Scalperology AI

By William Harris · Founder, FxRobotEasy

The strategy behind Scalperology AI

Scalperology is the purest expression of our AI engine, and the fastest bot in the EASY line. Where our other systems add a structural gate on top of the AI signal — Breakopedia checks institutional pivot levels, Trendopedia checks live tick momentum — Scalperology deliberately removes those gates. When the model detects a high-confidence opportunity in the tick stream, the trade opens immediately. In scalping the edge exists for seconds; hesitation is the enemy. That directness is the entire design philosophy: the AI model itself is the filter, and speed is the edge.

It is a genuinely multi-instrument system. Scalperology looks for micro-patterns and price inefficiencies across FX majors, crosses, metals (including gold) and crypto, and trades whichever symbols currently carry a scored edge. Positions last seconds to minutes. This breadth is why it is the highest-frequency strategy we publish — and why broker execution quality matters more here than anywhere else in the line.

Capturing the signal is only half the job. The harder half — the half that kills most scalping bots — is knowing where to place the stop, where to take profit, when to move to breakeven, how to size, and how to adapt all of that when conditions shift next week. Scalperology solves this with two adaptation layers running at once, and this dual-layer architecture is the reason it stays viable where static bots decay.

Layer 1 — Cloud SET optimisation

Behind every running instance is a fleet of dedicated optimisation servers running the MT5 Strategy Tester continuously — across every supported symbol, at multiple depths, for multiple risk profiles. Each candidate parameter set (stop distance, take-profit, breakeven trigger, early-exit threshold, signal sensitivity and more) is scored on a composite of six metrics — Profit Factor, Sharpe, Drawdown, Trade Count, Win Rate, Recovery Factor — then filtered through hard gates that reject anything with too few trades, too little edge, or too much risk. The winners are published to the cloud catalog and delivered into your bot automatically. Critically, a new configuration replaces the current one only if it is strictly betteron the composite score — this is the “Best-Wins” rule, so the parameters your bot uses can only improve, never regress. The optimisation runs at three depths: Quick (~10 min, health check), Standard (~60 min, the daily workhorse), and Deep (2–4 h, full genetic exploration after a regime shift).

Layer 2 — per-account live adaptation

Cloud parameters describe how to trade a symbol in general. Your account is not “in general” — it has a specific broker, specific spreads, a specific execution speed and a specific balance. So each running instance also maintains a persistent state file (a .datin the shared MT5 Common Files folder) per symbol per account, tracking execution-quality EMA, edge-strength averages, trade-frequency EMA, recent P&L and a self-assessed regime state. The bot reads these before every trade: it tightens the breakeven trigger when your broker fills cleanly and widens it when execution degrades, sizes each position from your live balance, tightens the stop on textbook-quality entries, and cuts losers early when a trade moves against it immediately. Because the state lives in Common Files, it survives restarts, VPS reboots and EA reloads — the bot picks up where it left off rather than re-learning your account from scratch.

Risk management around all of this is deterministic and inspectable: a hard stop-loss on every trade, dynamic stop-loss, breakeven, trailing stop, early exit and a news filter that pauses trading around scheduled high-impact releases. No martingale, no grid, no averaging into losers. I keep the exit logic auditable on purpose — every additional exit rule increases the surface area for broker-execution quirks to interfere.

Recommended setup

Three things move Scalperology’s realised performance away from the published reference account: broker cost, execution latency, and which risk profile you run. Here is what I recommend for each:

Broker

• Execution model: ECN or raw-spread account with explicit commission ($3–7 per round-turn lot). STP/market-maker accounts work but cost you a real share of the edge in spread overhead — and on a high-frequency system that cost compounds fast.
• Server location: near your broker’s MT5 server. London (Equinix LD4) covers the high-liquidity London/NY window well; US-server brokers also work for the NY session.
• Scalping & multi-symbol policy: confirmed in writing. Some retail brokers throttle scalpers via slippage or restrict certain symbols (crypto, exotics); verify before depositing.

