Our 6 Demo Forward Accounts Have Made 0.6% to 6.6% of Their Backtests' Trade Counts. How to Read a Sample That Small

We run public forward accounts for our EAs. All of them are demo accounts, and none of them is a real-money record. This article takes 6 of them and asks one narrow question: how many trades has each forward account made, as a share of the trade count in the published backtest of the same EA? Figures are as of 2026-10-08.

- The largest share is 6.6%: GOLD NEURON, with 39 forward trades against 595 in its backtest.

- The two smallest are both 0.6%: ETHEREUM TREND at 1 of 154, and BITCOIN COMET at 5 of 775.

- All 6 are under 10%. Added together, the 6 forward accounts have made 211 trades against 6,837 in the backtests, or 3.1%.

- On gain, 1 of the 6 forward accounts is up and 5 are down. With samples this small, that split says little in either direction.

EA | Preset | Test period and data | Profit factor | Drawdown as printed by the MT5 Strategy Tester | Trades

GOLD KING | Defensive, v6.24 | 2019.01-2026.06, real ticks | 8.26 | 50.0% (relative equity) | 3,549

GOLD NEURON | Ultra, v3.16 | 2022.01-2026.08, 1-minute OHLC | 1.19 | 36.85% (relative equity) | 595

BITCOIN COMET | Ultra, v1.44 | 2018-2025, 1-minute OHLC | 1.72 | 26.8% (maximum equity) | 775

ATLAS PORTFOLIO | Standard | 2019.01-2026.06 | 1.48 | 19.74% (maximum equity) | 1,193

MEGAMAX DONCHIAN (EURJPY) | as published | 2017.01-2026.06 | 1.51 | 17.66% (maximum equity) | 571

ETHEREUM TREND | Aggressive, v1.55 | 2019-2026, real ticks | 2.83 | 17.1% (maximum equity) | 154

Each row is the headline test from the product description or from our backtest summary. The drawdown is copied as the MT5 Strategy Tester printed it, and its type differs by row (relative or maximum equity drawdown).

EA | Gain | Drawdown as reported by the account-tracking service | Profit factor | Trades

GOLD KING | +15.57% | 3.76% | 4.47 | 108

GOLD NEURON | -0.55% | 39.38% | 0.99 | 39

BITCOIN COMET | -3.04% | 16.74% | 0.84 | 5

ATLAS PORTFOLIO | -13.7% | 13.7% | 0.16 | 34

MEGAMAX DONCHIAN (EURJPY) | -18.91% | 18.91% | 0.15 | 24

ETHEREUM TREND | -0.44% | 0.44% | not shown yet | 1

Drawdown in this table is the figure reported by the account-tracking service that follows each demo account. It is not calculated the same way as the Strategy Tester figure, so do not read the two drawdown columns against each other. Both are shown because a result quoted without its drawdown is half a result.

GOLD KING is the only row where we can confirm that the forward account runs the same risk preset as its backtest row (Defensive). For the other 5 we cannot confirm it, so gain and drawdown are not lined up across the two tables.

EA | Forward trades | Backtest trades | Share | Forward trades at 10% of the backtest count | Still to go

GOLD NEURON | 39 | 595 | 6.6% | 60 | 21

MEGAMAX DONCHIAN (EURJPY) | 24 | 571 | 4.2% | 58 | 34

GOLD KING | 108 | 3,549 | 3.0% | 355 | 247

ATLAS PORTFOLIO | 34 | 1,193 | 2.8% | 120 | 86

ETHEREUM TREND | 1 | 154 | 0.6% | 16 | 15

BITCOIN COMET | 5 | 775 | 0.6% | 78 | 73

The 10% line is our own rule of thumb for when a forward stops being an early sample. It is not a statistical threshold, and reaching it proves nothing by itself.

- A green result and a red result deserve the same caution. GOLD KING at +15.57% rests on 108 trades, 3.0% of its backtest count. MEGAMAX DONCHIAN (EURJPY) at -18.91% rests on 24 trades, 4.2%. Neither figure confirms or refutes the backtest behind it.

- A profit factor built on a handful of trades moves a lot with each new one. The 0.84 of BITCOIN COMET comes from 5 trades. For ETHEREUM TREND, with 1 trade, the tracking service shows no profit factor yet.

- The share tells you how much has been seen, not whether what was seen is good. A forward can reach any share and still be losing.

We chose the trade count because it depends less on position sizing than gain or drawdown do. It is still not a fixed number:

- Preset. The GOLD KING backtest has 3,549 trades on Defensive and 2,827 on Ultra over the same period. The forward account runs Defensive, so the row uses 3,549.

- Configuration. The 775 trades of BITCOIN COMET are the default with all 4 strategies, which needs a hedging account. Its original engine alone has 128 trades in the published table for the same period. Against 128, the 5 forward trades would be 3.9%. We have not confirmed for this article which of the two the forward account ran throughout; it is under 10% either way.

- Versions. All 6 forward accounts were updated to a newer EA version at least once since they started, so each forward count spans versions while each backtest is of one version.

- Counting. A tracking service and the Strategy Tester do not necessarily count a trade the same way, and we did not reconcile the two trade by trade for this table.

The table covers 6 of the 12 demo accounts in our tracking file. Counting all 12, 4 are up and 8 are down.

- Divide the forward trade count by the backtest trade count before you look at the gain. Under 10%, treat it as an early sample whether it is green or red.

- Check that the backtest row and the forward account use the same preset before you compare anything that scales with position size.

- Ask how each drawdown figure was measured. Two percentages with the same name can come from different formulas.

- These are demo accounts and historical simulations. Neither is a promise about future results.

The two EAs in this article: GOLD KING https://www.mql5.com/en/market/product/180879 · GOLD NEURON (AI) https://www.mql5.com/en/market/product/187329