AI Trading CampAI Trading Camp
RoboticEA

RoboticEA Features 2026: Complete Breakdown

As of March 2026

Visit RoboticEA
RoboticEA screenshot
RoboticEA
Feature Overview
FeatureStatus
34 strategy archetypes across 6 families, tuned by hand or by AI from a plain-English description
Batch generation: 10-50 candidate strategies per run, every one backtested
Around 20,000 named presets plus 74 tunable native strategy types
Visual schematic builder — every indicator, threshold, exit and sizing rule is an editable block
C++ backtest engine on real 2020-2026 data, 25 instruments across 6 timeframes
Fixed $10,000 starting account on every run, so results are directly comparable
Real costs charged and reported: commission, swap and average spread as line items
Per-year validation grid — every year re-run separately with an 'N of 7 profitable' tally and worst drawdown
Automatic validation of batch winners once they clear a return threshold
Variation refiner: up to 50 perturbed copies tested against a robustness invariant, fragile winners flagged not promoted
Bar-by-bar trade replay on real OHLC candles with SL/TP lines and a synced equity curve at 1x-8x
Automatic personal leaderboard, filterable by symbol, timeframe, year, cost preset and risk
MetaTrader 5 export as readable .mq5 source with a preview, not a compiled binary
Weekend-flatten and rollover-avoidance rules exported into the EA as inputs
Published engine-vs-MetaTrader-5 parity study with the disagreement quantified
MQL5 export included on the free tier
Features in Detail

34 strategy archetypes across 6 families, tuned by hand or by AI from a plain-English description

Batch generation: 10-50 candidate strategies per run, every one backtested

Around 20,000 named presets plus 74 tunable native strategy types

Visual schematic builder — every indicator, threshold, exit and sizing rule is an editable block

C++ backtest engine on real 2020-2026 data, 25 instruments across 6 timeframes

Fixed $10,000 starting account on every run, so results are directly comparable

Real costs charged and reported: commission, swap and average spread as line items

Per-year validation grid — every year re-run separately with an 'N of 7 profitable' tally and worst drawdown

Automatic validation of batch winners once they clear a return threshold

Variation refiner: up to 50 perturbed copies tested against a robustness invariant, fragile winners flagged not promoted

Bar-by-bar trade replay on real OHLC candles with SL/TP lines and a synced equity curve at 1x-8x

Automatic personal leaderboard, filterable by symbol, timeframe, year, cost preset and risk

MetaTrader 5 export as readable .mq5 source with a preview, not a compiled binary