VPS

Strongly recommended — see the FAQ below for the full reasoning. Specifics: 2 vCPU / 4 GB RAM / 60 GB SSD on Windows Server 2019 or 2022, located in the same data centre as your broker’s MT5 server. Sub-20 ms ping is the target.

Risk profiles (Cloud SET)

Scalperology ships five professionally-optimised risk profiles rather than raw .set files you tune by hand:

• Conservative — 2–5 trades/day, patient and selective, lowest drawdown. Best for small accounts, prop-firm challenges and cautious traders.
• Balanced — 5–15 trades/day, the recommended starting point for most traders.
• Active — 15–30+ trades/day, high-frequency with tighter stops and faster compounding. For experienced traders on a fast VPS.
• Protected — 3–8 trades/day with all safety layers enabled. For volatile markets and strict drawdown rules.
• Precision — 10–25 trades/day with AI Precision Mode on, for surgical tick-level accuracy on major pairs.

You switch profiles from the EA’s input dialog without restarting MT5. And if you don’t want to choose, set the profile to Best: the system automatically delivers whichever profile currently has the highest quality score for each symbol.

Scalperology AI in MetaTrader 5

Backtest results and caveats

Scalperology has been backtested and forward-run across its full multi-symbol universe (majors, crosses, metals and crypto) on multiple ECN brokers’ tick data. I don’t pin a profit factor or a drawdown band in this text. Both are running figures that change with every closed trade and with the active profile, so any range printed here would be out of date before you read it — and a stale number that flatters the system is exactly the kind of marketing I am trying not to do. The current profit factor, drawdown, win rate and trade count are on the live account itself, in whatever state it happens to be in today. I deliberately publish that account on /live-trading rather than a backtest equity curve, because tick backtests still under-represent realistic spread widening during news, and we have all seen marketplaces that publish backtests that look like financial perpetual-motion machines.

Use the MT5 Strategy Tester (Modeling: “Every tick based on real ticks”, 90%+ modelling quality) to reproduce the backtest on your own broker’s historical data — see the Strategy Tester guide for the procedure. Your numbers will differ from mine by the magnitude of the cost difference between your broker and the brokers I tested on.

Risk profile

Even on the conservative profile, Scalperology can produce losing weeks. The typical pattern when it loses is a sequence of stops in a row during a period of low volatility or unusual session behaviour (extended Asia session, holiday-thinned liquidity, choppy post-event ranges). What to expect on a properly-sized account, and where to read the current numbers instead of trusting a figure typed into a review page:

These are descriptive characteristics of historical and reference-account behaviour, not a forecast, and the reference account has losing periods like any other. Live performance on your own account will differ — sometimes materially — based on broker execution, your chosen profile, symbol mix, and account size relative to the EA’s tested operating range.

Pros & cons

Pros

• Verified performance on a public broker-API-connected account.
• Trades directly on the AI signal across many symbols — the highest-frequency, fastest-compounding system in the line.
• Dual-layer adaptation: cloud parameters can only improve (Best-Wins), and the bot personalises to your broker and balance in real time.
• Deterministic, inspectable exit rules. No martingale, no grid, no averaging.
• Built-in news filter and five ready profiles from prop-firm-friendly to scaling-account aggressive.

Cons

• Highest broker-execution sensitivity in the line — needs an ECN/raw-spread account to perform to spec.
• High frequency means a VPS is effectively mandatory; your laptop’s reliability becomes the bottleneck otherwise.
• $500 minimum deposit is a hard math floor, not a marketing floor — smaller accounts (and thin margin headroom for concurrent symbols) will under-perform.
• Most demanding in our line. If you want lower-attention trading, look at Trendopedia AI or SessionSniper AI.

Scalperology vs the rest of the EASY Bots line

Scalperology AI FAQ