Seven Years Waiting for an Unopened Market: ROLR, Spike Up Media, and the Missing Rails of U.S. Esports Betting

On the night of November 5, 2026, Chase Center in San Francisco was full. DRX and T1 went five games in the League of Legends World Championship final, and I sat high enough to watch the entire arena change color with every fight.

Three days later, on November 8, 2026, California voters went to the polls. Both Proposition 26 and Proposition 27 failed. Sports betting — which dozens of other U.S. states had legalized after PASPA was struck down on May 14, 2026 — still had no legal foothold in America's most populous state.

Which means that on the night the biggest esports final ever staged on American soil took place, nobody in that arena could legally stake a single dollar on it, in that state.

I wrote two numbers in my notebook. The first was how many people were in the room. The second was how much money could legally be wagered inside that room. The second number was zero.

This article is about the wall between those two numbers — and about a man who has stood in front of that wall for seven years, said exactly one thing, and kept counting.

Context: who is saying it

Seth Young is the CEO of ROLR. Before he took the executive chair, he competed as a professional CS2 player. That detail matters more than it looks. The person running an esports betting platform once sat on the other side of the monitor, knew what a 1v3 clutch in qualifiers feels like, and knew why people watch esports differently than they watch football. Nobody has to explain to him what his product actually is.

ROLR occupies an awkward middle. On one side sit traditional sportsbooks licensed state by state: DraftKings, FanDuel, Fanatics. On the other sit federally regulated event-contract platforms, of which Kalshi is the clearest name. ROLR chose neither edge. That is a decision, not an accident.

The most notable line in the material I have is not a number. It is an admission: the U.S. esports market is not there yet. And Young adds that he said the same thing seven years ago.

Seven years. In that span, Worlds has travelled through four countries, the CS2 circuit has changed owners and structures, and the VALORANT Champions Tour launched and became the esports industry's first closed-franchise league system. Americans still watch esports in enormous numbers. And Americans still do not bet on it proportionally.

On the business side, ROLR does not burn money. The company describes its approach as surgical: every dollar out must be measured in ROAS, return on ad spend. Its key partner is Spike Up Media, a lead-generation firm and a major ROLR shareholder. The two have worked together for five years, and across those five years the High Roller product delivered positive ROAS — in markets the CEO himself rates as far weaker than the United States.

ROLR's stated goal is not domination. It is to get its fair share of a large and growing pie. That sounds modest. But the phrasing contains a very specific assumption about market structure, and that assumption is what deserves dissection.

Viewership is not liquidity

In football, expected goals has been abused to the point where people use it to explain things it does not measure: player form, referee decisions, a team's nerve at minute 90. In esports, the metric abused in exactly the same way is viewership.

Peak concurrent viewers, hours watched, bodies in the arena — these numbers are presented as proof of an entire industry's commercial potential. They cannot do that job. They measure attention, and attention is only a necessary condition.

From my own experience watching matches at LAN events in North America, there is a physical gap that no dashboard ever shows. The person next to me at Chase Center that night screamed herself hoarse. She wore a T1 jersey, had flown in from Dallas, and had paid several hundred dollars for the ticket and the flight. But when I asked whether she had anything riding on the match, the answer was no — and not because she did not want to. The answer was that there was no legal door where she was sitting.

This is the first variable most analyses of the esports betting market skip: esports audiences are global, while betting licenses are state by state. The audience for a Worlds final is scattered across Vietnam, China, Korea, Brazil, Germany, Poland and the Philippines. The licenses needed to serve them with a regulated betting product sit in New Jersey, Pennsylvania, Colorado, Indiana — places where domestic esports viewership is a small fraction of the global total.

The stadium and the market are two different rooms, and the wall between them was built by law, not by demand.

There is a fair counterargument: if demand were truly large, it would find a way. It has found a way — through unlicensed offshore books, through black markets, through crypto rails used by players who have no other option. Which is exactly why I have to be careful about a trap: licensed-market data does not measure total demand, only the portion of demand the legal system permits you to see. When a CEO says the market is not there yet, he is talking about the licensed room. The unlicensed room may have been crowded for years, and nobody has the numbers.

That is both the industry's weakness and its reason to exist.

Fragmented supply: why the esports calendar cannot feed a liquid market

Sports betting lives on repetition. A book needs a steady stream of events to bring users back, to cycle capital, and to produce spreads thin enough for sharp bettors and wide enough for the house. The English Premier League plays 380 matches across nine months, almost weekly. The NFL plays eighteen weeks plus playoffs, with a fixed weekly rhythm everyone knows by heart.

Does esports have a repeating calendar? Partly. The LCS and LEC run weekly through most of the season. The CS2 circuit runs events year-round. But three structural differences make this flow of events hard to convert into liquidity.

The first is time zones and geography. An LEC group-stage match on a Saturday evening in Europe is midday on the U.S. East Coast — workable. But most professional League of Legends and VALORANT competition in Asia runs in hours when American bettors are asleep. For a U.S.-licensed book, the effective supply is cut down to a fraction of the global total.

