Longest vs Deepest Losing-Month Run, 8 EA Backtests: Different in 7; GOLD NEURON 6 Months −19,123 vs 2 Months −76,406

A losing-month streak is a row of calendar months that each closed below zero. We looked at two different runs inside the published MT5 backtests of 8 of our EAs: the longest run counted in months, and the run with the largest total loss in money. We wanted to see whether they are the same stretch of the test. This article uses backtests only. It contains no forward-account figures. Figures are as of 2026-11-06.

- In 7 of the 8 EAs the longest losing run and the deepest losing run by money are different stretches. Only ATLAS PORTFOLIO has one run that is both (4 months, 2025-01 to 2025-04, −52,227.90).

- GOLD NEURON: the longest run is 6 months (2015-08 to 2016-01) and lost −19,123.30. Its deepest run is only 2 months (2026-04 to 2026-05) and lost −76,406.27, which is 4.00 times as much.

- MEGAMAX DONCHIAN shows the widest gap: a 6-month run of −4,529.06 against a 3-month run of −27,739.35 (6.12 times). DONCHIAN ENGINE, the free engine behind it, has the same two stretches of dates: 2018-08 to 2019-01 and 2024-07 to 2024-09.

- The gap is smallest for ETHEREUM TREND (1.13 times): its deepest run is a single month, 2025-10, at −1,722.81, just ahead of its 4-month run of −1,521.00.

- Counting months in a row tells you how long a bad stretch lasted. It does not tell you how much it cost. In these tests the cost sat in different months from the length.

- Each closed trade is assigned to the calendar month in which it closed. Commission and swap are included. Open positions are not valued. Only months with at least one trade are counted, and a month with no trade ends a run. ETHEREUM TREND has 7 such months.

- A losing month is a month whose net result is below zero. A run is a row of consecutive losing months. The longest run is the one with the most months; if several runs tie, the table shows the one with the largest loss (ATLAS PORTFOLIO has 2 four-month runs, ETHEREUM TREND 3, GOLD VIPER 2).

- The deepest run by money is the run, of any length from 1 month up, with the most negative total. The ratio column is the deepest run's loss divided by the longest run's loss.

- Money amounts are not comparable across EAs. Most tests start with a 10,000 balance and several compound, so later runs are in larger sums; GOLD TRIPWIRE starts with 1,000. Compare the two runs inside the same EA, not between EAs.

- The last column is the maximum relative equity drawdown of the whole test (site settings): the largest fall as a percent of the equity peak. It is not a monthly figure and is shown for scale. Closed months do not value open positions.

- GOLD KING has no losing month in 89 months, so it has no run to compare; its maximum relative equity drawdown in the same test is 91.3%. 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 | Losing months / months with trades | Longest run: months, span, net (money) | Deepest run by money: months, span, net (money) | Deepest ÷ longest | Max relative equity DD (whole test)

MEGAMAX DONCHIAN (USDJPY) | H1, 2017.01.01-2026.06.01, 100% | 51/113 | 6, 2018-08–2019-01, −4,529.06 | 3, 2024-07–2024-09, −27,739.35 | 6.12x | 58.6%

GOLD NEURON | M30, 2004.01.01-2026.09.01, 100% real ticks | 124/270 | 6, 2015-08–2016-01, −19,123.30 | 2, 2026-04–2026-05, −76,406.27 | 4.00x | 55.5%

DONCHIAN ENGINE (free) | H1, 2017.01.01-2026.06.01, 100% real ticks | 52/113 | 6, 2018-08–2019-01, −526.02 | 3, 2024-07–2024-09, −1,699.12 | 3.23x | 10.8%

GOLD TRIPWIRE | M15, 2020.01.01-2026.10.14, 11% real ticks | 14/82 | 2, 2023-02–2023-03, −75.11 | 1, 2023-06, −147.53 | 1.96x | 12.7%

BITCOIN COMET | H4, 2018.01.01-2025.12.31, 98% | 41/94 | 5, 2022-08–2022-12, −42,186.80 | 3, 2024-07–2024-09, −78,796.27 | 1.87x | 31.2%

GOLD VIPER | M30, 2019.01.01-2026.09.01, 8% real ticks | 24/92 | 3, 2024-12–2025-02, −185,034.80 | 2, 2026-04–2026-05, −262,504.57 | 1.42x | 30.3%

ETHEREUM TREND | H4, 2019.01.01-2026.06.28, 100% real ticks | 38/78 | 4, 2026-03–2026-06, −1,521.00 | 1, 2025-10, −1,722.81 | 1.13x | 17.1%

ATLAS PORTFOLIO | H4, 2019.01.01-2026.06.28, 100% | 36/90 | 4, 2025-01–2025-04, −52,227.90 | 4, 2025-01–2025-04, −52,227.90 | 1.00x | 26.3%

- GOLD VIPER: its longest run (3 months, 2024-12 to 2025-02) lost −185,034.80, yet 2026-04 and 2026-05 together lost −262,504.57 in 2 months. May 2026 alone was −192,160.63, more than the whole 3-month run. The sums are larger because the account had grown by then.

- BITCOIN COMET: a 5-month run in 2022 (−42,186.80) against a 3-month run in 2024 (−78,796.27). Both lie inside a test with a 31.2% maximum relative equity drawdown.

- GOLD TRIPWIRE: its longest run is 2 months (−75.11); a single month, 2023-06, lost −147.53. It has the shortest longest run in the table, 2 months, and the fewest losing months, 14 of 82.

- The three EAs with the most losing months (MEGAMAX DONCHIAN 51, DONCHIAN ENGINE 52, GOLD NEURON 124) each have a longest run of 6 months. The run length says nothing about the size of the worst loss: GOLD NEURON's 6-month run cost less than its 2-month run.

- When a page shows the longest losing streak, ask also for the month with the largest loss and for the maximum relative drawdown of the same test. The longest streak can be the cheapest one.

- A short bad stretch is not a safe one. In 4 of the 8 tests the deepest run is 1 or 2 months long.

- Plan your risk around the amount you could lose in two or three months, not only around how many months in a row can be red.

- This article is about past tests. It does not promise any future result.

- All forward accounts we publish are demo accounts, and none is used in this article.

The two EAs discussed most in this article: GOLD NEURON https://www.mql5.com/en/market/product/187329 · GOLD VIPER https://www.mql5.com/en/market/product/182297