Packed Stands, Empty Order Books: Seven Years of Waiting for US Esports Betting

I still keep a photo on my phone of the stands in a North American arena during a League of Legends final. Eighteen thousand seats, not a single gap. The roar was loud enough that I had to cover my ears. Thirty minutes later, when game two began, I opened a prediction-market app on my tablet and saw liquidity for that game thin as paper — a few thousand dollars scattered across both sides. At the same moment, a no-name college basketball game on a Tuesday night pulled many times that.

Two screens side by side. The same crowd. The same emotion. Two entirely different markets.

When the stadium is empty, I can hear the ball breathing. This time the stadium was not empty. It was packed, and the silence was in the money.

Context: a CEO speaking straight into the wound

A recent conversation with Seth Young, chief executive of ROLR, offers an explanation for that paradox, and the explanation is not easy listening. Young competed in Counter-Strike 2 at professional level before moving into the executive chair. He says plainly that the esports betting market in the United States is not there yet. He said the same thing seven years ago. He still says it today.

ROLR positions itself in the land between two worlds. On one side sit the traditional sportsbooks — DraftKings, FanDuel and Fanatics — names that have dominated the US sports betting market since the federal ban of 2026 was struck down in 2026. On the other side sit event-contract markets under federal supervision, most prominently Kalshi, where users trade binary contracts on the outcome of a future event.

ROLR takes on neither camp head-on. Young's phrasing is precise: the goal is to get its fair share, not to swallow the whole pie. Running a prediction market rather than a fixed-odds book means ROLR does not have to beat DraftKings in a promotional war. It needs a user base loyal enough, literate enough and patient enough to trade regularly.

On distribution, ROLR leans on Spike Up Media, a lead-generation firm that is also a major shareholder. The relationship is long-running rather than a one-off purchase, and the acquisition partner carries both an equity incentive and a commission incentive to keep conversion costs low.

Young stresses repeatedly that ROLR spends surgically. It measures return on ad spend before it scales budget. Over five years, the predecessor product High Roller delivered positive ROAS in markets that Young himself concedes are not nearly as strong as the United States.

The United States also has no single legal framework for esports betting. Each state runs its own rulebook, while event-contract markets sit under a different layer of federal oversight. A platform seeking national coverage must stack compliance layers, each of which burns time and money without producing a single new user. ROLR chooses to move slowly here: present where it can be, measuring where it is permitted, and refusing to expand purely under media pressure.

I stopped reading at that point and wrote a question in my notebook: if a weaker market still turns a profit, what exactly is missing in the United States?

Core: liquidity is not volume

I do not predict the future; I only read the maps other people draw wrong. The most popular map in this industry is a straight line between two points: huge viewership leads to huge betting volume. That line has never existed in esports the way it once existed in football.

Based on my experience watching matches, I notice a structural feature that no revenue summary ever shows: esports has a great many matches, but each one is far too thin. According to Riot Games' public data, the World Championship final typically peaks above six million concurrent viewers, excluding regional platforms in China. An NFL Sunday has sixteen games, and each is a national event with hundreds of millions of dollars changing hands. A week of the Korean League of Legends circuit has dozens of matches, but each is only a small stream.

A rough comparison: if one NFL game draws trading volume equal to one hundred esports matches combined, then owning ten times as many matches helps the operator not at all. It only multiplies the number of times the cost must be paid.

Viewed as an aggregate, the picture looks attractive. Viewed as per-match liquidity, a market maker carries a near-fixed cost for every pair of sides, and that cost erodes the margin. In basketball, a help defence works by sliding one beat early, before the pass arrives. In a prediction market, market making runs the same way: you must stand on both sides of the book before the order flow arrives. If both sides are thin, you cannot help yourself.

This is why the large platforms do not pour money into esports, despite coveting a young audience. DraftKings and FanDuel can list esports markets, but listing a market is not the same as building one. A thin pair of sides does not produce a good experience; sophisticated players leave for lower slippage elsewhere. As they leave, liquidity thins further, and the spiral deepens.

ROLR understands this. Choosing a prediction market over fixed odds is a structural decision, not an aesthetic one. A prediction market lets the platform earn from trading fees rather than from betting margin, which means the risk sits with the counterparty trader and not on the platform's own book. With thin liquidity, that model is far safer. In exchange, the platform depends entirely on the quality of its first cohort of users.

