Is your backtest lying to you?
Get an honest verdict on your trading edge.
Drop in a trade log. Verdkt runs the checks most backtests skip: out-of-sample stability, sample-size significance, and whether the result is separable from noise. It tells you plainly whether the edge is real, fragile, or fictional.
Drop your trade log CSV here, or browse
What it checks
01
Expectancy in R
Not total profit, but average risk-adjusted result per trade. A high win rate with fat losers gets caught here.
02
Out-of-sample split
Splits your trades in half by time. If the edge flips sign across the halves, it's a regime artifact, not a strategy.
03
Significance
A t-test against zero. If the result isn't statistically separable from luck plus costs, you'll be told.
04
Sample size
Small samples routinely show fake edges. Verdkt is blunt about when you simply don't have enough trades.