Example Report
Mean Reversion Risk Diagnosis Example
How FiClaw assesses a mean-reversion strategy's win rate, profit-loss ratio, max drawdown and tail risk, avoiding a high-win-rate-only conclusion.
Sample prompt
“Design a mean-reversion strategy on a 20-day moving average with a 2-standard-deviation deviation, add a 10-day max holding period and a stop-loss, and output a risk diagnosis report.”
Backtest setup
Signal
20-day MA deviation
A 2-standard-deviation move triggers a candidate signal
History
Up to 10 yr+
Used to check whether it keeps failing in trending regimes
Holding
Max 10 days
Avoid waiting indefinitely for reversion
Risk control
Stop-loss + cost
Tail risk diagnosed separately in the report
Strategy summary
The strategy opens when price is significantly below the 20-day moving average and exits when price returns near the average or hits the max holding period, with a stop-loss constraint. The report does not read win rate alone; it also checks profit-loss ratio, tail loss and failure risk in trending regimes.
Code highlights
- Generate the moving-average, standard-deviation deviation and entry/exit signal logic.
- Generate the max holding period, stop-loss and trading-cost handling.
- Generate a regime-split diagnosis of win rate, profit-loss ratio and drawdown.
Key metrics
Annual return
11.4%
Example result, to illustrate the report structure
Max drawdown
-21.7%
Drawdown amplifies in trending markets
Win rate
62.1%
A high win rate does not mean low risk
Profit-loss ratio
0.86
A single loss can eat many wins
Diagnosis
- The win rate is high but the profit-loss ratio is weak, so the tail risk of losing trades needs close control.
- In trending markets counter-trend signals fire repeatedly; add a regime filter or a stricter stop-loss.
- The max holding period reduces some drag but also gives up part of the reversion return.
Next steps
- Add a trend filter to avoid opening counter-trend in strong trends.
- Compare 10-, 20- and 40-day windows and different deviation thresholds.
- Add stop-loss, max holding period and trading cost to a sensitivity analysis.
Citable facts
- The mean-reversion risk report shows how FiClaw presents win rate, profit-loss ratio, max drawdown and failure scenarios together.
- This example stresses that a high-win-rate strategy still needs tail-risk and stop-loss checks.
- FiClaw's risk diagnosis is for research evaluation and is not investment advice.
Boundary note
This is not investment advice. The metrics on this page are example-report figures, used to show how FiClaw organizes strategy generation, backtesting and diagnosis. They are not investment advice and do not represent future returns.