In most creative markets a small number of stars capture the bulk of the earnings, a pattern usually explained as an outcome of the market. Elite esports present a different case. Here the concentration of prize money is not produced by the market but written into a payout table by those who own the game, and the effect of that design is visible in a striking divergence: over fifteen years, real prize money has risen steeply in team formats while remaining broadly static in individual ones.

Inside the Pacific Coliseum for the 2025 Mid-Season Invitational, before the first games of the event

 

In the autumn of 2021, five Dota 2 players shared a first prize of $18.2 million on a single tournament stage. Most were in their early twenties. The runners-up, beaten in the final series, received $5.2 million between them, and the remaining teams divided what was left of a $40 million pool in shares that fell away steeply down the standings. In the same industry, and in the same year, the leading players in the individual disciplines (the fighting games, the racing titles, the one-versus-one strategy games) were competing for a small fraction of that sum. And they had been, more or less, for a decade. That disparity is the subject of our recent paper, and it proves informative about how creative labour markets operate more generally.

The standard explanation for skewed earnings in a creative field is Rosen’s economics of superstars. Where a performance can be reproduced and distributed at low marginal cost, and where lesser talent is an imperfect substitute for greater, demand concentrates on a handful of top performers, and small differences in ability are magnified into large differences in reward. Thus, the concentration is a market outcome: it emerges from the aggregation of dispersed consumer demand rather than from any deliberate allocation of the rewards themselves.

At first sight esports appears to be a textbook case. Prize earnings are extraordinarily concentrated. In our panel of 846 elite players, drawn from 62 countries and 25 years of tournament records, the Gini coefficient for annual prize money exceeds 0.7, higher than the salary distribution of any major professional sports league that we were able to identify, and it rises further towards the top of the ranking. A small minority accounts for the overwhelming share of recorded earnings.

Yet the superstar account fits esports poorly, and the mismatch is instructive. Here the concentration of prize money is not the aggregate of dispersed demand, but a magnitude fixed in advance, by the game’s publisher, in the payout schedule of each tournament. That the winners of The International receive roughly 45 per cent of the pool, and the runners-up around 13, is not a figure the market arrives at but a design decision, taken by the organiser, Valve. The inequality is less discovered by the market than engineered by the organiser.

A more precise framework comes from the same author. Well before the superstar paper, Lazear and Rosen (1981) showed that where an employer cannot readily observe each worker’s output but can rank workers against one another, payment by rank, that is, a tournament, provides an efficient means of eliciting effort. Rosen (1986) extended the argument to multi-stage elimination tournaments and derived a clear prediction: sustaining effort through to the final round requires that the gap between first and second place be disproportionately large. A finalist has no further round in which to advance, so the reward for victory must compensate for the loss of that option. The resulting payout schedule is deliberately and steeply top-heavy.

This is precisely the shape of an esports mega-event, and it is imposed by design rather than arising from demand. Rosen observed that in professional tennis the top four finishers typically receive half or more of the purse. The flagship esports tournaments exhibit the same pattern in more pronounced form. The International concentrates its purse on the first two placings, and the Counter-Strike Majors do likewise: the 2023 Paris Major allocated 40 per cent of its prize pool to the winners and roughly 14 per cent to the runners-up, with sharply tapering shares thereafter. These are structures designed to concentrate reward, because concentration is what sustains effort across successive rounds.

This returns us to the gap between the team players and the soloists. If prize concentration in esports is a matter of design, we should expect it to shift when the designers alter their choices; and it does, dramatically, in a single direction.

When we track real prize money across the full 1999–2024 panel, the two categories of game diverge markedly. In the team titles, the five-player disciplines such as Dota 2, Counter-Strike and League of Legends, real prize money has grown substantially from one period to the next, even after the general expansion of the industry is netted out. The leading Dota 2 championship pool rose from roughly $1.6 million in 2011 to $40 million a decade later. In the individual disciplines over the same interval, prize money has remained broadly flat, and in some cases has declined slightly. The timing of a team player’s peak years is therefore strongly associated with earnings, whereas for a soloist the period of competition makes little difference. The industry and its audience grew for both, yet their fortunes diverged, and the distinguishing factor is simply whether prizes are shared across a roster.

