Pipsgrowth EX12055 MultiIndicatorConfluence

EX12055 is a TrendCatcher v1.1 build out of the MultiIndicatorConfluence family, and what sets it apart from its EX12 siblings is the second confirmation tier sitting on top of the base four-condition confluence. The base chain is the same as elsewhere in the family: a Heikin-Ashi direction read on the just-closed bar, an EMA(21) over EMA(50) crossover, a sign-bounded MACD(12,26,9) cross that requires the main line to be on the correct side of zero, and an EMA-slope tendency check (at least two of the last three fast-EMA prints must be rising for a buy, falling for a sell). On top of that, EX12055 layers an explicit higher-timeframe (HTF) confirmation via two more indicator sets pulled at the user-selected HTF — default H1 — covering both EMA(21/50) regime and an optional MACD(12,26,9) sign-bounded read, and a fully configurable Heikin-Ashi pattern filter that issues both confirm and reject signals in parallel. The HTF gate is strict: if HTF_UseEMA is true and the H1 fast EMA is not above the slow EMA, the buy arm is dropped; same for the optional HTF MACD, which only validates a buy if the H1 main line is above the H1 signal line and above zero. The HA pattern filter, gated by EnableHA_Filters, is a separate evaluation pass that builds a small window of HA bars on a configurable timeframe (default H1) and runs body-percentage, shadow-percentage, doji, color-flip, inside-bar, long-opposite-shadow, and color-run checks; the confirm block and the reject block are independent toggles, so a trader can demand a strong-bull HA bar for buys while only blocking entries that have an HA doji — or stack both. The full 44-input surface is partitioned across CORE, DLP, INITIAL vs ADDITIONAL, PAUSE GUARDS, ENTRY & FILTERS, HTF EMA, MACD EMA (HTF MACD), RSI, three HA filter groups (CORE, CONFIRM, REJECT), RISK & STOPS, and EXECUTION, and every group has its own INIT-time range check so a misconfiguration fails the EA at startup rather than mid-trade.

Signal generation runs only on a new bar — OnTick pushes a spread sample into the 20-tick sliding window first, applies DLP to every owned position, and only then checks IsNewBar() to advance the signal pipeline. GenerateSignal() returns SIG_BUY when the just-closed bar's HA close is above its HA open, the fast EMA is above the slow EMA, the MACD main is above the signal and above zero, the EMA slope tendency confirms, the optional RSI is at or above the buy threshold (RSI_BuyMin=45, off by default), the HTF EMA and HTF MACD (if enabled) both agree, and the HA pattern filter clears the buy arm; SIG_SELL is the strict mirror. SL and TP are calculated at send time from the live ask/bid — StopLossPoints=300 means a 300-point hard stop, TakeProfitPoints=600 means a 1:2 risk-to-reward by default. Both values are validated against SYMBOL_TRADE_STOPS_LEVEL and pushed one tick wider if they would land inside the broker's minimum-stops band. Lots are pulled from LotsByRisk(StopLossPoints), which divides account balance × RiskPercent/100 by the per-point value at the chosen SL, snaps to SYMBOL_VOLUME_STEP, and clamps to the symbol's min/max lot; if UseRiskPercent is off (the default), the lot falls back to FixedLot=0.10. SendMarketOrder walks three filling modes in order — FOK, IOC, RETURN — and on REQUOTE or PRICE_OFF refreshes the price; if the broker drops the inline SL/TP, the function locates the newly opened position and calls ModifyPositionSLTP() once to re-attach the protective orders.

Management of open positions happens every tick via ApplyDynamicLockProfitToPosition(). For each owned position the function computes movedPts from entry to bid/ask, divides by InpLockProfitEveryXPoints=150, takes the floor, and sets the new stop to entry ± (steps × 150 − InpLockMinusYPointsBuffer=20). The ratchet only ever tightens — the buy check refuses to send the modify if newSL ≤ curSL+0.5pt, and the sell side is symmetric — and it respects both SYMBOL_TRADE_STOPS_LEVEL and SYMBOL_TRADE_FREEZE_LEVEL, refusing to issue the modify at all if the new stop would land inside the broker's freeze band. The initial-versus-additional scale-in policy uses g_addCountBuy and g_addCountSell per-direction counters, allows up to MaxOpenTrades=5 open positions on the same symbol, and gates every additional entry on IsAllSameDirProfitableAtLeast() — every existing same-direction position must be at least MinProfitPerTradeToAdd=5.0 points in the green before an Add#N ticket will fire. Order comments are tagged explicitly: the first position on each side carries "Initial BUY/SELL" in the comment, and every subsequent same-direction fill carries "Add#1 BUY/SELL", "Add#2 BUY/SELL", and so on, which makes deal-level post-trade analysis trivial in MT5's history view.

Risk control has two independent pause mechanisms wired into OnTradeTransaction. Every DEAL_ENTRY_OUT that closes an owned position at a loss increments g_lossStreakCount and adds the absolute pnl (after swap and commission) to g_lossStreakAmt; if PauseOnConsecutiveLosses is enabled and the streak hits PauseConsecutiveLossesCount=3, the EA sets g_tradePauseUntil to now + 60 minutes via MathMax (so back-to-back triggers stack rather than reset). PauseOnLossAmount is the parallel dollar cap, default 100 USD over 60 minutes, also stackable. Both pause switches ship in the OFF state by default, so traders who want the EA to keep trading through a drawdown simply leave the inputs at their defaults; those who want the EA to step aside after a rough sequence can flip either toggle without touching the strategy logic. PreflightAllowsEntry() also blocks new entries when the time falls outside the 7-22 server-time session window (with midnight-wrap support), when the current spread exceeds 3.0× the median over the last 20 ticks, and when CountAllOpenPositions() has reached MaxOpenTrades.

The expected deployment is a $100 minimum-balance ECN account on XAUUSD M5 with a GMT+2 or GMT+3 server clock, which is the regime the H1 HTF confirmation was sized for. Below M5 the spread cap and HTF MACD become noise; above H1 the session window compresses. Backtests should be evaluated over at least one full seasonal cycle so the HTF EMA filter and the loss-streak pause guards have time to fire under both trending and ranging conditions. Sizing the lot down to the broker's minimum is a reasonable first step for traders who want to observe the strategy live before scaling in, and the magic=22212055 plus the Initial/Add#N comment tags make it easy to filter this EA's trades cleanly from the account history.