And here the five-year High Roller data becomes interesting. Positive returns in markets weaker than the United States do not prove that the United States will work. They prove something narrower but more important: a segment of esports users exists that will trade regularly if the product matches their behaviour. That segment is small, but the cost of reaching it is far lower than the cost of prising a football bettor out of DraftKings' hands.

The unit economics here are simple and merciless. If the cost of acquiring one new trader is X and that trader's lifetime value is Y, the whole story rests on the ratio between Y and X. At traditional sportsbooks, X is driven very high by the advertising war, and they compensate with enormous volume. In an esports prediction market, X is lower but Y is lower too, because users trade with unpredictable frequency and churn faster. The difference is that ROLR does not need a large Y. It needs Y greater than X, consistently, for long enough.

Put another way, ROLR is not trying to make the pie bigger. It is trying to eat the slice everyone else considers too small to bother with.

People look at the scoreboard; I look at the gaps between the numbers. The gap here is the distance between viewers and traders. And that gap may not be a hole that needs plugging. It may be a permanent feature of this sport.

Contrarian: three structural drags nobody wants to name

Young's admission can be read in the opposite direction. He says the market is not there yet. He has said so for seven years. The popular reading is that he is patient and honest. The other reading is that this industry is stuck inside a promise that has never been tested, and every year another person stands up to repeat that promise in a more sober tone.

I once wrote forty-seven handwritten pages of analysis about a match in which a team with twenty-five percent possession still won. Forty-seven handwritten pages are never wrong, only the way we read them is. The same holds here. The US esports betting market not being mature does not mean it is on the road to maturity. There is another possibility: it has reached its own steady state, and that state is far smaller than everyone expects.

Three structural drags exist that no amount of viewership can break.

Prediction culture in esports has existed for a long time, but it is free. Fans predict in Discord, in group chats, in internal fantasy leagues, and they predict for honour rather than money. When a behaviour has already been satisfied for free, selling it for cash is far harder than selling a behaviour that never existed.

The esports fan is also a player. This is the fundamental difference from football. A football fan rarely plays at a level that allows them to judge a combination, so they lean on experts and on odds to shape judgement. An esports fan often understands the mechanics well enough to believe they read the game better than the bookmaker. That cohort does not need a platform to tell it what a market is worth. The paradox is that this same cohort is the hardest to retain, because it spots mispricing immediately.

Publishers control the entire data ecosystem. Riot Games and Valve own the schedules, the game versions and, more importantly, every channel that monetises fans directly: in-game cosmetics, items, tickets. A betting platform that wants to run smoothly needs accurate real-time data, stable scheduling and an integrity environment controlled by the publisher. When power concentrates on one side, the value that trickles down to the intermediary is always thinner than expected.

The race does not begin when the gun fires; it begins when you realise the lane has been switched. For US esports betting, the lane may have been switched before the race was even announced.

Three risks worth tracking

The biggest risk is timing. If the US market takes another five years to mature, ROLR's strategy survives, but investors face a long silence with no growth story to tell.

The second risk is regulation. Prediction markets sit under different supervision from sportsbooks, and any federal change could narrow the product surface within a single season.

The third risk is integrity. A single match-fixing case in any league is enough to collapse the trust of the sophisticated cohort, and that is precisely the cohort ROLR needs most.

ROLR's greatest support is Spike Up Media. If US esports betting grows more slowly than expected, a multi-vertical acquisition partner can still rotate into other verticals. That is a rare cushion, and it is the reason ROLR can talk about slowness without panic.

I once wrote about a 115 million euro transfer before it was signed. 115 million euro is the price of a prophecy; but a prophecy never pays the price. Markets always price promises about the future before there is proof, and in esports the distance between promise and proof has stretched for seven years.

Takeaway

What stands out about ROLR is not its ambition. It is its refusal of ambition. A company willing to tell investors its market is not mature, willing to say it only wants its fair share, and willing to spend as though every dollar must justify itself — that is a rare posture in an industry that lives on growth stories.

And if Young is right, if this market truly needs a few more years, then the question for everyone else is not who will win. It is who still has the patience to stand and wait when the lights in the stands have gone out and the roar faded long ago.