King of Gold Ucb Agent EA MT5 (Latest version) - Unlimited

King of Gold Ucb Agent EA MT5 Settings Guide

King of Gold Ucb Agent EA MT5 is an advanced adaptive Expert Advisor built around contextual decision-making, persistent learning, multi-bandit optimization, liquidity analysis, market-state evaluation, and dynamic risk management. The supplied configuration reveals an extensive architecture covering trading schedules, position sizing, trailing protection, break-even management, daily profit and loss controls, persistent learning memory, UCB decision bands, liquidity-zone execution, volume analysis, order-book signals, economic news, session quality, sentiment feedback, and adaptive performance management.

The official product description identifies King of Gold as a self-learning EA primarily built for XAUUSD while also supporting other market categories. Its architecture uses contextual multi-armed bandits to learn which entries, position sizing, and execution methods work best in different market contexts. The EA also supports persistent memory, profile selection, execution decisions, liquidity zones, order-flow protection, economic-calendar awareness, and London/New York session tracking.

Magic Number

🔹 Magic Number = 5052026 — Assigns 5052026 as the unique identifier for trades managed by the EA, helping separate its positions from other Expert Advisors or manual trades.

Trading Time Settings

🔹 Start trading hour (0-23) = 0 — Sets the earliest trading hour to 0, allowing the EA’s configured trading window to begin at midnight.

🔹 End trading hour (0-23) = 23 — Sets the trading window end hour to 23, covering the full configured daily schedule.

🔹 Trade on Monday = true — Allows trading on Monday.

🔹 Trade on Tuesday = true — Allows trading on Tuesday.

🔹 Trade on Wednesday = true — Allows trading on Wednesday.

🔹 Trade on Thursday = true — Allows trading on Thursday.

🔹 Trade on Friday = true — Allows trading on Friday.

🔹 Trade on Saturday = false — Disables trading on Saturday.

🔹 Trade on Sunday = false — Disables trading on Sunday.

Risk Management Settings

🔹 Risk % per trade (0-10) = 0.5 — Sets the primary percentage-based risk level to 0.5% per trade.

🔹 Default risk % if RiskPercent is set to 0 = 0.5 — Defines the fallback risk percentage as 0.5 when the main RiskPercent value is configured to zero.

🔹 Hard cap for risk-based lot size. 0 = disable cap and use broker = 0.0 — Sets the risk-based lot-size ceiling to 0.0; according to the displayed label, zero disables this additional cap and allows the broker’s limit to apply.

🔹 Take Profit distance (pips if InputsArePips=true, otherwise points) = 2000.0 — Sets the configured Take Profit distance to 2000.0, with the unit interpreted according to the InputsArePips setting.

🔹 Stop Loss distance (pips if InputsArePips=true, otherwise points) = 2000.0 — Sets the configured Stop Loss distance to 2000.0, using pips or points according to the corresponding unit mode.

🔹 When true distances (TP/SL/trailing/breakeven/spread) are in pips = false — Keeps the unit mode disabled, meaning the displayed distance values are interpreted as points rather than pips.

🔹 Maximum open positions (adjust for multiple orders) = 5 — Limits the total number of simultaneously open positions to 5 under the EA’s position-management rules.

🔹 Maximum open positions per symbol = 5 — Allows a maximum of 5 open positions for an individual symbol.

🔹 Maximum pending orders per symbol = 5 — Limits pending orders for each symbol to 5.

MaxSpread

🔹 Maximum allowed spread (pips if InputsArePips=true, otherwise points) = 3500 — Sets the maximum accepted spread threshold to 3500, with the actual unit determined by the configured pip/point mode.

Pair & Lot Settings

🔹 Fixed lot size (if 0, uses risk %) = 0.05 — Sets the fixed position size to 0.05 lots; when the fixed-lot mechanism is applicable, this value defines the trade volume.

🔹 Maximum slippage (pips if InputsArePips=true, otherwise points) = 20 — Limits permitted execution slippage to 20, interpreted as pips or points according to the configured distance mode.

Timezone Settings

🔹 Use broker server time (recommended). If false, uses local PC time = true — Uses the broker’s server time because the setting is true, keeping trading-hour calculations aligned with broker time.

🔹 Timezone offset in hours (only if UseBrokerTime=false). 0 = use = 0 — Sets the manual timezone offset to 0; this parameter becomes relevant when broker-server time is disabled.

