Repeating fixed-dollar trades made up most of Kalshi's crypto perp volume, CoinDesk data shows

A single recurring trade size accounted for the majority of activity on Kalshi's ether perpetual-futures market, and two recurring sizes did the same on bitcoin, according to a CoinDesk analysis of the exchange's own public trade API. The pattern is consistent with automated trading hitting a fixed dollar target — and it raises the question of how many distinct participants Kalshi's reported crypto volume actually represents.

On ether, trades valued within $2 of $5,499 accounted for $7.7 million, or 57%, of the $13.5 million CoinDesk sampled from Sept. 17–20. On bitcoin, recurring sizes of about $2,500 and $5,000 made up 54% of an $8.5 million sample over the same window. CoinDesk examined 3,450 ether-perp trades across 23 one-hour samples; 1,406 landed near $5,499.

The behavior predates the four-day snapshot. Across 46 one-hour samples between June 19 and Sept. 20, ether trades clustered at recurring dollar targets in 43 of them, per CoinDesk. As ether rose from roughly $1,700 to $2,500, the number of contracts per trade shifted while the dollar value stayed near-fixed — a July cluster held ~2,800 contracts versus ~2,200 in September. The target itself migrated over time: near $4,999 early on, $9,999 (72% of value) on June 28, then $3,999 from Aug. 10, $4,499 from Aug. 18 and $5,499 from Aug. 24. On bitcoin, the larger of the two recurring trades was almost exactly double the smaller in every sample containing both.

Kalshi, a CFTC-regulated exchange best known for prediction markets, added bitcoin perpetual futures in late May. Its ether contract showed a 24-hour volume-to-open-interest ratio of 61 on Monday — second-highest of 20 Kalshi perp markets against a median of about eight; bitcoin's was 26. CoinDesk notes high turnover alone doesn't indicate wrongdoing.

Key facts

• $5,499-sized ether trades: $7.7M / 57% of a $13.5M sample, Sept. 17–20 (CoinDesk).
• Bitcoin ~$2,500 and ~$5,000 sizes: 54% of an $8.5M sample (CoinDesk).
• Recurring targets appeared in 43 of 46 hourly samples, June 19–Sept. 20 (CoinDesk).
• A CFTC-filed rebate program cutting fees to 0.003% took effect Sept. 16 (CFTC filing, via CoinDesk).
• Kalshi did not respond to CoinDesk's questions by press time.

The real-world read

This is one outlet's read of public order-book data, and CoinDesk is careful to say it: the records don't name traders, and turnover this heavy doesn't by itself prove anything improper. But note the timeline. The rebate program that made trading nearly free landed Sept. 16 — a day before the four-day sample, but almost a month after the $5,499 trades first appeared, so it can't explain the pattern's origin, only sweeten its later economics. Pseudonymous trader "Beni" alleged on X that Kalshi "fakes" its crypto volume. Kalshi's crypto head (posting as IcoBeast) pushed back on one chart Beni used — saying it covered prediction markets, not perps — and said Kalshi pays no rebates on prediction markets. Notably, that response did not identify who is behind the repeating perp trades or why the fixed target keeps moving. Those questions remain open.

This is news coverage, not financial advice.

Sources

• CoinDesk (Shaurya Malwa, ed. Aoyon Ashraf), Sept. 22 — primary analysis of Kalshi's public trade API, the trade-size clustering, volume-to-open-interest ratios, and the Beni/IcoBeast exchange. Secondary reporting; CoinDesk performed and published its own dataset.
• CFTC filing (rules09022622952, effective Sept. 16), cited by CoinDesk — the 0.003% fee/rebate program.
• Kalshi public trade API and CFTC registration, via CoinDesk — contract pricing, launch date, and the underlying trade records.