The lab: what we tested, and what we found
Every trading idea we test runs on years of data and is compared with random entries that use the same stop, target, session and costs. If it does not clearly beat them, we write it here, even when it is our own idea.
Compared with luck
A thousand random traders per test, entering at random times under the same conditions as the idea.
Locked data
Part of the history stays closed until the very end, so an idea cannot be tailored to it.
Number of tries
The more ideas we try, the harder the bar, because luck alone makes good-looking results if you try enough.
Test log
| Idea | Data | Average trade after costs | Versus luck | Verdict |
|---|---|---|---|---|
| Trend following (8 daily rules) | Gold 2018–2026 | +0.85R | Better than 98.7% | Beat it |
| The same rules | EUR/USD 2018–2026 | −0.04R | Better than 0.2% | Failed |
| Trend following on every market | From November 2026 | — | Measured in the public test | Public test running |
| Scenario rule (liquidity sweep) | Gold 2018–2022 | −0.15R | Worse than most | Did not beat it |
| Scenario rule | EUR/USD 2018–2026 | −0.02R | No better | Did not beat it |
| Golden zone and OTE (H4 → H1) | Gold 2018–2026 | −0.07 / −0.10R | Better than 6–7% | Failed |
| Daily bias + liquidity ideas | Gold and EUR/USD 2018–2026 | — | None beat it (5 ideas) | Failed |
| Opening-range breakout | 3 markets, 16 months | −0.10 to +0.01R | None beat it | Failed |
| Strategy factory (1,584 ideas) | 13 markets 2018–2022 | — | All under the luck bar | Nothing passed |
| Multi-confirmation SMC bot (V4) | 6 markets 2018–2022 | −0.41R | Worse than random (−0.23R) | Failed |
R = the profit or loss divided by the trade's risk. Results come from historical tests and do not represent any real account.
What does this tell us?
Short chart patterns on their own did not beat luck after costs in our tests. That is why we offer the liquidity map as a tool to understand and learn the chart, not as an entry signal. When an idea passes the public test, it appears here first.
Follow the public test: trend following.