Cooldown Settings

🔹 Enable cooldown after losing trade = false — Disables the post-loss cooldown mechanism.

🔹 Number of bars to wait after a loss = 5 — Sets the cooldown duration to 5 bars if the losing-trade cooldown feature is enabled.

Trailing Stop Loss Settings

🔹 Enable trailing stop loss = false — Disables the standard trailing Stop Loss mechanism.

🔹 Activate trailing SL after X profit (same units as TP/SL) = 50 — Sets the trailing activation threshold to 50 units of profit when the trailing feature is enabled.

🔹 Trail stop by X from highest/lowest price (same units as TP/SL) = 50 — Sets the trailing distance to 50 units from the highest or lowest favorable price.

Breakeven Stop Settings

🔹 Enable breakeven stop = false — Disables automatic break-even Stop Loss adjustment.

🔹 Move SL to breakeven after X profit (same units as TP/SL) = 20 — Configures the break-even trigger at 20 units of profit when enabled.

Dynamic SL/TP Adjustment Settings

🔹 Enable dynamic SL/TP adjustment based on market analysis = false — Disables adaptive Stop Loss and Take Profit adjustment based on market analysis.

🔹 Maximum TP multiplier (2.0 = 2x original TP) = 1.5 — Limits the dynamic Take Profit expansion factor to 1.5× the original TP when the feature is enabled.

🔹 Minimum SL in profit (points above/below entry) = 60 — Sets the minimum profitable Stop Loss displacement to 60 points when dynamic protection is active.

🔹 Volume ratio threshold for strong momentum = 1.3 — Uses 1.3 as the configured volume-ratio threshold for identifying stronger momentum conditions.

🔹 Strong-DOM threshold. Values <=1.0 mark every book state strong = 1.0 — Sets the strong order-book threshold to 1.0, with the panel explicitly noting that values at or below 1.0 mark every book state as strong.

🔹 Distance to trail SL from current price when conditions are very strong = 30 — Sets the trailing Stop Loss distance to 30 under very strong market conditions when the applicable dynamic mechanism is active.

Auto-Close Settings

🔹 Enable auto-close when target profit reached = true — Enables automatic closure when the configured profit target is reached.

🔹 Close all trades when profit reaches X USD = 15.0 — Sets the basket or account-level auto-close trigger at 15.0 USD of profit.

🔹 Enable trailing close – continue profiting after target = true — Enables trailing-close behavior after the initial profit target has been reached.

🔹 Activate trailing close after X USD profit = 15.0 — Activates the trailing-close mechanism after 15.0 USD of profit.

🔹 Trail close by X USD from highest profit = 1.0 — Sets the allowed profit retracement to 1.0 USD from the highest recorded profit.

🔹 Continue trading after auto-close triggers (same session) = true — Allows the EA to continue opening new trades after an automatic close within the same trading session.

Daily Target Settings

🔹 Enable daily realized PnL target logic = false — Disables the daily realized-profit target logic.

🔹 Realized net PnL target for current day (USD) = 250.0 — Sets the daily realized net profit target to 250.0 USD when the daily target feature is enabled.

🔹 If false, pause new entries until next day StartHour = false — Keeps the pause-until-next-day behavior disabled.

🔹 Let memory and risk react to daily target progress = false — Prevents learning memory and risk behavior from dynamically reacting to progress toward the daily target.

🔹 Reduce lot risk as realized profit approaches daily target = false — Disables progressive lot-risk reduction as realized profit approaches the daily target.

Daily Loss Settings

🔹 Enable daily realized loss limit logic = false — Disables the primary daily realized-loss limit logic.

Session Settings

🔹 London session start hour (broker time) = 7 — Sets the London session start to 07:00 broker time.

🔹 London session end hour (broker time) = 16 — Sets the London session end to 16:00 broker time.

🔹 New York session start hour (broker time) = 13 — Sets the New York session start to 13:00 broker time.

🔹 New York session end hour (broker time) = 22 — Sets the New York session end to 22:00 broker time.

🔹 Periodic best-session report interval = 600 — Sets the periodic session-performance reporting interval to 600 seconds.

State Persistence (File-Based)

🔹 Save/load EA memory to file = true — Enables saving and restoring the EA’s learning memory through a file-based persistence system.

🔹 Autosave interval (seconds). 0 = disable autosave = 60 — Configures automatic state saving every 60 seconds.

