Pipsgrowth EX16030 Trend

Pipsgrowth EX16030 Trend is a 2-bar Harami continuation pattern trader that adds a Relative Strength Index confirmation gate and a 5-period simple moving average trend filter on top. The class is the inside-bar / containment branch of the BullishBearish family, paired here specifically with the iRSI(37) oscillator for confirmation. Harami — Japanese for 'pregnant' — is the geometry where the second candle's body sits fully inside the first candle's body. Where a Bullish Engulfing pattern requires the second body to swallow the first from outside, Harami is the inverted relationship: after a long, dominant candle in the prior direction, the next bar is small and the market has effectively paused or rejected the prior move inside its body. EX16030 looks for that exact stillness, then asks the RSI and the 5-SMA whether the pause is a real reversal or a one-bar pause before continuation.

The pattern detection in CheckPattern() enforces five conditions for each side. For a Bullish Harami on a downtrend: bar 1 must be bullish (Close(1) > Open(1)), bar 2 — the prior bar — must be a long bearish candle with body size larger than the 12-bar rolling average body size AvgBody(1), the bullish bar's close must be below the prior bearish bar's open, the bullish bar's open must be above the prior bearish bar's close (containment), and MidPoint(2) = (High(2)+Low(2))/2 must be below the 5-SMA CloseAvg(2) — confirming the broader trend is still down. The bearish mirror flips every comparison. That trend filter pulls wicks: a long lower wick on the prior bar softens the downtrend read even if the prior body is well above the SMA. This is the distinctive choice versus the body-midpoint MidOpenClose(2) filter used in the engulfing siblings, and it changes the actual setups that qualify.

Confirmation runs through CheckConfirmation() reading the iRSI handle at bar 1. A Bullish Harami only fires if RSI(1) < 40 — the prior momentum oscillator must still be in the lower half of its range, confirming that the prior bearish leg was strong and the inside-bar pause is a real exhaustion, not just a routine pullback inside an uptrend. A Bearish Harami only fires if RSI(1) > 60. Those 40/60 bands are deliberately inside the standard 30/70 oversold/overbought zone — they require the pattern to do most of the directional work, with RSI only validating that the prior move actually stretched the oscillator into the meaningful region. This is the same band layout the MFI confirmation uses in the Harami-MFI sibling, but RSI brings a pure price-momentum read with no volume weighting and no unbounded deviation like CCI.

Entry is a market order via ExtTrade.Buy() or ExtTrade.Sell() at the next available ask or bid, with both stop-loss and take-profit set inline. PositionOpen() builds the stop with a spread-override guard: if the current spread is wider than the configured InpSL or InpTP (in points), the EA falls back to a spread-priced exit instead of submitting an obviously-doomed order. That guard runs on both legs — a deliberate choice to keep the EA from sending the broker a position with negative expected distance to SL or TP at the moment of entry. Lot is fixed at 0.10 from InpLot. The OnInit() function routes the order object through ExtTrade.SetDeviationInPoints(InpSlippage) = 10 and SetExpertMagicNumber(22216030), so any orphan orders from a different instance or magic are ignored by the position-existence check.

Trade management follows the two-leg framework the family uses everywhere: a reversal-pattern close via CheckCloseSignal() reading RSI crosses at 70/30 (four combinations: long-overbought exit, long-oversold-plunge, short-oversold exit, short-overbought-recovery), plus a time stop via CloseByTime() checking BarsHold(open_time) >= InpDuration (10 bars). There is no trailing stop, no break-even step, and no martingale or grid layer — the EA exits either on pattern invalidation, on the 200-point SL, on the 200-point TP, or after 10 bars, whichever comes first. Every close path uses the 3-attempt TryClose_EX16030() retry wrapper, which re-submits on REQUOTE, TIMEOUT, PRICE_OFF, and PRICE_CHANGED with a 200ms backoff. TryClosePartial_EX16030() and TryModify_EX16030() are defined but never wired into the live call sites — they are residual scaffolding from the template.

The header claims a 12-layer architecture (REGIME, SIGNAL, ENTRY, CONFIRM, NO-TRADE, CAPITAL CAP, RISK, SIZING, MANAGE, EXIT, SCALING, OnTester), but the actual code wires only the operational subset: REGIME via the MidPoint-vs-SMA trend filter, SIGNAL via the 5-condition Harami detection, ENTRY via market order, CONFIRM via the RSI<40 / RSI>60 band check, NO-TRADE via the next-bar-open guard in OnTick(), RISK via the 200-point SL with the spread-override guard, SIZING via the fixed 0.10 lot, EXIT via the RSI cross + 10-bar time stop + SL/TP. CAPITAL CAP is not a balance-aware risk cap; it is just a per-trade lot input. MANAGE has no trailing/BE layer. SCALING and OnTester are not present in the file. Counting only the wired layers, this is closer to a 9-layer build dressed in the family template's 12-layer jacket.

Several template-sibling defects have been carried over. MapAppliedPriceInt() is defined four times in the source (lines 37-50, 58-71, 78-91, 95-108) with the matching g_InpPrice declarations also duplicated four times — the compiler accepts the redefinition but the redundancy is wasted source. The RSI init error message on line 151 still reads 'Error creating CCI indicator' despite the actual call being iRSI — a stale string from the CCI-template sibling. The MA init on line 157 validates ExtIndicatorHandle instead of ExtTrendMAHandle, so the MA's INVALID_HANDLE is never actually caught by its own check (it is only caught by accident if the RSI handle fails first). The OnDeinit release on line 172 only frees ExtIndicatorHandle and forgets ExtTrendMAHandle — the SMA handle leaks on EA unload. The two debug Print('InpSL=', InpSL) and Print('InpTP=', InpTP) lines on 139-140 fire on every EA init. These are real code-hygiene issues, not marketing defects, and any of them is worth a one-line patch.

This is the RSI leg of the Harami + oscillator family. The MFI leg is EX16029 (with the 40/60 confirmation bands shared). EX16024 through EX16027 form the parallel engulfing + oscillator cousins (CCI, MFI, RSI, Stoch), so EX16030 is the first Harami-RSI cross and the second Harami-family variant overall. Each runs on its own magic (22216030 here) and is intended to be deployed as one independent unit — different magics, different order books, different on-chart footprints, so an operator can run several siblings on the same symbol and compare setups without position-book interference. The RSI(37) period is deliberately slow for M5: a 37-period RSI on a 5-minute chart covers roughly 3 hours of tape, which makes the oscillator slow enough that a single spike cannot invalidate the read. The 200/200 SL/TP pair is a flat 1:1 risk-reward, so the EA is essentially a higher-probability-reversal scalper with time discipline — the inside-bar geometry is the higher-probability claim, the 1:1 R:R is the math, and the 10-bar cap is the discipline that keeps losing streaks from running away.