Pipsgrowth EX18117 TrendFollow

Pipsgrowth EX18117 is a USDJPY M5 multi-indicator scalper built around a closed-bar EMA(9)/EMA(21) crossover that has to clear a 13-gate consensus filter before it opens anything. The file ships with magic 22218117, a fixed 1.0 percent risk per trade, and fourteen MetaTrader 5 indicator handles, all of which are released in OnDeinit: two EMA handles on the current timeframe, eight Linear Weighted Moving Average handles used to reconstruct the True Hull MA on both the current timeframe and the chosen higher timeframe, and four filter handles (RSI 14, ADX 14, ATR 14, MACD 12/26/9).

The spine of the entry logic is the True Hull MA. Because MQL5 cannot express a WMA of an indicator buffer dynamically, the EA approximates the textbook Hull MA in CalculateHullMA: it computes rawHull = 2*LWMA(n/2) - LWMA(n) for each of the last sqrt(n) bars and then averages those values together. The result is cached per bar in UpdateHullValues for four series (fast current-TF, slow current-TF, fast HTF, slow HTF). On every closed bar the EA compares the closed EMA(9) and EMA(21) values from bar 1 and bar 2; if the fast line crossed above the slow line, the system runs ValidateBuyFilters, which demands True Hull fast > slow on both timeframes, an EMA slope of at least 0.05 points per bar (ruling out flat markets), no proximity within 50 points of the daily R1 or R2 pivot (so the entry is not parked at obvious resistance), tick volume at or above 0.8 times its 20-bar average, RSI under 75, ADX at or above 20, MACD histogram above zero, ATR(14) at or above 0.0005, current spread under 60 points, server hour between 2 and 21, and — if the news filter is enabled — no USD or JPY high-impact event within the configured 30 minutes before / 15 minutes after window. Sells run ValidateSellFilters, which is the mirror image (RSI above 25 instead of below 75, MACD histogram negative, S/R check against S1 and S2).

Sizing is done by CalculateLotSize. When InpFixedLot is zero, the EA divides account balance times InpRiskPercent by the notional stop distance in ticks, floors to the broker volume step, and clamps between min and max lot — a standard 1 percent risk formula. When InpFixedLot is non-zero, that value is used as-is. The 150-point stop and 250-point take profit are passed into the order when it is opened, and the original TP is never modified: only the stop is ratcheted.

The position-management loop in ManagePositions runs every tick. After 50 points of profit, the stop is moved to entry plus 10 points (the breakeven offset), and once the trade is in profit by 50 points or more, a 50-point trailing stop takes over and follows price forward. The ratchet is one-way — the stop only moves up for buys and down for sells, never backwards — and the trailing stop only tightens, never widens, the protective floor. Pyramid entries are handled separately in CheckPyramidOpportunity: the EA looks for an open position on the side it intends to add to, requires the most recent same-direction position to be in profit by at least 30 points, requires every open position on that side to still be in the green (AllPositionsProfitable), and re-runs the full filter stack before opening the next leg. Lots for the pyramid leg are scaled by MathPow(InpPyramidMultiplier, currentCount), so with the default multiplier of 1.0 every leg is the same size; setting it to 1.5 gives a geometric ramp. CheckPyramidOpportunity only fires for one side at a time and only up to InpMaxPyramid/2+1 legs per side, even though InpMaxPyramid is set to 5 by default — the practical pyramid cap per direction is 3, with the total exposure cap of 5 between both sides.

Exits are layered. CheckExitSignals runs after the pyramid check and looks for the reverse EMA crossover, which closes every open position on the opposite side (a bullish cross while sells are open closes all sells, and vice versa). RSI extremes are also exit triggers: if RSI climbs above 75 with the trade already 50 points in profit, the long is closed via ClosePositionsAtExtreme, and the mirror logic closes shorts at RSI under 25. The same ClosePositionsAtExtreme routine wraps the close in a 3x200ms retry via TryClose_EX18117 to handle requotes and price changes. The OnTick function is wrapped in a portfolio guard: IsDrawdownExceeded updates g_peakEquity whenever equity climbs to a new high (and never resets it) and short-circuits new entries when current equity is more than 20 percent below that peak — but ManagePositions keeps running, so trailing stops and breakeven adjustments continue to protect open positions.

The visual layer is two-tiered. CreateDashboard builds a 200-by-230 dark-slate panel pinned to the top-left with 13 labels (TITLE, SIGNAL, EMA, HULL, RSI, ADX, PIVOT, POS, PROFIT, SPREAD, VOLUME, SESSION, NEWS) that are redrawn on every tick by UpdateDashboard. The SIGNAL label takes its cue from the True Hull MA direction on the current timeframe plus the EMA ordering, not the crossover itself, so it shows a smoothed bias rather than the discrete event. The PIVOT label prints the daily PP, and SESSION / NEWS labels toggle between ACTIVE / PAUSED and OK / BLOCKED depending on the time-of-day and event filters. CreateChartVisuals and UpdateChartVisuals draw the daily pivot HLINEs (PP in gold, R1/R2 in red, S1/S2 in green) and a cyan / magenta dashed HLINE for the live True Hull fast and slow values; a CHART_SIGNAL_ARROW object is created at (0,0) and never moved, so the arrow itself does not function as a real signal marker despite the name.

There is a small but real quirk in the source: the helper functions MapTimeframeInt, MapAppliedPriceInt, and the global variable g_InpHTF are declared twelve times each in the input block, once per input group, because the file was assembled by pasting the same group scaffolding. MQL5 absorbs the duplicate declarations without error. The InpHTF input itself is parsed in OnInit, so the value set in the panel does drive the timeframe used for the higher-timeframe Hull — the redundant g_InpHTF assignments are dead.

Pyramiding is the clearest place where the risk profile diverges from a flat scalper: a single EMA crossover that clears the entire filter stack can turn into three open buys within a few bars if price keeps trending, each at the same lot size. The 20 percent drawdown cap is also a peak-equity-never-reset rule, so a long, slow grind back to the high is required before new entries resume after a drawdown pause.