief and whether, at this time, certification for purposes of injunctive relief is appropriate for the Class. 24 25 26 While Plaintiffs’ motion to certify class does not directly address certifying the class for the purposes of issuing injunctive relief, Plaintiffs pled that such certification is proper and previously requested that the Court so certify the class. Indeed, courts in this Circuit have certified 27 28 so-called “hybrid” class actions. See Buchanan v. Tata Consultancy Services, Ltd., 2017 WL 6611653, at *23 (N.D. Cal. 2017) (“District courts may certify both a 23(b)(2) class for the portion - 72 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 73 of 80 1 of the case concerning injunctive and declaratory relief and a 23(b)(3) class for the portion of the 2 case requesting monetary damages.”); West v. California Services Bureau, Inc., 323 F.R.D. 295, 3 307 (N.D. Cal. 2017); McMillion v. Rash Curtis & Associates, 2017 WL 3895764, at *10 (N.D. 4 Cal. Sept. 6, 2017); Raffin v. Medicredit, Inc., 2017 WL 131745, *10 (C.D. Cal. Jan. 3, 2017); cf. 5 Ubaldi v. SLM Corporation, 2014 WL 1266783, at *9 (N.D. Cal. Mar. 24, 2014) (noting the 6 possibility of a (b)(2) class for declaratory relief and a (b)(3) class for damages, but declining to 7 consider the hybrid option since the plaintiffs in that case only addressed the (b)(3) possibility). 8 Defendant argues that Plaintiffs do not have standing to bring a Rule 23(b)(2) claim for 9 injunctive relief because they are former UFC fighters. It argues that only current employees have 10 standing to seek injunctive relief. Plaintiffs reply that two of the named Plaintiffs, Mr. Fitch and 11 Mr. Vera, were current MMA fighters at the time of filing the consolidated complaint and the 12 motion for class certification. One additional named Plaintiff, Mr. Le, was under contract with the 13 UFC at the time the suit was filed. Plaintiffs argue that because these named Plaintiffs have 14 standing, certifying the class for injunctive relief is proper. 15 In order to demonstrate Article III standing, a plaintiff must show: (1) a concrete injury; 16 (2) fairly traceable to the challenged action of the defendant; (3) that is likely to be redressed by a 17 favorable decision. Lujan v. Defenders of Wildlife, 504 U.S. 555, 560-61 (1992). “In a class action, 18 standing is satisfied if at least one named plaintiff meets the requirements.” Bates v. United Parcel 19 Serv., Inc., 511 F.3d 974, 985 (9th Cir. 2007); see also Haro v. Sebelius, 747 F.3d 1099, 1108 (9th 20 Cir. 2014). “[A] plaintiff must demonstrate standing for each claim” and “for each form of relief 21 sought.” Daimler Chrysler Corp. v. Cuno, 547 U.S. 332, 352 (2006). “The standing formulation 22 for a plaintiff seeking prospective injunctive relief generally requires that the plaintiff’s concrete 23 injury be coupled with a sufficient likelihood that he will again be wronged in a similar way.” 24 Bates, 511 F.3d at 985. Further a plaintiff will have standing when the requirements of Article III 25 standing were satisfied at the time the complaint was filed. See Haro, 747 F.3d at 1108, see also 26 County of Riverside v. McLaughlin, 500 U.S. 44, 51 (1991). 27 The Court finds that Plaintiffs have standing to bring their claims of injunctive relief. As 28 provided above, Plaintiffs seek injunctive relief on behalf of the Bout Class due to Defendant’s - 73 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 74 of 80 1 violation of the antitrust law. At least three named Plaintiffs were subject to that action when the 2 Complaint was filed. Therefore, at that time, they met the standing requirements of a concrete 3 injury that was fairly traceable to the challenged action of the Defendant and that could be 4 redressed by a favorable decision. 5 The Court will next address whether certification for injunctive relief is appropriate for this 6 putative class. Federal Rule of Civil Procedure 23(b)(2) provides that “[a] class action may be 7 maintained if Rule 23(a) is satisfied.” As it found previously, the Bout Class satisfies the 8 requirements of Rule 23(a). 