Pipsgrowth EX16089 Trend

Pipsgrowth EX16089 Trend is a price-action + oscillator system that runs entirely on the candlestick Meeting Lines reversal pattern. The detector in CheckPattern() reads the last two finished bars, requires both to be long-bodied in opposite directions, and requires their closing prices to land within ten percent of a running 12-bar average body. The check runs only on the open of a new bar, so a single new-bar gate (next_bar_open) is the only timer the EA maintains inside OnTick().

The body length is compared against AvgBody(1), which is the 12-bar average of |Open - Close| taken on the current bar. That 12-period lookback is the InpAverBodyPeriod input. If the second bar's body is the opposite color from the first, both bodies exceed the running average, and the closes of bars 1 and 2 differ by less than 0.1*AvgBody(1), the pattern is considered a Meeting Lines reversal. Bearish Meeting Lines need bar 2 to be a long white candle followed by a long black candle at bar 1. Bullish Meeting Lines need bar 2 to be a long black candle followed by a long white candle at bar 1. When CheckPattern() returns true, the EA arms ExtSignalOpen to SIGNAL_BUY or SIGNAL_SELL and writes a short pattern info string.

The pattern alone is not enough. CheckConfirmation() then queries the RSI(37) indicator at bar 1 and demands the buy signal be accompanied by RSI below 40, or the sell signal be accompanied by RSI above 60. Only when ExtConfirmed is set does PositionOpen() get called on the next phase of OnTick(). The RSI uses the close price by default (InpPrice = 1 maps to PRICE_CLOSE through the MapAppliedPriceInt switch).

Position sizing is a single fixed lot: InpLot = 0.1. There is no risk percentage, no per-trade dollar cap, no per-day cap, no equity floor, no scaling, and no OnTester() pass. The fixed lot is the only money-management input the source exposes. PositionOpen() refreshes the symbol via CSymbolInfo, reads Ask/Bid, computes a stoploss and takeprofit from InpSL = 200 and InpTP = 200 points, and routes the order through CTrade.Buy or CTrade.Sell with the slippage tolerance set to 10 points via ExtTrade.SetDeviationInPoints.

There is one piece of safety built into the order router that the brief description does not flag. If the current spread is wider than InpSL points (or wider than InpTP points), the EA refuses to place a stop at a distance narrower than the spread. Instead, it installs the stop at price ± spread. This is a guard against getting stopped out instantly on a wide-spread candle (typical around news or session open on XAUUSD) and is logged via PrintFormat. The same guard is mirrored for both long and short, so the protective distance never collapses below the current spread.

Exit management runs as the third and fourth phases of OnTick(). Phase 3 handles CloseBySignal, which closes positions based on four RSI crossover conditions computed by CheckCloseSignal(): a long position is closed when RSI(2) > 70 and RSI(1) < 70 (overbought exit), or when RSI(2) > 30 and RSI(1) < 30 (momentum loss exit). A short position is closed when RSI(2) < 30 and RSI(1) > 30, or when RSI(2) < 70 and RSI(1) > 70. Phase 4 is the time exit: CloseByTime() walks open positions for the current symbol+magic, calls BarsHold() to count bars since the position was opened, and if that count is greater than or equal to InpDuration = 10, closes the position via TryClose_EX16089. The bar count comes from CopyRates, so a server-time vs bar-time mismatch on a weekend or holiday is handled gracefully (the function returns -1 in that case and the time exit still triggers).

All three trade-side close helpers (TryClose_EX16089, TryClosePartial_EX16089, TryModify_EX16089) use a 3-attempt retry loop with a 200ms sleep on TRADE_RETCODE_REQUOTE, TRADE_RETCODE_TIMEOUT, TRADE_RETCODE_PRICE_OFF, and TRADE_RETCODE_PRICE_CHANGED. TryClosePartial_EX16089 and TryModify_EX16089 are defined in the source but never called from the live OnTick path — they exist for future use or were left over from refactoring. Only TryClose_EX16089 is invoked from CloseBySignal and CloseByTime.

The source header advertises a 12-layer stack (REGIME, SIGNAL, ENTRY, CONFIRM, NO-TRADE, CAPITAL CAP, RISK, SIZING, MANAGE, EXIT, SCALING, OnTester). In the actual code, only the SIGNAL, ENTRY, CONFIRM, NO-TRADE, and EXIT layers are wired. There is no regime classifier, no daily P&L circuit breaker, no capital cap input, no risk-percentage input, no position-scaling module, no trailing stop, no breakeven, no partial close, no session filter, and no news filter. The EA does not implement OnTester() either, so optimizer runs can be set up, but the EA does not consume a custom optimization criterion.

Practically, this means the EA trades a fixed 0.1 lot on every confirmed Meeting Lines + RSI(37) signal, holds the position for at most 10 bars, and exits on SL, TP, RSI crossover, or the time gate — whichever comes first. The symmetric 200pt SL and 200pt TP (a one-to-one risk-reward profile) means the long-term expectancy of the strategy depends almost entirely on the pattern's win rate; there is no asymmetric payoff to compensate for sub-50% accuracy. The 200pt protective distance on XAUUSD is roughly 2.0 price points (since the symbol uses 2 decimal places on a 1-point scale: 200 points = $2.00 on a standard lot), which puts the typical risk per trade at $20 per 0.1 lot. The 12-bar holding window on M5 is one hour of price action; on H1 it is twelve hours.

For traders looking to use this EA live, the realistic expectations are: a small number of trades per day on M5, tight slippage and spread discipline matter because the SL/TP distance is only 200 points, the RSI(37) lag is significant on M5, and there is no global equity protection. If you want a global safety net, pair this EA with a separate equity-stop utility or a hard daily-loss limit enforced at the account level. Backtest the pattern detector with realistic spread assumptions — the spread-safety guard in PositionOpen() will reduce the effective SL in wide-spread environments, which is usually a feature but will look like a tighter stop in a backtest report.