A real run, 4 October 2026
The backtest made $678,847. The verdict was do not trust.
On the backtest alone, the strategy beat holding gold. After 49 checks, its edge could not be told apart from holding gold.
Not yet: do not trust this result
Deflated Sharpe ratio: Charged for all 13 variants tried on this strategy so far, the probability the Sharpe is above zero is 0.0%.
29 pass4 warn3 fail10 info3 skip
- Failed
- Deflated Sharpe ratioHaircut Sharpe and multiple testingRandom-entry benchmark
- Cautions
- Data qualityProbability of backtest overfitting (CSCV)Sub-period consistencyAlpha versus buy-and-hold
| Net profit | Sharpe | Max drawdown | Trades | |
|---|---|---|---|---|
| Volatility-targeting strategy | $678,847 | 0.87 | 50.7% | 462 |
| Buy and hold, same data and costs | $559,788 | 0.72 | 45.2% | n/a |
Read the 49 checks yourself
These come from the report the engine exported. Validation ran on the most recent 40,000 bars, where the baseline made $129,320 in 143 trades. The last 20% was sealed as a holdout and stayed unopened.
Is it more than luck? Significance
Charged for all 13 variants tried on this strategy so far, the probability the Sharpe is above zero is 0.0%.
Open the raw validation reportIt is the file the engine exported, unedited. It covers the validation window, and its signature is tied to the machine that ran it, so it shows the file is unaltered rather than who made it.