9 Furthermore, Rule 23(b)(2) also requires that “the party opposing the class has acted or 10 refused to act on grounds that apply generally to the class, so that final injunctive relief or 11 corresponding declaratory relief is appropriate respecting the class as a whole.” In their opposition 12 to class certification, Defendant rightly points out that “Plaintiffs’ Motion does not argue for class 13 certification pursuant to Rule 23(b)(2), but…[have] pled that certification is proper and have 14 previously requested that the Court so certify the classes.” Indeed, in their reply, Plaintiffs’ address 15 only the issue of standing with regard to injunctive relief. It is improper for the Court to guess the 16 nature of the 23(b)(2) elements or requested relief where Plaintiffs have not briefed it. See Ubaldi 17 v. SLM Corp., 2014 WL 1266783, at *9 (N.D. Cal. 2014) (declining to address injunctive relief 18 where “Plaintiffs alleged class actions under all three subsections of Rule 23(b) in the [complaint], 19 their motion for class certification only addresses Rule 23(b)(3)”). Because Plaintiffs do not 20 specifically address the 23(b)(2) requirements or the nature of their requested injunctive relief in 21 their motion for class certification, the Court declines to certify the class under Rule 23(b)(2), at 22 this stage. Instead, the Court will hold a status conference after the issuance of this Order and will 23 potentially allow parties to supplement the motion, for the Court to have a fulsome discussion 24 about whether certification of the class for the purposes of injunctive relief is appropriate here. 25 5. Bout Class: Conclusion 26 In sum, the Court holds that Plaintiffs have met the requirements of Rule 23(a) and Rule 27 23(b)(3), showing that, for Bout Class members, a class action is superior to individual proceedings 28 for the most prevalent questions of antitrust violations, impact, and the fact of damages. The Court, - 74 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 75 of 80 1 however, declines to reach the issue of whether certification of the Bout Class is appropriate under 2 Rule 23(b)(2). Rather, it proceeds on that issue as described above. 3 4 5 6 7 8 9 10 11 The Court therefore grants Plaintiffs’ motion for class certification as to the Bout Class pursuant to Rule 23(b)(3). C. The Identity Class The Court now addresses Plaintiffs’ request to certify a class identified as the “Identity Class” pursuant to Fed. R. Civ. P. 23(a), 23(b)(2), and 23(b)(3). The class is defined by Plaintiffs as containing: Identity Class: Each and every UFC Fighter whose Identity was expropriated or exploited by the UFC, including in UFC Licensed Merchandise and/or UFC Promotional Materials in the United States from December 16, 2010 to June 30, 2017. 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 For the reasons provided below, the Court holds that Plaintiffs fail to meet the requirements of Rule 23(b)(3) as to the Identity class. The Court therefore declines to certify the Identity class. As stated earlier, a class may be certified under Rule 23(b)(3) only if the Court finds that questions common to the class “predominate” over individualized questions, and that litigation by means of a class action is “superior” to the individual pursuit of claims. Fed. R. Civ. Proc. 23(b)(3). For plaintiffs to carry their burden of proving that a common question predominates, they must show that the common question relates to a central issue in the plaintiffs’ claim. See WalMart, 564 U.S. at 349-50. Therefore, “[c]onsidering whether ‘questions of law or fact common to class members predominate’ begins, of course, with the elements of the underlying cause of action.” Erica P. John Fund, Inc. v. Halliburton Co., 563 U.S. 804, 809 (2011) (quoting Fed. R. Civ. P. 23(b)(3)). Here, Plaintiffs allege that two common questions of law and fact predominate over individualized issues: first, whether Defendant expropriated or exploited the identities of members of the Identity Class in UFC Licensed Merchandise or Promotional Materials during the Class Period, and second, whether Defendant’s actions caused common injury to the Identity Class Plaintiffs and members of the Identity Class in the form of suppressed compensation. The Court finds that, because of the wide variety of identity rights at issue, Plaintiffs have failed to present a - 75 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 76 of 80 1 methodology that can demonstrate common impact through common evidence. 2 Plaintiffs break the Identity Class into two subgroups. The first subgroup consists of 3 fighters who, according to Zuffa’s records, received at least some sponsorship payments, video 4 game payments, merchandise royalty payments, or athlete outfitting policy payments from Zuffa 5 during the Class Period (“Identity Subgroup One”). To determine the effect of the antitrust conduct 6 on this group, Dr. Singer used data from an internal Zuffa database that reflected various line-item 7 identity-based payments Zuffa made to fighters. Dr. Singer’s research revealed that 826 fighters 8 were in this subgroup. Dr. Singer argues that the effect of the anticompetitive conduct can be 9 quantified by applying the results of the Bout Class’s regression analysis because Zuffa’s 10 foreclosure impaired other MMA promoters and their ability to provide competitive compensation 11 to fighters generally. He argues that “regardless of the exact level of foreclosure that would have 12 prevailed in the but-for world, the implied suppression in compensation for Identity Subgroup One 13 is substantial.” 