[Essay 01] ·

Notes from the desk

Notes on liquidity fragmentation across digital-asset venues

Digital-asset liquidity does not sit in one place. It is split across dozens of venues whose books overlap on the same instruments but differ in fee schedules, matching engines, tick structures, listed pairs, and the composition of participants who choose to quote there. Two venues advertising the same instrument are, for all practical purposes, two related but distinct markets.

Price discovery in such a landscape is a distributed process. There is no canonical mid; the best estimate of fair value is a weighted aggregate, with weights that depend on the depth, latency, and quality of each contributing venue. Per-venue mids drift around that aggregate on their own clocks, occasionally leading and occasionally lagging, in patterns that shift with time of day and with whichever participants happen to be active. The aggregate itself is an estimate, revised continuously as new prints arrive.

The implications for systematic liquidity provision follow naturally. Quoting on every venue reachable is rarely the right answer, because each venue carries its own adverse-selection profile and its own cost of being on the wrong side of a faster participant. The defensible posture is to quote where there is a structural reason to expect an information edge, and to hedge or warehouse the resulting exposure on the deeper, more correlated venues where flow is less idiosyncratic.

In that framing, the aggregate fair value is an input to the quoting decision, not the price at which any single venue trades. Treating it as the latter is one of the more common ways to lose money slowly in fragmented markets.

← Back to Insights