Worked example: the same Bitcoin backtest at 0% and 0.1% fees
Fees are charged on every entry and every exit, so a rule that trades often can turn a profit before fees into a loss after them. This page shows it with real numbers: two presets of the free Hunt4Coin backtester, run on the same 552 Bitcoin 4-hour candles (Binance BTC/USDT, 1 August 2026 to 31 October 2026), at 0% and at 0.1% per side.
The window is in the past, so the candles and the results on this page do not change. Every trade is listed in the ledger (CSV), so each result can be checked by hand. It is an illustration of how fees work, not a strategy recommendation, and past results say nothing about future ones.
Results
| Strategy | Fee per side | Trades | Net result | Fees paid | Win rate | Max drawdown |
|---|---|---|---|---|---|---|
| EMA 20/50 crossover | 0% | 3 | $2,120.38 (+21.20%) | $0.00 | 66.7% | 6.93% |
| EMA 20/50 crossover | 0.1% | 3 | $2,049.90 (+20.50%) | $71.24 | 66.7% | 7.22% |
| MACD crossover | 0% | 17 | $1,726.01 (+17.26%) | $0.00 | 35.3% | 12.48% |
| MACD crossover | 0.1% | 17 | $1,335.18 (+13.35%) | $420.62 | 35.3% | 13.96% |
Starting capital $10,000.00, 100% of the balance in each trade, long only, no stop-loss or take-profit. Buying and holding Bitcoin over the same candles (first close $63,004.48, last close $77,582.02) returned +23.14% before fees.
EMA 20/50 crossover
Rules: buy when EMA(20) crosses above EMA(50); sell when EMA(20) crosses below EMA(50). Signals are checked on each candle's close and filled at the next candle's open.
It made +21.20% before fees and +20.50% after them: still positive, but fees took 0.70 percentage points.
With 3 round trips, a 0.1% fee on each side costs about 0.6% of the traded amount in total, before any compounding. Fees paid at 0.1%: $71.24.
| # | Entry (open) | Entry price | Exit | Exit price | Result at 0% | Result at 0.1% |
|---|---|---|---|---|---|---|
| 1 | 2026-08-18 08:00 UTC | $64,198.97 | 2026-09-10 00:00 UTC | $78,306.43 | +21.97% | +21.75% |
| 2 | 2026-09-18 20:00 UTC | $81,204.77 | 2026-10-08 04:00 UTC | $82,757.00 | +1.91% | +1.71% |
| 3 | 2026-10-21 12:00 UTC | $83,500.01 | 2026-10-23 00:00 UTC | $81,416.09 | −2.50% | −2.69% |
MACD crossover
Rules: buy when the MACD line (12, 26, 9) crosses above its signal line; sell when it crosses below. Signals are checked on each candle's close and filled at the next candle's open.
It made +17.26% before fees and +13.35% after them: still positive, but fees took 3.91 percentage points.
With 17 round trips, a 0.1% fee on each side costs about 3.4% of the traded amount in total, before any compounding. Fees paid at 0.1%: $420.62.
| # | Entry (open) | Entry price | Exit | Exit price | Result at 0% | Result at 0.1% |
|---|---|---|---|---|---|---|
| 1 | 2026-08-15 12:00 UTC | $63,022.01 | 2026-08-22 20:00 UTC | $77,418.26 | +22.84% | +22.62% |
| 2 | 2026-08-31 20:00 UTC | $78,898.06 | 2026-09-01 16:00 UTC | $77,919.07 | −1.24% | −1.44% |
| 3 | 2026-09-03 12:00 UTC | $77,948.39 | 2026-09-05 08:00 UTC | $79,709.96 | +2.26% | +2.06% |
| 4 | 2026-09-12 04:00 UTC | $77,268.09 | 2026-09-15 16:00 UTC | $76,512.01 | −0.98% | −1.18% |
| 5 | 2026-09-17 08:00 UTC | $76,462.77 | 2026-09-20 16:00 UTC | $80,912.00 | +5.82% | +5.61% |
| 6 | 2026-09-21 12:00 UTC | $84,887.19 | 2026-09-23 12:00 UTC | $85,649.96 | +0.90% | +0.70% |
| 7 | 2026-09-27 12:00 UTC | $84,898.00 | 2026-09-27 16:00 UTC | $84,463.38 | −0.51% | −0.71% |
| 8 | 2026-09-27 20:00 UTC | $84,736.02 | 2026-09-28 00:00 UTC | $84,472.01 | −0.31% | −0.51% |
| 9 | 2026-09-30 12:00 UTC | $83,915.17 | 2026-10-03 08:00 UTC | $84,560.00 | +0.77% | +0.57% |
| 10 | 2026-10-04 16:00 UTC | $85,256.01 | 2026-10-06 04:00 UTC | $85,516.27 | +0.31% | +0.10% |
| 11 | 2026-10-09 16:00 UTC | $82,884.00 | 2026-10-13 00:00 UTC | $81,850.01 | −1.25% | −1.45% |
| 12 | 2026-10-13 08:00 UTC | $83,870.89 | 2026-10-14 16:00 UTC | $81,641.44 | −2.66% | −2.86% |
| …and 5 more in the CSV |
How a fee changes one trade
Take a trade of $10,000 that the price moves up by 0.15%. Without fees it makes $15. At 0.1% per side, the entry costs $10 and the exit about $10.02, so the same trade loses about $5. A round trip at 0.1% per side needs a move of a little over 0.2% just to break even, which is why rules that trade small moves are hit hardest.
What this test does NOT model
- Slippage and the bid-ask spread: every order fills at the candle's open price. Real market orders fill a little worse, which makes the after-fee results here optimistic.
- Partial fills and order-book depth: the whole order fills at once, whatever its size.
- Funding rates and leverage: none; the test buys and sells the coin itself.
- Fee discounts and maker rebates: one flat fee per side. Check your own exchange's fee on the exchange fees table.
- Taxes, withdrawal fees and downtime.
Method
- Data: 552 closed 4-hour candles of Binance BTC/USDT, opening from 2026-08-01 00:00 UTC to 2026-10-31 20:00 UTC. Indicators are computed from the first candle of the window, the way the backtester computes them from the first candle it loads.
- Engine: the same code as the in-app backtester (how it works). If a position is still open on the last candle, it is closed at that candle's close.
- Fees: charged on the amount bought at entry and on the amount sold at exit.
- Reproduce it: open the backtester, choose 4h and the preset, and set the fee to 0 and then 0.1. The backtester tests the latest candles, so its window (and numbers) will differ from this fixed one; the ledger lets you check this page's trades one by one.
Download and reuse
- Trade ledger (CSV): one row per trade for both strategies at both fee levels, with entry and exit times and prices, the amount, fees paid and the result.
- Licence: CC BY 4.0, with attribution to hunt4coin.com (for example: "Source: Hunt4Coin backtest fee example, hunt4coin.com").
- Related: Guide: how to backtest a crypto strategy with fees · Free crypto backtesting · Paper trading with fees · Facts and limits
Information only, not financial advice. Past results do not predict future results.