Figure 1. Within-format period effects on log annual prize earnings, relative to the 1999–2005 baseline. Period dummies from format-specific fixed-effects regressions (Table 6); Mixed format omitted owing to sparse early cells.

Why would publishers direct such sums into the team formats while leaving the individual ones largely untouched? Because the team formats host the marquee events, and it is in those events that the tournament logic operates most forcefully. The escalation of prize pools was not an incidental consequence of demand but the product of deliberate mechanisms. Valve allowed the fans of Dota 2 to crowdfund the championship pool through purchases of in-game content, converting the purse into a marketing instrument that expanded year on year. Riot organised League of Legends around franchised leagues with permanent slots and shared revenue. These are the instruments through which a publisher uses prize money to construct a season-long spectacle around its flagship title, exactly as elimination-tournament theory would recommend when the aim is to keep the strongest teams competing intensely through to a final. The individual disciplines, contested across a dispersed set of smaller events with no single mega-event to anchor the calendar, received no comparable treatment.

For cultural economists the case is instructive, because it separates two things that are ordinarily conflated. When we observe a top-heavy income distribution among novelists, musicians or visual artists, we tend to attribute it to the market: to talent, to taste, and to the winner-take-most dynamics of attention. Placed on the same scale, esports yield familiar figures. Its prize inequality lies above that of the traditional sports leagues, broadly in line with estimates for tournament golf and for writing, and below the extreme concentration documented for live-streaming platforms. It would be tempting to classify esports as one more winner-take-most creative field and to leave the matter there.

Figure 2. Gini coefficients across selected labour markets, with esports prize money highlighted. The measures are not strictly comparable: some are salaries, some prize money and some total income, covering elite subsets or whole occupations. The comparison is therefore illustrative rather than exact. Sources: study Tables 7–8.

Esports demonstrate, however, that a distribution may resemble the superstar effect while being generated by something quite different. The publisher of a competitive game occupies a position with almost no parallel in the older arts: it creates the medium, sets the terms under which everyone competes without directly employing them, stages the premier events, and writes the prize schedule. It is as if the Medici had not only commissioned the paintings but also laid down the rules of what counted as good painting and owned the only galleries in which it might be exhibited. That concentration of roles means that inequality in esports is, to an unusual degree, a parameter rather than an outcome: a level that a company can raise or lower from one season to the next.

This is the point worth carrying beyond esports. Wherever a single platform comes to control the medium, the marketplace and the reward structure simultaneously, and this now describes a growing share of cultural production, the distribution of income ceases to be a fact about talent or taste and becomes a fact about design. The pertinent question is then no longer merely why the distribution is so unequal, but who determined the setting, and what they intended to purchase by placing it where they did.

References

Lazear, E.P. and Rosen, S. (1981) ‘Rank-order tournaments as optimum labor contracts’, Journal of Political Economy, 89(5), 841–864. https://doi.org/10.1086/261010

Rosen, S. (1981) ‘The economics of superstars’, American Economic Review, 71(5), 845–858.

Rosen, S. (1986) ‘Prizes and incentives in elimination tournaments’, American Economic Review, 76(4), 701–715.

About the article

This post draws on Newham, T. and Fry, T.R.L., ‘The business of play: Careers and earnings in elite esports’, Journal of Cultural Economics. doi: 10.1007/s10824-026-09603-2

About the authors 

Thomas Newham is a Senior Lecturer in Personalisation and Experiential Learning in Nottingham Business School (NBS).

Tim R. L. Fry is a Professor and Visiting Fellow at Nottingham Trent University.

About the image

Yuet Man Lee, CC BY-SA 4.0, via Wikimedia Commons, https://commons.wikimedia.org/wiki/File:2025_Mid-Season_Invitational_at_Pacific_Coliseum.jpg

November 3, 2026
What Makes A Great Composer?
October 20, 2026
Superstars By Design
October 6, 2026
Does Public Funding Make Theatre Repertoires More Original? Evidence From Italian Opera And Drama
September 21, 2026
Family First? How Children Reshape Parents’ Cultural Participation

Become a member

Members of the ACEI will be part of a network of scholars, researchers and practitioners interested in advancing cultural economics.

Join Today