TradeHouse Fidelity Lab · Guides
Trading strategy audit service
A strategy audit is an independent check of whether a trading strategy's reported results hold up. It is most useful at two points: before you put real capital behind a backtest, and before you show results to someone else who will.
What an audit should cover
| Area | What is checked |
|---|---|
| Data integrity | Look-ahead stamps, duplicate legs, crossed books, sign errors, open positions counted as realised |
| Costs | Fees re-netted from the venue's schedule instead of the file's own fee column |
| Generalisation | A rule frozen on a training window and scored on later data |
| Sample adequacy | Independent days after autocorrelation, not trade count |
| Multiple testing | Correction across every rule or cohort submitted |
| Selection | Whether the cohort was chosen because it was already profitable |
What an audit cannot tell you
- Whether the strategy will make money in future. It can only remove reasons to believe a past result that is not real.
- Anything about data that is not in the file, such as instruments excluded before export.
- Whether the strategy suits your risk tolerance. That is a separate decision and this is not investment advice.
How the Fidelity Lab audit works
- Export your trades (legs) as CSV or JSON from your backtester or broker.
- Upload them. The report runs 11 contamination detectors, re-nets costs, runs a frozen-threshold walk-forward and applies Benjamini-Hochberg at q=0.05.
- You get one JSON document and the same report as HTML, with a verdict of SURVIVES, DISSOLVES or UNTESTABLE and the specific legs behind each finding.
- If something is flagged, fix it at the source and re-run.
| Option | Price | Best for |
|---|---|---|
| Free | $0 | One report a month up to 5,000 legs |
| One audit | $99 one-time | A single strategy before going live |
| Five-audit pack | $399 one-time | Fix and re-test loops |
| Starter / Pro | $49 / $149 per month | Ongoing validation of new strategies |
Try it on your own trades. The free Fidelity Lab tier runs one report a month on up to 5,000 legs: all 11 contamination detectors, cost re-netting and the walk-forward verdict. Run a free report or see pricing.
Frequently asked questions
Do I need to share my strategy code?
No. The audit grades the trades the strategy produced, so only the trade file is uploaded.
What if my file is too small?
The verdict will be UNTESTABLE, with the reason, for example too few independent days. That is a result, not a failure of the audit.
Related guides
- Backtest overfitting: a practical guide
- Walk-forward validation vs cross-validation
- Backtest validation tool: what it should check
- Probability of backtest overfitting (PBO)
- Overfitting in backtrader
- Trading strategy robustness tests
- Monte Carlo backtests
- Look-ahead bias in backtests
- Survivorship bias in backtests
- Data snooping in trading
- Out-of-sample testing in trading
- What is backtest overfitting?
- Deflated Sharpe ratio calculator
- Fidelity Lab
Educational content, not investment advice. Updated 2026-10-28.