Pipsgrowth EX07005 MA

Pipsgrowth EX07005 MA targets XAUUSD on the M15 chart with a stripped-down triple-EMA alignment engine. Instead of waiting for a single moving-average cross, the EA demands that three EMAs (Fast = 5, Mid = 10, Slow = 20) all sit in the same vertical order — and it only fires on the bar where the alignment first forms. The configuration uses close prices, MODE_EMA smoothing, and a single M15 timeframe hard-coded into the OnInit indicator handles. That compactness is the design choice: the EA has no regime filter, no higher-timeframe bias, no news blackout, no session window, and no partial-close or trailing-stop machinery. What it does, it does in one step.

The signal engine is a transition detector, not a level test. GetTripleEMASignal() compares the current and previous bar's three EMA values. A long setup requires emaFast > emaMid > emaSlow on the current bar while that strict ordering was NOT true on the prior bar. Symmetrically, a short setup requires emaFast < emaMid < emaSlow on the current bar, transitioning from a non-aligned previous bar. If the alignment is already in place and continues bar after bar, the EA does nothing — there is no re-entry on every aligned bar. That is what makes the entry count low and the trigger precise: only the bar that flips the stack into alignment produces a ticket.

The SL placement is the EA's most distinctive design choice. In OpenPosition(), the code first proposes the slow-EMA price (emaSlow) as the stop anchor, then compares the implied stop distance to two fallbacks: the 15-pip minimum from StopLossPips and 1.5x the broker's minimum stop distance. The largest of the three values becomes the actual stop distance, so the SL is effectively the wider of (slow-EMA distance, 15 pips, broker min x 1.5). On XAUUSD at 15-pip resolution this usually means the slow-EMA anchor wins on most bars, and the stop sits behind the slow EMA rather than at a fixed pip count. TP is set at exactly twice the chosen stop distance (1:2 R:R) — not ATR-derived, not a multiple of price, just a hard 2:1 reward-to-risk on whatever stop the EA ends up using.

Risk sizing is the only branch in the EA that takes user-configured numbers seriously. CalculatePositionSize() reads RiskPercent = 2% of the account balance, divides it by the actual stop distance in pips times the instrument's per-pip tick value, floors to the symbol's lot step, and clamps to the symbol's min/max lot. When UseMoneyManagement is false, the EA falls back to FixedLotSize = 0.01 lots — a single micro-lot regardless of stop distance or balance. There is no equity-floor guard, no daily-loss cap, no drawdown circuit breaker in the source. The header advertises a '12-layer architecture' but the actual code only implements SIGNAL, ENTRY, NO-TRADE (just the spread filter), RISK (the 2% sizing), SIZING, and EXIT (SL/TP only). The REGIME, CONFIRM, CAPITAL CAP, MANAGE, SCALING, and OnTester layers from the brief are not actually wired up as functions in the source.

Trade management is intentionally minimal. The EA opens a single position per signal — CountPositions() returning any ticket with the EX07005 magic number blocks the next entry until the trade closes by SL or TP. There is no break-even ratchet, no trailing stop, no partial close, no time stop, and no opposite-signal exit. SafePositionModify() exists as a defensive wrapper for the broker's stop-level constraints, but the EA does not actually call it in the live flow because OpenPosition() sets the SL and TP at order entry and never modifies them again. The TryClose_EX07005(), TryClosePartial_EX07005(), and TryModify_EX07005() helpers do exist, but in this build they sit unused after the initial send.

The new-bar gate is the main brake on trade frequency. OnTick() first checks IsNewBar() — it reads iTime(_Symbol, PERIOD_M15, 0) and compares against lastBarTime; if the bar timestamp has not changed, the EA returns immediately. So all evaluation happens on the M15 bar open, not on every tick. Combined with the single-position cap, this means an aligned run on XAUUSD will typically produce one entry per several hours at most, with stretches of bars where the alignment never transitions. Backtests will show fewer trades than a typical M15 cross EA — expect low trade counts relative to the test period.

The no-trade filter is a single spread check. CheckFilters() reads SYMBOL_SPREAD, divides by pip size, and compares to MaxSpreadPips = 25 (configurable, default 25 pips for XAUUSD — generous, reflecting gold's typically wider spread). If the current spread is wider, the EA simply does not evaluate signals that bar. There is no time-of-day filter, no day-of-week filter, no economic-calendar awareness, no margin filter, and no minimum-balance guard. The EA will trade through news spikes, through the Asian-session lull, and through the broker's rollover window as long as spread is below 25 pips.

The order-execution layer uses MT5's CTrade wrapper with slippage = 3 points, filling mode auto-detected via SymbolInfoInteger(SYMBOL_FILLING_MODE) — preferring FOK, falling back to IOC, then RETURN. TryModify_EX07005() and TryClose_EX07005() both retry up to 3 times at 100-200ms on REQUOTE / TIMEOUT / PRICE_OFF / PRICE_CHANGED before giving up. Magic number 22207005 is set via SetExpertMagicNumber() and the trade comment is 'Psgrowth.com Expert_07005', so every order is identifiable in the journal and history.

What to expect in backtest. With M15 bars, 5/10/20 EMA alignment, and the slow-EMA-anchored stop, EX07005 produces a small number of high-conviction trades per week on XAUUSD. Most weeks will see 0-3 entries. The 1:2 R:R means winners need to be roughly half as frequent as losers for the strategy to be net positive. The 2% risk per trade is high for a scalping-style EA; consider setting RiskPercent = 0.5-1.0 for live use, and consider FixedLotSize = 0.01 for a flat-size test run. The lack of any daily-loss cap means a string of losers will compound — combine with a broker-level equity stop or a manual daily shutdown if you run it unattended.