Demo Forward Drawdown vs Backtest Max Drawdown in 7 EAs: 5.0% to 118.6% Reached; BITCOIN COMET's 37% Tops Its 31.2%

21 November 2026, 00:00
Kenichiro Sakamoto
0
5
How deep has each of our demo forward accounts already fallen, compared with the deepest drawdown in the published backtest of the same EA? We run public forward accounts for our EAs. All of them are demo accounts, and none of them is a real-money record. For 7 EAs we divided the forward account's drawdown by the backtest's maximum relative equity drawdown. Result: the forward accounts have reached 5.0% to 118.6% of their backtest drawdown, with a median of 56.6%. One account, BITCOIN COMET, is already past its backtest: 37% on the forward account against 31.2% in the backtest. GOLD KING is at the other end: 4.6% against 91.3%. Figures are as of 2026-11-20.

How it is counted
- Forward drawdown, gain and trade count are the values shown on each account's public tracking page, as captured on 2026-11-20. Gain and trades are counted from the day the account started, through every version it has run. They are not the result of the current version alone.
- Backtest drawdown is the maximum relative equity drawdown of the MT5 Strategy Tester report for the settings we publish. It is the only backtest drawdown used in this article.
- The two drawdowns are not computed by the same tool, on the same broker feed, or over the same length of time. The ratio is a rough gauge, not an exact comparison.
- Ratio = forward drawdown / backtest maximum relative equity drawdown.

The table: forward drawdown as a share of backtest drawdown
EA | Backtest period | History quality | Backtest max relative equity DD | Forward start | Forward trades | Forward gain (since start) | Forward DD | Forward DD / backtest DD
BITCOIN COMET | 2018.01.01–2025.12.31 | 98% | 31.2% | 2026-08-21 | 14 | −25.93% | 37% | 118.6%
ATLAS PORTFOLIO | 2019.01.01–2026.06.28 | 100% | 26.3% | 2026-08-21 | 88 | −19.53% | 19.79% | 75.2%
AIFW | 2023.01.01–2026.09.01 | 99% | 4.6% | 2026-10-13 | 148 | +1.24% | 2.82% | 61.3%
GOLD NEURON | 2004.01.01–2026.09.01 | 100% real ticks | 69.6% | 2026-07-17 | 58 | −7.32% | 39.38% | 56.6%
GOLD VIPER | 2019.01.01–2026.09.01 | 8% real ticks | 30.3% | 2026-07-25 | 155 | +0.42% | 10.9% | 36.0%
ETHEREUM TREND | 2019.01.01–2026.06.28 | 100% real ticks | 17.1% | 2026-08-21 | 3 | −4.68% | 4.68% | 27.4%
GOLD KING | 2019.01.01–2026.06.01 | 5% real ticks | 91.3% | 2026-07-25 | 222 | +21.61% | 4.6% | 5.0%
Rows are sorted by the last column. The order is not a ranking of the EAs.

BITCOIN COMET: past the backtest in 91 days
The BITCOIN COMET forward account started on 2026-08-21. In 91 days and 14 trades it reached a 37% drawdown, more than the 31.2% maximum of a backtest that runs from 2018.01.01 to 2025.12.31. The account was at −25.93% on 2026-11-20, so it has come back part of the way from its low. With 14 trades we do not read a win rate or profit factor from it. What the row does show is that the backtest maximum is not a ceiling: a forward account can go past it, and here it did so within the first months.

ATLAS PORTFOLIO and AIFW: over half in a short window
- ATLAS PORTFOLIO reached 19.79% in 91 days, 75.2% of its 26.3% backtest maximum. Its gain was −19.53%, close to the low.
- AIFW reached 2.82% in 38 days, 61.3% of a backtest maximum of 4.6%. Its gain was +1.24% over 148 trades. A small backtest drawdown leaves little room: a few losing trades already make a large share of it.

GOLD KING: a low ratio against a very deep backtest
GOLD KING has the lowest ratio, 5.0%, with +21.61% over 222 trades in 118 days and a forward drawdown of 4.6%. But its backtest maximum relative equity drawdown is 91.3%. A low ratio here means the account has not met the kind of period that made that 91.3%, not that such a period will not come. Read it next to the backtest figure, not instead of it.

Left out
- GOLD TRIPWIRE: the forward drawdown is not available in our data, so it is not compared.
- MEGAMAX DONCHIAN and NASDAQ DIP BUYER: 0 forward trades so far.
- TRINITY: the backtest period of the report behind its drawdown is not on file, so it is not compared.
- ETHEREUM TREND is kept in the table, but with 3 trades its forward drawdown says very little.

Limits
- The forward accounts are 38 to 126 days old. The backtests cover several years each. A short window can only show how fast a drawdown can arrive, not how deep it can get.
- All 7 accounts have changed version since they started, so gain and drawdown include earlier versions. The backtest is of the settings we publish now and may not match every version the account has run.
- Brokers, leverage and tick data differ between the forward accounts and the backtests. GOLD KING and GOLD VIPER ran on 5% and 8% real ticks.
- These are demo accounts and historical simulations. None of these figures says what an account will do next.

For buyers
- When a seller shows a forward account, put its drawdown next to the backtest's maximum drawdown, not only its gain.
- A forward drawdown past the backtest maximum means the test did not cover what the account met. Ask the seller what changed and whether the backtest is still the right reference.
- A forward drawdown far below the backtest maximum is not proof of safety. The backtest figure is still the one to size the account for.

The two EAs discussed most in this article: BITCOIN COMET https://www.mql5.com/en/market/product/181831 · ATLAS PORTFOLIO https://www.mql5.com/en/market/product/182751

Our published backtests state period, balance and settings, and our forward accounts are labelled as demo accounts wherever we show them. The full list: fxea365.com/ea/ranking