🔹 Limit trades persisted = 10000 — Limits the number of stored trade-learning records to 10000.

🔹 Limit regime history persisted = 10000 — Limits stored market-regime history to 10000 records.

🔹 Limit dynamic patterns persisted = 10000 — Allows up to 10000 persisted dynamic-pattern records.

🔹 Limit persisted setup buckets = 10000 — Limits stored setup-bucket learning records to 10000.

🔹 Limit persisted per-symbol learning memory = 10000 — Allows up to 10000 persisted per-symbol learning-memory records.

🔹 Limit persisted open-position learning contexts = 10000 — Limits stored open-position learning contexts to 10000.

🔹 If true, delete state file on init = false — Keeps the existing persistent state file instead of deleting it when the EA initializes.

🔹 In Strategy Tester always start with a clean state file = false — Does not force a clean learning-state file at the start of every Strategy Tester run.

🔹 Delete state file on unload/stop = false — Preserves the state file when the EA is unloaded or stopped.

Bandit v1.05 Improvements

🔹 Reduce UCB exploration during recovery / high drawdown = true — Reduces exploratory UCB behavior during recovery or elevated drawdown conditions.

🔹 UCB C multiplier when recovery mode active (0.10..1.00) = 0.5 — Sets the UCB exploration multiplier to 0.5 while recovery mode is active.

🔹 UCB C multiplier when DD >= 60% of recovery start (0.10..1.00) = 0.7 — Uses 0.7 as the UCB C multiplier when drawdown reaches at least 60% of the configured recovery-start level.

🔹 Per-bandit reward attribution (slip->exec, MFE-gap->profile) = true — Enables separate reward attribution for different bandit components, including execution slippage and profile-related MFE gaps.

🔹 Slippage/spread penalty weight for execution bandit (0..1) = 0.3 — Sets the execution-bandit penalty weight for slippage and spread conditions to 0.3.

🔹 MFE-vs-actual gap penalty weight for profile bandit (0..1) = 0.4 — Sets the profile-bandit penalty weight for the difference between maximum favorable excursion and realized outcome to 0.4.

🔹 Track MFE/MAE per position and shape decision-bandit reward = true — Enables tracking of Maximum Favorable Excursion and Maximum Adverse Excursion for individual positions.

🔹 Bonus when exit captured >= 70% of MFE = 0.15 — Adds a 0.15 reward bonus when the exit captures at least 70% of MFE.

🔹 Penalty when MAE during life exceeded 1.5x risk reference = 0.1 — Applies a 0.1 penalty when the position’s MAE exceeds 1.5× its risk reference.

🔹 UNUSED – the extraHOLD bonus biased the bandit toward skipping = true — Retains this legacy/unused feature flag at true, as shown in the Inputs panel.

🔹 UNUSED – see UseEnhancedHoldEval = 0.5 — Keeps the displayed unused/legacy parameter at 0.5.

🔹 Compact decision context: 640 cells, false = 12800 cells and mu… = true — Enables the compact decision-context mode, using the smaller context representation described in the parameter label.

Bandit Entry & Decision Controls

🔹 Include confluence bandit report in learning telemetry (now act… = true — Includes confluence-bandit information in the learning telemetry.

🔹 Bandit owns entries and learns the entry rule itself (false = f… = true — Allows the bandit system to control entry decisions and learn the entry rule.

🔹 Minimum composite decision score before bandit can trigger = 5 — Requires a minimum composite score of 5 before the bandit can trigger a trade.

🔹 Minimum regime quality required for bandit entries = 0.45 — Requires a minimum regime-quality score of 0.45 for bandit entries.

🔹 Move/delete pending liquidity limits when signal changes = true — Allows pending liquidity limit orders to be moved or deleted when the trading signal changes.

🔹 Minimum pending-price move before modifying a limit order = 3 — Requires a minimum price movement of 3 before modifying a pending limit order.

🔹 Tighten moved limit SL/TP when market gets this close = 0.35 — Sets the proximity threshold to 0.35 for tightening SL/TP when a moved limit order becomes sufficiently close to the market.

🔹 SL distance factor applied (to the ORIGINAL distance, never com… = 0.82 — Applies a 0.82 multiplier to the original Stop Loss distance under the relevant adaptive adjustment.

🔹 TP distance factor applied (to the ORIGINAL distance, never com… = 0.92 — Applies a 0.92 multiplier to the original Take Profit distance.

