Hub Power and Hub(uses): Power Dynamics in Platform Ecosystems

• Pages: 327-390
• Date: 01 January 2026
• Published date: 01 January 2026
• Author: Raz Agranat,Michal S. Gal
• Subject Matter: Administrative & Public Law

HUB POWER AND HUB(USES):

POWER DYNAMICS IN PLATFORM ECOSYSTEMS

R A

M S. G*

INTRODUCTION ......................................... 328

I. HUB POWER ........................................ 332

A. O N, L,  H ......................... 332

B. F M D  H P ............. 335

1. Hub Attractiveness .............................. 336

2. Platform Dependence on the Hub ................... 337

3. The Hub’s Switching Feasibility .................... 341

4. Countervailing Power Interplays ................... 343

C. S M  H-P

A ....................................... 344

1. Framework .................................... 344

2. Application .................................... 346

D. I  S M P I ... 348

E. T E  H P  P P ....... 349

II. THREE MANIFESTATIONS OF HUB POWER ............. 350

A. U C .............................. 351

1. Overt Exercises of Hub Power ..................... 351

2. Legal Limitations: The Case Against Carlsen

and Chess.com ................................. 354

3. Ripple Effects: A Stylized Example .................. 361

* Raz Agranat is Assistant Professor, Harry Radzyner Law School, Reichman University, and

a member of the International Association of Competition Law Scholars (ASCOLA). Michal S.

Gal (LL.B., LL.M., S.J.D) is Professor, University of Haifa Faculty of Law and Honorary Doctor,

University of Zurich, and the past president of ASCOLA (2016–2023). Our sincere thanks to

Baruch Barzel, Omri Ben-Shahar, Inge Graef, Giorgio Monti, John Newman, Stephen Ross,

D. Daniel Sokol, Volker Stocker, Robby Robertson, Nick Grimmer, Michal Shur-Ofry, partici-

pants of ASCOLA and CRESSE conferences and Tilburg, NYU, Cornell-Tech, Bar-Ilan, Haifa,

and Weizenbaum Institute workshops, and the two anonymous reviewers for superb comments,

as well as to Noy Benisti, Tomer Stern, and Almog Zackry for excellent research assistance. Any

mistakes or omissions remain our own.

327

B. C H P ........................... 365

1. United States v. Apple: Facts and Court Decisions ...... 365

2. United States v. Apple: Applying Network Science ...... 368

C. U C H A P ... 377

1. GDS Platforms: Industry Characteristics ............. 377

2. GDS Platforms: A Stylized Case .................... 379

3. GDS Platforms: Applying Network Science ........... 380

4. Antitrust Implications ............................ 384

CONCLUSION ........................................... 388

INTRODUCTION

Magnus Carlsen, the world’s premier chess player and arguably the greatest

of all time, commands a global audience of millions who scrutinize his every

move. In a professional tournament in 2022, Carlsen faced Hans Niemann, a

lesser-known and lower-ranked opponent. In a shocking upset, Carlsen suffered

a crushing defeat in the match.1 Although he lacked direct evidence to sup-

port his claim, Carlsen accused Niemann of cheating by using an AI-powered

chess engine. He then took the dramatic step of refusing to participate in tour-

naments featuring Niemann.2 Just a day into this controversy, Chess.com, the

world’s largest online chess platform, banned Niemann from accessing their

site.3 The platform cited indirect evidence of prior misconduct as justification

for their decision, although this evidence was notably unrelated to the tourna-

ment in question.4 This sequence of events sent shockwaves through the chess

community, raising questions about fair play, the power dynamics within the

1 With his stunning win, Niemann ended Carlsen’s 53-game unbeaten streak in classical

chess with white, which had lasted for more than two years. Niemann is one of only five players

in five years to defeat Carlsen while he was playing the white pieces, with three of Carlsen’s

losses occurring before 2018. See C., The H N R 18 (2022), www.

chess.com/blog/CHESScom/hans-niemann-report; L, Magnus Carlsen’s Classical

Losses, lichess.org/study/ptWMvZDH/GkOeAj9l (detailing Magnus Carlsen’s losses in clas-

sical chess in 2011 through 2013). Niemann was notably the lowest-rated player at the tourna-

ment and ranked approximately 40th among the top players at the time, well below the other

players who defeated Carlsen. See Greg Keener, The Chess World Isn’t Ready for a Cheating

