Grid profitability is driven by spacing geometry, not by predicting direction. For an arithmetic grid with lower bound L, upper bound U, and N levels, the step size is s = (U − L) / N. Each completed round trip (buy at one level, sell at the next) captures roughly s minus fees.

Per-cycle gross profit on a fixed quote allocation Q per level is approximately Q × s / P_buy, where P_buy is the entry price. Because P_buy rises across the grid, the same dollar step yields fewer percentage points at higher prices — a subtle drag on upper-grid efficiency.

Wider spacing → fewer fills, larger profit per fill, more idle capital
Tighter spacing → more fills, thinner margin, higher fee sensitivity
Geometric grids keep constant percentage spacing, arithmetic grids keep constant dollar spacing

Grid type	Spacing	Best when
Arithmetic	Constant $ step	Range-bound, low volatility
Geometric	Constant % step	Trending or wide ranges