The second is fragmentation. Esports has no single governing body in the mould of a national federation. Every title has its own publisher, every region its own organizer, every tournament the ability to change format mid-season. For a bookmaker, each format change means rewriting pricing models, retraining bettors, and re-explaining everything to compliance.

The third, and the most important, is the structure of the money. In mature football betting, most revenue comes from in-play: people staking money while the match unfolds, continuously, minute by minute. That is where the best margins are and where retention is strongest. Pre-match is just the doorway.

In-play needs something esports does not yet have at industrial scale: an official, standardized, legally sanctioned, latency-committed real-time data feed. Without it, a book cannot price continuously, cannot take risk safely, and is forced back to crude markets: who wins, who advances, who lifts the trophy.

That is why most esports betting today looks more like a lottery counter than an exchange.

The rails nobody is selling

In football, bookmakers buy data from professional suppliers: firms with staff in the stadium, their own cameras, and contracts with the leagues. Data exploitation rights are a revenue line for organizers, and they exist as a mature ancillary market three decades old.

Esports has no comparable ancillary market, and I believe this is the real bottleneck — not demand, not marketing, not user awareness.

Picture the betting machine as a railway. The bookmaker runs the trains. Bettors are passengers. But the rails, the rolling stock and the signalling must come from a third party: data. Whoever owns real-time match data controls the speed and the ticket price of the entire system.

In esports, the rails belong to publishers. Riot Games owns League of Legends and VALORANT data. Valve owns CS2 and Dota 2 data. For years, both have been cautious about commercializing data for wagering, for sound reasons: they do not want their names tied to match-fixing, to suspicion, to scandals a product aimed partly at teenagers should never touch.

That caution is ethically correct. It also creates a vacuum that anyone building a serious esports book must fill on their own — with self-collected data, relationships with organizers, informal agreements. None of that produces durable competitive advantage, because nobody owns the source.

This is the single point I want to stress most in this article: the problem with the U.S. esports betting market is not immature demand, it is that nobody is selling the rails. And when the rails have no owner, everyone in the industry is operating in standby mode — including the best of them.

Seven years is data, not emotion

People laughed at my predictions, but nobody laughed at how I recounted every number.

I read Young's line the way a counter reads. A CEO repeating the same assessment for seven years is either someone holding a correct thesis or someone stuck inside an expired one. Those are not mutually exclusive. The way to tell them apart is to ask: what compounded in those seven years?

If seven years produced only seven more years of waiting, that is the signature of a rotting thesis. If seven years produced compounding capability — lower user acquisition costs, a more stable product, deeper user data, tighter partner relationships — that is the signature of a machine that works, running in a small room.

With ROLR, I lean toward the second. Five years of positive ROAS with High Roller in markets weaker than the U.S. is concrete evidence. It shows the company is not waiting for the market to open before learning how to operate. They have finished their unit-economics homework and are waiting on their scale homework.

But there is another reading I am obliged to include, because it weakens the argument above: positive ROAS in a small, less competitive, low-compliance-cost market cannot be extrapolated into positive ROAS in the United States. In the U.S., user acquisition costs are set by the giants. DraftKings and FanDuel spend hundreds of millions of dollars a year on marketing. They do not need to care about esports to bid up ad prices in every segment ROLR might want.

So the right question is not whether ROLR is good. The right question is how long ROLR's skill retains its value when input costs are bid up by people who are not even playing on the esports field.

The economics of the middle player

I noticed ROLR's word choices. They talk about their share of the pie, not about seizing the pie. They talk about measured spending, not growth at all costs. They talk about differentiating from the big books rather than beating them on scale.

In many contexts, that is the language of a latecomer who knows they cannot win on weight. Here, I think it is accurate — and more importantly, honest about the structure of the game.

Esports is a segment the big books have not bothered to optimize. Not out of ignorance, but out of opportunity cost. A product manager at DraftKings has two options: throw resources at a Thursday night NBA game with hundreds of millions in handle, or at a CS2 quarterfinal with a fraction of that handle and a far higher knowledge requirement. Both choices are rational, and the result is almost always the same.

That gap is ROLR's living room.

There, competitive advantage is not capital and not brand. It is being the only party willing to do the work of pricing a product category the big books do not think is worth pricing.

That is a real advantage, but it is the kind that exists only inside a narrow band. And that narrow band is what the next section is about.

Contrarian angle: this is not a bet on esports

A good hot take is not about daring to be wrong, but about daring to be right ahead of the world.

If I had to compress ROLR's strategy into one sentence, I would not say they are betting on the maturation of the U.S. esports market. I would say this: they are betting on a Goldilocks band — large enough to be profitable, small enough that the giants cannot be bothered to optimize it, and stable enough that nobody fills the gap for five more years.

