The free course

The Backtest Trap

Every strategy looks brilliant in a backtest — that is what backtests are for. This course teaches the difference between testing a system and flattering one: the biases that fake an edge, why small samples lie, and the honest protocol that survives contact with the future. Unusually, every mistake in it is illustrated from our own published research — because we made most of them in public.

5 lessons · ~32 minutes · free forever · no sign-up, no upsell


6 min01

The Seductive Curve

A backtest is not evidence that a strategy works. It is evidence that a strategy can be made to have worked.

7 min02

The Five Ways a Test Lies

Overfitting is only the most famous liar in the room. Four quieter ones corrupt tests that were never even tuned.

6 min03

Twenty Trades Prove Nothing

The most common backtest error isn't a bias at all. It's believing a number that hasn't earned belief yet.

6 min04

Choose Plateaus, Not Peaks

When a parameter sweep crowns a winner, the winner is usually the luckiest cell in the spreadsheet — not the best rule.

7 min05

Pre-Registration: The Honest Protocol

Science solved this exact problem decades ago. The solution transfers to trading completely — and almost nobody uses it.

The receipts behind this course are all public: the research ledger that lost pips, the rules rebuilt from it, and the forward test now running live on the record and the real account. Related: Trade Like a Machine.