Choosing Grid Count by Backtesting Spacing, Trade Size, and Fees

Summary

This article examines how the number of price levels affects a grid strategy. Fewer grids create wider spacing and larger allocations per trade, but may miss smaller price swings. More grids create more trading opportunities while shrinking each allocation and potentially increasing fee drag. The suggested choice depends on market volatility, account capital, and desired trading frequency.

A sample backtest defines logarithmically spaced price levels between a historical low and high, allocates capital evenly across them, and buys when a bar reaches a level before selling at the next one. It accounts for fees and marks positions to market, then compares net profit and observed profit extremes across several grid counts for one selected crypto pair. The reported result peaks at 15 grids in that particular experiment, then falls as grid count rises. This is a narrow historical example, not a general optimum: the method depends on the selected range and data, and the document does not establish robustness across assets, periods, or execution conditions.

Key ideas

• Grid count sets the spacing between levels and the capital allocated to each level.
• Sparse grids may miss smaller swings, while dense grids can increase turnover and fee costs.
• The example uses logarithmic spacing, equal notional allocation, and bar-based entry and exit checks.
• The backtest includes estimated fees and marks open inventory to market when calculating net profit.
• The tested configuration performed best at 15 grids in the stated sample, which does not establish a broadly optimal setting.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.