Scandal, N.Y. T (Sep. 13, 2022), www.nytimes.com/2022/09/13/crosswords/hans-niemann-

magnus-carlsen-cheating-accusation.html; see also Top 100 Players November 2022 - Archive,

I’ C F’, web.archive.org/web/20240423053511/https://ratings.fide.com/toparc.

phtml?cod=713. Niemann later commented “[Carlsen] was just so demoralized because he’s

losing to such an idiot like me... . It must be embarrassing for the world champion to lose to

me.” GC, Hans Niemann Beats Magnus Carlsen, at 25:10 (YouTube, Sep. 4, 2022),

www.youtube.com/watch?v=GSLM1K6O6aU.

2 See Niemann v. Carlsen, No.4:22-cv-01110, 2023 WL 4198227, at *4–6 (E.D. Mo.

June27, 2023).

3 See id. at *4.

4 While Chess.com stated that it was never pressured by Carlsen to remove Niemann from the

platform or disinvite him from organized tournaments, it cited his “unprecedented withdrawal”

as a reason for its response. See C., supra note 1, at1–4.

328 A L J [Vol. 87

sport, and the role of digital platforms in arbitrating disputes. It also triggered

an antitrust lawsuit by Niemann against Carlsen, Chess.com, and others.5

Although the lawsuit ultimately failed,6 our analysis shows that the case

possessed greater substantive merit than the court’s limited understanding of

network dynamics suggested.

Chess.com’s swift alignment with Carlsen’s unsubstantiated accusations

offers an instructive case study of the asymmetric influence that certain

participants—such as Carlsen—can wield within platforms by virtue of their

high connectivity and centrality in the platform’s network topology.7 These

participants, known as hubs in network-science literature,8 possess orders of

magnitude more connections than the average participant. As a result, they

enjoy significantly greater influence over platform-governance decisions and

can markedly shape the potential trajectories of other market participants.

This phenomenon, which we term hub power, represents a distinct form of

influence that arises from structural positioning within networked environ-

ments, rather than from traditional market dominance or formal authority. It

is a nuanced form of power that warrants careful examination and carries pro-

found implications for our understanding of competitive dynamics in platform

economies.

Platform competition has captivated antitrust academics, decision-makers,

and industry leaders alike, spurring groundbreaking research9 and landmark

cases involving some of history’s most successful firms.10 Yet while platforms

have been thoroughly studied,11 the competitive dynamics within platforms

5 Niemann attempted to sue Carlsen and Chess.com, including on antitrust grounds, but these

claims were ultimately dismissed. See Niemann, 2023 WL 4198227, at *13. The dispute was

subsequently settled. See Andrew Beaton & Joshua Robinson, The Chess Cheating Scandal

Ends with an Uncomfortable Handshake, W S. J. (Aug. 28, 2023), www.wsj.com/sports/

chess-cheating-magnus-carlsen-hans-niemann-be3b9c5; see also infra Section II.A.2.

6 See Beaton & Robinson, supra note 5.

7 The term “topology” refers to the arrangement of nodes and links in the network, that is, the

network architecture.

8 See infra Section II.A.

9 See, e.g., Howard A. Shelanski & J. Gregory Sidak, Antitrust Divestiture in Network

Industries, 68 U. C. L. R. 1 (2001); Erik Hovenkamp, Platform Antitrust, 44 J. C. L.

713, 714–15 (2019); Herbert Hovenkamp, Antitrust and Platform Monopoly, 130 Y L.J.

1952 (2021); Daniel A. Crane, Defining Relevant Markets in Digital Ecosystems, 7 J.L. &

I 10 (2024).

10 See, e.g., United States v. Google LLC, No.1:20-cv-3010, 2024 WL 3647498 (D.D.C.

Aug. 5, 2024) (finding Google liable for monopolization under Section 2 of the Sherman Act);

Complaint, FTC v. Facebook, Inc., No.1:20cv03590 (D.D.C. Dec. 9, 2020); Complaint, United

States v. Apple Inc., No.2:24cv04055 (D.N.J. Mar. 21, 2024).

11 For an overview of the research, see Hsing Kenneth Cheng, D. Daniel Sokol & Xinyu Zang,

The Rise of Empirical Online Platform Research in the New Millennium, 33 J. E. & M.

S 416 (2024).

2026] H P  H() 329