Pipsgrowth EX16087 Trend

A two-bar Japanese-candlestick pattern called Meeting Lines sits at the heart of this EA. The Meeting Lines setup is the classical Western interpretation of an Asian candlestick pair where two consecutive candles close at almost the same level, signalling that the previous directional move has stalled and the market is now turning. EX16087 is built around the bullish and bearish variants of that pattern, gated by a single momentum filter (CCI) and exited on a fixed bar-count deadline.

The EA scans every new bar of the running timeframe (M5 to H1, default M5) and looks for two adjacent completed candles whose closes sit within a tight tolerance of each other. Specifically, CheckPattern() first tests for the bullish variant: bar 2 must be a long black body (its open-to-close range exceeds the 12-bar average body length) and bar 1 must be a long white body of equal magnitude, with the two closes within 0.1 × AvgBody(1) of each other. The bearish mirror is the second test: a long white body on bar 2 followed by a long black body on bar 1, with the two closes again almost identical. The word 'long' here is measured relative to AvgBody(), which averages the absolute open-minus-close difference across the 12 most recent bars, so a candle has to be visibly larger than the recent noise to count as part of the pattern. This is what makes the pattern rare — a pair of large opposing bodies with near-identical closes does not appear every bar, even on XAUUSD.

When the pattern fires, the EA flags a tentative signal but does not open a position until the CCI(37) confirmation passes on the same bar. CheckConfirmation() pulls the most recent completed bar's Commodity Channel Index value using the applied price selected through InpPrice (default close, but configurable from 1=close through 7=weighted). A bullish Meeting Lines needs CCI(1) below −50 to be considered a confirmed buy, while the bearish variant needs CCI(1) above +50 to be confirmed as a sell. Those thresholds filter out pattern occurrences that appear in low-impulse environments and keep entries restricted to bars where momentum has actually been running against the prior trend. A pattern without confirmation simply clears the signal — there is no 'maybe' state carried into the next bar.

Once a pattern is detected and confirmed, PositionOpen() is called on the next tick. The order is sent at the current ask for a buy or current bid for a sell with a fixed 0.10 lot (InpLot). Risk is fully configurable by the trader — there is no automatic risk-percent sizing in this EA, no balance-based scaling, and no margin-utilisation check. Position sizing is whatever the lot input specifies, so a trader upgrading from 0.01 to 1.0 needs to think about contract value themselves.

The protective stop and target are both fixed in points through InpSL (default 200 points, which is 20 pips on a 5-digit broker) and InpTP (also 200 points). Both are inserted into the order ticket at entry, not maintained on a separate schedule. PositionOpen() applies a useful safety check: if the current spread is wider than the configured stop or target, the SL or TP distance is set to the spread itself rather than the requested value. This keeps the EA from sending obviously bad tickets during news spikes or low-liquidity transitions, although the trade-off is that the protection is then reduced to whatever the spread is at the moment of the order.

There is no trailing stop, no break-even function, and no partial-close logic wired into the live trading path. Three retry helpers — TryClose_EX16087, TryClosePartial_EX16087, and TryModify_EX16087 — exist at the bottom of the file with the standard 3-attempt, 100–200 ms sleep pattern for handling requotes, timeouts, and price changes. Only TryClose_EX16087 is invoked, from CloseBySignal() and CloseByTime(). The partial-close and modify helpers are defined but never called from the live OnTick path, so they would only fire if a developer wired them up manually.

A dedicated time-based exit is what separates this EA from a pure price-action reversal bot. After the position is opened, every tick checks how many bars have elapsed since entry. BarsHold() returns the number of completed bars between the position's open time and now using CopyRates on the chart symbol, and once that count reaches InpDuration (default 10 bars) the position is force-closed via CloseByTime(), regardless of whether price is in profit or loss. On M5 this works out to roughly 50 minutes of maximum exposure, on M15 to two and a half hours, and on H1 to ten hours. A second independent close path runs through CheckCloseSignal(): if CCI swings back across ±80 (for example, CCI(1) drops below 80 while CCI(2) was above 80, or crosses back through −80), the EA closes the open position. So a position can exit via four distinct routes — the CCI extreme unwind, the fixed TP, the fixed SL, or simply running out of bars.

Risk is one-sided: there is no portfolio-level drawdown cap, no daily loss limit, no news filter, no session filter, and no equity-flooring function. The closest safety gates are the spread-versus-SL/TP check at entry and the magic-number filter at every close loop, so the EA will only ever manage its own positions on the chart symbol with magic 22216087. The lot input is the only capital-allocation lever; everything else is deterministic from the bar structure and the CCI value.

A practical backtest expectation: the equity curve depends almost entirely on how often the 12-bar average body definition catches a real reversal rather than a coincidental near-close pair. The pattern is rare, so the EA will typically open only a handful of trades per day on M5. On a 1:1 reward-to-risk ratio with a 10-bar time stop, expect roughly half of closed trades to be small winners, with the remainder split between time-stop losses and SL hits. Traders looking to optimise further should treat the SL/TP pair and the time-out as a single tunable, because changing the timeframe quietly changes how long a position is allowed to live.