Penerangan
XAUUSD H1 Adaptive Breakout cBot - Built Specifically for Gold's Post-$4,000 Market Regime
IMPORTANT β PLEASE READ BEFORE BACKTESTING
The XAUUSD H1 Adaptive Breakout cBot was developed specifically for the Gold market environment observed from October 8, 2025 onward, following Gold's move through the $4,000 price level.
Recommended evaluation/backtesting period: October 8, 2025 β Present
The strategy was not developed or calibrated around Gold's pre-$4,000 market behaviour. Users are free to test earlier historical periods; however, results from those periods should not be expected to represent the market regime for which this strategy was designed.
Why October 8, 2025 Matters
Gold has not traded with one consistent character throughout its modern history. Over time, XAUUSD has moved through distinctly different market regimes β from inflation-driven momentum and extended periods of consolidation to crisis-driven volatility, structural bull markets and increasingly rapid macro-driven price movements.
Gold moved through the $4,000 level on October 8, 2025, we identified a market environment displaying characteristics particularly relevant to this strategy, including larger price movements, faster reversals and increased breakout-driven activity.
Rather than assuming that a breakout strategy developed around older Gold behaviour would remain equally suitable under these newer conditions, we took a regime-specific approach.
Working with our team of specialists, the XAUUSD H1 Adaptive Breakout cBot was built, tested and refined specifically around the price behaviour observed in this newer Gold regime.
What This Means for Backtesting
When independently evaluating the cBot, users should therefore place primary emphasis on results generated from October 8, 2025 onward.
Testing substantially earlier periods may produce materially different results. This does not necessarily indicate that the strategy is malfunctioning; those periods represent market conditions outside the regime around which the current strategy was developed and calibrated.
Earlier data can still be useful for research, comparison and understanding how the strategy behaves under different historical conditions.
Designed for the Current Regime
A key principle behind this cBot is that market behaviour can change.
The strategy was created in response to an observed change in Gold's behaviour rather than on the assumption that historical market characteristics remain permanent.
For this reason, the strategy may continue to be reviewed as the Gold market evolves. Where material changes in market structure justify an adjustment, future versions may be refined accordingly.
How the Strategy Identifies Breakouts
The cBot does not simply enter whenever price moves above a previous high or below a previous low.
It uses a proprietary price-action trigger based on short-term candle structure to identify when current price behaviour deviates meaningfully from a recent reference period.
Directional and momentum confirmation are derived from price action at the potential entry point.
The entry model works from standard OHLC price data. ATR is not used as the entry trigger. Volume-derived measurements are used within the system's filtering and regime-management architecture.
The exact comparison logic and combination of conditions are intentionally undisclosed because they form part of the strategy's proprietary methodology.
Confirmation Before Capital Deployment
A key characteristic of the system is that it does not simply detect a breakout and enter immediately.
When a valid setup is identified, the cBot places a pending order that requires price to continue in the anticipated direction before the position is opened.
Price-action setup β directional confirmation β pending order β continuation β trade execution
The intention is to filter out a portion of initial breakout attempts that fail to follow through. In the reference backtest, 37 of 159 signals (23%) were cancelled without ever being filled, because the required continuation did not occur.
Calibrated From Bar One
The system's filters are statistical and require history to be calibrated correctly. On startup the cBot loads additional market history before trading begins, so an instance launched today behaves the same as one that has been running for months.
This matters for evaluation: without it, a freshly started instance can behave differently from the backtest over the same period. Users should therefore see consistent behaviour whether they start the cBot at any point in the market cycle.
Adaptive Risk Management
Position sizing is percentage-based and dynamic rather than a fixed lot size. The cBot calculates position size from the account balance and the individual trade's dynamically determined stop distance.
Risk parameters
Position sizing
Dynamic, balance- and structure-based
Maximum simultaneous positions:
3
Maximum holding time:
8 hours
Minimum practical account:
$1500
Recommended starting balance:
$3,500
Leverage:
Determined by the trader's broker/account, not set by the cBot
Please read this carefully. At its default setting this is an aggressive, high-variance system. Position size is calculated from a structural reference distance, while the protective stop is placed considerably wider than that reference. The practical consequence is that an individual losing trade can cost substantially more than the nominal risk percentage suggests.
Measured as a share of account balance, the largest single losing trade was -6.74% at the 2% risk setting. At the 10% default the same effect scales up proportionally, and the largest single loss in that run was -$175,059.81 against a peak balance that was millions. Most losing positions are closed by the system's time-based management well before the protective stop is reached, but users must be prepared for occasional individual losses. The risk percentage is fully configurable β see "Choosing Your Risk Level" below, and the backtest section for figures at both 2% and 10%.
