Winning EWC and Still Looking for a New Owner: Where Esports Money Flows in 2026

In October 2026, I sat in stand B of My Dinh Stadium, my left hand gripping a worn mechanical stopwatch, my right hand holding a sheet of graph paper lined with ballpoint ink. It was the men's 4x400m relay final at the national youth athletics championships. Hanoi finished second. The gap to the winners was 0.8 seconds. I recorded every baton exchange across all four legs, then replayed the footage three times in one night. On the third leg, Hanoi's receiver started 2.1 metres earlier than the standard mark. He had to decelerate half a step to match hands, and that half step travelled along a straight trajectory for the remaining 200 metres. Nobody fell. Nobody dropped the baton. The trajectory simply broke at a point the stands could not see. 0.8 seconds is never just 0.8 seconds; it is where the trajectory breaks. Nine years later, in August 2026, I sat in front of three monitors watching a different kind of baton exchange. This time the runners wore no numbers and the track had no lanes. But the mechanism was identical: a world champion team, a handover half a beat late, and a trajectory broken somewhere the scoreboard does not display. CONTEXT: FROM FORTY MILLION DOLLARS DOWN TO A FEW MILLION Before discussing any team, I need to rebuild the field they stand on. Every figure below comes from a table I compiled myself from numbers published year by year, with notes marking which entries are hard data and which are qualitative descriptions I could not verify against a second source. The International prize pool, Dota 2's world championship: roughly 40 million USD in 2026. It fell to 18.9 million USD in 2026. In 2026 it dropped to about 3.4 million USD. The most recent level mentioned is "a few million USD". From the 2026 peak to now, the decline is approximately 91 percent. The 2026, 2026 and 2026 marks match the tournament's public record, so I classify them as cross-checkable data. The "few million USD" phrase for the current period is a qualitative description; I flag it as pending verification and will not use it for arithmetic. What built the 40 million USD figure in 2026 was not publisher money. It was player money. Valve tied The International prize pool to Battle Pass revenue: players bought in-game items, and a share of that revenue flowed directly into the tournament prize pool. This mechanism turned spectators into shareholders without equity. The more excited the community, the larger the number on the board, and that number in turn became the measuring stick for the tournament's own prestige. Then Valve changed the Battle Pass model. The link between in-game purchasing behaviour and the prize pool was cut. From that season onward, The International prize pool shifted from a community-pumped growth metric into a reward determined by the publisher. I begin with a self-counted data table, because memory does not know how to make room for error. At the opposite pole of the map, money flowed the other way. The Esports World Cup 2026 announced a total prize pool of 75 million USD spread across dozens of titles. The Saudi eLeague 2026 gathered 37 clubs with total prizes above 4 million SAR. This is a third-party funding model, tied to multi-title infrastructure and national ambition, not dependent on item purchases in a single game. At the same time, in South Korea, the LCK imposed a salary cap with a luxury tax. This is an intervention at league level, aimed at two goals at once: competitive balance and long-term viability. In other words, the LCK organisers conceded that the player salary market had outgrown the market's own ability to control it. Those three data points, the TI prize pool collapse, the EWC prize pool expansion and the LCK salary cap, do not sit apart. They are three faces of the same process. ANALYSIS: THE MONEY DID NOT DISAPPEAR, IT CHANGED PLACES I built a simple comparison table to answer one question for myself: is there more or less esports money in 2026 than in 2026? The answer depends on where you count. If you count at the level of single-title tournaments run by publishers, the number is shrinking hard. The International prize pool lost about 91 percent from its 2026 peak. That is a contraction any ecosystem would feel, because it drags the entire chain of dependency with it: teams living on prize money, coaches living on team contracts, and a whole layer of analysts and media living on the attention the tournament generates. If you count at the level of multi-title events funded by third parties, the number is rising. The Esports World Cup 2026 with 75 million USD is a year-round money distribution machine, not a single-season one. The Saudi eLeague 2026 with 37 clubs is a domestic league system funded with intent. But this is where I want to pause a little longer. These two money flows are not the same in nature. The International prize pool at its peak was community money, generated from player emotion and cyclical by season. Multi-title event money tied to state capital and long-term strategy is investment money, deliberate in both economic and political terms. One is rainwater, the other is tap water. Both wet the ground, but only one depends on whether it rains. And here is the consequence I consider the most important in the whole story: when the prize pool is decided by the community, fans hold indirect influence over the scale of the tournament. When the prize pool is decided by the publisher or a sponsor, that voice disappears. Audience interest no longer automatically converts into prize money for players. Two quantities assumed to be locked together for a decade were separated by a single product