Weekend-flatten and rollover-avoidance rules exported into the EA as inputs

Published engine-vs-MetaTrader-5 parity study with the disagreement quantified

MQL5 export included on the free tier

Key Strengths
  • Per-year validation is built in and automatic — the single best defence against a robot that had one lucky year
  • The robustness invariant in the refiner actively rejects fragile high-scoring variations instead of promoting them
  • Real spread, commission and swap are the default and are reported as charged amounts, not hidden in assumptions
  • Exports readable .mq5 source you can audit and edit, rather than a sealed .ex5
  • Grid and martingale exist as explicitly labelled archetypes rather than undisclosed behaviour
  • Fixed $10,000 account on every run makes every result on the platform directly comparable
  • Publishes a parity study against MetaTrader 5 and quantifies its own disagreement
  • Genuinely usable free tier that includes MQL5 export and no card requirement
Limitations
  • Steeper learning curve than a signal tool — you are expected to understand drawdown, profit factor and out-of-sample testing
  • Token-metered billing means heavy batch generation and multi-year validation consume allowance quickly
  • Single-year backtests on Starter; multi-year runs in one job need the Unlimited plan
  • Focused on MetaTrader 5 export — no MT4, cTrader or TradingView output
  • Backtesting and export only: it does not execute or manage live trades for you
  • Support is email-based; no live chat or phone
RoboticEA vs Competitors
Feature
R
RoboticEA
C
ChartSnipe
T
TrendSpider
T
TradingView
34 strategy archetypes across 6 families, tuned by hand or by AI from a plain-English description
Batch generation: 10-50 candidate strategies per run, every one backtested
Around 20,000 named presets plus 74 tunable native strategy types
Visual schematic builder — every indicator, threshold, exit and sizing rule is an editable block
C++ backtest engine on real 2020-2026 data, 25 instruments across 6 timeframes
Fixed $10,000 starting account on every run, so results are directly comparable
Real costs charged and reported: commission, swap and average spread as line items
Per-year validation grid — every year re-run separately with an 'N of 7 profitable' tally and worst drawdown
Starting PriceFreeFree$59/moFree
FAQ
What is RoboticEA?
RoboticEA is a no-code platform for building, backtesting and exporting your own MetaTrader 5 expert advisor. You compose a strategy from 34 archetypes or pick from around 20,000 presets, the C++ engine backtests it on real 2020-2026 data with spread, commission and swap charged, validation re-runs it on every year separately, and the winner exports as readable .mq5 source you compile in MetaEditor. It does not trade for you — it builds the robot that does.
Do you need to know MQL5 to use RoboticEA?
No. The entire build is visual: indicators, entry conditions, exits and sizing are editable blocks in a schematic, and the export generates the MQL5 for you. That said, you do receive readable source rather than a compiled binary, so anyone who can read code is free to audit or modify the EA before running it — which is the main practical advantage over buying a sealed .ex5 from a marketplace.
How does RoboticEA validate a strategy?
It re-runs the strategy on every year of its instrument as separate, replayable backtests and presents a grid showing mean percent per year, an 'N of 7 profitable' tally, the range from worst to best, the worst drawdown, and a green or red chip per year. This runs automatically on batch winners that clear your return threshold. The variation refiner adds a second layer: up to 50 perturbed copies tested against a robustness invariant, with variations that break the invariant flagged FRAGILE rather than promoted.
Does RoboticEA include real trading costs in backtests?
Yes, and this is one of its strongest points. Real spread is the default cost preset rather than a frictionless zero, and the results panel reports commission, swap and average spread as charged amounts alongside return, CAGR, Sharpe, Sortino, Calmar, max drawdown, profit factor, win rate and expectancy. Every run starts from a fixed $10,000 account, which makes results directly comparable across strategies.
Can you export RoboticEA strategies to MetaTrader 5?
Yes, and MQL5 export is available even on the free tier. You get a readable .mq5 source file with a preview before download. Installation is four steps: place the file in your terminal's MQL5\Experts folder, open it in MetaEditor, compile with F7, then attach it to a chart and enable AutoTrading. Account-level rules such as flatten-before-weekend and avoid-daily-rollover are written into the exported EA as inputs.
Will the MetaTrader 5 results match RoboticEA's backtest?
Closely, but not exactly, and the vendor says so directly — your broker's feed, spreads and swaps are not the ones in the backtest. RoboticEA publishes a parity study on 100 EURUSD strategies run through both its engine and MetaTrader 5 on identical bars: 95% land within 5% on trade count (97% for ordinary entry/exit strategies), with a median trade-count gap of 0.3% and a median return gap of 1.1 percentage points. Position-stacking strategies such as grids and martingales account for most of the remaining disagreement.
How much does RoboticEA cost?
There is a free tier with a one-time grant of 250 tokens and no card required, which includes building, backtesting, validation and MQL5 export. Paid plans are Starter at $49/mo (20,000 tokens), Pro at $89/mo (50,000 tokens, batch generation of 100, priority queue) and Unlimited at $129/mo (unlimited tokens, multi-year backtests in single runs). Billing is metered at one token per backtest and one per validation year.
How does RoboticEA compare to buying an EA from the MQL5 Market?
A marketplace EA is a compiled binary: you cannot see whether it martingales, cannot check what costs its backtest assumed, and cannot re-run it on your years and your broker's spreads. RoboticEA inverts that — you build the rules, you see every one of them, you run the tests yourself, and you receive source. The trade-off is effort: a marketplace EA is a download, whereas this is a workflow you have to actually work through.
Is RoboticEA suitable for beginners?
Partly. The no-code interface removes the programming barrier entirely, and the preset library gives you 20,000 working starting points. But the platform assumes you understand what max drawdown, profit factor and out-of-sample testing mean, and its most valuable features are the ones that tell you a strategy is worse than it looks. A complete beginner will get more out of it after learning to read a backtest than before.