🔹 Adapt initial TP/SL by entry profile = true — Enables adaptive initial TP/SL selection according to the selected entry profile.

🔹 Score threshold for LONG_TERM profile = 6 — Requires a score of 6 for the LONG_TERM profile.

🔹 Score threshold for SHORT_TERM profile = 5 — Requires a score of 5 for the SHORT_TERM profile.

🔹 Initial TP multiplier for SHORT_TERM = 1.8 — Sets the initial Take Profit multiplier for SHORT_TERM trades to 1.8.

🔹 Initial SL multiplier for SHORT_TERM = 1.8 — Sets the initial Stop Loss multiplier for SHORT_TERM trades to 1.8.

🔹 Initial TP multiplier for LONG_TERM = 1.5 — Sets the initial Take Profit multiplier for LONG_TERM trades to 1.5.

🔹 Initial SL multiplier for LONG_TERM = 1.5 — Sets the initial Stop Loss multiplier for LONG_TERM trades to 1.5.

Asset Tracking

🔹 Track forex pairs in learning memory (the multi-symbol scanner) = true — Enables learning-memory tracking for forex pairs.

🔹 Track metals like XAU/XAG in learning memory = true — Enables learning-memory tracking for metals such as XAU/XAG.

🔹 Track indices in learning memory = true — Enables learning-memory tracking for index symbols.

🔹 Track stocks/equities in learning memory = true — Enables learning-memory tracking for stocks and equities.

🔹 Track crypto symbols in learning memory = true — Enables learning-memory tracking for cryptocurrency symbols.

Order Flow

🔹 Order-flow engine: ms-based spike entries AND the protective close = true — Enables the order-flow engine for millisecond-based spike-entry logic and protective basket-close behavior.

Alerts: Sound, News & Future Events

🔹 Enable PlaySound alerts = false — Disables sound alerts.

🔹 Minimum seconds between same sound event = 10 — Sets a 10-second minimum interval between repeated identical sound events.

🔹 Sound file for BUY execution = buy.wav — Uses buy.wav for BUY execution alerts.

🔹 Sound file for SELL execution = sell.wav — Uses sell.wav for SELL execution alerts.

🔹 Sound file for blocked order/event = blocked.wav — Uses blocked.wav for blocked-order or event alerts.

🔹 Sound file for risk/stress warnings = warning.wav — Uses warning.wav for risk and stress warnings.

🔹 Sound file for profitable close = profit.wav — Uses profit.wav for profitable-close notifications.

🔹 Sound file for losing close = loss.wav — Uses loss.wav for losing-close notifications.

🔹 Sound file for upcoming economic news = news.wav — Uses news.wav for upcoming economic-news alerts.

🔹 Sound file for future session events = event.wav — Uses event.wav for future session-event notifications.

🔹 Enable economic calendar alerts = true — Enables economic-calendar alert processing.

🔹 Calendar scan interval = 60 — Checks the economic-calendar information every 60 seconds.

🔹 Alert for events within next X minutes = 30 — Generates alerts for events occurring within the next 30 minutes.

🔹 Cooldown between news alerts = 300 — Sets a 300-second cooldown between economic-news alerts.

🔹 If false: only symbol currencies = true — Enables the displayed symbol-currency filtering behavior.

🔹 Add upcoming high-impact news context to liquidity bandit memory = true — Adds upcoming high-impact economic-news context to liquidity-bandit learning memory.

🔹 Lookahead window for liquidity-bandit news context = 60 — Uses a 60-minute forward-looking window for news context.

🔹 Interpret released high-impact news as bullish/bearish bias for… = true — Enables interpretation of released high-impact news as directional market bias.

🔹 How far back to analyze released important news = 120 — Looks back 120 minutes when evaluating recently released important news.

🔹 Alert before session start/end = true — Enables alerts before configured trading-session starts or ends.

🔹 Minutes before start/end to alert = 15 — Generates session alerts 15 minutes before the configured session boundary.

Telegram Alerts

🔹 Enable Telegram notifications via Bot API = false — Disables Telegram notifications in the supplied configuration.

🔹 Telegram bot token from BotFather = blank — No Telegram BotFather token is configured.

🔹 Telegram chat_id (user/group/channel) = blank — No Telegram user, group, or channel ID is configured.

🔹 Cooldown between same Telegram event messages = 15 — Sets the minimum interval between identical Telegram events to 15 seconds.