decision. In South Korea, the reaction at league level ran against that separation. The LCK set a salary cap and a luxury tax, accepting administrative intervention to keep costs from outpacing revenue. The luxury tax mechanism is not purely a savings tool. It is a redistribution channel: the biggest spenders pay extra, and that money cycles back into the system to raise the overall floor. In traditional sport this is a model with precedent in US basketball and football leagues, where salary caps and luxury taxes exist as a balancing pair. As an observer, I note that the LCK is ahead in self-correction. That is a positive structural signal, not a sign of weakness. A THREE BILLION WON BATON EXCHANGE Against that backdrop, two organisational events appeared, and they stand in sharp contrast. Dplus KIA won the League of Legends title at the Esports World Cup 2026. The organisation's predecessor, DAMWON Gaming, won Worlds in 2026. In other words, this is an organisation with a tradition of winning at the highest level, capped most recently by a world championship as a club in 2026. Alongside that title, Dplus KIA delayed salary payments to players and sought a new owner. Their League of Legends roster is cited at a cost of roughly 3 billion won, approximately 2 million USD. I want to be blunt about this comparison, because it breaks an assumption the entire industry operates on. That assumption is: win and you will be saved. Winning a major event means prize money, sponsorship, and renegotiating contracts from a position of strength. Dplus KIA shows that chain breaking in the middle. A championship does not automatically convert into enough cash flow to pay a 3 billion won roster. Looking closer at that 3 billion won figure: it covers a single roster, excluding administration, facilities, and any other titles the organisation maintains. When a fixed cost at that level meets a revenue stream that does not flex with performance, the outcome does not depend on whether you win or lose. It depends on how many long-term sponsorship deals you hold, how much jersey revenue you generate, how much league distribution you receive. And across those items, Dota 2 and League of Legends sit in very different states. Every match is a countable wager. You just have to be willing to observe. A WITHDRAWAL THAT WAS CALCULATED On the other side sits Falcons. The organisation won The International 2026, and in 2026 it entered 18 tournaments at the Esports World Cup. At its competitive peak, it withdrew from Dota 2. Read only the headline and it looks like a distress signal. I do not read it that way, and I have a reason from my old job. In 2026, during the World Cup in Russia, a sports outlet invited me to write a guest piece. I picked the Russia versus Spain round-of-16 match and did something nobody asked for: I counted every Russian corner. The match had 12 corners in total. Among them, one near-post header routine repeated exactly seven times. Two of those seven produced genuinely dangerous chances. In extra time, Spain's defence broke at exactly that point. When a team repeats the same routine seven times, they are not hoping for luck, they are carving tactics into muscle. Apply that spirit to Falcons. An organisation entering 18 tournaments in one season is running at maximum intensity, with a dense calendar and high travel, staffing and recovery costs. Their withdrawal from one title, immediately after winning it, reads under the logic of repetition as a portfolio decision, not an emotional one. They did not leave because they lost. They left because they calculated. The lesson from my athletics work sits here. In 2026, when domestic competition paused, I built a database tracking the results of 40 Vietnamese track and field athletes, logging injury recovery times and competition frequency for each. A sports medicine doctoral researcher helped me with the physiology, and I built an index I called record reproducibility. In early 2026, that index predicted Nguyen Thi Oanh would break the national 3000m steeplechase record. It happened, with a time of 10:05.23. What I learned was not that I predicted correctly. What I learned is that decisions at the peak of performance are almost always made on data about the ability to sustain intensity, not on the euphoria of a single win. Falcons left Dota 2 in exactly that state. THE CONTRARIAN ANGLE: FOUR BLIND SPOTS Here I want to step away from the crowd, because four common readings strike me as technically wrong. First, reading The International prize pool falling to "a few million USD" as proof that Dota 2 is dying. That is an arithmetic error, not a difference of opinion. The prize pool collapsed because the mechanism tying it to item revenue was cut. Player interest could hold steady, or even rise, while the prize pool falls. I have no concurrent player data for this period, so I draw no conclusion in either direction. What I can conclude is this: using prize pool as a vitality indicator for a game, in a period when the publisher just removed its automatic mechanism, is structurally invalid reasoning. Second, reading Dplus KIA as a story of weak governance. Salary delays are a contract-performance issue, not a competitive-integrity issue. There is no match-fixing, cheating or sporting violation alleged anywhere in this story. If a world champion organisation still cannot pay wages, the fault does not lie in a small operational detail but in the cost structure relative to the title's revenue ceiling. Third, reading the Saudi eLeague and Esports World Cup as the industry's salvation. Third-party capital is badly needed, but it comes with conditions. When funding concentrates into a few giant event