LEARN · EN · easyquanttrading.com
How EasyQuant validates strategies — evidence you can filter
A beautiful backtest proves nothing by itself. Any optimizer can find a strategy that looks perfect on history — the question is what evidence supports it. EasyQuant applies baseline technical screening before listing and shows each additional validation separately. You can require the checks that matter to you, then paper-test in your own trading software before making any live-trading decision.
By the EasyQuant Research Team·Published 2026-08-25·We publish the tests our own strategies fail. Nothing here is a return promise.
- Baseline screening keeps technically broken candidates off the shelf
- Signal health: empty shells and always-on signals are rejected first
- Look-ahead-free engine: bars computed strictly in order, no future data
- OOS, walk-forward, Monte Carlo and statistical checks are disclosed independently
- Green means passed, amber means attention, and grey means pending — not a return forecast
- Marketplace filters let you require the evidence you consider essential
- Robust certification highlights candidates with stronger combined evidence
Why show several checks instead of one verdict
A single held-out test is statistically weak: one lucky or unlucky segment can flip the verdict. Different checks measure different failure modes, so EasyQuant keeps them separate rather than hiding them behind one opaque score. Baseline screening determines whether a candidate is technically valid to list; advanced evidence helps you decide how selective to be.
Gate by gate
- 1. Signal health — rejects empty shells (no signals) and always-on signals (no timing). A strategy that never fires, or fires every bar, is filtered before anything else.
- 2. Discovery replay — the authoritative re-run checks that the discovery record can be reproduced. Divergent numbers are kept out of the public shelf.
- 3. Out-of-sample (OOS) — tests a held-out period. A severe deterioration is an important warning, while the exact result remains visible for you to filter.
- 4. Walk-forward — repeatedly trains on one period and tests on the next. A chain of healthy out-of-sample segments is stronger evidence than one full-sample curve.
- 5. Monte Carlo — shuffles trade sequences to test whether results depend heavily on trade order.
- 6. Statistical checks such as DSR / PBO — estimate selection bias and the chance that a result comes from repeated searching rather than a persistent edge.
- 7. CPCV, where available — combinatorial purged cross-validation creates a distribution of time-ordered test outcomes while avoiding information leakage between folds.
What 'listed' and 'robust certified' mean
Listed means the candidate passed baseline technical and replay screening; it is not a live-profit promise. Screened strategies meet the platform's stronger baseline, while robust certification marks stronger combined evidence. Every card shows its validation status, and you can filter the library by the checks you require. Download to your own trading software, observe it in simulation, then decide whether it deserves any live capital.
Current platform facts
Read live from the strategy library when this page was generated. These are the same counts published on our transparency page, and they change as strategies are added and rejected.
| Strategies in the audited library | 3795 |
|---|---|
| Flagged by the audit | 2129 |
| Flag rate | 56.1% |
| Checks still pending | 1656 |
| Passed the DSR overfitting check | 1 |
| Passed the significance check | 513 |
| DSR threshold used | 0.90 |
FAQ
- What is CPCV?
- Combinatorial Purged Cross-Validation (from Lopez de Prado's Advances in Financial Machine Learning). Instead of one train/test split, the data is cut into time-ordered blocks and every combination of train/test is evaluated, with purge and embargo periods to stop information leakage. The result is a distribution of test performances, not a single lucky number.
- Why do you also run Monte Carlo if you already test out-of-sample?
- They test different things. Out-of-sample tests whether the strategy generalizes to unseen time. Monte Carlo shuffles the trade order to test whether the profit depends on a lucky sequence. A strategy can pass one and fail the other — both matter before listing.
- Can a strategy be listed before every advanced check is complete?
- Yes. Listing requires baseline technical and replay screening. Advanced checks such as OOS, walk-forward, Monte Carlo and statistical tests remain visible as pass, attention or pending evidence. You can use marketplace filters to require any checks that matter to you.
- Does passing the pipeline guarantee live profits?
- Never. Passing means the strategy survived honest statistical scrutiny on historical data — it does not predict future returns. We promise inspectability, not returns.
More guides
- Overfitting detection: catch it before you deploy
- Walk-forward analysis: testing a strategy the way you would actually trade it
- When a larger backtest holdout leaks training data
- Walk-forward analysis: the only backtest that fights overfitting
- Maximum drawdown: the number that decides if you can stay in (2026)
- Minimum account size for trading (how to work out what you need)
Not investment advice. Historical results do not guarantee future performance. EasyQuant is a research factory — you execute on accounts you control.