ROLR Between Two Rulebooks: Why the US Esports Prediction Market Hasn't Blown the Whistle

In 2026, at the height of the pandemic, I sat in the office of a regional football federation in Marseille drafting a 38-point checklist for matches played without spectators. The biggest lesson from the 23 friendlies that used it was not on the checklist at all. It was this: when the stands are empty, you finally see what actually controls a match. What controls it is usually a rulebook.

Six years later, in a different time zone, I ran into the same paradox, inverted. North American esports arenas are packed. The ledger behind them is empty.

Seth Young, founder and operator of ROLR and a former competitive CS2 player, said plainly in an interview that the US esports betting market is not there yet. He added that he had said exactly the same thing seven years earlier. To someone who works with rules, that reads less like patience and more like a confession about structure.

Context: a product standing between two rulebooks

ROLR does not position itself as a sportsbook. Its product belongs to the prediction-market category, where users trade on event outcomes rather than stake at fixed odds. That legal boundary is the single most important detail in the whole story, and the one most often skipped when people talk about esports betting.

A sportsbook such as DraftKings or FanDuel operates under state gaming commission licences. An event-contract exchange such as Kalshi operates under the supervision of the US Commodity Futures Trading Commission. Two rulebooks, two licensing authorities, two sets of compliance standards. ROLR chose to stand in between.

Young is explicit that he does not want to become a copy of DraftKings. He lists DraftKings, FanDuel, Fanatics and Kalshi as four powers he must differentiate from. Strategically that is sound: attacking those four head-on is suicide on user-acquisition cost.

ROLR's capital structure follows the same logic. Spike Up Media is both a major shareholder and the user-acquisition partner. Young describes his spending as surgical, with every advertising dollar measured by return on ad spend. He has data to back it up: for five consecutive years the predecessor product, High Roller, delivered positive ROAS in markets he himself calls weaker than the United States.

That is respectable evidence. But what does it prove? It proves ROLR knows how to buy users cheaply. It does not prove there are enough users to buy. Those are different questions, and the gap between them is this entire article.

Seventeen years covering this industry taught me a habit: read the match report before reading the news, because a match report does not lie. There is no match report here. There is an equivalent: the spending and return data series. Read that series and you see a company buying a ticket to a stadium that has not been built yet.

Analysis: the bottleneck sits in the rule layer, not the product layer

The gap between viewership and trading volume in the US is not a new phenomenon. Young admits he has watched it for seven years. The number carries its own meaning: it is longer than a typical venture cycle, longer than a patch cycle, and longer than the lifespan of several titles that rose and fell inside that window.

When a problem persists that long while demand remains present, the cause is rarely on the demand side. It sits in the legal infrastructure.

Three layers of infrastructure are stacked on top of each other in the United States.

The first is state-level gambling law. After the Professional and Amateur Sports Protection Act was struck down in 2026, authority to legalise betting moved to individual states. The result is a patchwork map. The same product can be legal in one state and illegal in the next. For an exchange, every state means another licence application, another dossier, another timeline, another set of compliance conditions.

The second is federal event-contract law. The CFTC supervises exchanges such as Kalshi, and every new contract must demonstrate it does not fall into the prohibited games-of-chance category. ROLR sits at the intersection of these two layers, dependent on how both the state and the federal government define its product.

The third layer, the decisive one in esports, is publisher law.

Riot Games, Valve, Epic and other studios own the games, the schedules, the match data and the commercial exploitation rights. No independent federation stands above them to write common rules. In football, IFAB owns the laws of the game and FIFA runs the competition system. In esports, that role sits with the studios.

This is the variable every esports prediction model must price. An event-contract exchange may secure licences in three states, but if a publisher changes its data-exploitation terms, alters the schedule, or launches a competing in-house product, those licences lose most of their practical value.

In esports, the real referee of the betting market is not a state or the CFTC. The real referee is the game publisher. They do not hold a whistle, they do not wear black, and they are never questioned on television. But every time they update their terms, they have just redrawn an offside line. And the offside line has never been straight; it is only that today I can see the curve.

This explains why Young's seven-year number is more worrying than it looks. If the bottleneck were demand, time would solve it. If the bottleneck is law, time solves nothing. A disallowed penalty can be corrected; a legal vacuum cannot.