🔹 Send BUY/SELL execution alerts to Telegram = true — Enables Telegram BUY/SELL execution messages when Telegram integration is activated.

🔹 Send close-profit/loss alerts to Telegram = true — Enables Telegram notifications for profitable and losing closes.

🔹 Send risk/stress alerts to Telegram = true — Enables Telegram notifications for risk and stress events.

🔹 Send economic news alerts to Telegram = true — Enables Telegram economic-news notifications.

🔹 Send future session event alerts to Telegram = true — Enables Telegram notifications for future session events.

🔹 Send blocked-order alerts to Telegram = true — Enables Telegram notifications when orders are blocked.

Best Sessions Engine

🔹 Learn and report session quality (London/NY/Overlap) = true — Enables learning and reporting of session quality for London, New York, and their overlap.

🔹 Block new entries when current session is weak = false — Does not block new entries solely because the current session is classified as weak.

🔹 Feed learned session quality into the decision score = true — Incorporates learned session-quality information into the trading decision score.

🔹 Minimum trades before a session is considered reliable = 5 — Requires at least 5 trades before session statistics are considered reliable.

🔹 Session must keep at least this win rate to be optimal = 55.0 — Sets the minimum session win rate for an optimal classification at 55.0%.

Adaptive Performance Settings

🔹 Winning mode: risk = RiskPercent x this (risk % mode only, Lots… = 2.5 — Sets the winning-mode risk multiplier to 2.5× the base RiskPercent when the percentage-risk mode applies.

🔹 Losing mode: risk = RiskPercent / this (1.8 -> 0.5% becomes 0.2… = 1.8 — Sets the losing-mode divisor to 1.8, reducing percentage-based risk during losing conditions according to the displayed formula.

🔹 Minimum win rate % to be considered “winning” mode = 55.0 — Requires a minimum win rate of 55.0% before the system classifies conditions as winning mode.

🔹 Adjust SL/TP based on performance = false — Disables performance-based Stop Loss and Take Profit adjustment.

🔹 Enable AGGRESSIVE/MODERATE/CONSERVATIVE based on signal strength = true — Enables automatic risk or behavior classification into AGGRESSIVE, MODERATE, or CONSERVATIVE modes according to signal strength.

Heartbeat Settings

🔹 Periodic alive/status log = true — Enables periodic alive/status logging.

🔹 Seconds between heartbeat logs = 300 — Sets the heartbeat interval to 300 seconds.

🔹 Print heartbeat in Journal = true — Enables heartbeat messages in the MT5 Journal.

🔹 Periodic learning/reward telemetry in Journal = true — Enables periodic learning and reward telemetry in the Journal.

🔹 Learning telemetry interval (seconds) = 60 — Sets the learning telemetry interval to 60 seconds.

Memory Decision Engine

🔹 Normal mode entry threshold (1..10) = 5 — Sets the normal decision-entry threshold to 5 on the 1–10 scale.

🔹 Stress mode entry threshold (1..10) = 7 — Sets the stress-mode entry threshold to 7, requiring stronger evidence under stressed conditions.

🔹 Allow conditional HOLD/watch after losing streak = true — Enables conditional HOLD/watch behavior following a losing streak.

🔹 Start HOLD/watch after this many consecutive losses = 1 — Starts HOLD/watch behavior after 1 consecutive loss.

🔹 Allow 1 in N below-threshold HOLD/watch signals = 1 — Allows 1 in N below-threshold HOLD/watch signals with N configured as 1.

🔹 Allow HOLD for scores [threshold-1]=1 — Allows HOLD decisions for scores 1 point below the configured threshold.

🔹 Lot multiplier for memory-system HOLD-promoted entries only = 1.5 — Applies a 1.5× lot multiplier to entries promoted through the memory-system HOLD mechanism.

🔹 Market-state refresh interval (seconds) for fast reaction = 1 — Refreshes market-state information every 1 second.

🔹 Reward increment for good decisions = 0.07 — Adds 0.07 to the learning reward for a good decision.

🔹 Penalty decrement for bad decisions = 0.05 — Applies a 0.05 penalty for a bad decision.

🔹 Minimum setup reward score = -0.01 — Sets the minimum setup reward score to -0.01.

🔹 Maximum setup reward score = 0.9 — Caps the setup reward score at 0.9.

🔹 Contextual UCB filters/boosts signal execution = true — Enables contextual UCB filtering and signal-execution boosting.

