
Grid Step Calculator: Every Grid Step Has a Cost
A free grid step calculator: set the range, levels, investment and fee, and see the step, capital per level, net profit per cycle and how many levels your range can take before fees eat every trade.
Last updated: October 6, 2026.
The grid step is the price range divided by the number of levels. Each completed buy-sell cycle earns that step minus two trading fees. When the step shrinks below the round-trip fee, every cycle loses money: on a $82,000–$88,000 BTC range with 0.1% fees per side, that happens at about 34 levels. To keep fees under a third of each cycle, make the step at least three times the round-trip fee.
Enter your own range, levels, investment and fee. The numbers update as you type.

What each number means
| Output | What it tells you |
|---|---|
| Grid step | Distance between two levels, in dollars (arithmetic) or percent (geometric) |
| Step, % of price | The same step as a share of price. It's smallest at the top of the range, so the calculator uses that for the profit math |
| Capital per level | Investment divided by levels: how much each order trades |
| Fees per cycle | One buy plus one sell, at your fee per side |
| Net profit per cycle | Step % minus fees, and the same in dollars for one level |
| Fees eat of gross | What share of each cycle's gross profit goes to the exchange |
The verdict line turns green when fees take less than half of each cycle, yellow between half and all of it, and red when a cycle loses money.
The formulas behind it
- Step: step = (upper price − lower price) ÷ levels
- Step in percent: step % = step ÷ (upper price − step), the smallest step in the range
- Net per cycle: net % = step % − 2 × fee per side
- Net in dollars: net $ = (investment ÷ levels) × net %
- Break-even levels: max levels = range ÷ (2 × fee × upper price)
For a geometric grid, each level sits the same percentage above the last: ratio = (upper ÷ lower)^(1 ÷ levels), and step % = ratio − 1.
One BTC range, six grid counts
$1,000 on BTC/USDT, range $82,000–$88,000, 0.1% maker fee per side.
| Levels | Step | Step % | Net per cycle | Net per cycle, $ | Fees take |
|---|---|---|---|---|---|
| 10 | $600 | 0.69% | +0.49% | +$0.49 | 29% |
| 20 | $300 | 0.34% | +0.14% | +$0.07 | 58% |
| 30 | $200 | 0.23% | +0.03% | +$0.01 | 88% |
| 34 | $176 | 0.20% | 0.00% | $0.00 | 100% |
| 50 | $120 | 0.14% | −0.06% | −$0.01 | 146% |
| 100 | $60 | 0.07% | −0.13% | −$0.01 | 293% |
More levels mean more fills, and that's why dense grids look busy on the chart. Past about 34 levels, each of those fills costs more than it earns. A grid with 100 levels on this range would trade constantly and lose a little every time.
The counterweight is fill frequency. A wide step fills less often, so 10 levels earn more per cycle but complete fewer cycles than 20. The calculator answers "is each cycle worth it." How often price crosses your levels is what backtesting and demo answer.
Your exchange's fee changes the answer
Same range, same $1,000. Only the maker fee changes.
| Venue, base tier | Maker fee per side | Break-even levels | Levels with fees under a third |
|---|---|---|---|
| Bybit spot, Binance spot without BNB | 0.10% | ~34 | ~11 |
| OKX spot | 0.08% | ~42 | ~14 |
| Binance spot, fees in BNB | 0.075% | ~45 | ~15 |
| Perpetual futures, all three | 0.02% | ~170 | ~56 |
A quarter less in fees lets the same range carry about a third more levels. Bybit, Binance, OKX: Where Your Bot Pays Less has the full fee comparison.
Arithmetic or geometric spacing
Arithmetic grids space levels by the same dollar amount. On a narrow range the step barely changes as a percentage, so it's the simpler choice.
Geometric grids space levels by the same percentage. On a wide range, an equal dollar step becomes tiny in percent near the top and large near the bottom. A $500 step is 0.5% at $100,000 and 1% at $50,000. Geometric spacing keeps every cycle's profit the same, which matters once the range is wider than about 20%.
Switch the "Spacing" field in the calculator to compare both on your range.
A rule of thumb that holds up
Keep the step at least three times the round-trip fee. At 0.1% per side, that's a step of 0.6% or more, and fees take no more than a third of each cycle. Below 2× the fee, half of every cycle goes to the exchange. At 1×, the bot works for nothing.
Then check the step against how the coin actually moves. A step much smaller than the typical hourly candle trades noise. A step much larger than the typical daily move sits idle. How Many Grids Should a Grid Bot Use? covers matching the step to volatility.
What the calculator leaves out
- Spread and slippage. Limit orders on a liquid pair rarely slip, but a wide spread on a thin pair adds to every cycle's cost.
- How often price crosses each level. That depends on the market, not the math.
- Unrealized PnL. A grid can complete profitable cycles while the coin it holds loses value. Bot profit vs portfolio profit explains the gap.
- Taker fees. Market entries and stop losses pay the taker rate.
Found a grid that clears its fees? Backtest it on real price history before it trades real money.
Frequently asked questions
How do you calculate the grid step? Divide the price range by the number of grid levels. A $82,000 to $88,000 range with 30 levels has a step of $200, about 0.23% of price at the top of the range.
How many grid levels is too many? When the step falls below the round-trip fee, every cycle loses money. At 0.1% maker fees per side, a $82,000 to $88,000 BTC range breaks even at about 34 levels. To keep fees under a third of each cycle, the step needs to be at least three times the round-trip fee.
How much does a grid bot make per grid? Net profit per completed cycle equals capital per level times (step percent minus two times the fee). With $1,000 over 10 levels on a $6,000 BTC range and 0.1% fees, that's about 0.49%, or $0.49 per cycle.
Arithmetic or geometric grid: which is better? Arithmetic grids use equal dollar steps and suit narrow ranges. Geometric grids use equal percentage steps and suit wide ranges, where an equal dollar step would be tiny in percent at the top and large at the bottom.
Do lower fees allow more grid levels? Yes. On the same BTC range, break-even rises from about 34 levels at 0.10% maker fees to 42 at 0.08% and 45 at 0.075%. On perpetual futures at 0.02% maker, it's about 170.