providing liquidity in a prediction market is the most misread trade in the space.

on screen it looks like market making but in risk and P&L it is the bookmaker's trade.

the LP earns a house edge on ordinary flow, pays adverse selection to informed flow and is forced to carry the irreducible residual to settlement, because a single binary's payoff is not spanned by any liquid hedge.

you don't survive that contract by contract. you price the edge, eat the selection, and underwrite the residual in a broad, imperfectly correlated, well capitalized book.

so the binding constraint on durable depth is loss-absorbing capital reserved against the correlated tail, not spread.

today's bootstrapping playbook is mis-targeted. it scores displayed size and tightness, a spread-capture proxy, instead of capital placed at risk.

it buys depth that is thin and mercenary. gone the moment informed flow arrives.

so the real race was never tighter spreads. it's who can underwrite the tail and still be standing.

that's the prediction market that doesn't thin out the moment it matters.

https://x.com/embrron/status/2067009649946382753