Advanced

Edge Validation

A backtest is one draw, and this stage is how you find out what it's worth. You'll put a range on a win rate, test rules on trades they've never seen, and shuffle the order of the trades to see the losses you should expect. Then you'll score a strategy, combine two, and write the rule that tells you when to stop. Eleven lessons, and nothing here takes a strategy's word for it.

0 / 11 done

StartLesson 1 · An Edge Is a Distribution

What you’ll leave with11 lessons · 7-line playbook

A win rate from thirty trades is a range, and I trade the low end of the range.
An edge that survives walk-forward is real, and one that only works in-sample is a story.
I pick the parameter in the middle of a plateau, never the one on a spike.
The drawdown to plan for is the 95th percentile of the fan, not the one in the backtest.
Four numbers score a strategy, and profit factor and MAR are the two I trust first.
Two strategies with uncorrelated equity curves are worth more together than either alone.
I stop a strategy when its drawdown leaves the fan, not when it hurts.

Edge Validation

All lessons