Risk exposure is also reduced automatically when the system detects that current market conditions have become less favourable. The exact risk-adjustment methodology is proprietary.
Choosing Your Risk Level
Risk is controlled by two parameters in the cBot's settings panel:
ParameterPurpose
Risk % Per Trade
In this setting you adjust. It determines how large each position is relative to account balance.
Max Risk % Cap
A hard ceiling. The cBot will never size a position above this figure, whatever the first parameter is set to.
To reduce position size, lower Risk % Per Trade. For instance Setting it to 2 reproduces the 2% results shown in the backtest section.
To increase position size, raise Risk % Per Trade. Because Max Risk % Cap acts as a ceiling, you must raise it as well if you want to exceed its current value β this is deliberate, so that a mistyped figure cannot silently produce an oversized position.
No other settings need to be changed. Position sizing is calculated automatically from the account balance and each trade's individually determined stop distance, so the cBot adapts lot size on every trade without further configuration. Changing the risk percentage does not alter which trades are taken; it changes only how much capital is committed to each.
One effect worth understanding: several of the cBot's protective mechanisms respond to movements measured as a percentage of account balance. At larger position sizes, ordinary market fluctuations cross those thresholds more readily, and positions may be closed earlier than the strategy would otherwise intend. This is why the lower risk setting produced a higher profit factor in testing.
The cBot ships with Risk % Per Trade set to 10. Users may adjust the cBot's risk/exposure settings according to their account size and individual risk tolerance. Changing the Risk % Per Trade affects position sizing rather than the underlying entry logic. Both settings are documented in the backtest section so the choice can be made on the figures.
Adaptive Market-Regime Detection
The strategy contains a dedicated regime-adaptation layer.
Rather than relying solely on conventional indicators such as ATR or a fixed trend-strength threshold, the system monitors a statistical signal derived from its own recent trading behaviour. It examines how positions have been behaving shortly after entry over a rolling window.
When the monitored behaviour deteriorates relative to its recent baseline, the system automatically reduces exposure. When conditions normalise, exposure is automatically restored.
This makes the risk adjustment continuous and self-correcting rather than a manually programmed seasonal rule. Importantly, the regime layer responds only to the system's own realised trading outcomes, not to estimates β so it never engages on assumption.
The exact statistical calculation, thresholds and adjustment logic are proprietary.
Trade Management & Exits
The strategy does not use a conventional fixed-pip Take Profit.
Stop Loss. Determined dynamically from the price structure present at the time of entry rather than one fixed pip distance for every trade.
Take Profit. There is no fixed target. Positions are managed through a combination of dynamic stop placement, adaptive profit protection and time-based position management.
Profit Protection. The system does not trail the stop on every tick. When a trade reaches a defined favourable state, the cBot makes a selective protective adjustment to the real Stop Loss, securing part of the gain already achieved. The exact trigger conditions remain proprietary.
Maximum holding time. A position can remain open for a maximum of 8 hours, preventing a breakout position from remaining exposed indefinitely when the expected continuation does not develop.
Backtesting Results
Backtested on Pepperstone M1 historical data, with the strategy operating on the H1 timeframe.
Testing period: 9 October 2025 β 20 August 2026 (~10.4 months) Initial balance: $10,000 Signals generated: 159 β of which 37 (23%) were cancelled unfilled because the required continuation did not occur, leaving 122 executed positions.
Results are shown at two risk settings. Both runs use identical strategy logic on the identical period and take the same signals; only position sizing differs.
Risk 10% β the shipped default
Ending balance
$1,599,604.72
Net profit
$1,589,604.72
Ending-balance multiple
160Γ
Profit factor
Total trades
123
Winning / losing trades
67 / 56 (54.5% win rate)
Maximum consecutive losses
4
Average trade
$12,923.61
Largest winning trade
$283,558.30
Largest losing trade
-$175,059.81
Max balance drawdown
Net swap
+$5,025.69
Risk 2%
Ending balance
$48,797.33
Net profit
$38,797.33
Ending-balance multiple
4.88Γ
Profit factor
Max balance drawdown
Net swap
+$47.05
Comparing the Two Settings
Risk 2% Risk 10%
Ending balance $48,797 $1,599,605
Multiple 4.88Γ 160Γ
Profit factor all trades 2.43 2.08
Max balance drawdown 8.82% 28.01%
The higher setting produced the larger return. The lower setting produced the higher profit factor and shallower drawdowns.
