Pipsgrowth EX02029 Breakout

Pipsgrowth EX02029 is a gap-fade Expert Advisor — a mean-reversion specialist that buys into overnight gaps down and sells into overnight gaps up on the assumption that the unfilled portion will be closed during the next session. Where most breakout EAs in the EX02 family ride a clean range break, EX02029 deliberately fades the move that gap traders try to monetise. The execution path is intentionally minimal: a single gap threshold, fixed stops, fixed targets, and a stepped trailing stop — there are no oscillators, no ATR, no Donchian, and no Heikin-Ashi overlays. The whole signal resolves from two candles and one parameter.

The core detection lives in OnTick(). After the standard new-bar guard (a static datetime PrevBars that compares iTime to the previous tick's value) and the IsSafeToTrade_EX02029 gate, the EA pulls the most recent two rates with CopyRates on the configured g_tf (default M5) and runs a single inequality check on each side:

if(rates[0].open < rates[1].low - ExtGap) → OpenBuy at ask, SL = ask - ExtStopLoss, TP = ask + ExtTakeProfit if(rates[0].open > rates[1].high + ExtGap) → OpenSell at bid, SL = bid + ExtStopLoss, TP = bid - ExtTakeProfit

The gap magnitude is set by InpGap (default 1 pip, scaled by the m_adjusted_point helper that compensates for 3-digit and 5-digit broker pricing), and only the most recent closed bar is referenced — shift-1 by construction. Both entry paths run sequentially inside the same tick, so the EA can in principle open a buy and a sell within one bar if the gap condition somehow fires both ways, though in practice the open/close structure of the previous bar usually prevents that. There is a deliberate sanity check before the buy fires: if the computed SL is at or above the current bid, the position would be in violation of the broker's stop-level minimum, and the EA bails by setting PrevBars=0 to force a re-evaluation next tick rather than send a malformed order.

Position management is built around three pips-based inputs. InpStopLoss defaults to 50 pips, InpTakeProfit defaults to 50 pips (so the EA ships at a 1:1 risk-reward out of the box), and InpTrailingStop + InpTrailingStep default to 5 pips each. The Trailing() function runs on every tick — not just new bars — and for each open position belonging to this EA's magic (22202029) and symbol, it checks whether price has moved more than ExtTrailingStop+ExtTrailingStep beyond the entry in the favourable direction. If yes, and the existing stop is still behind that new threshold, it modifies the SL to ExtTrailingStop behind current price via TryModify_EX02029, which retries the modify up to three times at 200ms intervals on REQUOTE, TIMEOUT, PRICE_OFF, and PRICE_CHANGED. Trailing only ever tightens in the profitable direction — there is no BE+lock step, no partial close, and no break-even pad. Once price has covered the trailing window the stop ratchets forward; if the SL never moves before TP, the trade closes at the fixed target.

The OnInit routine also enforces a strict invariant: InpTrailingStop!=0 with InpTrailingStep==0 returns INIT_PARAMETERS_INCORRECT with a localised Alert — trailing without a step would cause the modify call to fire on every tick and overload the trade server. That single guard catches the most common configuration mistake before any live tick. The remaining hardening runs in a separate module:

IsSafeToTrade_EX02029 is the all-in-one gate that wraps seven checks. IsMarketOpen_EX02029 blocks Saturday, Sunday, the first two hours of Monday (post-weekend rollover noise), and the last two hours of Friday. InActiveSession_EX02029 applies the optional London 7-16 / New York 12-21 GMT windows (gated by InpUseGMTSessions, which defaults to false so the EA is permissive by default). IsNewsTime_EX02029 queries MqlCalendarValue history for any HIGH-importance event in the ±30 minute window and blocks entries if found. The capital-protection check applies when InpCapAmount > 0 and refuses to trade if effective capital drops below InpCapFloor (1000 default). The equity-floor check refuses to trade if current equity is below g_initialBalance × InpMinEquityPct/100 (70% default). The daily-trade counter (g_tradesToday >= InpMaxTradesPerDay, default 10) and the cooldown timer (g_cooldownUntil > TimeCurrent()) round out the gate.

PnL bookkeeping happens in OnTradeTransaction. Every DEAL_ADD with the magic 22202029 on the current symbol updates g_realizedToday, g_realizedWeek, and g_tradesToday. Losing deals increment g_consecLosses; when that hits InpMaxConsecLosses (5 default), g_cooldownUntil is set to TimeCurrent() + InpCooldownHours*3600 (4 hours). UpdateDailyCounters_EX02029 rotates the daily and weekly counters when the day-of-month or month changes. The OnTester fitness function returns profit × profitFactor / maxDD, gated on a minimum of 10 trades — so the strategy optimiser rewards both absolute return and smoothness, with a hard floor on sample size.

The intended symbol per the runtime context is XAUUSD on M5 with a $100 minimum account, but the source is symbol-agnostic — the brief explicitly notes "FX Majors, Gold/Metals (portable)". Because the entire strategy resolves from open-to-open gap math with no volatility scaling, the only meaningful input for a different symbol is recalibrating InpGap (a 1-pip gap on EURUSD is small, on XAUUSD it is essentially noise, on NAS100 it is a major move). Traders who want fewer signals should raise InpGap; traders who want more should lower it. The EA is also timeframe-agnostic via the InpTimeFrame input, which routes through MapTimeframe_EX02029 — set it to 5 (PERIOD_H1) on less-liquid pairs where 1-pip gaps appear several times an hour on M5 and would otherwise over-trade.

Two practical caveats from the source. First, there is no spread filter and no slippage guard beyond the fixed 10-point m_slippage used in trade requests — a high-spread broker will eat most of a 1-pip edge before the order even lands, so this is strictly an ECN/RAW-spread deployment. Second, the 1:1 default RR means the strategy needs a fill rate above 50% to be net profitable, which a fade strategy on a noisy instrument may not achieve. Backtest on a tick-data-quality account before going live, and consider raising InpTakeProfit to 75 or 100 pips once the gap-fill behaviour is confirmed on your broker's data — the inputs are deliberately minimal so the trader can tune the RR without having to compensate for other built-in assumptions.