🔹 UCB exploration coefficient (used as-is; rewards are in R, so… = 1.25 — Sets the UCB exploration coefficient to 1.25.

🔹 Optional |reward| clamp in R. 0 = no clamp = 0.0 — Sets the optional absolute-reward clamp to 0.0, which represents no clamp according to the label.

🔹 Weight kept on the OLD mean after warm-up (used as-is; <0 or… = 0.97 — Retains 0.97 weight on the previous mean after the warm-up phase.

🔹 Pulls served by a true running average before the EWMA (UcbRewa… = 20 — Uses 20 pulls before the running-average/EWMA transition.

🔹 Context pulls required before the HOLD arm is explored (HOLD bl… = 8 — Requires 8 context pulls before the HOLD arm can be explored.

🔹 Lot multiplier for the SMALL arm (used as-is, applied to fixed… = 0.6 — Applies a 0.6× lot multiplier to the SMALL decision arm.

🔹 Lot multiplier for the STRONG arm (used as-is, applied to fixed… = 1.25 — Applies a 1.25× lot multiplier to the STRONG decision arm.

🔹 STRONG arm is not considered until the context has this many pu… = 8 — Requires 8 context pulls before the STRONG arm becomes eligible.

🔹 Max closed bars a HOLD counterfactual is followed. 0 = until th… = 0 — Sets the HOLD counterfactual follow-up limit to 0, according to the displayed configuration.

🔹 UNUSED – HOLD reward is now the pure counterfactual in R (kept… = 0.2 — Retains the displayed unused HOLD-reward parameter at 0.2.

Persistent Bandit Context Limits

🔹 Limit persisted UCB contexts (payout allows 640 max, so this ne… = 10000 — Allows up to 10000 persisted UCB contexts.

🔹 Limit persisted execution-bandit contexts = 10000 — Allows up to 10000 persisted execution-bandit contexts.

🔹 Limit persisted profile-bandit contexts = 10000 — Allows up to 10000 persisted profile-bandit contexts.

🔹 Limit persisted liquidity-zone contexts = 10000 — Allows up to 10000 persisted liquidity-zone contexts.

Execution, Profile & Liquidity Bandits

🔹 Include contextual bandit report in learning telemetry = true — Includes contextual-bandit reports in learning telemetry.

🔹 Separate bandit for market vs limit execution choice = false — Keeps market and limit execution choices within the same bandit rather than separating them.

🔹 Exploration coefficient for execution bandit = 0.9 — Sets execution-bandit exploration to 0.9.

🔹 Include execution bandit report in learning telemetry = true — Enables execution-bandit reporting in learning telemetry.

🔹 Separate bandit for SHORT_TERM/NORMAL/LONG_TERM profile choice = true — Enables separate bandit handling for SHORT_TERM, NORMAL, and LONG_TERM profile selection.

🔹 Exploration coefficient for profile bandit = 0.85 — Sets profile-bandit exploration to 0.85.

🔹 Include profile bandit report in learning telemetry = true — Includes profile-bandit results in learning telemetry.

🔹 Learn which historical volume zone works best for liquidity = true — Enables learning of which volume/liquidity zones perform best.

🔹 Exploration coefficient for liquidity-zone arm selection = 0.8 — Sets liquidity-zone bandit exploration to 0.8.

Timeframe & Pattern Learning

🔹 Learn which timeframe (M1/M5/M15/H1/H4) gives best edge per con… = true — Enables learning of which timeframe among M1/M5/M15/H1/H4 provides the best contextual edge.

🔹 Exploration coefficient for timeframe arm selection = 0.85 — Sets timeframe-selection exploration to 0.85.

🔹 Lot multiplier nudge when chosen TF aligns with trade direction = 0.08 — Applies a 0.08 lot-size nudge when the selected timeframe aligns with trade direction.

🔹 Limit persisted timeframe-bandit contexts = 10000 — Limits persisted timeframe-bandit contexts to 10000.

🔹 Include timeframe bandit report in learning telemetry = true — Includes timeframe-bandit information in learning telemetry.

🔹 Learn which candle pattern type performs best per context = true — Enables learning of which candle-pattern type performs best in each market context.

🔹 Exploration coefficient for pattern arm selection = 0.85 — Sets pattern-selection exploration to 0.85.

🔹 Weight applied to chosen pattern arm mean-reward (lot nudge) = 0.12 — Applies a 0.12 weighting to the selected pattern arm’s mean reward for the lot-size adjustment.