14 The second identity subgroup consists of all fighters who executed a PAR Agreement 15 during the Class Period, regardless of whether these fighters received compensation for the use of 16 their identity in the actual world or in the databases to which Dr. Singer had access (“Identity 17 Subgroup Two”). Dr. Singer argues that in the but-for world, all fighters who executed PAR 18 Agreements would have received at least nominal compensation in exchange for signing away 19 these rights in perpetuity. 20 The Court finds that Plaintiffs have failed to prove common impact or injury for the Identity 21 Class that can be traced to the alleged anticompetitive conduct. Plaintiffs’ theory of liability as to 22 this Class relies on a few assumptions that the Court finds to be too speculative to serve as a basis 23 for certification at this point. 24 First, for example, Plaintiffs do not indicate the specific mechanism by which foreclosure 25 share impacts the various forms of identity right payments that fighters receive. Contrary to the 26 extensive statistical data and qualitative data which directly supported a connection between 27 foreclosure share and fighter compensation/revenue share for the Bout Class, there is a paucity of 28 data or expert analysis that supports this connection for the Identity Class. In the context of the - 76 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 77 of 80 1 Bout Class, expert and industry evidence pointed to a connection between the wage or revenue 2 share and the but-for compensation that fighters would receive. Plaintiffs have provided no such 3 evidence for this connection regarding the Identity Class. 4 Second, the underlying nature of the contractual rights and their manner of imposition 5 differ. More specifically, the contractual rights around identity are different than the exclusionary 6 contract rights that apply to the Bout Class. For instance, the merchandising rights are voluntary 7 and non-exclusive. Additionally, fighters in the Identity Class significantly vary in notoriety. The 8 difference in the subjective interpretations of Zuffa personnel as to the “identity” value of fighters 9 and the objective basis for such a valuation cannot be so robustly captured with controlling 10 variables as was done with the Bout Class, since there is no parallel evidence of identity 11 compensation moving in uniformly across the class. Plaintiffs’ experts simply did not have the 12 breadth of data as to potentially controlling variables to “identity value” or MRP like they did with 13 respect to the Bout Class. 14 Third, while Plaintiffs provide evidence sufficient to find that there was an internal pay 15 structure for participation in bouts, Plaintiffs have not provided adequate evidence to support a 16 finding that there was a consistent internal structure with respect to royalties or other payments 17 based on a fighter’s identity. The lack of evidence supporting such an internal pay structure for 18 identity payments or compensation, and the dearth of evidence as to the linkage between identity 19 compensation and bout-level data, prevents this Court from simply importing or incorporating its 20 Bout Class analysis to the Identity Class. This means that Plaintiffs cannot establish a means of 21 tracing any alleged anticompetitive conduct in the form of restrictive contracts to any purported 22 suppressed identity compensation. Therefore, the Court does not find that Plaintiffs can rely upon 23 the detailed and robust analysis for the Bout Class to establish predominance for the Identity Class. 24 The Court rejects Plaintiffs’ argument that common injury can be demonstrated through a 25 common pay structure for the Identity Class. For example, as to Identity Subgroup Two, Dr. Singer 26 argues that “in the but-for world, all fighters who executed PAR Agreements would have received at 27 least some nominal compensation in exchange for signing away these rights in perpetuity…. To 28 estimate the amount of this nominal payment, I relied on a Zuffa document that includes a schedule of - 77 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 78 of 80 1 nominal, one-time, formulaic recommended payments in exchange for the use of their Identity in the 2 UFC Electronic Arts video game series… In the but-for world, all fighters in Identity Subgroup Two 3 would have received at least the minimum (lowest-tier) payment, which is $2,500.” Importantly, 4 however, as noted above, there is a lack of evidence indicating that these “recommended” 5 payments were consistently applied across the class. There is no regression analysis as to this 6 specific point and not qualitative evidence identifying an internal identity compensation structure. 