Kalshi’s Crypto Perpetuals Face ‘Fake Volume’ Questions — What Wash-Trading Risk Means for Cryptocurrency Exchange Development
Quick Answer
A CoinDesk analysis published this week found that a single repeating trade size near $5,499 accounted for 57% of sampled Ethereum perpetual-futures volume on Kalshi, a CFTC-regulated exchange, with a similar pattern on Bitcoin perpetuals. A separate researcher flagged a 174-to-1 ratio between 24-hour trading volume and open interest on the ETH contract — a pattern typically associated with wash trading — and tied it to a rebate program Kalshi filed with the CFTC on September 16. Kalshi has pushed back on parts of the analysis but hasn’t fully explained the repeating trade sizes. For anyone working on Cryptocurrency Exchange Development, it’s a real-time case study in how rebate program design can accidentally create incentives for manufactured volume.
What the Data Actually Shows
CoinDesk analyzed 3,450 Ethereum perpetual-futures trades on Kalshi across 23 one-hour samples between September 17 and 20, pulling from the exchange’s public API. Of those trades, 1,406 landed within $2 of $5,499 — a single repeating trade size accounting for $7.7 million, or 57%, of the $13.5 million in transactions reviewed. Bitcoin perpetuals showed a similar pattern, with recurring trades near $2,500 and $5,000 making up 54% of sampled volume.
This kind of publicly reproducible analysis is exactly why volume integrity has become such a central concern in Cryptocurrency Exchange Development. Anyone with access to an exchange’s public API can run the same kind of trade-size clustering check CoinDesk did, which means suspicious patterns rarely stay hidden for long once a platform reaches meaningful scale.
This wasn’t a one-off finding. Looking back across 46 hourly samples between June 19 and September 20, 43 showed trades clustering repeatedly around specific dollar targets, averaging about 45% of sampled value and exceeding half the traded value on 15 separate dates. The target dollar amount itself shifted over time — $9,999 in late June, $4,999 and $3,999 at various points, settling near $5,499 by September — while the number of contracts adjusted to keep the dollar value roughly constant as Ethereum’s price moved from around $1,700 to $2,500 over the same period. That pattern is consistent with an automated trading program executing fixed-dollar orders, something traders commonly call “clips.”
The Wash-Trading Allegation Goes Further
Separately, a researcher going by Beni, co-founder of research firm Stealth Neolab, raised a more pointed concern on September 20: Kalshi’s Ethereum perpetual contract logged roughly $539 million in 24-hour trading volume against just $3.1 million in open interest — a 174-to-1 ratio. That kind of gap between reported volume and actual outstanding positions is widely treated as a textbook signal of wash trading, where the same participant effectively trades against themselves to inflate reported activity without changing real market exposure.
Beni connected this to a CFTC-certified rule filing Kalshi made on September 16, updating a temporary rebate program for its crypto perpetuals: market makers earn 0.3 basis points on trades, while takers pay 0.3 basis points, a structure designed to roughly net to zero. His argument was straightforward — under that fee structure, an entity trading against its own orders repeatedly costs very little, making it economically feasible to manufacture large volume figures at minimal expense.
Kalshi didn’t stay silent. A staffer working on the exchange’s crypto products responded publicly the same day, pointing out that part of the original comparison had conflated a chart measuring prediction-market share with actual perpetual-contract volume, and explaining that Kalshi’s volume reporting follows the same convention as Polymarket — counting the maximum potential payout of a contract rather than the actual cash traders spent. That explains part of why headline volume figures look inflated relative to cash outlay, but it doesn’t, on its own, account for the specific pattern of repeating trade sizes CoinDesk identified.
Why This Matters Beyond Kalshi
This episode is a useful case study for anyone working on Cryptocurrency Exchange Development, and not because it definitively proves wrongdoing — Kalshi’s public data doesn’t identify the traders involved or establish that any rule was broken. What it demonstrates clearly is how a seemingly reasonable maker-taker rebate structure can create an unintended incentive for manufactured volume if the numbers aren’t stress-tested against that specific risk before launch. A near-zero net cost for trading against yourself is exactly the kind of loophole that sophisticated participants, whether market makers or automated bots, will find and exploit if it exists.