🔹 Limit persisted pattern-bandit contexts = 10000 — Limits persisted pattern-bandit contexts to 10000.

🔹 Include pattern bandit report in learning telemetry = true — Includes pattern-bandit results in learning telemetry.

🔹 Learn which TF-confluence threshold (>=2/>=3/>=4) is best per c… = true — Enables learning of which multi-timeframe confluence threshold, such as >=2, >=3, or >=4, performs best for each context.

🔹 Exploration coefficient for confluence arm selection = 0.85 — Sets confluence-bandit exploration to 0.85.

🔹 Lot multiplier when current confluence is below chosen threshold = 0.75 — Applies a 0.75× lot multiplier when current confluence is below the selected threshold.

🔹 Limit persisted confluence-bandit contexts = 10000 — Limits persisted confluence-bandit contexts to 10000.

Daily Loss Protection

🔹 Stop trading if realized daily PnL falls below -X USD = 120.0 — Sets the realized daily PnL stop threshold to -120.0 USD.

🔹 If true, pause new entries until next day StartHour after loss = false — Keeps the next-day entry pause disabled after the daily loss threshold is reached.

🔹 Measure the daily loss from the day’s profit PEAK, not from the… = true — Measures the daily giveback from the day’s realized-profit peak rather than only from the day’s starting value.

🔹 Max USD allowed to be given back from the daily peak (0 = use D… = 0.0 — Sets the maximum USD giveback from the daily peak to 0.0, meaning this optional threshold is not configured with a positive value.

🔹 Max % of the daily peak allowed to be given back (0 = off; tigh… = 0.0 — Sets the percentage-based daily-peak giveback limit to 0.0, leaving that optional percentage guard inactive.

🔹 Arm the trailing guard only after the daily peak reaches this U… = 0.0 — Sets the daily-profit threshold required to arm the trailing guard to 0.0.

🔹 Count open-position floating PnL in the daily PnL used by the l… = false — Excludes floating open-position PnL from the daily PnL calculation used by this protection mechanism.

🔹 When PauseAfterDailyLoss=false, re-arm the guard after X minutes = 30.0 — Sets the re-arm delay to 30.0 minutes when the daily-loss pause feature is disabled.

Recovery Auto-Close

🔹 Enable DD-based recovery mode = false — Disables the drawdown-based recovery mode.

🔹 Enter recovery mode when DD >= this % = 3.0 — Sets the recovery-entry drawdown threshold to 3.0% if recovery mode is enabled.

🔹 Exit recovery mode when DD <= this % = 1.0 — Sets the recovery-exit drawdown threshold to 1.0%.

🔹 In recovery: close all when total profit >= this small positive = 10.0 — Sets the recovery close-all profit target to 10.0 USD.

🔹 In recovery: use trailing close instead of direct close = false — Keeps recovery-mode trailing closure disabled.

🔹 Recovery trailing start (USD) = 10.0 — Sets the recovery trailing-close activation level to 10.0 USD if enabled.

🔹 Recovery trailing stop (USD) = 1.0 — Sets the recovery trailing distance to 1.0 USD.

🔹 Risk multiplier applied in recovery mode = 1.2 — Sets the recovery-mode risk multiplier to 1.2×.

🔹 In recovery: allow assist entries to help close baskets faster = false — Disables additional assist entries during recovery.

🔹 Start assist entries when symbol basket <= -X USD = 4.0 — Sets the assist-entry activation level at a basket loss of -4.0 USD if assist entries are enabled.

🔹 Minimum seconds between assist entries per symbol = 60 — Requires at least 60 seconds between assist entries for the same symbol.

🔹 Minimum decision score for assist entry = 3 — Requires a minimum decision score of 3 before an assist entry can qualify.

Safe Mode Settings

🔹 Enable safe mode when balance is low = false — Disables the low-balance safe mode.

🔹 Balance threshold to enter safe mode = 100.0 — Sets the balance threshold for entering safe mode to 100.0.

🔹 Multiply risk by this in safe mode (e.g., 0.5 = half risk) = 0.5 — Configures safe mode to multiply risk by 0.5, equivalent to half the normal risk when active.

🔹 Limit max open trades in safe mode = 3 — Limits simultaneous open trades to 3 under safe mode.

🔹 Volume ratio multiplier for entries in safe mode = 1.0 — Keeps the safe-mode volume-ratio multiplier at 1.0.

