Remove the 5 Best Months From 9 EA Backtests: Each Keeps 4.8% to 53.8% of Its Compounded Return

We took the published MT5 backtests of 9 of our EAs, split the closed trades by month, and turned each month into a percentage return on the closed balance at the start of that month. Then we asked one question: if the 5 best months had returned 0% and every other month had stayed the same, how much of the total return would be left? This article uses backtests only. It contains no forward-account figures. Figures are as of 2026-10-25.

- Without its 5 best months, each of the 9 tests keeps between 4.8% (BITCOIN COMET) and 53.8% (GOLD TRIPWIRE) of its total return. All 9 still end above the opening deposit.

- 5 months are 1.9% to 6.4% of the months with a closed trade in these tests. In 6 of the 9, those few months account for more than three quarters of the total return: only 4.8% to 23.5% is left without them.

- The number of best months you would have to remove before the test ends at or below its opening deposit ranges from 9 (ETHEREUM TREND) to 42 (GOLD TRIPWIRE). GOLD KING never gets there: all 89 of its months are positive on closed balance, while the same test has a 42.8% equity drawdown.

- MEGAMAX DONCHIAN and DONCHIAN ENGINE have nearly the same trades but a different position size. Without the 5 best months, MEGAMAX DONCHIAN keeps 7.0% of its return and DONCHIAN ENGINE keeps 44.5%.

- Each closed trade is assigned to the month in which it closed. Commission and swap are included. Open positions are not valued, so a month's return is on closed balance only.

- The return of a month is its closed result divided by the closed balance at the start of that month. The balance starts at the 10,000 opening deposit (1,000 for GOLD TRIPWIRE). Only months with at least one closed trade are counted. Multiplied back together, the monthly returns of every test give exactly the total return printed in its MT5 report, which we checked for all 9.

- "Without the 5 best months" means: set the 5 highest monthly returns to 0% and multiply the rest. This assumes that the other months would have earned the same percentage on a smaller balance, which holds only for an EA whose lot size follows the balance. It is a stress test of how concentrated the return is, not a result anyone could have traded: the best months are only known afterwards.

- The drawdown in the table is the equity drawdown from the same MT5 report for the whole test. It includes open trades and is a different measurement from the closed-balance months above.

- TRINITY and AIFW are not in this article: the original report files behind their published backtests are not in our breakdown, so we could not split them by month.

EA | Test period and data | Months with a closed trade | Best month (return) | Total return (MT5 report) | Return without the 5 best months | Share of the total return kept | Best months to remove until the test ends at or below the deposit | Equity drawdown (MT5 report, whole test)

GOLD TRIPWIRE | M15, 2020.01.01-2026.10.14, 11% real ticks | 82 | 2020-11 (+13.46%) | 760.4% | 409.4% | 53.8% | 42 | 7.08%

GOLD KING | M1, 2019.01.01-2026.06.01, 5% real ticks | 89 | 2020-03 (+16.64%) | 6177.7% | 3045.1% | 49.3% | never (no losing month) | 42.8%

DONCHIAN ENGINE (free) | H1, 2017.01.01-2026.06.01, 100% real ticks | 113 | 2021-02 (+10.59%) | 165.2% | 73.5% | 44.5% | 17 | 10.34%

ATLAS PORTFOLIO | H4, 2019.01.01-2026.06.28, 100% | 90 | 2022-06 (+44.63%) | 4941% | 1159.8% | 23.5% | 27 | 19.74%

GOLD VIPER | M30, 2019.01.01-2026.09.01, 8% real ticks | 92 | 2019-12 (+42.94%) | 19565.4% | 4182.3% | 21.4% | 31 | 24.09%

GOLD NEURON | M30, 2004.01.01-2026.09.01, 100% real ticks | 270 | 2013-04 (+64.88%) | 4160.4% | 683.3% | 16.4% | 17 | 28.5%

ETHEREUM TREND | H4, 2019.01.01-2026.06.28, 100% real ticks | 78 | 2024-02 (+55.96%) | 323.8% | 49.2% | 15.2% | 9 | 10.33%

MEGAMAX DONCHIAN (USDJPY) | H1, 2017.01.01-2026.06.01, 100% | 113 | 2021-02 (+104.74%) | 2155.4% | 150.7% | 7.0% | 10 | 16.27%

BITCOIN COMET | H4, 2018.01.01-2025.12.31, 98% | 94 | 2021-04 (+184.32%) | 12164.8% | 587.6% | 4.8% | 16 | 26.76%

Sorted by the share kept, largest first. The share kept is the return without the 5 best months divided by the total return. Total return and equity drawdown come from the same MT5 Strategy Tester report as the monthly split. The tests cover different periods and data quality, so read each row on its own and do not rank the returns against each other.

The single best month of the BITCOIN COMET test returned +184.32% on the closed balance it started with. With only that month set to 0%, the total return falls from 12164.8% to 4213.8%, so one month out of 94 holds about two thirds of the result. Without the 5 best months, 587.6% is left. The test is still positive after that, and it takes 16 best months before it ends at or below the deposit. The equity drawdown of the same report is 26.76%.

The two tests run on the same period, H1 2017.01.01-2026.06.01, with 629 and 631 trades, and their yearly win rates in the breakdown are the same in most years (for example 29.0% in 2017 and 48.3% in 2021 for both). Both have their best month in 2021-02. What differs is the size: that month returned +104.74% for MEGAMAX DONCHIAN and +10.59% for DONCHIAN ENGINE.

With the larger size, the result leans much more on a few months: 7.0% of the return is left without the 5 best months, against 44.5%, and 10 best months are enough to bring MEGAMAX DONCHIAN to the deposit, against 17. The equity drawdown is 16.27% against 10.34%. We do not claim size is the only difference between the two, but the trades themselves are close.

All 89 months of the GOLD KING test closed with a profit, so removing best months never brings it below the deposit. GOLD KING adds positions while the price moves against it and closes the basket when the price comes back, and the monthly split only sees closed baskets. The equity drawdown of the same test, which includes open trades, is 42.8%. For this EA the "never" describes closed balance only.

- A short live run can miss the few months that carry a backtest. For ETHEREUM TREND, 9 of 78 months hold the whole gain on closed balance: a year without one of them could look flat or red and still be consistent with the test.

- Read the share kept next to the drawdown. They do not line up: GOLD TRIPWIRE keeps 53.8% with a 7.08% equity drawdown, GOLD KING keeps 49.3% with 42.8%.

- If an EA offers a larger risk setting, expect the result to depend more on its best months, as the two Donchian tests show.

- A backtest value is a measurement of the past on one history. Future months can be distributed differently. Nothing here is a promise of future results.