Every few years the American esports industry contracts. Organisations lay off staff, leagues restructure, teams that raised large rounds sell for a fraction of what they raised, and a wave of commentary declares the whole thing a bubble. Then it grows again, and a wave of commentary declares it the future of sport.
Both waves are wrong in the same way. They treat esports as one thing with one trajectory, when what actually exists is four different money flows with four different reliabilities, stacked on top of an enormous participatory hobby that is in no danger at all.
Here is each flow, what it depends on, and what happens when it stops.
1. Publisher money
The largest and least discussed. Publishers fund leagues, prize pools, production, venue costs and — directly or indirectly — a good deal of team revenue through revenue shares and slot arrangements.
Why they do it is the important part. For a publisher, competition is marketing. A visible professional scene sells copies, sells cosmetics, retains a player base, and gives a title cultural permanence. That is a perfectly rational spend, and it does not require the league to be profitable in its own right.
Which is exactly the risk. A marketing budget is reviewed annually against alternatives. When a publisher decides its money produces more return elsewhere — or the title's population declines, or the company itself is restructured or acquired — the subsidy is reduced or withdrawn. Teams, staff, players and third-party organisers built on that subsidy discover their revenue was a line item in somebody else's budget.
This is not hypothetical: the largest single-title contractions in American esports have followed publisher strategy changes, not audience collapse. Understand this and the structure of the leagues makes immediate sense.
2. Sponsorship
The main revenue line for almost every team, and much larger than prize money, which is not really revenue at all. Two very different types.
Endemic sponsors sell to gamers: peripherals, components, chairs, monitors, energy drinks, and increasingly gambling and crypto products. They understand the audience, they buy readily, and their budgets are small, because the sponsorship is a customer-acquisition cost aimed at people who are already customers. Historically, endemic money kept the entire Western scene alive.
Non-endemic sponsors sell to everyone: cars, banks, telecoms, snacks. Their budgets are the ones that would change the industry's scale, and they were the promise underlying most of the growth forecasts. They have arrived in smaller numbers and for shorter terms than expected, for a specific reason: they buy measured reach, benchmarked against other media, and esports has struggled to produce audience measurement that survives that comparison. Concurrent viewers on a free stream, counted with varying methodology, does not sit comfortably next to a rated broadcast.
Sponsorship is also the first line cut in a downturn, in every industry. When endemic budgets tighten and non-endemic experiments end at the same time, a team dependent on sponsorship loses most of its revenue in a single quarter. That correlation is the mechanism behind the contractions.
3. Media rights
In traditional American sport this is the largest line, and it exists because access is scarce: the game happens once, and the right to show it can be sold.
Esports gave that away, for an excellent reason — free streaming is what finally solved its distribution problem after a decade of not having one. But an audience acquired on the expectation of free access cannot be converted to paid without losing the scale that made it commercially interesting in the first place. Attempts to move esports broadcasts behind a wall have consistently traded away more reach than they gained in revenue.
What exists instead is platform exclusivity: a rights-holder is paid by a streaming platform for exclusive carriage. This is real money and it has been paid at meaningful scale. It is also weaker than it looks. The platform, not the rights-holder, owns the viewer relationship, the advertising inventory, the data and the recommendation surface. And the deals reset — when a platform's strategy changes, the line vanishes, and the audience has to be re-acquired somewhere else.
4. Direct sales
Merchandise, tickets, memberships, in-game items sold on behalf of teams, coaching, and content businesses run by players and organisations.
This is the smallest line and the most durable, because it is the only one where the buyer is the fan rather than an intermediary buying access to the fan. It also has an uncomfortable implication the industry has been slow to accept: it works best for brands people want to wear and personalities people want to follow, not for rosters people watch. A team whose value is entirely its current five players has almost nothing to sell directly, because the players can leave and take the relationship with them.
The organisations that have weathered contractions best are consistently the ones with real direct revenue — an audience they own, merchandise with actual brand equity, or a content operation that would survive losing the roster.
Why the industry keeps contracting
Put the four lines together and the cycle is mechanical rather than mysterious.
Investment arrives on a growth forecast. Teams raise capital and spend it on salaries and slots, because talent is the visible competitive lever. Costs become fixed — multi-year contracts, guaranteed league fees, facilities — while all four revenue lines remain variable. Non-endemic sponsorship does not scale as projected, media rights do not materialise, and publisher spending is reviewed. Costs cannot be reduced as fast as revenue falls, because the costs were contracted. Layoffs, roster releases, league restructuring, and a round of obituaries.
Then the underlying activity — millions of people playing competitively, which never went anywhere — reasserts itself, growth resumes, and the cycle begins again.
Note what is not in that description: the audience abandoning esports. It generally has not. What contracts is the business built around the audience, not the audience.
Why the forecasts were wrong
The "esports will be bigger than traditional sport" pieces of the late 2010s share a small number of identifiable errors, all of which are still being made.
Participation was counted as spectatorship. Hundreds of millions of people play competitive games. Far fewer watch other people play them, and fewer again watch in a way that monetises. Player counts made spectacular slides and predicted nothing.
Audience size was assumed to convert like a television audience. Revenue per viewer-hour in esports is a fraction of the figure for major American sports, because distribution is free, the demographic is young with less disposable income, the geography is international and fragmented across many currencies and markets, and the ad relationship largely belongs to a third-party platform.
Growth rates were extrapolated from a base of nearly zero. Anything trebling from a small base produces an enormous number when projected a decade forward. Almost every widely circulated market-size forecast from that period overshot.
Ownership was ignored. Comparisons to football and basketball skipped the fact that no company can revoke the right to play those sports, and that a game's competitive life is finite in a way a sport's is not.
And prize pools were read as revenue. They are marketing expenditure denominated in payments to competitors — as the split calculator makes concrete, the headline pool and the money reaching a player differ by more than an order of magnitude.
What is actually stable
The parts of American competitive gaming that support themselves have a common shape: they sell something to somebody directly, and their cost base matches their revenue.
- Creator and content businesses. A direct audience relationship, sold as memberships, sponsorship and merchandise, without an intermediary owning the connection.
- Regional and community competition run at a matching cost base. Local events where ticketing, entry fees and local sponsorship cover the venue. Unglamorous and remarkably durable — the fighting-game scene has run on close to this model for two decades and survived multiple industry winters intact.
- Collegiate esports. Funded by institutions for institutional reasons — recruitment, retention, engagement — which is a very different and much steadier funding logic than sponsorship.
- Tools, coaching and services. Selling to players, who are numerous and already spending, rather than to advertisers who want access to players.
The conclusion most people resist is worth stating plainly: competitive gaming may be an enormous participatory activity with a modest spectator business attached, and the last decade has been an expensive attempt to force the second to match the size of the first. That would not make esports a failure. It would make it something other than the NFL — which, given that nobody owns football and somebody owns every video game, it was always going to be.