Entry Signal: Consecutive Bars

🔹 Master switch for the only setup detector. false = no entries a… = true — Enables the consecutive-bar setup detector.

🔹 Allow SELL entries at all (reversal arms fire on bullish streak = true — Allows SELL entries, including reversal-arm behavior following bullish streak conditions.

🔹 Allow BUY entries at all (reversal arms fire on bearish streaks = true — Allows BUY entries, including reversal-arm behavior following bearish streak conditions.

🔹 BASELINE consecutive BULLISH bars; the entry-rule bandit shifts = 5 — Uses 5 consecutive bullish bars as the baseline bullish streak threshold.

🔹 BASELINE consecutive BEARISH bars; the entry-rule bandit shifts = 4 — Uses 4 consecutive bearish bars as the baseline bearish streak threshold.

🔹 Build synthetic tick-bars for the streak instead of timeframe b… = false — Disables synthetic tick-bar construction for the streak detector.

🔹 Ticks per synthetic bar. Smaller = far more frequent entry oppo… = 50 — Sets 50 ticks per synthetic bar if synthetic bars are enabled.

Volume Bandit Signals

🔹 Enable volume-based bandit signal logging and scoring context = true — Enables volume-based signal logging and scoring context for the bandit system.

🔹 Use tick volume (works in tester) instead of real volume = true — Uses tick volume rather than real exchange volume.

🔹 Reference minimum volume ratio for bandit scoring and telemetry = 0.7 — Sets the reference minimum volume ratio to 0.7.

Volume Signal Settings

🔹 Use tick volume to build bullish/bearish momentum signals for t… = true — Enables tick-volume analysis for bullish and bearish momentum signals.

🔹 Number of bars used to calculate average tick volume = 1000 — Uses 1000 bars to calculate average tick volume.

🔹 Current volume >= factor * average volume => “strong” bar = 0.3 — Sets the displayed strong-bar volume factor to 0.3.

DOM Order Book Bandit Signals

🔹 Enable Order Book (DOM) as a bandit signal source = true — Enables Order Book/DOM data as a signal source for the bandit system.

🔹 Hard bypass DOM. false = auto: use the book if the broker serve… = false — Keeps hard DOM bypass disabled, allowing the EA to automatically use DOM when supported by the broker.

🔹 Min imbalance ratio to confirm direction. Values <=1.0 filter n… = 1.3 — Requires a minimum DOM imbalance ratio of 1.3 to confirm directional conditions.

🔹 Minimum total liquidity (lots) required to trade = 1.4 — Requires at least 1.4 lots of total liquidity under the configured DOM filter.

🔹 Number of DOM levels to analyze (each side) = 30 — Analyzes up to 30 DOM levels on each side of the order book.

Bandit Smart Entry + Liquidity

🔹 Enables liquidity-zone execution. Required together with UseLim… = true — Enables liquidity-zone execution and activates the corresponding limit-order functionality.

🔹 Enables volume-profile liquidity zones used by bandit scoring a… = true — Enables volume-profile liquidity zones for bandit scoring and execution decisions.

🔹 Bars analyzed to build the liquidity profile used by the bandit = 2000 — Uses 2000 bars to construct the liquidity profile.

🔹 HARD max distance in points for liquidity-zone scoring AND for… = 4000.0 — Sets the maximum liquidity-zone distance to 4000.0 points.

🔹 Place liquidity-based BUY_LIMIT/SELL_LIMIT instead of market… = true — Enables liquidity-based BUY_LIMIT and SELL_LIMIT placement instead of relying exclusively on market execution.

🔹 Deletes stale liquidity limit orders when momentum weakens = true — Removes stale liquidity limit orders when momentum conditions weaken.

🔹 Max age for stale liquidity limit orders before renewal can rem… = 160 — Sets the maximum stale-order age to 160 units before renewal can be considered.

🔹 Minimum seconds between liquidity-limit renewal passes = 30 — Requires 30 seconds between liquidity-limit renewal passes.

Sentiment & Adaptive Settings

🔹 Enable sentiment/emoji feedback on wins/losses = true — Enables sentiment or emoji feedback based on winning and losing trade outcomes.

🔹 Show balance changes after each trade = true — Displays balance changes after each completed trade.

🔹 Enable adaptive trading based on sentiment/performance = true — Enables adaptive trading behavior based on sentiment and observed performance.