ROLR's spending discipline should be read for what it is. It is a risk-management instrument. When a company chooses to be surgical about every advertising dollar, it is saying it is not certain about the size of the market ahead. That is the measured caution of someone who played professionally and knows what losing an all-in feels like.

There is another detail worth examining. Young says his goal is to take a fair share of a large and growing pie, not to swallow the whole pie. It sounds modest. In investment language, though, a claim to a fair share is a defensive claim: it assumes the pie will grow by itself and the company's job is merely to be standing in the right place when it is baked. If the pie does not grow by itself, that strategy has no plan B.

High Roller delivered positive ROAS in markets weaker than the United States. That detail deserves scrutiny because it raises a question of geography: if the model worked where there are fewer players, why has it not worked where there are the most? The most reasonable answer still sits in infrastructure. In many non-US markets, esports betting either has a clear legal framework or exists in a grey zone operators are willing to accept. In the US, both routes are blocked by a supervisory system that is transparent but fragmented.

And this is where the ROAS data begins to show its limits. ROAS measures user-acquisition cost. It does not measure order-book depth, the bid-ask spread, or whether users return after the first month. A prediction exchange can buy users extremely cheaply and still die from a thin order book. Five years of positive ROAS in weak markets is evidence of distribution capability, not yet evidence of retention capability.

The same holds for event-integrity risk. In traditional sport, a match-fixing case is handled by the federation, the police and the courts, with a public record. In esports, most fixing cases are handled internally by publishers, with wildly varying degrees of disclosure between studios. Sportsbooks priced this risk into their odds long ago. Prediction exchanges have no such shield: the risk sits directly on the order book, and the final bearer is the trader.

The counter-intuitive point: caution is itself part of the problem

There is another way to read the line about saying the same thing seven years ago.

The first reading is compressed optimism: the market is not ripe, but it will be, and ROLR is holding a good position.

The second reading is less comfortable. A prediction repeated for seven years can become a self-fulfilling prophecy. If the whole industry agrees the market is not there yet, nobody builds infrastructure for it. Nobody invests in event-integrity standards. Nobody lobbies to have esports law written on its own terms rather than folded into the traditional sports-betting mould. And because the infrastructure is missing, the market stays not there.

VAR is not wrong. The people operating VAR are only people. Here, the operators are platform executives, and they are waiting for a signal that only they can send.

I want to place one more point on the table, even if it costs me some readers.

Esports fans respond to the notion that the market is not there yet with a sense of being condescended to. That feeling is valid data. They filled the arenas, paid for tickets, stayed up all night. From the stands, the market has clearly arrived.

But the habit of watching and the habit of wagering are two different things, and they are not born at the same time. European football took nearly a century to build a culture in which people remember league tables, memorise head-to-head records, and argue over statistics. That record-keeping culture is what turns a match into a market. Esports built its audience in fifteen years but has not finished building the record-keeping culture. That is why I do not believe the explosion scenario, and equally why I do not believe the collapse scenario.

This matters to someone who works with data. Distance covered and sprint counts get packaged as effort metrics, but running without effect also produces pretty numbers. Likewise, viewership gets packaged as a potential metric, but watching without trading also produces pretty numbers on a fundraising slide. A viewership table is not a liquidity table.

Takeaway: a concrete proposal, and an open question

If I sat in the room drafting rules for this market, I would not propose scrapping the current framework. I would propose a separate licence tier for esports event contracts, with three measurable conditions.

First, publishers publish an event-integrity protocol: who monitors, what the fraud investigation process is, and how long before findings are disclosed. Low cost, because most of the data already exists inside their own systems. Second, standardise real-time match data feeds into a single format so every exchange receives the same source stream and data error stops being a competitive advantage. Third, mandate periodic liquidity disclosure, requiring exchanges to publish minimum order-book depth, so regulators can tell platforms with real trading apart from platforms with only an advertising campaign.

None of those three requires Congress. All three are achievable within eighteen months. And all three aim at the actual bottleneck: legal infrastructure.

Anyone who writes rules needs someone standing outside the line to check their signature. In this market, that person has not appeared. A match does not end with the whistle; it ends when the report has been read. For esports betting in the United States, the report is still blank. The real question is not when the market arrives. The real question is who will hold the pen.