Earn the spread on OBX swaps.

Model & assumptions: fills follow a demand curve acceptance = e^−s/s₀ with taker price-tolerance s₀ = 3%, so the profit-optimal spread is s* = s₀. Capital turnover assumes a ~10-min happy-path lockup and a ~20-h lockup on the ~8% of swaps that abort (⇒ ~14×/day, not the naïve 180×). We show net OBX/day as a ±25% band (proceeds are held in the paired coin) rather than an annualized APR, because annualizing daily market-making returns overstates them. The monthly figure is an early-provider rate: it's high because you'd be one of the only makers, and it falls fast as others add depth (raise "competing depth" to see it). And note: one-sided selling depletes your OBX — a sustained daily rate needs two-sided making or continuous replenishment. Capture-share by depth and s₀ are heuristics, not yet calibrated to live fills.