One Step Up the Risk Dropdown: How Much Return and How Much Drawdown It Added in Four of Our EAs

11 October 2026, 01:00
Kenichiro Sakamoto
0
29
Our EAs have a four-step risk dropdown: Defensive, Standard, Aggressive, Ultra. The signal is identical on every step and only position size changes. We have published the backtest of each step per product. This post puts four products side by side and asks one question: what did each step up actually cost in drawdown for what it added in return?
All figures are MT5 backtests on $10,000, historical, not a live record.

NASDAQ DIP BUYER (74 trades, about seven years): a straight line
Step | Return | Profit factor | Max equity drawdown | Return per point of drawdown
Defensive | +7.8% | 1.42 | 5.6% | 1.4
Standard | +15.7% | 1.40 | 11.0% | 1.4
Aggressive | +31% | 1.36 | 21.3% | 1.5
Ultra | +51% | 1.30 | 35.1% | 1.5
Each step roughly doubles both columns. Return per point of drawdown stays at 1.4 to 1.5 on every step. Nothing is gained by going up except a bigger version of the same curve, and the profit factor slips from 1.42 to 1.30 on the way.

BITCOIN COMET, original engine (128 trades, 2018-2025): return runs ahead
Step | Return | Profit factor | Max equity drawdown | Return per point of drawdown
Defensive | +104% | 2.57 | 5.5% | 19
Standard | +324% | 2.37 | 8.3% | 39
Aggressive | +1,328% | 2.03 | 16.1% | 82
Ultra | +3,700% | 2.19 | 18.2% | 203
Here the return grows much faster than the drawdown. That is compounding: a trend follower with percentage risk over eight years reinvests its early wins. It is real arithmetic and it is also the most path-dependent number on the page. The drawdown column, 5.5% to 18.2%, is the one to plan around.

ETHEREUM TREND (154 trades, 2019-2026): the top step adds nothing
Step | Return | Profit factor | Max equity drawdown
Defensive | +73% | 2.53 | 6.5%
Standard | +171% | 2.70 | 11.0%
Aggressive | +324% | 2.83 | 17.1%
Ultra | +324% | 2.83 | 17.1%
Ultra and Aggressive are the same result, because the maximum-lot cap is reached at about twice the base risk. Selecting the highest step changes nothing in that test.

AUDCAD PEACE (averaging grid, 2015-2024): the lowest step is not the safest
Step | Return | Profit factor | Max margin drawdown
Defensive | +32% | 1.40 | 30.4%
Standard | +313% | 1.74 | 13.7% (relative 30.8%)
Aggressive | +608% | 1.64 | 15.9% (relative 60.4%)
Ultra | +725% | 1.53 | 36.2% (relative 76.0%)
Defensive halves the lot but keeps the same dollar basket stop, so the baskets behave differently and its percentage drawdown came out deeper than Standard's, for a tenth of the return. On a grid, "lower risk setting" and "lower drawdown" were not the same thing.

What the four tables say together
- In three of the four, profit factor is not highest on the highest step. More size never improved the edge.
- Drawdown scaled close to the size step on the non-compounding index EA (5.6%, 11.0%, 21.3%, 35.1%). That is the honest default expectation: double the risk, double the worst case.
- The dropdown label is not a risk measure. The same word, Defensive, meant a 5.5% drawdown on one EA and 30.4% on another.

For buyers
Do not choose a step by its name or by the return column. Run the EA on each step over the full period on your own broker's data, write down the maximum equity drawdown of each run, and pick the step whose drawdown you could sit through. Then expect the live drawdown to be able to exceed it.

The two EAs in this article: NASDAQ DIP BUYER https://www.mql5.com/en/market/product/181839 · BITCOIN COMET https://www.mql5.com/en/market/product/181831

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