US Esports: Packed Arenas, Empty Order Books

Seth Young played competitive CS2 before taking the CEO chair at ROLR, a prediction market platform built for esports. In his latest interview he repeated a line he first said seven years ago: the US esports betting market is "not there yet."

What stands out is that the line has not changed. Seven years is long enough for a young roster to go from qualifiers to a world title, long enough for an entirely new meta to arrive and be broken, long enough for a tournament to grow from nothing into a hundred-million-view event. Yet a financial market attached to that exact same audience has stayed still.

I know this feeling. It is the moment you watch a team hold a ten-thousand-gold lead, control the entire map, take three dragons in a row, and still fail to close the game. Every advantage is in hand, and the win stays out of reach. In esports we call that a team with no conversion path. The US esports betting market has exactly that condition.

ROLR is not trying to become DraftKings. That is the single most important positioning statement in the conversation, and it determines every tactical choice behind it.

The US sports betting market has two clear poles. On one side sit traditional sportsbooks — DraftKings, FanDuel, Fanatics — where bettors wager at fixed odds set by the house. On the other sits Kalshi, an event-contract exchange operating under the oversight of the CFTC, the US Commodity Futures Trading Commission. ROLR picked the prediction market: users buy and sell positions on match outcomes, prices move with supply and demand, and no bookmaker stands in the middle setting odds.

That choice is not avoidance. It is a structural decision. When you cannot win a full teamfight against a lineup that is stronger on every metric, you change how you play — split the map, attack the economy, control vision, force the opponent into reacting. ROLR is attacking the gap the giants leave open: an esports community with a huge viewing audience but no matching financial product.

Its partner is Spike Up Media, a multi-vertical lead generation firm that is also a major shareholder. Over five years the High Roller product, ROLR's predecessor, delivered consistently positive ROAS in markets the CEO himself concedes are not as strong as the United States. ROAS, return on ad spend, is a metric that is hard to fake. Five years of positive data is a long enough sample to argue the model does not survive on luck.

The problem sits in a simple paradox: US esports viewership is high, but the rate at which that viewership converts into betting activity is abnormally low compared with traditional sports.

Young describes the image of crowds packing an arena to watch a League of Legends match. That scene is real, and I have stood in those stands myself. But most of those same people, walking out the arena doors, never open an order book.

In meta language, this is a team that wins the fight but cannot push a tower. You have the advantage, the resources, the tempo, but no conversion path. In esports, a team without a conversion path loses to a team that turns kills into towers — and it loses quietly, with no highlight to blame.

Several layers of cause are woven together.

Regulation is the first and hardest layer. US sports betting opened up after the 2026 PASPA ruling, but the power to decide sits at state level. Esports betting falls into grey space: some states classify it as sports betting, some treat it as a game of skill, some have no definition at all. Prediction markets such as Kalshi sit under an entirely different legal framework. ROLR stands between two frameworks, which means every state is a new match it must play from scratch.

Product culture is the second layer. US esports viewers grew up with Twitch and YouTube, where free content is the default and interaction happens through chat, memes and clips. They did not grow up with fantasy leagues and betting slips the way a generation of football viewers did. Moving a community from watching for fun to watching to hold a position is a behavioural problem, not an interface problem. And behavioural problems always take longer than technical ones.

The final layer is data. Betting needs reliable real-time data: scores, timings, minute-by-minute metrics, updates with zero latency. Esports moves far faster than football. A League of Legends match can flip in thirty seconds, while football needs several minutes for the picture to change. That demands a much lower-latency data infrastructure, and current esports data providers have not reached the maturity the financial market requires.

One detail gets little attention but matters: trading volume per esports match is far lower than at major traditional sports events, even though viewership is not inferior. A single traditional team-sport match can attract tens of millions of dollars in wagers. In esports the figure is an order of magnitude smaller. That gap does not reflect the appeal of the discipline — it reflects liquidity depth, and liquidity only forms when enough people believe the market is worth participating in over the long run.

The calendar is another variable. Esports runs almost year-round: LPL, LCK, LEC and LCS at regional level, MSI and Worlds at international level, plus countless minor events. Abundant event supply is an advantage for product builders, but it also fragments liquidity across markets. A team playing map control needs vision in every lane, and esports liquidity is currently spread too thin across too many lanes at once.

Every collapse starts with a bug the team was too casual to fix. Here the bug is not a single technical fault but a prolonged delay in building synchronised data infrastructure and regulation. The industry has known it needs both for years. But it is the kind of cost that produces no highlights and no views, so nobody wants to be the one who pays first.

ROLR chose not to wait. It spends in a measured way, focuses on verifiable metrics, and uses Spike Up Media as a multi-vertical acquisition channel. Young sums up the strategy in a line worth keeping: they are not trying to take the whole pie, only their fair share.

In draft terms, this is a composition with no single carry. No superstar, no solo play to close out the opponent. Only vision control, tempo retention, and towers taken one step at a time. That style does not create fanatical crowds, but it wins across seasons.

Here I want to pause and doubt the "market is not there yet" frame that both interviewer and interviewee accept so comfortably.

That frame implicitly assumes the US esports betting market will follow the same path other markets took, just a few years later. The evidence does not fully support it. US esports viewership has been high for years. If "not there yet" means "it will arrive," then seven years is far too long for a market that is merely slow.

There is another reading, and it is less comfortable: US esports fans are not slow, they are different. They consume esports as a social entertainment product, not a financial one. If so, waiting for them to convert is like waiting for a control-oriented team to switch to hyper-carry — possible, uncertain, and slow enough to cost you the season.

In the sports rights business, this pattern has repeated many times. Streaming platforms poured money into rights on the assumption that viewership would convert itself into subscription revenue. Most deals failed because that assumption was wrong — viewers kept watching, but did not pay more. US esports betting may be walking the same road, just at a different layer of the value chain. The sports rights bubble has already peaked; and if it has peaked, expecting a new revenue layer to grow without limit is an assumption requiring proof, not a truth to be assumed.

Fate never favours anyone; it only rewards those who know how to read RNG. ROLR, with five years of positive ROAS in weak markets, is a team that reads RNG well. But reading RNG does not mean the RNG will reward you on a different stage. Probability does not move with you when you change maps.

What deserves credit in Young is that he does not sell the dream. He said "not there yet" in an interview where saying "almost there" would have served him best. That honesty is rare in this industry, and it is also the thing most likely to cool short-term investor interest.

What I take from this conversation has nothing to do with ROLR.

It concerns how the esports industry measures itself. For years we have used viewership as the yardstick of the discipline's strength: concurrent viewers, packed arenas, video plays. But viewership is an input metric, not an output metric. It is like gold earned during the laning phase — impressive, but it does not decide who wins the final fight.

Esports has proven it can draw an audience. The next task is proving it can convert that audience into stable financial value — through betting, paid content, digital goods, or a model nobody has invented yet.

The stands are empty, but the heart of the match still beats — only now we hear it more clearly. In the United States the stands are not empty. The heart of the match is still beating. The problem for US esports is not a lack of viewers, but a lack of conversion paths. That is a far harder problem than filling an arena — and it will not be solved by adding one more tournament to the calendar.