One clip size is more than half of Kalshi's ether tape

CoinDesk spent four days reading Kalshi's public trade records, and what carried the ether perpetual was one dollar figure, a concentration that tells you how orders arrive rather than how many hands are there to receive them. Trades valued within $2 of $5,499 accounted for $7.7 million of the $13.5 million in ether transactions sampled between Sept. 17 and Sept. 20, 57% of everything that traded. On the bitcoin perpetual, two recurring sizes worth about $2,500 and $5,000 made up 54% of the $8.5 million sampled over the same four days, so the pattern holds beyond one product.

The behavior of the clip is worth studying because ether climbed from around $1,700 to $2,500 between June and September, and as it did the number of contracts needed to reach the target changed while the dollar value of each trade held nearly steady; an order sized in contracts would have left a drifting dollar value on the tape, and this one left a fixed one. CoinDesk reads the pattern as consistent with one or more automated programs executing predetermined dollar amounts — clips, in trader shorthand.

Four days is a sample, so CoinDesk widened the window, and in 43 of 46 one-hour ether samples taken between June 19 and Sept. 20, trades clustered around recurring dollar targets; the prevailing size accounted for about 45% of value across that stretch and more than half the value on 15 dates. Inside the shorter window, 1,406 of the 3,450 ether trades analyzed, drawn from 23 one-hour samples and the exchange's public API datasets, landed within $2 of $5,499, which is roughly 41% of the count against 57% of the value and puts the recurring ticket above the venue's average.

Volume is the first metric a trader reaches for when asking whether a market can absorb an order without moving the price, and it is the number a young venue leads with, which is why the composition of a tape matters as much as its size. CoinDesk's own framing is careful on this point: volume alone does not reveal how many different traders generated the activity, and the recurring sizes mean the answer for Kalshi's perpetuals is not in the headline number.

A fixed-dollar clip is what a market maker's execution engine produces, and it is also what a points or incentive program produces; both send the same instruction and leave the same footprint, and nothing in the sampled records separates them. CoinDesk argued on Sept. 21 that an incentive budget, not organic demand, wrote the ether tape; the bitcoin leg carrying a comparable signature widens that argument past a single product, though it does not settle it either way. The number the data cannot produce is the one that would: how many distinct accounts stand behind the recurring ticket.

An institution reading the print should treat it the way CoinDesk treats a stock's daily volume — a starting point that says nothing on its own about how many buyers and sellers sit on the other side. Perpetual contracts track an asset without expiring, and Kalshi, a CFTC-regulated U.S. derivatives exchange best known for prediction markets, added bitcoin perpetual futures in late May and divides the exposure into small contracts that traded near $2.70 apiece on Monday; the notional on the screen is the venue's advertisement, while the counterparty count is the venue's depth, and only one of the two appears anywhere in the sample.

The commercial stakes are not abstract. Venue selection in institutional digital assets is being sorted on questions of this kind — what a print reflects, how a venue earns its liquidity, whether the number a desk reports is a number a desk can trade. Kalshi holds a federal license, and a federal license is the scarce input in this market; a volume print is not.

Kalshi's durable asset is the venue license that lets the perpetual book exist rather than the book itself, and that license became harder to replicate when the Clarity Act died at 49-50 and market-structure definitions moved to the SEC and CFTC. A venue operating inside that perimeter holds something most offshore perpetual venues cannot buy, and it does not need a tape in which one clip size is more than half the sampled value to prove anything. Leaning on that tape converts a regulatory advantage into a volume claim the data will not carry, and perpetual volume is cheap to generate at scale because a single program can generate all of it. The right thing to ask a venue selling perpetual access to institutions is a counterparty count, not a notional.

Nothing in the sampled records alleges an improper trade, and the coverage does not allege one. What the records show is that a print is a claim about liquidity, and claims of that kind get tested the first time an allocator routes real size. CoinDesk found recurring dollar targets in 43 of the 46 one-hour samples it examined from June into September, and the venue's clearest route to institutional perpetual flow runs through the federal perimeter it already holds rather than through a volume number one program can write. The next desk that looks at Kalshi's perpetuals will ask who is sending the clips, and that answer, rather than the tape, is what decides whether the book is worth routing to.