Kalshi and $1.73 billion in esports trading: what the data says, and what it doesn't

In 2026, aged 26, I sat in the third row of the Sungui Arena stand and counted. Across three consecutive Incheon United training sessions, the squad repeated one corner-kick routine 47 times. Nobody in the press room asked me about those 47 repetitions. A year later, when my analysis of Incheon's corner-kick decoy pattern passed 200,000 reads, I understood what has kept me in this job since: a season looks like a single summary line from outside, and like hundreds of quiet repetitions from inside. The training grass at Incheon still remembers every step I stood waiting on.

This week the esports world read one of those summary lines. Kalshi, a US-licensed prediction market, reported $1.73 billion in esports trading volume over twelve months. Four titles sit in the tracked set: CS2, League of Legends, Valorant and Dota 2. The unit of trading is the individual single-match market.

If I do what I always do on a training pitch — break one summary line into its repetitions — the $1.73 billion resolves into four very unequal quarters: $44 million in Q4 2026, $173 million in Q1 2026, $549 million in Q2 2026 and $967 million in Q3 2026. The sum is $1,733 million, or $1.73 billion. The arithmetic is exact. The headline is not wrong by a single dollar. It is simply a sum nobody has read closely, and almost the whole story lives there.

An event-contract exchange, not a sportsbook

Kalshi operates as a designated contract market licensed by the US Commodity Futures Trading Commission. Participants buy and sell binary contracts on event outcomes; the contract price moves around the probability the market assigns. Volume here counts matched contracts, both sides combined. That structure differs fundamentally from a sportsbook, where one-directional stakes flow into the book and out again at payout odds.

Three details in the original report shape how I read the whole dataset. First, the audience scope is described as largely American. Second, Polymarket — an offshore, crypto-accessible venue — has traditionally been the home of most League of Legends trading. Third, Kalshi supplies data to tournament organisers, publishers and sponsors so it can sit alongside viewership data.

Put those three together and the picture stops being tidy. A US venue reports volume across four titles, while the title with the largest global audience trades mostly somewhere else. The tracked set contains no Honor of Kings, no Peace Elite, no mobile title at all. That is consistent with a platform serving a US audience, in a world where the largest mobile esports market is largely unreachable for legal reasons.

The legal ground under this sector is not settled either. Prediction-market operators are suing and being sued by state gambling regulators in the United States. A venue contesting jurisdiction with state authorities still publishes real data, but the floor beneath it is moving.

Based on my experience covering matches, I keep one habit: before trusting a table of numbers, ask what the table counts. Here it counts matched contracts on single-match markets, across four titles, for a mostly US audience. That is a very narrow definition, and any reading that goes beyond it needs more evidence.

Four quarters, three speeds, one quarter holding more than half

Quarter-on-quarter growth ran at +293%, +217% and +76%. Absolute volume is still rising, but the growth rate has now decelerated for three consecutive quarters. Anyone who has followed a team through a season knows the feeling: the team keeps scoring, but the rate of scoring slows, and the people inside the dressing room feel it before the table shows it.

Q3 2026 alone accounts for $967 million of the $1,733 million total — 55.8% of the entire twelve months. The $1.73 billion headline is stale on arrival: it overstates the past with quarters that have already decayed while understating the present with a run-rate that is far higher. Annualising Q3 2026 gives roughly $3.87 billion, about 2.2 times the headline figure. An article using $1.73 billion to describe the present is describing an average point that has already slipped behind.

This is where prediction-market data diverges from revenue data, and where most readers go wrong. Volume counts matched contracts. Someone who opens a position and closes it immediately adds volume twice while adding no new cash to the system. Turnover generates volume by itself. On single-match markets — the highest-variance, shortest-life product available — that turnover is the main engine, not a side effect.

Real money flowing into the esports ecosystem from this story is, so far, approximately zero. No integrity fee is shared with clubs. No data-licensing payment goes to teams. No revenue share exists. The only relationship named is that Kalshi supplies data for organisers, publishers and sponsors to place alongside viewership numbers. Supplying data and sharing revenue are two different things, and the original report only describes the first.

The real value is optionality, not cash. If prediction-market volume can validate attention and thereby support sponsorship pricing, media-rights negotiation and slot valuation, the benefit flows to whoever sells attention — organisers, publishers, and indirectly clubs. That is a plausible chain, but it remains a chain of assumptions.