machines, mid-tier teams will depend increasingly on guaranteed participation payments rather than on competitive performance. That means part of a team's income no longer comes from playing well, but from showing up. This is a structural shift, and it needs to be counted in numbers, not felt as reassurance. Fourth, the biggest blind spot in how this story is told: China, Europe and North America are almost entirely absent. A picture labelled global that omits the three largest regions by capital and player base is only a picture of two poles. I have no data on those regions in this cycle, so I draw nothing extra. But I note it: what is not counted will not be explained. On governance, there is one point I consider the most underrated risk. Valve is both the rule-maker and the commercial beneficiary of the ecosystem it governs. A single product decision of theirs redefined the entire revenue equation of a title. No cross-publisher counterweight exists to absorb a shock of that kind. That is a structural gap, not the fault of any individual. ON THE TWO POLES OF THE MAP The regional picture resolves into two poles, correcting themselves in opposite ways. South Korea is maturing and self-adjusting. The LCK imposes a salary cap, adds a luxury tax, and prioritises competitive balance and long-term sustainability over open spending. This is a designed intervention with a target and a redistribution mechanism. Saudi Arabia is expanding and injecting capital. The Esports World Cup 2026 with 75 million USD, the Saudi eLeague 2026 with 37 clubs, a multi-title model. This is the capital injection phase, not yet the cost control phase. If both trends continue, talent flow will follow money flow. South Korea develops talent; the Gulf buys talent. The two models can coexist for a while, but they produce two very different kinds of organisation: one a player developer, the other a portfolio investor. I watch my events with one fixed habit: for each team, I count how many times a specific routine repeats across matches. I do not trust the best play of a match. I trust the most repeated play. The same holds at the economic layer. A team can win on one miraculous play. An organisation only survives on recurring monthly revenue. THIS IS NOT A WINTER A 91 percent drop in The International prize pool, a champion team still seeking a new owner, another world champion withdrawing from a title. Place those three facts side by side and they suggest an easy story: esports is dying. But if I count at both ends of the map, the story changes. The money did not vanish. It concentrated. Into a small number of large-scale events. Into titles with durable commercial revenue. Into multi-title organisations with controlled cost structures. And out of single-title organisations living on prize money, with payrolls exceeding their revenue-generating capacity. This is a distribution problem, not a volume problem. When I sat in front of the screen for the men's 1500m final at the Tokyo Olympics in 2026, I recorded the last 200 metres of the Norwegian champion: 24.7 seconds, 1.2 seconds faster than the runner-up. On television, that 1.2 seconds looks like it was created in the kick. It was not created there. It was collected across thousands of recovery sessions nobody broadcast, from a starting position on the inside lane, from a change-of-rhythm technique honed in empty sessions. A national record is not born in the final second, it is gathered across thousands of recovery sessions. Esports organisations are the same. Their survival is not decided on the big match day. It is decided in the cost structure table, in long-term sponsorship contracts, in whether they hold a revenue indicator independent of competitive results. Titles are a consequence, not a cause. I built a forecast frame for the rest of 2026, and I present it as probabilities with uncertainty bands, not assertions. For Falcons, the probability they return to Dota 2 within 12 months is low, which I place around 15 to 25 percent, since the portfolio logic has already been set and the cost of returning would exceed the cost of staying out. For Dplus KIA, the probability of completing a sale and cost restructuring in the first half of 2027 is moderately high, around 55 to 70 percent, since selling a winning asset under cash pressure is a familiar scenario in any market. For the LCK salary cap, the probability that at least one other major league adopts a similar mechanism within two seasons is low, around 20 to 35 percent, because the mechanism requires an organiser with enough power to force big teams to accept it. Those three numbers are not prophecy. They are uncertainty bands I use to remind myself that the model can be wrong. WHAT IS MOST WORTH WATCHING If I had to pick one indicator to track over the next six months, I would not pick any tournament's prize pool. I would pick a simple ratio: roster cost divided by each organisation's recurring monthly revenue. When that ratio is below one, a team can lose and still survive. When it is above one, a team can win a world title and still have to sell itself. Dplus KIA is the second case, and it is no longer an exception. In that 4x400m relay leg at My Dinh, Hanoi did not lose because they ran slower. They lost on a decision at the exchange point, executed 2.1 metres earlier than the standard mark. The entire remaining leg was just the consequence of that error. In-game item revenue was once a stable law for an entire esports scene. It has just been changed. The question I keep for next season is not which team will win. It is how many organisations are running with a cost-to-revenue ratio above one, and whether they will notice before entering the final leg.