43 Gold EAs Tested Over 11.5 Years | $1,999 Lost It All

43 Gold EAs Tested Side by Side Over 11.5 Years — Price and Quality Were Unrelated

We ran a mass backtest of 43 popular paid gold EAs from the MQL5 Market ($50–$1,999, median price $299), all in the same environment, over the same period, with default parameters. This is a side-by-side comparison of "what would have happened over 11.5 years if you had bought it and used it as is."

Test Conditions (Full Disclosure)

Distribution of Results (43 EAs)

Of the 37 EAs we could measure, 7 (19%) ended in a blow-up or a loss. Even among popular paid EAs, roughly 1 in 5 lost the account when run for 11.5 years on default settings.

The Biggest Finding: No Correlation Between Price and Quality

• Median price of the 17 excellent EAs: $299
• Median price of the 7 blown-up EAs: $499

There is an inversion where the more expensive group did worse. Two EAs at opposite extremes illustrate this.

The Most Expensive EA, at $1,999, Lost Everything — XG Gold Robot MT5

The most expensive EA in the test, XG Gold Robot MT5 ($1,999, 109 reviews), posted PF 0.23, maximum DD 98.99%, and a net loss of -$9,898.61 (3,271 trades). The $10,000 starting capital was almost entirely wiped out. A large number of reviews was no guarantee of quality either. Likewise, products with many reviews such as Aurum AI ($499, 47 reviews) and ArtQuant Gold ($700, 24 reviews) also landed in the blown-up group (ArtQuant Gold was confirmed to blow up on 7.5 years of real ticks from another broker as well).

The Top Spot Went to a $199 EA — Vex Gold EA

First place went to the low-priced Vex Gold EA ($199), with PF 11.35, maximum DD 5.30% and 874 trades. It was followed by Xauusd Aether ($189) with PF 10.25/DD 1.66%, One Gold ($195, 43 reviews) with PF 4.84/DD 1.38%, and the $99 Smart Gold Impulse with PF 2.10/DD 0.95% — inexpensive EAs dominated the top.

Some Popular EAs Lived Up to Their Reputation

Some expensive products do deliver. The Gold Reaper ($849, 101 reviews) at PF 2.24/DD 17.8%, Goldwave ($999, 69 reviews) at PF 5.73/DD 4.1% and Gold House ($749, 56 reviews) at PF 2.93/DD 1.1% posted results that matched their reputation. It is not that "expensive = bad"; the accurate conclusion is simply that price carries no information about quality.

Real-Tick Retest of the Top EA, Vex — Is PF 11 Genuine?

This test used Model=1 (OHLC), which has the known limitation of potentially overrating thin-TP scalpers (our explanation of the difference between real ticks and OHLC). So we retested the original Vex Gold EA binary on 100% real ticks (2020–2026, MetaQuotes).

• Result: PF 3.00, equity DD 5.60%, 703 trades, with all 7 years from 2020 to 2026 profitable (annual PF 1.35–4.53)
• The OHLC PF of 11.35 was optimistic, but the EA remained solidly profitable on real ticks, and its top-tier verdict was not overturned

There is an important caveat, however. In our behavior analysis (using a reproduction implementation built for testing), this EA's edge depends heavily on the spread. With the same logic, a measured spread of $0.074 (MetaQuotes) gave PF 2.2, while $0.474 (Dukascopy) gave PF 0.53 — the edge disappeared as spreads widened. Running it requires a tight-spread ECN account.

Update: Sales Status as of October 2026

We rechecked the MQL5 product pages on 2026-10-02. The results are for the versions tested at the time; we have not retested any versions updated since then.

Across all 43, 11 had been delisted (the product page redirects to the sales listing). The current status of all 43 can be checked in the table in the ranking of all 43 by name.

Limitations of This Test (Please Read)

• Model=1 (OHLC) can overrate thin-TP scalpers. We recommend a real-tick retest for products with PF>10 (in this article, only Vex was retested). Conversely, the blow-up verdicts are more certain, since those EAs blew up even under optimistic conditions.
• Absolute net profit balloons into the millions of dollars for compounding-lot EAs, so comparisons are made on PF and DD.
• This test used default parameters and a single feed. Results may differ with vendor-recommended settings.
• The 5 unmeasurable EAs (Axiom Gold / EA Gold Stuff / Helios Gold AI / Prop Firm Gold EA / The Gold Phantom) could not be judged, and we make no assessment of their merits.

Summary

• Of 43 popular paid gold EAs, 19% of the 37 we could measure blew up or lost money on default settings.
• Price and quality are unrelated (median $299 for the excellent group vs. $499 for the blown-up group). Review counts are no guarantee either.
• What to base your decision on is neither price nor popularity but measured values from long-term tests with disclosed conditions (PF, margin-based DD, spread dependence). We also explain the danger of choosing on win rate alone in The Pitfalls of 99% Win Rate EAs.

Our own EAs are tested to the same standard (real ticks, long term, margin DD disclosed), with all figures published. See our Testing Methodology for the method and the EA Ranking for the list.

Notes on Naming and Rights

• All EA names in this article are trademarks or product names of their respective owners. This article uses them descriptively for the purpose of comparison and critique, and we have no affiliation with any seller.
• All figures in this article are our own measured backtest data. No official images or official performance records are reproduced. The spread-dependence analysis is based on measurements from a reproduction implementation our site built for testing purposes.
• If you find any factual error, please let us know via the contact page. We will verify it and correct it promptly. The original data (reports, result CSVs, settings) has been preserved.