7 The Court, thus, agrees with Defendant that Dr. Singer’s assumptions here are not supported by 8 evidence in the record. 9 In conclusion, because of the multiplicity of potential revenue streams encompassed in the 10 Identity Class, the lack of evidence as to what funding streams applied to which fighters, the lack 11 of evidence creating a measurable connection between the alleged antitrust conduct of Zuffa and 12 the suppressed identity compensation, as well as the lack of evidence entitling fighters to any of 13 these revenue streams, the Court finds that common questions of fact will not predominate as to 14 the Identity Class. Plaintiffs have additionally failed to identify a convincing methodology for 15 quantifying this class harm. The dearth of analysis and evidence as to the Identity Class stands in 16 strong contrast to the ample analysis and evidence of impact and damages as to the Bout Class. As 17 a result, the Court declines to certify the Identity Class pursuant to Federal Rule of Civil Procedure 18 23(b)(3). The Court simply does not have enough evidence to understand the nature and degree of 19 the putative class’s injury. 20 Further, to the extent that Plaintiffs are requesting class certification pursuant to Federal 21 Rule of Civil Procedure 23(b)(2) to obtain injunctive relief for the Identity Class, the Court declines 22 to do so. The Court’s Rule 23(b)(3) analysis leads it to conclude that this Class would not even 23 meet the Federal Rule of Civil Procedure 23(a) prerequisite of commonality. 24 The Court therefore denies Plaintiffs’ motion for class certification as to the Identity Class. 25 26 D. Appointment of Class Counsel and Class Representatives 27 28 The Court previously granted Plaintiffs’ unopposed motion to appoint interim class - 78 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 79 of 80 1 counsel. In their instant motion for class certification, Plaintiffs’ interim class counsel seek 2 appointment as class counsel under Rule 23(g) of the Federal Rules of Civil Procedure. The Court 3 now considers this request. Interim class counsel, through the course of this litigation, have 4 demonstrated that they have extensive experience and knowledge with antitrust, class action 5 litigation. They have done extensive work in identifying and investigating the claims in this action 6 and have already expended considerable resources to represent the class. The Court also finds that 7 interim class counsel will adequately and fairly represent the interest of the class. Accordingly, the 8 Court appoints the law firms Berger & Montague, P.C., Cohen Milstein Sellers & Toll PLLC, 9 Joseph Saveri Law Firm, Inc., and Wolf, Rifkin, Shapiro, Schulman & Rabkin, LLP as Co-Lead 10 Class Counsel to represent the Bout Class. 11 12 V. 13 IT IS ORDERED that Plaintiff’s Motion to Certify Class (ECF No. 518) is GRANTED 14 15 in part and DENIED in part. The Court hereby certifies the following class of persons: All persons who competed in one or more live professional UFC-promoted MMA bouts 16 17 18 19 20 21 22 taking place or broadcast in the United States from December 16, 2010 to June 30, 2017. The Bout Class excludes all persons who are not residents or citizens of the United States unless the UFC paid such persons for competing in a bout fought in the United States. IT IS FURTHER ORDERED that the Court will set a status conference for the parties to address Plaintiffs’ claims for injunctive relief pursuant to Federal Rule of Civil Procedure 23(b)(3) and invite additional briefing if required. IT IS FURTHER ORDERED that the Court appoints named Plaintiffs as Class 23 24 CONCLUSION Representatives and appoints Interim Class Counsel as Class Counsel. 25 26 27 28 / / / - 79 - Case 2:15-cv-01045-RFB-BNW Document 839 Filed 08/09/23 Page 80 of 80 1 2 IT IS FURTHER ORDERED that a status conference regarding this Order is set for August 21, 2023, at 3:30 p.m. in courtroom 7C. 3 4 DATED: August 9, 2023. 5 6 7 _________________________ RICHARD F. BOULWARE, II UNITED STATES DISTRICT JUDGE 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 - 80 -

Some case metadata and case summaries were written with the help of AI, which can produce inaccuracies. You should read the full case before relying on it for legal research purposes.

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.