A grid bot places buy and sell orders at set levels inside a price range. Each time the price falls to a level it buys, and each time it rises to the next level it sells. It earns from the price moving up and down, so it needs a market that moves sideways.
A worked example
This is a sample with round numbers: a range of $60,000 to $70,000 split into 10 grids, so one level every $1,000.
- Buy at $65,000, sell at $66,000. The move is about 1.54%.
- Take a fee of 0.1% per order as an example. A buy and a sell cost about 0.2%.
- What is left is about 1.34% of the money in that one grid.
Each grid holds only a small part of the budget, so a 1.34% gain on one grid is a small gain on the whole bot. Income depends on how many times the price crosses levels.
Why fees decide it
Narrower grids trade more often and keep less each time. With 0.2% between levels and 0.2% in fees, nothing is left. Work out the profit per grid after fees before you choose the number of grids.
The settings
Binance’s spot grid asks for the same things as most platforms: an upper price, a lower price, the number of grids, arithmetic or geometric spacing, and the investment.
- Arithmetic spacing keeps the same dollar gap between levels.
- Geometric spacing keeps the same percentage gap, which suits wide ranges.
What breaks a grid
The price falls below the range. Binance says orders stop once the price drops under the lower limit. By then the bot has bought all the way down and holds coins worth less than it paid. It’s the same position as buying and holding from a higher price, and the small grid profits rarely cover it.
The price rises above the range. The bot has sold everything and sits in cash while the coin keeps climbing. You don’t lose money, but you can earn far less than holding would have.
A sudden move. Binance warns that grid orders keep executing as levels are reached, even in a rapid collapse or surge, and recommends setting stop-trigger orders in advance.
How to test one
Test over a period when the price left your range, and not only one where it stayed inside. Compare the result with holding the coin over the same dates. A grid that earned 8% while the coin rose 40% was a poor trade, even though it shows a profit.