A handful of repeating trade sizes accounted for more than half the value changing hands on Kalshi bitcoin and ether perpetual futures, in a CoinDesk review of trade-level data covering September 17 to September 20. On the ether book, a single size near $5,499 represented $7.7 million of $13.5 million sampled, or 57% of the total. The bitcoin book showed the same shape, with two sizes at $2,500 and $5,000 making up 54% of an $8.5 million sample. The exchange responded that the pattern comes from one market maker running a fixed-size order programme, with hundreds of distinct traders on the other side. For an allocator reading exchange volume as a liquidity signal, the episode is a reminder that headline notional and tradeable depth are not the same number.
The Read
- One trade size made up 57% of sampled ether perpetual volume over four days.
- Recurring fixed-dollar trades appeared on 43 of 46 sampled dates since June.
- Kalshi attributes the pattern to a single market maker quoting fixed size.
Get This Analysis Summarized by AI
ChatGPTWhat the Kalshi Trade-Level Sample Actually Shows
The Kalshi finding rests on clustering rather than on any single print. Across the September 17 to September 20 window, the dominant ether ticket sat between $5,499 and $5,500 before shifting to roughly $5,425, and the dollar value stayed nearly fixed while the contract count moved with price.
That signature points to automated execution of a predetermined notional. A discretionary trader sizes in contracts and lets the dollar value float. A fixed dollar amount with a moving contract count is machine behaviour, which narrows the range of plausible explanations considerably.
The pattern runs well beyond one week on Kalshi. Recurring fixed-dollar clustering showed up on 43 of 46 sampled one-hour windows between June 19 and September 20, with the dominant size carrying roughly 45% of value across those samples.
One number frames the rest. At a point on September 21, the Kalshi book printed $612.5 million of 24-hour notional against $9.7 million of open interest, a ratio near 63 times. High turnover relative to open interest is normal in perpetuals, but that multiple sits far outside what a positioning-driven book produces.
Market Making or Wash Trading on Kalshi, and Why the Distinction Is Thin
Kalshi set out its own account of the flow in a post detailing how its liquidity programme pays market makers to hold fixed size on the book, describing repeated sizes that originate from one market maker holding fixed liquidity, with hundreds of separate traders taking the opposite side. On that description the volume is genuine: real risk changed hands, and the taker side was diverse.
The critique runs on a different axis. Takers were consistently faster and profitable while the market maker repeatedly traded at a disadvantage as prices moved elsewhere, which is what adverse selection looks like when a passive quoter is slower than the market it quotes into.
Here is why the label matters less than the structure. Whether the flow is called market making or something harsher, a book where one participant sits on half the volume has one point of failure. If that quoter widens or steps away, the visible liquidity vanishes from the book rather than reappearing on another venue.
There is also a measurement problem for anyone building a venue scorecard. Most liquidity screens rank exchanges on reported notional, a field that treats a thousand independent tickets and a thousand copies of the same ticket as identical. Trade-size dispersion is the cheap correction, and almost nobody runs it before allocating flow.
The concentration question is not specific to prediction venues. Crypto derivatives as a whole passed through a thin stretch this year, with monthly volume hitting a twelve-month low at $2.9 trillion in May, and thin markets are exactly where a single programme can dominate a tape.
Also on CFinance:
- Circle Opens USDC Borrowing Against Bitcoin to Institutions
- Saylor’s Strategy Resumes Buying With 950 Bitcoin
- Bitcoin ETF Holders Back in Profit as Inflows Near $1B
The Case for Reading This as Growing Pains
A young Kalshi contract needs a committed quoter, and that is the ordinary path. Listed products routinely launch with one designated market maker carrying most of the two-sided flow until natural interest builds, and the arrangement is disclosed rather than hidden.
The taker diversity supports that reading. Hundreds of distinct counterparties trading against one quoter describes a market finding its users. A closed loop recycling the same position between two accounts would show a far narrower set of participants.
Timing also sits inside a regulatory push toward exactly this kind of product. The CFTC chair told a Treasury market conference this week that regulators must prepare for mass tokenization and round-the-clock trading, which places venue experimentation on the encouraged side of the ledger.
Prediction venues have been building institutional credibility for a while on the back of that framing, to the point where their pricing now gets quoted as a market view, as when Clarity Act passage odds slipped to 16% on prediction markets.
What Breaks If the Quoter Steps Back
The Kalshi bear case is operational before it is regulatory. An institution sizing a hedge off a $612.5 million daily notional is implicitly assuming it can exit into that depth. Remove the fixed-size programme and the executable size collapses toward the $9.7 million of open interest, which is a different market entirely.
Position limits make that worse rather than better. A desk that sized its book against visible depth cannot scale out gradually if the quoter leaves, because the remaining participants were takers by construction and have no obligation to bid. Exits in that configuration happen in one move, at whatever price the remaining bids allow.
The regulatory leg then compounds it. The CFTC is already reviewing the activity on the Kalshi ether contract, and the same division issued an advisory this week flagging heightened manipulation risk on a neighbouring product line. Two open files on one venue rarely resolve in parallel.
Reputational spillover across the category is the other side of this. Ether perpetuals on a regulated US venue such as Kalshi were meant to be the clean instrument, the one a compliance team could approve without a memo about offshore counterparties. A volume dispute in the first months of trading pushes that approval further out.
Asymmetry here is straightforward to price. Confirmation that the flow is ordinary market making changes very little, because the concentration stays regardless. A finding that goes the other way, or a quoter that simply withdraws, removes the liquidity case for the contract in a single session. Institutional interest in ether exposure keeps building through regulated wrappers, including the moment Fidelity filed to stake its $898M Ethereum ETF, and those vehicles are where the volume question gets answered with audited numbers instead of sampled tape.
More to come.
About the Author
