Some EAs pass. Some show warnings. Some fail. That is the point.
This is not a promotional ranking and not an investment recommendation. It is a public technical lab showing how tested EAs behave under the AntiOverfit robustness framework.
A Public Lab Result is not always a Public Certificate. Clean, suitable results may become certificates. Warning, limited or failed results remain public robustness test records.
Score measures stability of the tested performance. Warnings flag suspicious structural patterns that may override blind trust. A high score with a warning is not a clean pass.
| EA Name | Score | Result | Test Page |
|---|---|---|---|
| Goldwave EA MT5 XAUUSD · M1 · AOA-2026-0001 |
94.5/100 A — High Robustness |
Warning | View Test |
| TwisterPro Scalper XAUUSD · M15 · AOA-2026-0002 |
64.9/100 C — Limited Robustness |
Limited | View Test |
| BB Return mt5 XAUUSD · M5 · AOA-2026-0003 |
32.7/100 D — Insufficient Robustness |
Failed | View Test |
| Aurum AI mt5 XAUUSD · M5 · AOA-2026-0004 |
26.0/100 D — Insufficient Robustness |
Failed | View Test |
| Quantum King EA AUDCAD · M5 · AOA-2026-0005 |
40.0/100 D — Insufficient Robustness |
Failed | View Test |
| Wave Rider EA MT5 XAUUSD · H1 · AOA-2026-0006 |
46.1/100 D — Insufficient Robustness |
Failed | View Test |
| Gold House MT5 XAUUSD · Any · AOA-2026-0007 |
76.1/100 B — Solid Robustness |
Passed | View Test |
| Gold Zilla AI MT5 XAUUSD · M5 · AOA-2026-0008 |
62.6/100 C — Limited Robustness |
Limited | View Test |
| AnE XAUUSD · M15 · AOA-2026-0009 |
29.0/100 D — Insufficient Robustness |
Failed | View Test |
Scores marked in red include warning signals and must be interpreted with caution. A high score with a warning is not a clean pass.
Lab result ≠ certificate
A lab result documents what happened under the tested conditions. A public certificate is reserved for clean, suitable results that can be used as a trust asset.
Why can a high score still show a warning?
Some systems may appear highly stable because their trade count or timing remains unusually constant across synthetic worlds. This can indicate time anchoring or preprogrammed behavior. AntiOverfit flags these cases separately from the score.
What this lab is not
Public results are technical robustness tests, not commercial endorsements, not investment advice and not guarantees of future performance.