Bitcoin, Ether Perpetual Trading on Kalshi Shows Unusual Repetitive Activity

CoinDesk identified a recurring $5,499 trade size that represented 57% of the sampled ether-perpetual volume. In bitcoin perpetuals, recurring trades of approximately $2,500 and $5,000 together represented 54% of the sampled volume.

A CoinDesk review of Kalshi’s publicly available trade data found that a relatively small number of repeated dollar amounts made up more than half of the trading value in its bitcoin and ether perpetual-futures markets.

For ether, trades falling within $2 of the $5,499 level represented $7.7 million, or 57%, of the $13.5 million in transactions examined between Sept. 17 and Sept. 20. Bitcoin showed a similar concentration, with trades around $2,500 and $5,000 accounting for 54% of the $8.5 million sampled during the same period.

Trading volume is among the primary measures used to assess activity and liquidity in a market. Higher volume generally suggests that traders can enter or exit positions without having a major impact on prices.

Large volumes can also create the impression that a market has broad participation. When a significant portion of that activity repeatedly comes from the same dollar amounts, however, identifying what is generating those trades becomes important when interpreting the underlying market activity.

For instance, millions of dollars in daily trading volume for a publicly listed stock could suggest that numerous buyers and sellers are actively participating. But the volume figure itself does not show how many independent traders generated it.

The same issue applies to Kalshi’s perpetual markets. A considerable portion of the observed volume was concentrated in a recurring group of trade sizes, raising questions about the mechanism or participants behind the activity.

Pattern Extends Beyond Four Days

The repeated trade sizes were not limited to the Sept. 17-20 period.

CoinDesk reviewed 46 hourly samples between June 19 and Sept. 20 and found recurring dollar-value clusters in 43 of them for ether. The dominant trade size represented roughly 45% of the traded value across those samples and exceeded 50% on 15 separate dates.

While the number of contracts changed as ether’s price moved, the dollar value associated with the recurring trades remained relatively consistent. That behavior resembles automated trading systems programmed to execute trades at predetermined dollar values, often referred to by traders as “clips.”

Kalshi is a U.S.-based derivatives exchange regulated by the Commodity Futures Trading Commission and is primarily known for its prediction markets. The platform introduced bitcoin perpetual futures, which track the underlying asset without an expiration date, in late May.

What the Trading Data Reveals

Kalshi divides its perpetual exposure into relatively small contracts, which were trading at around $2.70 each on Monday.

For its analysis, CoinDesk examined 3,450 ether-perpetual transactions across 23 one-hour samples during the four-day period using Kalshi’s publicly available API data. A total of 1,406 trades fell within $2 of the $5,499 target.

The dollar target remained remarkably consistent even as the number of contracts required to reach it changed with ether’s price. ETH climbed from approximately $1,700 to $2,500 between June and September, meaning traders needed fewer contracts to reach the same notional amount.

For example, a trade in one July cluster contained roughly 2,800 contracts, while a comparable September trade involved about 2,200 contracts.

The recurring dollar target also changed over time.

Earlier CoinDesk samples showed activity centered near $4,999. On June 28, trades close to $9,999 represented 72% of the sampled value. A $3,999 target appeared on Aug. 10, followed by approximately $4,499 on Aug. 18 and $5,499 on Aug. 24.

On June 19, about three weeks after Kalshi launched its cryptocurrency perpetual futures, trades worth almost exactly $4,999 represented 37% of the ether contract value in the hourly sample examined by CoinDesk.

Bitcoin displayed a comparable structure.

Two recurring trade sizes moved together as bitcoin’s price changed, with the larger position consistently close to twice the smaller one. In nine of the 22 samples containing both sizes, the larger trade was exactly double the smaller trade. In the remaining 13 samples, it was one contract above twice the smaller amount, a difference that can be explained by rounding.

When bitcoin traded around $76,300, the two recurring positions contained 327 and 655 contracts. By Monday, the corresponding sizes had shifted to 307 and 614 contracts.

Kalshi’s ether perpetual also recorded unusually high turnover compared with the amount of open interest remaining in the market.

A Monday snapshot showed roughly 93 million ether-perpetual contracts in 24-hour volume against approximately 1.5 million contracts of open interest, which represents outstanding positions. That produced a volume-to-open-interest ratio of about 61, meaning around 61 contracts changed hands for every contract that remained open.

Among Kalshi’s 20 perpetual markets with open interest, that was the second-highest ratio. The median was approximately eight. Bitcoin’s corresponding ratio was 26.

High turnover alone, however, does not establish that trading activity is improper.

CoinDesk contacted Kalshi to ask whether the recurring bitcoin and ether trade sizes were generated by one participant or multiple participants, whether any of the activity was connected to market-making or incentive programs, and whether the exchange had identified self-matching or common ownership between accounts. Kalshi had not responded by publication time.

Recurring trades based on fixed dollar amounts are commonly associated with algorithmic strategies that adjust contract quantities as prices change. Similar dynamic position-sizing techniques are discussed in quantitative trading research, including the work of Cartea, Jaimungal and Ricci on algorithmic and high-frequency trading.

Automated systems can continuously adjust quotes and position sizes in response to market movements, helping traders manage exposure and hedge against adverse price changes. The principle is also reflected in established market-making models such as the Avellaneda-Stoikov framework.

For Kalshi, the progression from roughly $4,999 to $3,999, $4,499 and eventually $5,499 indicates that the notional parameters used by a trading strategy may have been adjusted periodically.

Public order-book and trade data, however, cannot determine whether the activity represented ordinary algorithmic trading, incentive-driven activity or another form of market behavior.

Questions Around Incentives

Trading costs for certain firms that directly settled transactions with Kalshi became significantly lower shortly before CoinDesk’s four-day sample began.

A rebate program filed with the CFTC took effect on Sept. 16. It reduced fees for participating firms to 0.003% while providing market makers with a rebate at the same rate.

The program began one day before the Sept. 17-20 sample and nearly a month after the $5,499 trading pattern first appeared. As a result, it does not explain when the recurring trade sizes initially emerged, although it may have influenced the economics of trading later in the period.

Pseudonymous trader “Beni” highlighted the repeated trade sizes on X over the weekend and alleged that Kalshi was overstating its cryptocurrency trading volume.

Kalshi’s crypto chief, who uses the pseudonym IcoBeast, challenged part of Beni’s analysis. He said a volume-share chart referenced by Beni related to prediction markets rather than perpetual futures.

He also said Kalshi does not offer rebates on its crypto prediction markets and that incentives offered through its regulated exchange must be disclosed publicly.

The response did not specify which participants generated the recurring ether-perpetual trades or explain why the fixed dollar amounts shifted from one target to another over time.

About The Author