AI Algorithmic Trading Solution

FxMath Elite Pro — XAUUSD H2 Optimised: +59.43% on Gold, Confirmed Out-of-Sample

FxMath Financial Solution

FxMath Elite Pro — XAUUSD H2 Optimised: +59.43% on Gold, Confirmed Out-of-Sample

The optimised XAUUSD H2 preset for FxMath Elite Pro closed +$5,943.02 (+59.43%) on a $10,000 account across 969 trades — a 91.3% win rate, 2.43 profit factor and just 12.92% maximum drawdown. It stayed profitable on the untouched out-of-sample slice.

Executive Summary

969 trades · 91.3% win rate (885W / 84L) · +$5,943.02 · Profit Factor 2.43 · Max Drawdown 12.92% · Out-of-sample confirmed +17.69%

The same EA with stock defaults lost $2,539.22 across 248 trades at a 31.69% drawdown — the entire difference was the settings.

FxMath Elite Pro XAUUSD H2 Optimised — profit factor 2.43, 59.43% return, $5,943 net profit, 12.92% max drawdown
FxMath Elite Pro · XAUUSD H2 Optimised Preset — headline metrics.

Headline Performance

+$5,943.02
Net Profit (+59.43%)
2.43
Profit Factor
91.3%
Win Rate
12.92%
Max Drawdown
969
Trades (885W / 84L)
2.09
Sharpe · Sortino 2.05
$6.13
Expectancy / Trade
76%
Bar Exposure

Equity Curve

The equity curve climbs from $10,000 to $15,943.02, while a simple buy-and-hold over the same window returned just +39.15%. 17 of the 18 monthly periods closed positive; the single red month lost $3.12.

Equity curve XAUUSD H2 optimised — $10,000 to $15,943.02
Figure 1: Equity curve · deposit $10,000, buy & hold +39.15% for reference.

In-Sample vs Out-of-Sample

This is a walk-forward result — the preset was fitted on the in-sample window, then run untouched on data it had never seen. It stayed profitable.

Metric In-Sample Out-of-Sample
Net profit $3,504.95 $1,769.18
Return 35.05% 17.69%
Max drawdown 10.45% 13.42%
Profit factor 2.48 1.99
Trades 569 344
Win rate 91.9% 90.1%
Sharpe 2.49 1.91

Monthly Breakdown

Eighteen months, seventeen of them green. February 2026 was the strongest at +$821.03; September 2026 was the only negative month, at -$3.12.

Monthly P&L XAUUSD H2 optimised — 17 of 18 months positive
Figure 2: Monthly P&L across the full test window (Apr 2025 – Sep 2026).

Same EA, Better Settings

The headline is not the profit — it is the gap between the same engine left on defaults and the same engine configured correctly.

EA defaults versus optimised preset — net profit, profit factor, trades and win rate
Figure 3: EA defaults versus the optimised XAUUSD H2 preset.
Metric EA Defaults Optimised
Net profit -$2,539.22 $5,943.02
Return -25.39% 59.43%
Max drawdown 31.69% 12.92%
Profit factor 0.43 2.43
Trades 248 969
Win rate 67.3% 91.3%

Capital Requirement

Running 0.01 lot at 1:500 leverage, the minimum viable balance is $3,271.14. The binding constraint is the drawdown budget — twice the $1,635.57 worst backtest drawdown — not margin. Because this is a grid with no stop loss, size with headroom.

Item Amount
Leverage assumed 1:500
Peak simultaneous position (grid stack) 0.08 lots
Margin for the whole stack $82.47
Worst floating loss -$1,707.08
Margin + floating loss $1,789.55
Drawdown budget (×2.0) $3,271.14

Shipped Settings — XAUUSD_H2.set

The grid

  • Take-profit distance: 7.7 pips · trailing: 11 pips
  • Grid distance: 38.5 pips · maximum grid: 5 levels
  • Fixed base lot: 0.01 · stop loss: 0 (closed by basket TP or the protection switch)

The learning engine

  • Model: fast · window 12 / back 15 · label SL 14 / TP 26.25 pips
  • Train bars: 400 · retrain every 8 bars · 20 trees × 4 forests · depth 8
  • Protection enabled · 30% maximum drawdown cap · strong-trend auto-exit

How the Engine Works

FxMath Elite Pro trains four Random Forests directly inside MetaTrader 4 and MetaTrader 5 — no DLLs, no Python, no external data. Every retrain cycle recomputes 120 features, the regime detector reads ADX, Bollinger width and momentum, and a consensus vote from the Buy, Sell, Trend and Hedge models gates each entry. An ATR-scaled grid manages the basket, and a three-layer protection suite closes everything if a trend runs away.

Full Disclosure

  • 5,328 configurations evaluated · 0 finalists · 9 robust
  • This is a grid with no stop loss at 76% exposure — the downside is unbounded in principle
  • Worst losing streak: 6 consecutive losses
  • The $3,271.14 minimum is deliberately padded to twice the tested drawdown

Pricing & Payment

Every plan runs on up to 3 accounts, supports both MT4 and MT5, and includes all features. The optimised preset report and the free signals channel are open to everyone. Payment is cryptocurrency only — BTC, ETH, USDT, BNB, SOL, LTC, DOGE. You receive the compiled .ex5 and .ex4 files instantly.

1 Month
$69
per month

Buy Now

6 Months
$199
save $215

Buy Now

Most Popular

12 Months
$299
best value

Buy Now

24 Months
$399
save $1,257

Buy Now

Lifetime
$799
one-time payment

Buy Now

Payment: Cryptocurrency only — BTC · ETH · USDT · BNB · SOL · LTC · DOGE. Instant delivery of the compiled .ex5 and .ex4 files after payment. Up to 3 accounts per license.
Risk Warning: Past performance is not indicative of future results. This is a grid strategy with no stop loss at 76% exposure, so the downside is unbounded in principle. Trading forex, metals, indices and cryptocurrencies carries substantial risk of loss. Only risk capital you can afford to lose. Always test automated systems on a demo account before live deployment.

Tags

FxMathElite ProXAUUSDGoldOptimised PresetRandom ForestMachine LearningExpert AdvisorMetaTraderAlgorithmic TradingForex


Discover more from FxMath Financial Solution

Subscribe to get the latest posts sent to your email.

Follow me on Twitter

About FxMath

FxMath Financial Solution is a financial software team developing end-to-end algo trading systems for quantitative hedge funds and institutional trading groups. Our system based on profitable mathematical models with highest possible profit and lowest drawn down.

Subscription

Subscribe to Trade Signals via Email

Enter your email address to subscribe to our free trading signals by email

Discover more from FxMath Financial Solution

Subscribe now to keep reading and get access to the full archive.

Continue reading