Esports Is Not Dying: The Money Just Changed Owners

On 6 September 2026, I reopened the Excel file I built back in March 2026. Salaries on the left, sponsorship revenue on the right, and the gap in the middle, the column I marked red. Six years ago I drew it in pencil inside a subsidised apartment in Incheon. Tonight that same gap explains why Dplus KIA won the League of Legends title at the Esports World Cup 2026 and, in the same week, apologised to staff for delayed wages.

Nothing new happened. A number simply came due.

To make sense of this, step back. The International prize pool, Valve's Dota 2 world championship, peaked at roughly $40 million in 2026. It fell to $18.9 million in 2026 and about $3.4 million in 2026. Most recently it sits in the low millions. That is a decline of roughly 91 percent from peak.

Here is the part worth noting: this is not evidence that Dota 2 lost players. Valve removed the Battle Pass crowdfunding mechanism, under which players bought in-game items and a share of revenue flowed directly into the prize pool. When the pipe is unscrewed, the tank drains. Anyone who reads $3.4 million and concludes esports is dying is reading the arithmetic wrong.

Meanwhile, the Esports World Cup 2026 in Saudi Arabia announced $75 million spread across dozens of titles. Saudi eLeague 2026 gathered 37 clubs with more than 4 million riyals in prize money. In Korea, the LCK imposed a salary cap and a luxury tax for the first time in league history.

Three events, three different directions for money. The problem is not that money vanished. The problem is that money switched lanes.

Before the analysis, one honest note about how I work. I was wrong three times in 72 hours, and only the final correction is worth your time. In 2026 I published that Lee Myung-joo was moving to Japan without verifying the source. He extended with Incheon until 2026. Three straight corrections. My editor cut every piece of mine for a month. Since then every deal I write carries a timestamp, a primary source, and a contract clause.

So when I speak about Dota 2, I do not speak in feelings. I speak in structure.

In-game item purchase terms are a real contract. The Battle Pass is not an entertainment feature. It is a fundraising channel, in which players pay against a promise that a share of revenue returns to the tournament. When Valve changed the model, it unilaterally amended a clause no outsider had the right to negotiate.

That is the governance lesson of the year. A single product decision, with no press conference and no vote, erased a funding channel worth tens of millions. No protective mechanism exists between a publisher and its own competitive ecosystem.

For Dota 2 organisations the consequence follows a familiar sequence: prize pools shrink, player contracts cannot rise, and star rosters become absurdly expensive relative to revenue. A multi-million-dollar lineup without matching commercial value turns into a burden.

Dplus KIA is the clearest case of the paradox: champion, and still short of cash. Its League of Legends roster consumes about 3 billion won, close to $2 million a year in salary alone. It won the Esports World Cup 2026 LoL title. And it is searching for a new owner.

Based on my experience tracking matches and transfer windows, a champion team that still sells itself rarely loses on form. It loses because its cost structure was set during the growth phase, when everyone believed revenue would catch up to wages. That race has no finish line.

Whoever buys Dplus KIA inherits a winning roster attached to an unprofitable payroll. That is a distressed valuation, not a glamour deal.

Falcons withdrawing from Dota 2 is not a sign of weakness. They won The International 2026. They entered 18 tournaments at the Esports World Cup 2026. And they chose to leave one title.

Read the words carefully. In the Korean of the transfer world, withdrawal does not mean surrender. It means budget reallocation. Falcons kept many other titles. They did not leave esports; they left an economic model.

This is the early indicator most reporters skip. When a world-championship organisation cuts its portfolio instead of expanding, the signal is not about that game. It is about the entire strategy. Maximising title count is no longer rational.

The LCK salary cap is not a punishment; it is surgery. During the growth phase, player prices rose faster than revenue generation. When a market inflates without matching productivity, there are two options: let it break, or cap it before it breaks.

The LCK chose the second. A cap plus a luxury tax does not merely limit spending; it creates a sharing mechanism, where the biggest spender subsidises the rest. This is what I wrote during the football-less summer of 2026, when I built my own FFP spreadsheet to prove that people only cry when the spreadsheet has not been opened. Football's FFP never controlled the owners; the LCK cap, being league-level rather than federation-level, has a better chance of enforcement.

One side effect to price in: if other leagues do not cap, Korea risks losing stars to uncapped markets. That is a long-term equilibrium problem nobody has solved.

Saudi Arabia is doing the opposite: injecting money while others tighten. $75 million for EWC 2026, 37 clubs in the Saudi eLeague, no spending cap. Two poles separate clearly in the data: one stabilising, one pumping capital.

In my reports I usually split money into three tiers: prize money, sponsorship money, and owner money. Dota 2 lost tier one. The LCK is squeezing tier three. Saudi is pumping all three. That is why I do not use the word collapse when I talk about the esports winter. I use the word reallocation.

A successful deal has three versions: the rumour version that excites you, the closing version that disappoints you, and the liquidation version that teaches you about life. Dota 2 is on its third version, except it is not a deal at all. It is an entire model.

The biggest blind spot in the official story sits here: almost every data point I just laid out comes from unnamed sources. Only one statement, Falcons' remarks about long-term sustainable operations, is attributed to an identified source. The rest are unsourced figures or explicitly labelled opinion.

I do not say this to deny the picture. I say it for a professional reason: when data has not been cross-verified, the only way to read it is to look for the logic underneath. The TI pool at $40 million in 2026 and $3.4 million in 2026 matches the actual record, which gives medium confidence to the numbers around them.

But one gap remains unfilled: other regions. China, Europe and North America are almost absent from this picture. An article about global esports that leaves the three largest markets blank is something I will not accept in my own spreadsheet.

And this is what I want to stress: the risk this cycle is not evenly distributed. It punishes single-title, high-salary organisations with low commercial value. It rewards multi-title organisations with solid capital, tied to major events. One market, two fates.

The next question is not whether Dota 2 is dying. The question is: if the EWC and Gulf capital become the new centre, who prices a game's commercial value, the publisher or the party paying for the tournament?

And when a world champion still has to sell itself, my job is no longer reporting who wins. My job is reading the balance sheet before the scoreboard. This time, I will let the spreadsheet speak first.