Pipsgrowth EX12024 MultiIndicatorConfluence

Pipsgrowth EX12024 is a deliberately thin EA: a wizard-generated MQL5 expert that wires the standard CExpert framework around a single time-of-day signal class (CSignalITF) and a fixed-lot money manager. There are no custom indicators in the source, no regime classifier, no multi-vote confluence logic, and no trailing stop. Every entry decision flows through the IntradayTimeFilter class, whose only inputs are hour-of-day, day-of-week, and a single confidence weight. The point of the EA is to be a clean, minimal skeleton — the kind of baseline you compare more elaborate EAs against, or the seed you grow a session-based strategy from.

The signal class is the entire strategy. CSignalITF is a CExpertSignal-derived class from the MQL5 standard library. It evaluates whether the current moment is inside or outside a user-defined trading window and emits a directional vote based on price action during that window. The window is configured by four inputs: Signal_ITF_GoodHourOfDay (default -1, meaning any hour qualifies), Signal_ITF_BadHoursOfDay (default 0, a bitmap where each bit represents an hour to exclude), Signal_ITF_GoodDayOfWeek (default -1, any day), and Signal_ITF_BadDaysOfWeek (default 0, day bitmap). The Weight input (default 1.0) scales how strongly the time signal contributes to the overall confidence score that CExpert uses to decide entries and exits.

Confidence is computed against two thresholds: Signal_ThresholdOpen (default 10) and Signal_ThresholdClose (default 10), both expressed on a 0-100 scale. When the accumulated weighted signals from CSignalITF cross the open threshold in a bullish direction, the framework submits a long. When they cross bearish, a short. When the reverse signal reaches the close threshold, the framework exits. Because the only signal source in this EA is CSignalITF and the only signal CSignalITF produces is its hour/day vote, the thresholds effectively become a debounce: the EA only acts when the time signal points cleanly in one direction and stays there. A threshold of 10 (out of 100) is the CExpert default and is intentionally permissive — raise it to 30-50 if you want fewer, higher-conviction entries.

Three additional signal inputs are exposed but do nothing in this configuration. Signal_PriceLevel (default 0.0) is the price level a pending order would use, Signal_StopLevel (default 50.0) sets the SL distance in points, and Signal_TakeLevel (default 50.0) sets the TP distance in points. Because no filter in this EA produces pending-stop or pending-limit orders, Signal_PriceLevel and Signal_Expiration (default 4 bars) are dormant parameters — they exist because CSignalITF inherits the full CExpertSignal interface, not because EX12024 uses them. The 50-point SL and 50-point TP, however, are wired into every market order the framework submits.

Money management is the second fixed component. CMoneyFixedLot is the simplest MQL5 money class: it always trades Money_FixLot_Lots lots (default 0.1), modulated by Money_FixLot_Percent (default 10.0). The percent value is the proportion of free margin the framework will commit to a single position; if the lot is too large for the account, the framework will scale it down automatically. There is no per-trade risk-percent calculation, no equity curve smoothing, and no scaling in or out of positions. Every entry is exactly the same lot size, which is the property that makes this EA predictable in backtest.

There is no trailing stop. CTrailingNone is explicitly included from the standard library, and the source confirms no trailing logic is attached. The only stop movement is the original 50-point SL set at entry, and the only TP movement is the original 50-point TP. On a 5-digit XAUUSD broker where a point is 0.01, 50 points equals $0.50 of adverse move per lot — a tight stop. On a 3-digit point broker, 50 points equals 5 pips. Either way, the 1:1 risk-to-reward geometry means the EA is implicitly a higher-win-rate, lower-payoff system. Expect many small wins offset by occasional 50-point losses.

Trade infrastructure is the third layer. The source includes three helper functions: TryClose_EX12024, TryClosePartial_EX12024, and TryModify_EX12024. Each uses the CTrade wrapper g_trade_wrappers_EX12024 and retries up to three times on transient errors. TryClose and TryClosePartial retry on TRADE_RETCODE_REQUOTE, TRADE_RETCODE_TIMEOUT, TRADE_RETCODE_PRICE_OFF, and TRADE_RETCODE_PRICE_CHANGED with 200ms between attempts. TryModify is asymmetric: it only retries on REQUOTE and TIMEOUT, sleeping 100ms between attempts. The asymmetry reflects an editorial choice — SL/TP modifications are less time-sensitive than position closes, so a slower retry is acceptable. There is no explicit SetTypeFillingBySymbol call; the EA relies on whatever filling mode MT5 auto-negotiates with the broker.

The brief header describes a 12-layer stack: regime, signal, entry, confirm, no-trade, capital cap, risk, sizing, manage, exit, scaling, OnTester. The honest read of the source is that only four of those twelve layers are wired: signal (CSignalITF), entry/exit (CExpert thresholds), sizing (CMoneyFixedLot), and manage (3-retry close/modify wrappers). There is no regime classifier, no separate confirmation filter, no news guard, no drawdown cap, no position-scaling logic, and no custom OnTester function. The magic number 22212024 is used by CExpert as the order tag.

In backtest, expect a high trade frequency because the time filter is the only gate and its default is to admit every hour of every weekday. The framework will fire whenever CSignalITF produces a 10+ confidence vote, which on a liquid symbol like XAUUSD with no other filters happens often. On M5 bars the EA can produce dozens of round-trip trades per day; on H1 the cadence drops by a factor of twelve. Position lifetime is bounded by the 50-point TP/SL on a 1:1 ratio, so most trades close within the same hour they open unless the SL/TP is widened. The combination of fixed lot, 1:1 R:R, and CSignalITF-only signal makes this EA a useful baseline — a sanity-check reference, not a finished strategy. To turn it into one, you would need to layer a regime filter (ADX or ATR-percentile), a real confirmation (RSI or trend EMA), and a real trailing stop, all of which the source does not include.