This difference is structural rather than incidental. Several of the cBot's protective mechanisms respond to movements measured as a percentage of account balance. At larger position sizes, ordinary market fluctuations cross those thresholds more readily, and positions may be closed earlier than the strategy would otherwise intend. At smaller position sizes the same trades are allowed to develop further.
Both sets of figures come from identical strategy logic taking identical signals over the identical period. Only position sizing differs.
Long and Short Performance (Risk 10% run)
TradesWin rateProfit factorNet profitShare of total
Long
65
61.5%
$1,489,545.12
93.7%
Short
58
46.6%
$100,059.60
6.3%
Short trades were profitable overall, but with a profit factor of 1.11 they were close to breakeven and contributed a small fraction of the total. This is disclosed because it is visible in cTrader's own trade statistics and because it is relevant context: the tested period contained a strong directional bias in Gold, and results in a differently-trending market may not follow the same pattern.
Commission and Costs
Both runs used a fixed 1-pip spread with no commission applied. Traders on raw-spread accounts pay a per-lot commission on XAUUSD, which across 122 positions would reduce the result. Net swap was positive in both runs, though swap rates vary by broker and change over time.
Users should re-run the backtest with their own broker's commission and spread settings to see a result reflective of their own trading conditions. This is a deliberate disclosure: the headline figures are not net of all real-world costs.
Drawdown
At the 10% setting, maximum balance drawdown was 28.01% . At the 2% setting these fall to 8.82% .
Distribution of Returns β Concentration Analysis
Momentum systems typically derive a large share of their result from a minority of trades. The figures below show exactly how concentrated this strategy's returns were, measured as each group's share of total gross profit.
Winners removedShare of gross profit at Risk 1; 10%Share at Risk 2; 2%
Top 1
9.3%
6.6%
Top 3
24.1%
17.8%
Top 5.
35.9%
26.5%
Gross profit and loss for the period:
Gross profit
Risk 10% - $3,034,351
Risk 2%- $65,415
Gross loss
Risk 10%- $1,456,811
Risk 2%- $27,898
Net
Risk 10%- $1,589,604
Risk 2%- $37,517
Profit factor
Risk 10%- 2.08
Risk 2%- 2.34
The practical implication is that results depend heavily on capturing a small number of large continuation moves.
What Makes This Different
This is not presented as a universal gold strategy designed to work identically in every market environment. Its development philosophy is regime-aware.
Gold has demonstrated substantially different behaviour across different historical periods. A strategy that performs well during one type of market structure can behave very differently when volatility, liquidity, macroeconomic conditions and price behaviour change.
The system combines:
Proprietary price-action breakout detection β directional momentum confirmation β pending-order execution β dynamic position sizing β adaptive regime-based risk adjustment β selective profit protection β time-based exit control
The objective is to participate systematically in genuine continuation movements while reducing exposure when the system's own recent market evidence indicates that conditions have become less favourable.
What this cBot does not do: no martingale, no grid, no averaging into losing positions, no trading without a stop loss. Every position carries a protective stop from the moment it opens.
Account Requirements
The cBot can technically be deployed on accounts from approximately $1,500, depending on broker specifications, symbol contract specifications, margin requirements and chosen settings. For practical operation we recommend a starting balance of $3,500 or more.
A larger account does not eliminate trading risk. Traders should select exposure appropriate to their own risk tolerance and account conditions.
Continuous Strategy Updates
The strategy is actively maintained. Each month we review the system's performance, market behaviour and the evolving characteristics of XAUUSD to identify opportunities for refinement. Where validated improvements are identified, we update the cBot so users can continue working with the latest tested version.
Updates focus on maintaining methodology, execution quality, risk management and adaptability while preserving the core principles on which the system was developed.
Users therefore receive an actively maintained strategy rather than a one-time, static version.
Important Performance Disclosure
The historical results above were produced using the stated backtesting assumptions and parameters.
Backtests are simulations of historical market data. They can differ materially from live trading because of execution conditions, spread variation, slippage, liquidity, broker specifications, commissions, swaps and market conditions.
The results shown cover a single ~10-month period on one instrument, exclude commission, and assume a fixed 1-pip spread.
Past performance does not guarantee future results. Trading XAUUSD involves substantial risk, and losses can occur. The historical ending balances shown are backtest results under the stated conditions β not an expected or guaranteed return for a live account.