That bet has its own logic and two clean ways to be destroyed.

Scenario one: the market never grows. ROLR thrives in a small box, but founders do not build companies to stay small. Shareholders do not invest in a market that never opens. If in five years the pie is the same size, the value of ROLR's machine is capped, and the cap was not set by them.

Scenario two: the market grows too fast. When a segment gets big enough, the giants enter. They have capital, customer data, licenses in every state, and the ability to lose money to win share. For them, esports becomes a customer-acquisition line item, and they can outbid anyone. ROLR, with its surgical spend, gets squeezed between rising ad prices and shrinking margins.

In other words, the condition for ROLR to win is narrow: the market must grow, but not too fast, and in segments that remain unattractive for the giants to land in.

I do not think that is a bad strategy. Plenty of good companies are built in narrow bands, and surviving inside one while the window opens is itself a capability. But it is different from the story told to the press: that esports is an untapped gold mine and all you need is patience.

Patience is not a strategy. Patience plus low user acquisition cost is a strategy. And low user acquisition cost is an advantage only until someone decides it is worth paying more.

There is one more variable I want on the table, because it rarely gets said. Esports moves faster than football because esports is not afraid to be wrong. Publishers change the rules, the maps, the tournament formats, even the entire season, in far less time than a football federation needs to amend its statutes. That speed benefits players and viewers. But for a product that financializes fan belief, structural churn is operational risk.

Every time a publisher changes a map, any pricing model built on historical data loses part of its value. Every time a qualification format changes, the price curve of futures contracts has to be rewritten. Without official rails, those shifts make the operating cost of a small book far higher than it looks from outside.

Where my assumption could be wrong

In May 2026, when the Bundesliga returned to empty stadiums, I noticed home win rates fall sharply and wrote a piece arguing that home advantage is a con. A month later the Premier League restarted and the numbers reversed completely. I had to publish a correction. The lesson I took, and still use, is not to avoid conclusions, but to ask before publishing: what exception could falsify my numbers?

For this argument, there are four exceptions I must name.

The first is that the audience is ageing. The person in that arena in 2026 is now thirty. The first generation of League of Legends and CS2 viewers is entering the age of disposable income and of spending on risk-based entertainment. If the conversion threshold from watching to wagering happens on an age cycle, then seven years of waiting were not seven wasted years — they were seven years of marinating. In that scenario the market opens not because the law changes, but because the audience changes age. And I will have to rewrite this article.

The second is the maturation of event contracts. If the regulatory framework for event contracts extends to esports at the federal level, a platform like ROLR could serve users in more states without a state-by-state gaming licence — a complete change in operating conditions, potentially favourable. But it could also cut the other way: if every platform can list esports contracts, specialist expertise gets flattened and the game shifts to whoever has the deepest liquidity.

The third is official data. If a major publisher announced an official real-time data feed for wagering, the industry would turn a page. In-play becomes viable at scale. But when that happens, whoever controls data distribution sets the rules, and platforms without formal relationships become tenants on rails someone else owns.

The fourth, and the one that bothers me most, is that licensed-market data may be counting wrong. If most esports betting demand flows through unlicensed channels, then the line about the market not being there yet describes only half the picture — the half the law lets you see. Nobody has numbers for the other half, and what has no numbers cannot be priced.

I am putting these four exceptions on the record so that, if the data flips, the correction needs no long preamble. An empty stadium does not make the away team stronger; it only strips the mask off the home team. An unopened market works the same way: it does not make the latecomer smarter, it only strips the mask off the missing rails.

What I will recount myself

If I am wrong, I want to be wrong in a way that can be checked. So I am setting three markers to check myself against.

Marker one: within the next two years, will at least one major publisher announce an official data framework or integrity mechanism tied to wagering on a flagship title? If so, the industry's biggest knot has been partly untied, and every growth calculation has to be redone.

Marker two: will esports be broken out as its own line in any U.S. state's betting revenue report? As long as it sits inside another category, the market is not yet big enough for regulators to name — and a market nobody names rarely gets its own law.

Marker three, and the one I believe most: the real maturation signal for U.S. esports betting will not be a sold-out arena, and not an industry report. It will be a major book buying esports-targeted advertising at a budget large enough to move prices. When the cost of acquiring one esports bettor is bid up by people who do not understand esports, that is when the market really opens. And that is also when the surgical-spend model has to prove it is still worth something.

In 2026 I stood alone in front of the whole world and predicted Croatia would reach the World Cup final. It turned out to be the most valuable position I have ever held. Eight years later, I keep the same habit: make a falsifiable call, publish the exceptions that could kill it, and write down the date I have to come back and read myself.

The last thing I want to say is not for investors. It is for the fan sitting in row eleven, where I once sat. The fervour inside your room and the demand outside the market are two different things. A packed arena is proof of love, not proof of cash flow. As long as the rails have no owner, people will keep counting the people in the room — and forget to count the doors that are open.