Join The Writer's Circle event
It’s also a reminder that reported trading volume is one of the most heavily scrutinized, and most easily gamed, metrics in this industry. Volume drives an exchange’s perceived liquidity, its ranking on aggregator sites, and ultimately its credibility with both regulators and users. A pattern like this — however it’s eventually explained — invites exactly the kind of public scrutiny that damages trust even before any formal finding is made.
What This Means for Exchange Operators
Stress-test rebate and incentive programs against self-trading scenarios before launch, modeling what it would cost a sophisticated participant to manufacture volume under the exact fee structure being proposed.
Monitor the ratio between reported volume and open interest as an ongoing internal health metric, not just something external researchers check after the fact.
Watch for repeating trade-size clustering in your own market data, since that pattern is visible in public trade records and will eventually be noticed by outside analysts if it exists.
Be transparent about volume reporting conventions from the start, since ambiguity between cash-value and maximum-payout reporting creates exactly the kind of confusion that fuels public disputes like this one.
Treat market-microstructure review as a required step in Cryptocurrency Exchange Development, not an optional refinement — rebate programs, matching-engine behavior, and fee structures all interact in ways that are hard to predict without deliberate modeling.
The Regulatory Backdrop Makes This Riskier
This scrutiny lands at a particularly sensitive moment for Kalshi and its prediction-market peers. On September 10, the EU’s top markets regulator, ESMA, warned that Kalshi and Polymarket — now valued in the tens of billions of dollars combined — hold no EU authorization anywhere in the bloc. Combined with the CFTC’s own recent advisory flagging risks in behavior-based “mention markets,” the regulatory spotlight on this category is already unusually bright. A public wash-trading dispute, even one that ultimately gets resolved as a misunderstanding, adds friction at exactly the wrong time for a sector working to establish institutional credibility, and reinforces why market-integrity safeguards need to be built into Cryptocurrency Exchange Development as core infrastructure rather than treated as a reputational afterthought.
Frequently Asked Questions
What did CoinDesk’s analysis find?
That a single repeating trade size near $5,499 accounted for 57% of sampled Ethereum perpetual-futures volume on Kalshi, with a similar pattern of recurring trade sizes making up 54% of sampled Bitcoin perpetual volume, consistent with automated fixed-dollar trading.
Is this proof of wash trading?
No. Kalshi’s public data doesn’t identify the traders involved or establish wrongdoing. A separate researcher’s 174-to-1 volume-to-open-interest ratio is a commonly cited red flag for wash trading, but it isn’t conclusive on its own, and Kalshi has disputed parts of the broader analysis.
How could a rebate program create an incentive for wash trading?
If maker and taker fees roughly net to zero, an entity trading against its own orders repeatedly can generate large reported volume figures at very low cost, since the near-zero net fee removes the usual economic disincentive against self-trading.
What should businesses working on Cryptocurrency Exchange Development take from this?
That incentive structures like maker-taker rebates need to be modeled specifically against self-trading and wash-trading scenarios before launch, and that volume-to-open-interest ratios and trade-size clustering should be monitored as ongoing internal health signals.
Has Kalshi fully explained the repeating trade pattern?
Not entirely. Kalshi’s response clarified a comparison error in part of the original allegation and explained its volume-reporting convention, but it hasn’t specifically accounted for why trade sizes cluster so consistently around a shifting but narrow dollar target.
Why does volume integrity matter so much for Cryptocurrency Exchange Development?
Trading volume drives how liquid a platform appears, its ranking on aggregator sites, and its credibility with regulators and users. Inflated or manufactured volume, even if unintentional, undermines the trust that makes an exchange viable long-term.
Final Thoughts
Whatever the eventual explanation for Kalshi’s trade-size pattern turns out to be, the episode is a concrete reminder that market integrity isn’t a box to check once at launch — it’s an ongoing engineering and monitoring discipline. Rebate structures, volume reporting conventions, and matching-engine behavior all interact in ways that create real incentives, intended or not. For anyone serious about Cryptocurrency Exchange Development, building the internal monitoring to catch these patterns before a researcher on social media does is far cheaper than managing the reputational fallout afterward.
Designing a maker-taker rebate program that rewards genuine liquidity without creating a cheap path to manufactured volume takes real market-microstructure expertise. That’s exactly the kind of detail we get right in every Cryptocurrency Exchange Development engagement, so incentive structures strengthen your market instead of becoming a liability the moment a researcher starts digging into your trade data.
Top comments (0)