The model the original piece offers is that volume spikes around the biggest events, using the FIFA World Cup as the demonstrated comparator. If that model holds, Q3 2026 is not a new baseline. It is a peak quarter containing the Dota 2 season climax, the Valorant Champions window, the CS2 circuit, and a League of Legends Worlds build-up at the tail. Q4 2026 therefore becomes the decisive test: a fall back toward the Q1–Q2 trend line confirms event-drivenness, while holding near $967 million confirms structural growth.

There is a systematic title-region mismatch. League of Legends and Dota 2 draw their largest audiences from China, the CIS and Korea — places a US-facing prediction market barely reaches. CS2 and Valorant have comparatively stronger North American participation. A US-centric volume series therefore likely over-weights CS2 and Valorant against their global audience share and under-weights League of Legends and Dota 2, inverting the global viewership ranking.

Spectators watch the scoreline. I watch how they tie their laces before kickoff. Here, the laces are that the market is counted per match, not per tournament.

What the report never tests

The structural claim — strong underlying growth plus event spikes on top — is asserted, not tested. The only supporting evidence is a soccer analogy. There is no esports event-level series showing that a Dota 2 major or a Champions final produces a matching volume spike. The analogy may be right. It has not been demonstrated, and a data analysis should be read at exactly that level of confidence.

The phrase 'climbed almost every month' implies at least one down month already occurred, and the original piece does not identify which month or why. A down month inside a twelve-month series for a new product is the most newsworthy detail available, because it is the only seed of seasonality. I once held a story for six months because nobody was ready to hear it, so I understand holding a detail that has not ripened. But there is a difference between waiting for a subject to be ready and simply forgetting a falsification test.

Choosing single-match markets as the tracked unit is not neutral. A one-map market in a short series carries far higher variance than a long series, and variance is the raw material of a trading product. The dataset is built around the highest-variance, highest-churn product on offer. Anyone using this series to measure esports attention is measuring a deliberately selected slice of attention — a slice that leans toward traders rather than spectators.

The biggest hole is the absence of a per-title breakdown. The reader cannot tell whether $1.73 billion is a CS2 phenomenon, a League of Legends phenomenon, or a diversified basket. For a report aimed at publishers and tournament organisers — parties that need to know exactly which title generates attention flow — that omission matters more than any rounding error.

Integrity risk concentrates at the lower tiers, and the original piece never mentions event tier. Single-match markets at tier-two and tier-three events carry liquidity thin enough that a modest stake moves the price. CS2 has a documented public enforcement history on match-fixing in semi-professional tiers — precisely where liquidity is thin and player incomes are low. A per-match trading product creates a per-match, per-player financial incentive structure.

Roster-level information asymmetry is unaddressed. Substitutions, stand-ins and last-minute swaps are the highest-value non-public information in esports. A market pricing individual matches creates direct value for anyone holding that information before announcement. The report says nothing about information-handling rules, and that silence is more notable than any growth line.

Finally, there is a jurisdiction gap. Riot Games and Valve hold unilateral integrity authority over their own ecosystems. No mechanism is described that extends that authority to third-party trading venues. A publisher can ban a player, but it has no equivalent instrument against a venue operating outside its ecosystem. The gap is not hypothetical — it is the direct consequence of a US venue reporting volume across four titles it does not govern.

What I will wait for

The Q4 2026 print is the only test that can adjudicate the growth thesis. I will wait for it, and I will read it the way I read a team's form table: not one match, four consecutive matches.

Three other things are worth watching. Whether Kalshi publishes a per-title breakdown — without it, every claim about individual disciplines is speculation. Whether publishers write integrity rules covering third-party venues. And whether clubs ever receive a share of this value chain; so far, the answer is no.

My job is to keep the beat so others can walk in step. That beat is not a $1.73 billion headline; it is four uneven beats added together, the last of which weighs more than half the song. I write slowly, because I believe the ball never needs to be rushed.

What matters next quarter is not whether volume clears $1.73 billion. It is whether people start reading this data as a structured time series rather than a poster — and whether the people who produce that attention, from tier-two players to third-tier clubs, are counted in whatever sum comes next.