SurfBot Bidirectional Forex Grid
- Experts
-
Joao Prata Silva Yglesias
Trader and algorithmic developer focused on risk-managed automation for the Forex market. I build Expert Advisors that prioritize capital protection and survivability over hype — real strategies, transparent results, and live monitoring. - Versione: 2.7
- Aggiornato: 14 ottobre 2026
- Attivazioni: 5
SurfBot — one-direction Forex grid with hard capital caps. No stop-loss per trade, by design.
🔴 Live performance (verified, real account, this exact engine): https://www.mql5.com/en/signals/2383326
WHAT IT DOES
Runs one grid per pair, on ONE side only (long or short). The side is chosen by the better swap (GridSide=AUTO) or forced by you. Each layer has its own take-profit one spacing away; there is no stop-loss per trade. Spacing comes from a built-in table per pair (percent of price), widened after the 4th layer; a pair outside the table runs with FixedSpacingPct and a warning in the log. Pending orders trail the price when the grid is empty.
WHAT KEEPS IT ALIVE (defense layers)
- Per-pair cap: max 12 layers (MaxTotal).
- Per-pair loss cap: floating loss beyond N% of starting balance freezes adds and cancels pendings (PairLossFreezePct). It never closes positions.
- Per-currency cap: net layers per currency across the whole account (MaxNetLayersPerCcy).
- Account floor — the one hard stop: equity below N% of starting balance (EquityFloorPct, default 90 = -10%) closes the grid of the pair at market, cancels its pendings and stays frozen until you re-attach the EA.
- Margin freeze (MarginFreezePct), free-margin check before every order, spread filter (MaxSpreadPts).
WHAT IT IS NOT
- It is not a "safe" EA. With no stop per trade, a long one-way trend accumulates up to 12 losing layers per pair until the account floor closes them at a loss.
- It does not predict direction. The edge is the swing plus the swap of the chosen side.
PUBLISHED STRESS TEST (MT5 Strategy Tester, demo, 7 live pairs, 0.04 lots on USD 16,000)
Five crisis windows 2021–2025 (Aug–Dec 2021, Aug–Nov 2022, Jan–Apr 2023, Jun–Sep 2024, Mar–Jun 2025):
- No stop-out in any window.
- Worst equity drawdown: 60% (2025 tariff shock); 54% (2024 JPY unwind).
- Longest recovery: 116 days.
- Calm years profitable, crisis years negative. 2015 (SNB) and 2020 (COVID) are NOT covered.
Size for that: 0.01 lot per USD 10,000 keeps the same test under 25% drawdown.
RECOMMENDED SETUP (what the author runs, published live as signal 2383326)
Pairs: EURGBP (sell), USDSGD, USDCAD, GBPCAD, USDJPY, USDCHF (buy), EURUSD (sell). Timeframe: any (the grid is price-based); H1 charts. Hedging account, leverage 1:100 or higher, low-swap broker. Minimum: USD 4,000 for 0.01 lots on 6–7 pairs. Runs on the MQL5 VPS. Always start on a demo account.
FREE COMPANION DASHBOARD (web + Garmin watch)
SurfBot Export + surfbot.trade show your real result, net of commission and swap, in the browser and on your watch. Setup: https://surfbot.trade/dashboard-setup (access key provided after purchase).
INPUTS YOU MUST UNDERSTAND BEFORE GOING LIVE
GridSide, LotSize, MaxTotal, PairLossFreezePct, MaxNetLayersPerCcy, EquityFloorPct, UsePairTable / AllowTier2 / AllowUnlisted, FixedSpacingPct, WidenAfter / WidenFactor. All inputs are open and documented. No black box.
Trend hedge (HedgeEnabled) and TP promotion (PromoTPEnabled) are OFF by default and are not part of the stress test above.
Version 2.06 is a new engine (magic 880001). Positions opened by v1.x (magic 770001) are NOT managed by it: close or finish them with v1.x before updating a live account.
⚠️ Risk warning: Trading leveraged instruments involves a high risk of loss. Grid strategies can accumulate significant adverse positions in prolonged trends. Past or backtested results do not guarantee future performance. Use only risk capital you can afford to lose.
Website and the full stress-test report: https://surfbot.trade
